Biological Control of Postharvest Diseases of Pome Fruit Using Yeast Antagonists

Biological control of postharvest diseases of pome fruit using yeast antagonists Davide Spadaro UNIVERSITÀ DEGLI STUD...

Author: Davide Spadaro

143 downloads 835 Views 3MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Biological control of postharvest diseases of pome fruit using yeast antagonists

Davide Spadaro

UNIVERSITÀ DEGLI STUDI DI TORINO DiVaPRA – Patologia vegetale DOTTORATO DI RICERCA IN SCIENZE AGRARIE, FORESTALI ED AGROALIMENTARI CURRICULUM DIFESA INTEGRATA E BIOLOGICA DELLE COLTURE XVI CICLO

BIOLOGICAL CONTROL OF POSTHARVEST DISEASES OF POME FRUIT USING YEAST ANTAGONISTS Tesi presentata da: Davide SPADARO

Tutor:

Coordinatore del ciclo:

Prof. Maria Lodovica GULLINO

Prof. Elisabetta BARBERIS

Anni accademici: 2000/2001 – 2001/2002 – 2002/2003

to Manuela and Marta

Content Chapter 1...........................................................................

1

General introduction – State of the art and future prospects of the biological control of postharvest fruit diseases Chapter 2................................…........................................

23

Mechanisms of action and efficacy of four isolates of the yeast Metschnikowia

pulcherrima

active

against

postharvest

pathogens on apples Chapter 3..............................................…............…...........

47

Control of Penicillium expansum and Botrytis cinerea on apple combining a biocontrol agent with hot water dipping and acibenzolar-S-methyl, baking soda, or ethanol application Chapter 4...........................................................................

71

Metschnikowia pulcherrima: a promising species isolated from different food matrixes for biological control of postharvest diseases in apple Chapter 5...........................................................................

101

General Discussion Acknowledgements.............….…........................................ Curriculum vitae………………………………………………………… Publications……………………………………………………………...

119 121 123

CHAPTER 1

State of the art and future prospects of the biological control of postharvest fruit diseases

D. Spadaro and M.L. Gullino, International Journal of Food Microbiology (2004), in press (available online 25 September 2003).

Biological Control of Postharvest Diseases ________________________________________________________________

Abstract Synthetic fungicides are the primary means to control postharvest diseases of fruits. Biological control has emerged as one of the most promising alternatives to chemicals. During the last twenty years, several biological control agents have been widely investigated for use on

different

pathogens

and

fruit

crops.

Many

biological

control

mechanisms have been suggested for use on fruit including antibiosis, parasitism, induced resistance in the host tissue and competition. With the aim of extending the use of the biofungicides, there have been many studies on the application of various combinations of control agents, and on the application integrated with chemical and physical means of protection. The formulation and application methods are key issues for the efficacy and successful outcome of the commercial product. Genetic engineering may provide a useful tool in the enhancement of the biological control efficacy. Since biofungicides are usually not as effective as pesticides, this approach should be viewed as an important component of an integrated disease management scheme given that a significant and permanent reduction of pesticide use is our goal. Keywords: antagonist, fruit, mechanism of action, postharvest disease. 1. Introduction Fruits are an important part of the human diet because they supply essential nutrients such as vitamins and minerals and they are also considered important to human health and well-being because they contain other necessary compounds such as antioxidants. Increased consumer

awareness that diet and health are linked has therefore

resulted in a greater consumption of fruits. At the same time, however, consumers are equally concerned about the safety of the fruits they eat, and are wanting foods free from pesticide residues, toxins and harmful microorganisms. Losses due to pests and diseases in the field, during storage, as well as in transit and commercialisation can amount up to 25% of the total 3

Chapter 1 ____________________________________________________________________________________________

production in industrialised countries (Harvey, 1978) and in developing countries damage is often higher, exceeding 50%, because of the lack of adequate storage facilities (Eckert and Ogawa, 1985). There are two principal factors which make plant products more susceptible to spoiling: the high water content in fruit which allows pathogen attack (Harvey, 1978) and the wounds present on the plant organs during storage, often as a result of harvesting and transportation, which give microorganism easy access. When permitted, synthetic fungicides are the primary means to control postharvest diseases. However, several reasons, such as the public’s growing concern for the human health conditions and the environmental pollution associated with pesticide usage in orchards (Wilson and Wisniewski, 1994), the development of fungicide-resistant strains of postharvest pathogens (Romano et al., 1983; Spotts and Cervantes, 1986) and the lack of continued approval of some of the most effective fungicides (Gullino and Kuijpers, 1994) have motivated the search for alternative approaches. Biological control fits in well with the concept of sustainable agriculture because it mostly exploits natural cycles with reduced environmental impact. Among the biological strategies applicable to postharvest, the induction of resistance in the fruit, the use of plant or animal products with a fungicidal activity, and, above all, the application of antagonistic microorganisms can be considered. Biological control using antagonists (Wilson and Wisniewski, 1994) has proved to be one of the most promising alternatives, either alone or as part of an integrated pest management policy to reduce pesticide use. The postharvest environment represents a particular sector for the development of biological measures. Specific attention to the following points should be given to the control of postharvest diseases (Chalutz and Droby, 1998): i) the disease control level required is high (i.e. 9598%); ii) food safety requires special care in the direct use of living microorganisms on food products; and iii) the potential market to implement a biofungicide expressly developed for postharvest use is relatively small. It was also mentioned that the storage conditions, such as temperature and humidity, can be controlled to switch the host4


pathogen-antagonist equilibrium towards the antagonist. Lastly, the high value of the fruit can justify treatment with a relatively expensive product (Chalutz and Droby, 1998). During the past 20 years, several biocontrol agents have been exploited and widely investigated against different postharvest fungal pathogens, like Aspergillus spp., Botrytis, spp., Monilia spp., Penicillium spp. and Rhizopus spp. (Droby et al. 2002; Lima et al. 1997; Northover and Zhou, 2002; Viñas et al. 1998; Zahavi et al. 2000). Many of the early studies were aimed at the study of the mode of action and evaluation of the efficacy of some potential biocontrol bacteria, such as Brevibacillus subtilis, producers of antibiotics (Pusey et al., 1986), however, the application of such bacteria on fruit did not prove to be commercially acceptable. Wilson and Wisniewski (1994) indicated the following characteristics of an ideal antagonist: genetic stability, efficacy at low concentrations and against a wide range of pathogens on various fruit products,

simple

nutritional

requirements,

survival

in

adverse

environmental conditions, growth on cheap substrates in fermenters, lack of pathogenicity for the host plant and no production of metabolites potentially toxic to humans, resistance to the most frequently used pesticides

and

compatibility

with

other

chemical

and

physical

treatments. Yeast seem to possess a good number of the abovementioned features and, during the last few years, research has been focused on the selection and study of yeast (Chalutz and Droby, 1998). At present, a class of products containing Pseudomonas syringae Van Hall, active against the genera Botrytis,

Penicillium,

Mucor and

Geotrichum, (Janisiewicz and Jeffers, 1997) are available, as well as a product containing Candida oleophila Montrocher, effective against Botrytis and Penicillium spp. (Hofstein et al., 1994). Other yeast species have been extensively tested and could be registered relatively soon. Among the yeast under development, antagonistic strains belonging to the species Aureobasidium pullulans (Ippolito et al., 2000), Candida saitoana (El-Ghaouth et al., 1998), Candida sake (Teixidó et al., 1998) and Metschnikowia pulcherrima (Spadaro et al., 2002a) should be mentioned.

5

Chapter 1 ____________________________________________________________________________________________

2. Mechanisms of action for biocontrol agents Information on the mechanisms of action for most of the antagonists investigated is still incomplete because of the difficulties encountered during the study of the complex interactions between host, pathogen, antagonist

and

other

microorganisms

present.

However,

a

good

understanding of the mode of action is essential before developing appropriate formulations and methods of application, and in order to obtain official approval. Several possible biocontrol mechanisms have been suggested to be effective

against

post-harvest

rots

on

fruit

including

antibiosis,

competition for nutrients and space, parasitism or direct interaction with the pathogen, and induction of resistance in the host tissue (Droby and Chalutz, 1994). 2.1. Antibiosis Some of the most active biocontrol agents (BCAs) are bacteria producing antibiotics, whose action, at least partially, determines their effectiveness. For example, Brevibacillus subtilis produces iturin, a powerful antifungal peptide (Gueldner et al., 1988), as well as gramicidin S (Edwards and Seddon, 2001), Pseudomonas cepacia synthesises pyrrolnitrin, which used alone can control Botrytis cinerea and Penicillium expansum attacks on pome fruit (Janisiewicz et al., 1991). Often other mechanisms than antibiosis may be involved: e.g. Penicillium digitatum strains resistant to pyrrolnitrin are still inhibited by P. cepacia (Smilanick and Denis-Arrue, 1992). The main concern, related to the use of antibiotics in food products, is the development of human pathogens resistant to these compounds and the possible development

of

resistance

in

fruit

pathogens.

Even

if

antibiotic

producers appear to be able to control wound infections established before antagonist application, at the moment there are not such BCA’s registered for use on fruit.

6


2.2. Competition for nutrients and space Other selected microorganisms, particularly yeast, act mainly competing for space or for the utilisation of some nutrients with the pathogen (Piano et al., 1997; Filonow, 1998; Spadaro et al., 2002a). Yeast can successfully compete with the pathogen, inhibiting its growth but leaving it alive. Also some bacteria are competitive, for example Enterobacter cloacae against Rhizopus stolonifer on peaches (Wisniewski et al., 1989). In the competition for space, yeast are helped by the formation of an extracellular polysaccharide capsule that can promote adhesion to the fruit surface (Andrews et al., 1994). Competition for nutrients was demonstrated for Pichia guilliermondii against P. digitatum cocultivated on synthetic media (Droby et al., 1989): the addition of exogenous nutrients resulted in a reduced efficacy because the antagonists offered better results when nutrients were scarce. A rapid multiplication and colonisation by antagonist cells in the wound was elucidated in various interactions (Droby et al., 1989; Smilanick and Denis-Arrue, 1992; Piano et al., 1997). Studies on the distribution of radiolabelled glucose between the antagonistic yeast Sporobolomyces roseus or Cryptococcus laurentii and the pathogen B. cinerea show a strong sugar assumption by the BCAs that blocks fungus conidial germination due to the deprivation of nutrients (Filonow, 1998). In fruit wounds, competition for nutrients is probably extended to other nutrients, such as nitrogen compounds present in low concentration. Janisiewicz et al. (2000) have recently developed a non-destructive method using tissue culture plates: a defusing membrane at the lower end of cylindrical inserts is used to study the competition for nutrients separated from the competition for space. 2.3. Parasitism Antagonist and pathogen can interact also through a direct parasitism. Wisniewski et al. (1991) observed a strong adhesion in vitro of P. guilliermondii antagonist cells to B. cinerea mycelium, perhaps due to a lectin link. Moreover, P. guilliermondii shows a high activity of β-1,37

Chapter 1 ____________________________________________________________________________________________

glucanase enzyme that could result in the degradation of the fungal cell walls (Jijakli and Lepoivre, 1998). Aureobasidium pullulans in apple wounds produces extracellular exochitinase and β-1,3-glucanase which could play a role in the biocontrol activity (Castoria et al., 2001). Through ultrastructural and cytochemical studies, El-Ghaouth et al. (1998) found that Candida saitoana yeast cells, when cultivated together with B. cinerea mycelium, are associated with fungal hyphae showing cytological damage, such as papillae and other protuberances in the cell wall, and degeneration of the cytoplasm. Lastly, some yeast, such as S. roseus and C. laurentii, when applied, are able to reduce the conidial adhesion and germination of B. cinerea on apples, that is favoured by butyl acetate, a volatile aroma produced by the fruit (Filonow, 2001). 2.4. Induced resistance in the host tissue Some BCAs can interact with the host tissue, particularly wounds, increasing the cicatrisation processes (Droby and Chalutz, 1994). Several antagonistic yeast are as effective if applied before pathogen inoculation. This observation has suggested that yeast cell application induced resistance processes in the fruit skin. Some Candida strains are able to cause chemical and osmotic changes in apple tissues, favouring antagonist settlement (McLaughlin et al., 1990). A P. guilliermondii strain has been shown to stimulate the production of ethylene, a hormone in grapefruit able to activate the phenylalaninammonium-lyase (Wisniewski et al., 1991), an enzyme involved in the synthesis of phenols,

phytoalexins

and

lignins.

A

phytoalexins

accumulation

(scoparon and scopoletin) was noted in citrus fruits treated with yeast cells (Rodov et al., 1994). In addition to controlling decays, A. pullulans can cause a transient increase in β-1,3-glucanase, chitinase, and peroxidase activities on apple fruit, starting 24 h after treatment (Ippolito et al., 2000).

8


3. Extension of use of biocontrol agents Potential biocontrol agents often have some significant limitations: they have a narrow range of activity, because they act on specific hosts against

well-defined

pathogens

under

particular

environmental

conditions. A method to select antagonists with a broader spectrum of activity, preferably for commercial development, includes efficacy tests for various pathogens and fruit species (Wilson et al., 1993; Lima et al., 1999). 3.1. Antagonist mixtures In the enhancement of a biocontrol system, work could focus on a promising approach, which is the development of antagonist mixtures. An effective biological control based upon a mixture of several complementary and non-competitive antagonists is more likely than a control based upon microorganism alone. Such mixtures have several advantages (Janisiewicz, 1998): apart from a wider spectrum of activity (different fruits, cultivars and ripening stages), they can increase the efficacy (less biomass necessary), be more reliable and allow a reduction in application times and treatment costs. Moreover, they permit the combination of different genetic characteristics, minimising the need for genetic engineering. The evaluation of combinations of antagonists isolated from apple fruits and leaves resulted in the choice of a mixture of P. syringae and S. roseus, more effective than

both

BCAs when applied alone in the control of P. expansum (Janisiewicz and Bors, 1995). 3.2. Preharvest use One of the major obstacles to the development of postharvest BCAs is their inability to control previously established infections, such as latent infections. Field application of the BCAs may enable early colonisation of the fruit surfaces, protecting them from these infections (Ippolito and Nigro, 2000). To be successful in preharvest application, potential 9

Chapter 1 ____________________________________________________________________________________________

antagonists should be able to tolerate low-nutrient availability, UV rays, high temperature and dry conditions. Leibinger et al. (1997) applied a preharvest mixture of the yeast A. pullulans and the bacterium B. subtilis obtaining a level of control equivalent to fungicides for P. expansum and B. cinerea on apples. Teixidó et al. (1998) applied unmodified and low water activity tolerant cells of Candida sake before harvest to control blue mould on apples. 3.3. Integrated use Biological means cannot at the moment solve all the problems of postharvest rots during fruit storage and they must be considered instruments to be used in combination with other methods in an integrated vision of disease management. For example, biocontrol agents can be combined with waxes and fungicides applied not only in post but also in pre-harvest (Pusey, 1994). In some laboratory and semi-commercial trials the efficacy of the BCA P. guilliermondii was consistently increased by the addition of small concentrations of imazalil or thiabendazole, reaching a control level similar to the use of fungicides (Droby et al., 1993). Yeast are generally tolerant to many of the fungicides used in postharvest: Metschnikowia pulcherrima (Spadaro et al., 2002a) is tolerant to relatively high concentrations of benzimidazoles (benomyl and thiabendazole) and dicarboximides (vinchlozolin and procymidone). Moreover, microorganisms need to survive under the commercial temperature, humidity and atmosphere storage conditions. The efficacy of M. pulcherrima against B.cinerea and P. expansum is increased when fruits are stored at low temperatures optimal for hosts, as compared with storage at 23°C, because the antagonist is more adaptable than the pathogens to cold temperatures (Spadaro et al., 2002a). Among the strategies evaluated during the last few years, the combination of biological treatments with other alternative techniques to chemicals

should

thermotherapy

be

mentioned:

(Barkai-Golan

and

the

treatments

Phillips,

1991),

used

could

ultraviolet

be rays

(Chalutz et al., 1992), animal and plant natural products (Aharoni et al., 10


1993), calcium infiltrations (Janisiewicz et al.,1998), sodium bicarbonate (Teixidó et al., 2001) or ethanol (Spadaro et al., 2002b). Calcium chloride infiltrations combined with antagonist application on apples increases control of P. expansum after 3 and 6 months of storage at 1°C, compared to biological treatment alone (Janisiewicz et al., 1998). An even more integrated approach, experienced with “Gala” apples (Conway et al., 1999) consists of a heat treatment (4 days at 38°C), followed by a calcium chloride (2%) infiltration and application of a Pseudomonas syringae cell suspension. The result of these combined treatments was a reduction of blue mould which proved to be significantly better than the use of calcium treatment, antagonist application, or heating alone. The efficacy of Pantoea agglomerans for the control of green mould was improved when combined with sodium bicarbonate, or baking soda (Teixidó et al., 2001), a common food additive, allowed without restrictions for many applications in Europe and North America and listed as an approved ingredient on organic food products. The combination of sodium bicarbonate, or ethanol, or acybenzolar-S-methyl with M. pulcherrima cell suspension and heat treatment was also reported by Spadaro et al. (2002b). 4. Improvement of biocontrol agents 4.1. Formulation and application Unlike soilborne or open field pathogens, where a 70-80% margin of disease control is acceptable, postharvest disease control requires a higher level of efficacy and more consistent results. Considering even the most effective BCAs studied until now, they are rarely as effective as fungicides. In order to justify their practical use, greater antagonistic activity is required. The mass production by rapid, efficient and inexpensive fermentation of the antagonist is a key issue. The efficacy of many antagonists of wound pathogens is directly related to the number of antagonist cells inoculated (Hofstein et al., 1994), so that a really simple way to increase the 11

Chapter 1 ____________________________________________________________________________________________

effectiveness is the application of a higher number of cells. To scale-up a laboratory fermentation process to an industrial level, it is fundamental to find the nitrogen and carbon sources that provide maximum biomass production and minimum cost of media, whilst maintaining biocontrol efficacy. Costa et al. (2001) have studied yeast extract, dry brewer’s yeast, sucrose, and molasses as a possible substrates for the production of the biocontrol agent P. agglomerans. An accurate formulation can be decisive in the improvement of the efficacy and extension of the product shelf life, facilitating storage for commercially acceptable periods of time (Janisiewicz and Jeffers, 1997). The application of adjuvants can protect and stimulate the establishing of the antagonist on the host surface. The addition of calcium salts increases the activity of several antagonistic yeast (Janisiewicz et al., 1998). The addition of glycerol and trealose to the culture medium augmented osmotic tolerance and control capability of C. sake against P. expansum

on

apple

(Janisiewicz,

1998).

Sodium

alginate,

carboxymethylcellulose and chitosan are adhesion promoters and can be added to yeast cell suspension, to increase the activity of the formulation. These substances were added to a strain of M. pulcherrima (Piano et al., 1998) significantly increasing the efficacy against grey rot on apple. Chitosan has a fungistatic activity demonstrated against the main postharvest pathogens of strawberry (El-Ghaouth et al., 1992). Recently, El-Ghaouth et al. (2000a) developed a biocontrol product called "bioactive coating" consisting of a unique combination of an antagonist with glycolchitosan, a chemically-modified chitosan. The bioactive coating made it possible to exploit the antifungal property of glycolchitosan and the biological activity of the antagonist. Moreover, when applied as a pretreatment, sodium carbonate enhanced the efficacy of the bioactive coating (El-Ghaouth et al., 2000b). Application systems also affect the effectiveness. The coatings most often applied to citrus fruit contain “shellac”, which is a purified product of the hardened resinous secretion of the scale insect Kerria lacca. It is important to test the suitability of such products with antagonist applications. McGuire (2000) found that after 3 or 4 months of storage in a warehouse, the biocontrol agent C. oleophila had a higher survival 12


rate when the cell suspension was applied inside the shellac than when it was applied by firstly dipping and then drying the fruits prior to shellacking. Conversely, Chalutz and Droby (1998) noted that the application of a water suspension of antagonistic yeast before waxing proved to be more effective compared to a single application of antagonists and wax. Another issue involved in the commercial production of biocontrol agents is shelf life, that should be as long as possible. A biofungicide should be effective for at least 6 months and preferably for 2 years (Pusey, 1994). Abadias et al. (2001a) found that freezing at –20°C was the best method to preserve the viability of C. sake cells. Survival of the cells was higher using 10% skim milk as a protection, and further increased by using other appropriate protections, such as lactose, glucose, fructose or sucrose. Moreover skimmed milk with 1% peptone was the rehydration medium that kept the highest viability of the antagonist cells. In any case, freeze-dried cells were significantly less effective than fresh cells (Abadias et al., 2001b). 4.2. The nutritional environment In addition to optimising temperature, humidity and gas composition in warehouses to guarantee a high fruit quality during storage, it is also possible

to

manipulate

the

chemical

environment

to favour the

antagonist. The addition of nutrients, preferably metabolised by the antagonist and not likely to be metabolised by the pathogen, was suggested in several antagonist-pathogen interactions (Janisiewicz, 1998). The application of nitrogen compounds, L-asparagine and Lproline (Janisiewicz et al., 1992), or 2-deoxy-D-glucose, a sugar analogue, showed a consistent increase of the control of P. expansum on apples (Janisiewicz, 1994). Both aminoacids stimulated the germination of P. expansum but slowed mycelial growth. Also L-glutamine added to M. pulcherrima showed a direct influence on the yeast, because its application with the antagonist contributed to a reduction of Botrytis rot, while it was ineffective without yeast (Piano et al., 1998). The sugar analogue 2-deoxy-D-glucose could be a useful additive to antagonistic 13

Chapter 1 ____________________________________________________________________________________________

microorganisms, provided that it has a fungicidal action on the major postharvest pathogens of apple and peach fruit (El-Ghaouth et al., 1997) and that the antagonist is resistant to its toxic effects. Recently the sugar analogue was found to be compatible with the antagonistic yeast C. saitoana and effective against apple and citrus fruit decay (ElGhaouth et al., 2000c). Nutritional composition can also influence the production of metabolites important to many control systems, such as antibiotics (Gueldner et al., 1988) and cell wall degrading enzymes (Wisniewski et al., 1991). The type and concentration of nitrogen and carbon sources can therefore be important factors in the synthesis and secretion of key compounds of the biocontrol mechanism. 4.3. Manipulation of antagonists One of the major problems with the use of antagonists is their insufficient and inconsistent performance under commercial conditions. It is possible to increase the ecological competence and the capability to control the pathogens manipulating antagonists with techniques of conventional mutagenesis (ionising radiations, mutagenic chemicals, fungicide or antibiotic exposure) or of sexual recombination, through protoplast fusion or genetic transformation. With genetic engineering techniques, features for the exploitation of plant products or additives applied together with the antagonist, for fungicide resistance, for carposphere colonisation in storage conditions, for the synthesis of compounds favouring antagonism (antibiotics or siderophores) could be transferred into the potential antagonistic microorganisms (Pusey, 1994). Since one mechanism involved in the biocontrol of postharvest fruit pathogens is mycoparasitism, cell wall degrading enzymes, such as chitinases (Chernin et al., 1997), proteases and glucanases (De la Cruz et al., 1995) produced by bacterial and fungal microorganisms, could be inserted into the potential antagonists to improve the degradation of the pathogen cell walls, resulting in death or growth inhibition of the antagonised fungus. For the biocontrol of soilborne pathogens, for example, a chitinase gene isolated from 14


Serratia marcescens, was introduced into the endophytic bacterium Pseudomonas fluorescens to improve the control of Rhizoctonia solani on beans with an effective reduction of the pathogen damage (Downing and Thomson, 2000). Another idea could be the insertion of the gene for amylase under constitutive promoter in some BCAs to allow an effective utilisation of the fruit carposphere starch as a carbon source with a substantial advantage. Moreover, biocontrol strains with a higher capability of exploitation of the nitrogen compounds present, or with a higher transport or metabolism rate of the limiting factor could be developed, because nitrogen is often a limiting substance when the biocontrol mechanism of action is competition for nutrients (Janisiewicz, 1998). Mutants that use new substrates, not metabolised by the pathogen, to provide a nutritional advantage, could also be induced (Janisiewicz, 1998).

Lastly

chlorogenic

some

acid,

phenolic

present

at

compounds, the

wound

including sites,

benzoic

adversely

or

affect

colonisation by yeast; for this reason it is conceivable to attempt to obtain strains resistant to these phenolic compounds (Bizeau et al., 1989). Early experiments of transformation for marker genes have been successful: M. pulcherrima was transformed with the green fluorescent protein gene (Nigro et al., 1999), C. oleophila was transformed with the β-glucuronidase

gene

(Chand-Goyal

et

al.,

1998),

and

histidine

auxotrophs of C. oleophila were transformed with HIS3, HIS4 and HIS5 genes (Chand-Goyal et al., 1999). In all cases the transformed antagonists maintained their biocontrol capability and there were no detectable differences between the wild-type and the transformants. All these studies were accomplished only to obtain variants of the antagonistic strains with a genetically stable marker to expedite studies on the ecology of the yeast antagonists on the fruit surface, but are highly effective for subsequent insertions of useful genes. Jones and Prusky (2002) have investigated the possibility of expressing a DNA sequence in Saccharomyces cerevisiae to allow the production of a cecropin A-based antifungal peptide. Yeast transformants inhibited the growth of germinated Colletotrichum coccodes spores and inhibited 15

Chapter 1 ____________________________________________________________________________________________

decay developments caused by the pathogen in tomato fruit. The lack of activity toward nontarget organisms by the peptide and the use of S. cerevisiae as a delivery system suggest that this method could provide a safe alternative for postharvest disease control. 5. Conclusions From this review, it appears that some significant progress has been made toward biological and integrated control of postharvest diseases on fruits. Some biofungicides are already on the market in a few countries, and will probably become more widely available as they are registered in more areas. Other BCAs should reach the market soon. Postharvest conditions provide an ideal niche for BCAs since they are less subject to sudden weather changes, and are often equipped with a sophisticated climate control system. It is unrealistic to assume that perfect conditions for the development of BCAs will always prevail in the warehouse, and as a result, biofungicides will rarely stand alone as a complete measure of disease control under all conditions. For this reason, scientists, growers and consumers alike must accept the fact that BCAs are usually not as effective as pesticides. The success of biological control greatly depends on influencing the consumer to prefer inner quality to outward appearance. At the moment, biological control should be viewed as an important if not essential component of an integrated disease management scheme if a significant and permanent reduction of pesticide use is our goal. Acknowledgements Work supported by a grant from the Italian Ministry for Environment and Territory. References

Abadias, M., Benabarre, A., Teixidó, N., Usall, J., Viñas, I., 2001a. Effect of

16

freeze drying and protectants on viability of the biocontrol yeast Candida sake.


Int. J. Food Microbiol. 65, 3, 173-182. Abadias,

M.,

Benabarre,

Teixidó, A.,

N.,

Viñas,

Usall,

I.,

Wisniewski, J.,

2001b.

Viability, efficacy, and storage stability of

M.E.,

1992.

UV-induced

resistance to postharvest diseases of citrus fruit. Phytochem. Phytobiol. B.: Biol. 15, 367.

freeze-dried biocontrol agent Candida

Chand-Goyal T., Eckert J.W., Droby, S.,

sake

Atkinson,

using

different

protective

and

K.,

1998.

A

method

for

rehydration media. J. Food Prot. 64, 6,

studying the population dynamics of

856-861.

Candida oleophila on oranges in the

Aharoni, Y., Copel, A., Fallik, E., 1993. Hinokitiol (b-thujaplicin), for postharvest decay control on “Galia” melons. N. Z. J. Crop Hortic. Sci. 21, 165-169.

G.W.,

Nordheim,

E.V.,

Morphogenesis

and

adhesion

Aureobasidium

pullulans.

1994.

Can.

of J.

Microbiol. 40, 6-17. Barkai-Golan, R., Phillips, D. J., 1991. Postharvest

heat

treatment

using

a

selective

isolation

medium and PCR technique. J. Microbiol. Res. 153, 3, 265-270. Chand-Goyal, T., Eckert, J.W., Droby, S.,

Andrews, J.H., Harris, R.F., Spear, R.N., Lau,

grove,

of

fresh

fruits and vegetables for decay control. Plant Dis. 75, 1085-1089.

Glickmann,

E.,

Atkinson,

K.,

1999.

Transformation of Candida oleophila and survival of a transformant on orange fruit under field conditions. Curr. Genet. 35, 51-57. Chernin, L.S., De La Fuente, L., Sobolev, V., Haran, S., Vorgias, C.E., Oppenheim, A.B., Chet, I., 1997. Molecular cloning, structural analysis, and expression in Escherichia coli of a chitinase gene from

Bizeau, C., Ponchant, F., Moreau, M.,

Enterobacter

1989. Evolution of yeast flora on cider

Environm. Microbiol. 63, 834-839.

apples of Britanny. Proc. 7th Int. Symp. on Yeasts, Perugia, Italy. Yeast, vol. 5. John Wiley & Sons, Ltd., Chichester, UK, 491-494.

agglomerans.

Appl.

Conway, W.S., Janisiewicz, W.J., Klein, J.D., Sams, C.E., 1999. Strategy for combining

heat

infiltration,

and

treatment, biological

calcium

control

Castoria, R., De Curtis, F., Lima, G.,

reduce

Caputo, L., Pacifico, S., De Cicco, V.,

apples. HortScience 34, 4, 700-704.

2001. Aureobasidium pullulans (LS-30) an antagonist of postharvest pathogens of fruits: study on its modes of action. Postharvest Biol. Technol. 22, 7-17.

postharvest

decay

E., Viñas I., 2001. Production of the biocontrol agent Pantoea agglomerans strain CPA-2 using commercial products and

control

Biotechnol. 56, 367-371.

postharvest

disease.

to

“Gala”

Costa, E., Teixidó, N., Usall, J., Atares,

Chalutz, E., Droby, S., 1998. Biological of

of

In:

Boland, G.J., Kuykendall, L.D. (Eds.), Plant-Microbe Interactions and Biological Control Marcel Dekker, New York, USA, pp. 157-170. Chalutz, E., Droby, S., Wilson, C.L.,

by-products.

Appl.

Microbiol.

De La Cruz, J., Pintor-Toro, J.A., Benitez, T., Llobell, A., Romero, L.C., 1995. A novel endo-1,3-glucanase, BGN 13.1, involved

in

the

mycoparasitism

of

Trichoderma harzianum. J. Bacteriology

17

Chapter 1 ____________________________________________________________________________________________

177, 6937-6945. Downing,

Microbiol. 91, 4, 652-659. 2000.

El-Ghaouth, A., Arul, J., Grenier, J.,

Introduction of the Serratia marcescens

K.J.,

Asselin, A., 1992. Antifungal activity of

chiA

chitosan on two postharvest pathogens

gene

Thomson, into

J.A,

an

endophytic

Pseudomonas

fluorescens

biocontrol

phytopathogenic

of

for

the fungi.

Can. J. Microbiol. 46, 4, 363-369.

of strawberry fruits. Phytopathology 82, 398-402. El-Ghaouth, A., Smilanick, J.L., Brown,

Droby, S., Chalutz, E., 1994. Mode of

G.E.,

action

Wilson,

of

biocontrol

agents

of

Ippolito,

A.,

C.L.,

2000a.

Wisniewski,

M.,

Application

of

postharvest diseases. In: Wilson, C.L.,

Candida saitoana and glycolchitosan for

Wisniewski,

the control of postharvest diseases of

M.E.

(Eds.),

Biological

Control of Postharvest Diseases. Theory

apple

and Practice. CRC Press, Boca Raton,

commercial conditions. Plant Dis. 84,

USA, pp. 63-75.

243-248.

Droby, S., Chalutz, E., Wilson, C.L.,

El-Ghaouth, A., Smilanick, J.L., Wilson,

Wisniewski, M.E., 1989. Characterization

C.L.,

of

of

performance of Candida saitoana by the

Debaryomices hansenii in the control of

addition of glycolchitosan for the control

Penicillium digitatum on grapefruit. Can.

of postharvest decay of apple and citrus

J. Microbiol. 35, 794-800.

fruit. Postharvest Biol. Technol. 19, 103-

the

biocontrol

activity

Droby, S., Hofstein, R., Wilson, C.L.,

and

citrus

2000b.

fruit

under

Enhancement

of

semi-

the

110.

Wisniewski, M.E., Fridlender, B., Cohen,

El-Ghaouth,

L., Weiss, B., Timar, D., Chalutz, E.,

Wisniewski, M., Wilson, C.L., 2000c.

1993.

Improved control of apple and citrus fruit

Pilot

guilliermondii:

testing a

of

Pichia

J.L.,

decay with a combination of Candida

postharvest diseases of citrus fruit. Biol.

saitoana and 2-deoxy-D-glucose. Plant

Control 3, 47-52.

Dis. 84, 249-253.

Droby, S., Vinokur, V., Weiss, B., Cohen,

El-Ghaouth, A., Wilson, C.L., Wisniewski,

L., Daus, A., Goldschmidt, Porat, R.,

M.,

2002.

to

cytochemical aspects of the biological

Penicillium digitatum in grapefruit by the

control of Botrytis cinerea by Candida

yeast

saitoana in apple fruit. Phytopathology

biocontrol

of

agent

Smilanick,

of

Induction

biocontrol

A.,

resistance agent

Candida

1998.

Ultrastructural

and

oleophila. Phytopathology 92, 393-399.

88, 282-291.

Eckert, J.W., Ogawa, J.M., 1985. The

El-Ghaouth, A., Wilson, C.L., Wisniewski,

chemical

M., 1997. Antifungal activity of 2-deoxy-

control

of

postharvest

diseases: subtropical and tropical fruits.

D-glucose

Annu. Rev. Phytopathol. 23, 421-454.

Penicillium

Edwards, S.G., Seddon, B., 2001. Mode of antagonism of Brevibacillus brevis against Botrytis cinerea in vitro. J. Appl.

18

stolonifer.

on

Botrytis

expansum,

and

Ultrastructural

cinerea, Rhizopus and

cytochemical aspect. Phytopathology 88, 772-779.


Filonow, A.B., 2001. Butyl acetate and

Ippolito, A., Nigro, F., 2000. Impact of

yeasts

preharvest

interact

in

adhesion

and

application

of

biological

germination of Botrytis cinerea conidia in

control agents on postharvest diseases

vitro and in fungal decay of golden

of fresh fruits and vegetables. Crop Prot.

delicious apple. J. Chem. Ecol. 27, 831-

19, 610-619.

844.

Janisiewicz, W.J., 1994. Enhancement of

Filonow, A.B., 1998. Role of competition

biocontrol of blue mold with nutrient

for sugars by yeasts in the biocontrol of

analog 2-deoxy-D-glucose on apples and

grey mould of apple. Biocontrol Sci. Te-

pears.

chnol. 8, 243-256.

2671-2676.

Appl.

Environ.

Microbiol.

60,

Gueldner, R.C., Reilly, C.C., Pusey, R.L.,

Janisiewicz, W.J., 1998 – Biocontrol of

Costello,

postharvest diseases of temperate fruits

C.E.,

Arrendale,

R.F.,

Cox,

R.H., Himmelsbach, D.S., Crumley, F.G.,

–

Cutler,

and

Boland, G.J., Kuykendall, L.D. (Eds.),

antifungal

Plant-Microbe Interactions and Biological

peptides in biological control of peach

Control. Marcel Dekker, New York, USA,

brown rot with Bacillus subtilis. J. Agric.

pp. 171-198.

H.G.,

identification

1988. of

Isolation

iturins

as

Food Chem. 36, 366-370

Challenges

Janisiewicz,

and

W.J.,

opportunities.

Bors,

R.,

In:

1995.

Gullino, M.L., Kuijpers, L.A.M., 1994.

Development of a microbial community

Social

of bacterial and yeast antagonists to

and

political

implications

of

managing plant diseases with restricted

control

fungicides

pathogens. Appl. Environ. Microbiol. 61,

in

Europe.

Annu.

Rev.

wound-invading

postharvest

Phytopathol. 32, 559-579.

3261-3267.

Harvey, J.M., 1978. Reduction of losses

Janisiewicz, W.J., Conway, W.S., Glenn,

in fresh market fruits and vegetables.

D.M.,

Annu. Rev. Phytopathol. 16, 321-341.

biological control and calcium treatment

Hofstein, R., Friedlender, B., Chalutz, E., Droby, S., 1994. Large scale production

for

Sams,

C.E.,

controlling

1998.

postharvest

Integrating decay

of

apples. HortScience 33, 1, 105-109.

and pilot testing of biocontrol agents of

Janisiewicz, W.J., Jeffers, S.N., 1997.


Efficacy of commercial formulation of

Wisniewski, M. (Eds.), Biological Control

two biofungicides for control of blue

of Postharvest Diseases – Theory and

mold and gray mold of apples in cold

Practice. CRC Press Inc., Boca Raton, FL,

storage. Crop Prot. 16, 629-633.

USA, pp. 89-100.

Janisiewicz,

W.J.,

Tworkoski,

T.J.,

Ippolito, A., El-Ghaouth, A., Wilson, C.L.,

Sharer, C., 2000. Characterizing the

Wisniewski,

M.,

of

mechanism

postharvest

decay

by

postharvest diseases on fruits with a

Aureobasidium pullulans and induction of

simple method to study competition for

defense

nutrients.

2000. of

responses.

Technol. 19, 265-272.

Control

apple

fruit

Postharvest

Biol.

of

biological

Phytopathology

control

90,

of

1196-

1200.

19

Chapter 1 ____________________________________________________________________________________________

Janisiewicz, W.J., Usall, J., Bors, B.,

Mcguire,

R.G.,

1992.

dynamics

of

Nutritional

biocontrol

of

enhancement

blue

mold

on

of

apples.

Phytopathology 82, 1364-1370.

J.,

Mahoney,

N.,

1991.

Postharvest

Population decay

antagonists growing epiphytically and within

Janisiewicz, W.J., Yourman, L., Roitman,

2000.

postharvest

wounds

on

grapefruit.

Phytopathology 90, 1217-1223. Mclaughlin,

R.J.,

Wisniewski,

M.E.,

control of blue mold and gray mold of

Wilson, C.L., Chalutz, E., 1990. Effects

apples and pears by dip treatments with

of

pyrrolnitrin,

solutions

a

metabolite

of

inoculum on

concentration

and

biological

control

salt of

Pseudomonas cepacia. Plant Dis. 75,

postharvest

490-494.

Candida sp.. Phytopathology 80, 456-

Jijakli,

M.H.,

Lepoivre,

Characterization

of

an

P.,

1998.

exo-β-1,3-

diseases

of

apple

with

461. Nigro, F., Finetti Sialer, M.M., Gallitelli,

glucanase produced by Pichia anomala

D.,

strain K, antagonist of Botrytis cinerea

Metschnikowia

on apples. Phytopathology 88, 335-343.

biocontrol agent of storage rot, with the

Jones,

R.W.,

Prusky,

D.,

2002.

Expression of an antifungal peptide in

1999.

Transformation

of

pulcherrima

320,

green fluorescent protein gene. J. Plant Pathol. 81, 3, 205-208.

Saccharomyces: a new approach for

Northover, J., Zhou, T., 2002. Control of

biological

postharvest

rhizopus rot of peaches with postharvest

Colletotrichum

treatments of tebuconazole, fludioxonil,

disease

control caused

of

the

by

and

coccodes. Phytopathology 92, 33-37. Leibinger, W., Breuker, B., Hahn, M., Mengden,

K.,

1997.

Control

of

Pseudomonas

syringae.

Can.

J.

Plant. Pathol. 24, 144-153. Piano,

S.,

Cerchio,

F.,

Migheli,

Q.,

postharvest pathogens and colonization

Gullino, M.L., 1998. Effetto di diverse

of the apple surface by antagonistic

sostanze

microorganisms

Metschnikowia pulcherrima 4.4 contro

in

the

field.

sull'attività

Botrytis

Phytopathology 87, 1103-1110. Lima, G., Arru, S., De Curtis, F., Arras, G., 1999. Influence of antagonist, host fruit and pathogen on the biological

cinerea

dell'antagonista e

sulla

sua

sopravvivenza sulle mele. Atti Giornate Fitopatologiche, 3-7 maggio 1998, 495500.

control of postharvest fungal diseases by

Piano,

yeasts. J. Ind. Microbiol. Biot. 23, 223-

Gullino, M.L., 1997. Characterization of

229.

the

Lima, G., Ippolito, A., Nigro, F., Salerno, M.,

1997.

Effectiveness

of

Aureobasidium pullulans and Candida

S.,

Neyrotti,

V.,

biocontrol

Metschnikowia Botrytis

Migheli,

Q.,

capability

pulcherrima

postharvest

rot

of against

of

apple.

Postharvest Biol. Technol. 11, 131-140.

oleophila against postharvest strawberry

Pusey,

rots. Postharvest Biol. Technol. 10, 169-

biocontrol

178.

diseases and their integration with other control

20

P.L.,

1994.

agents

strategies.

Enhancement for In:

of

postharvest Wilson,

C.L.,


Wisniewski,

M.E.

(Eds.),

Biological

Control of Postharvest Diseases. Theory and Practice. CRC Press, Boca Raton, WV, USA, pp. 77-88. Pusey,

P.L.,

apples. Postharvest Biol. Technol. 24, 123-134. Spotts,

R.A.,

Cervantes,

L.A.,

1986.

Population, pathogenicity, and benomyl

Wilson,

C.L.,

Hotchkiss,

resistance of Botrytis spp., Penicillium

M.W., Franklin, J.D., 1986. Compatibility

spp.,

of Bacillus subtilis for postharvest control

packinghouses. Plant Dis. 70, 106-108.

of peach brown with commercial fruit waxes,

dicloran

and

cold-storage

conditions. Plant Dis. 70, 622-626.

and

control of green and blue molds of oranges

agglomerans

Accumulation

Farg,

of

the

D.,

1994.

phytoalexins,

scoparone and scopoletine in citrus fruit subjected

to

various

in

A., Nunes, C., Viñas, I., 2001. Improving

G.,

T.,

piriformis

Teixidó, N., Usall, J., Palou, L., Asensio,

Rodov, V., Ben-Yehoshua, S., D’hallewin, Castina,

Mucor

postharvest

treatments. Int. Symp. Natural Phenols in Plant Resistance, Acta Hortic. 381, 517.

by

combining (CPA-2)

Pantoea

and

sodium

bicarbonate. Eur. J. Plant Pathol. 107, 7, 685-694. Teixidó, N., Viñas, I., Usall, J., Magan, N., 1998. Control of blue mold of apples by preharvest application of Candida sake grown in media with different water

Romano, M.L., Gullino, M.L., Garibaldi, A., 1983. Evaluation of the sensitivity to several

fungicides

of

post-harvest

pathogens in North-western Italy. Mede. Fac. Landoubouww. Gent 48, 591-602.

activity. Phytopathology 88, 960-964. Viñas I., Usall, J., Teixidó, N., Sanchis, V., 1998. Biological control of major postharvest pathogens on apple with Candida sake. Int. J. Food Microbiol. 40,

Smilanick, J.L., Denis-Arrue, R., 1992. Control of green mold of lemon with Pseudomonas species. Plant Dis. 76, 481-485.

9-16. Wilson, C.L., Wisniewski, M.E., Droby S., Chalutz E., 1993. A selection strategy for microbial

antagonists

Gullino,

vegetables. Scientia Hort. 53, 183-189.

microorganisms,

2002b. heat

Use

of

treatment,

and

natural compounds against Botrytis rot on

apple.

2nd

Proc.

International

Conference on the alternative control methods

against

plant

pests

and

of

control

postharvest

M.L.,

disease

to

Spadaro, D., Piano, S., Duverney, C.,

fruits

and

Wilson, C.L., Wisniewski, M.E., 1994. Biological Diseases.

Control Theory

of and

Postharvest Practice.

CRC

Press, Boca Raton, USA.

diseases, Lille, France, 4-7 March 2002,

Wisniewski, M.E., Biles, C., Droby, S.,

446-453.

Mclaughlin, R., Wilson, C.L., Chalutz, E.,

Spadaro, D., Vola, R., Piano, S., Gullino, M.L., 2002a. Mechanisms of action and efficacy of four isolates of the yeast Metschnikowia against

pulcherrima

postharvest

active

pathogens

on

1991. Mode of action of the postharvest biocontrol yeast, Pichia guilliermondii. I. Characterization

of

attachment

to

Botrytis cinerea. Physiol. Mol. Plant P. 39, 245-258.

21

Chapter 1 ____________________________________________________________________________________________

Wisniewski,

M.E.,

Wilson,

C.L.,

Hershherger, W., 1989. Characterization of

inhibition

of

Rhizopus

stolonifer

germination and growth by Enterobacter cloacae. Can. J. Bot. 67, 2317-2323. Zahavi, T., Cohen, L., Weiss, B., Schena, L., Daus, A., Kaplunov, T., Zutkhi, J., Ben-Arie, R., Droby, S., 2000. Biological control

of

Botrytis,

Aspergillus

and

Rhizopus rots on table and wine grapes in Israel. Postharvest Biol. Technol. 20, 115-124.

22

CHAPTER 2

Mechanisms of action and efficacy of four isolates of the yeast

Metschnikowia

pulcherrima

active

against

postharvest pathogens on apples

D. Spadaro, R. Vola, S. Piano and M.L. Gullino, Postharvest Biology and Technology (2002), 24 (3), 123-134.

Mechanism of Action and Efficacy ________________________________________________________________

Abstract The mechanisms of action and efficacy of four isolates (GS37, GS88, GA 102, and BIO126) of the yeast Metschnikowia pulcherrima against Botrytis cinerea, Penicillium expansum, Alternaria sp., and Monilia sp., all postharvest pathogens of apple fruits, were studied in vitro and on apples, in controlled and semi-commercial conditions. An application of a cell suspension (108 cells ml-1) of the antagonists in artificial wounds of apples permitted to reduce the growth of B. cinerea and P. expansum after storage at 23°C. A complete suppression of the pathogen was obtained against Monilia sp., storing at 23°C, and against B. cinerea and P. expansum, storing at 4°C. The results against Alternaria sp. were more variable. Applications of culture filtrates and autoclaved cells of the isolates were ineffective in reducing the diameter of the lesions on the fruits, supporting the hypothesis that living cells are necessary for biocontrol. In experiments of antagonism in vitro, on different solid substrates, a reduction of the micelial growth of the pathogens emerged, so that, at least in vitro, the antagonists could produce some diffusible toxic metabolites. Co-cultivating in vitro on a synthetic medium, B. cinerea spore (105 ml-1) germination was completely inhibited by the presence of 108 cells of the antagonists, while culture filtrates and autoclaved suspensions were not able to reduce germination. Dipping boxes of apples cv Golden delicious in a suspension of 107 antagonist cells ml-1 and storing for 8 months in controlled atmosphere at 1°C, the isolates showed control capability against B. cinerea and P. expansum similar to thiabendazole. Keywords: Alternaria sp.; antagonism; apple; biocontrol; Botrytis cinerea;

inegrated

pest

management,

Metschnikowia

pulcherrima;

Monilia sp.; Penicillium expansum; postharvest rot; storage. 1. Introduction Apple Postharvest rots, caused by Penicillium expansum, Botrytis cinerea, and Alternaria sp., are particularly severe even in production 25

Chapter 2 ________________________________________________________________________

areas where the most advanced storage technologies are available (Eckert and Ogawa, 1988). In Northern Italy they can cause serious losses, also due to the presence of populations of B. cinerea and P. expansum resistant to fungicides (Romano et al., 1983). In this area during the last years, Monilia sp. has become an increasing problem (Trevisan et al., 1992), also due to integrated pest management techniques, which avoid treatments at flowering. Currently, fungicides, when admitted, are the major means to control postharvest diseases (Eckert and Ogawa, 1988). However, the growing public concern over the health and environmental hazards associated with pesticide use in orchards (Wisniewski and Wilson, 1992), the development of fungicide resistant strains of postharvest pathogens (Spotts and Cervantes, 1986) and the deregistration of some of the most effective fungicides (Gullino and Kuijpers, 1994) have generated interest in the development of alternative non chemical methods. Biological control using microbial antagonists (Wilson and Wisniewski, 1994) has emerged as one of the most promising alternatives, either alone or as part of an integrated pest management to reduce pesticide use.

Several

biocontrol

agents

have

been

exploited

and

widely

investigated against different postharvest fungal pathogens (B. cinerea, Penicillium, Monilia, and Alternaria spp.) and tested on apple fruits (Mc Laughlin, 1990; Roberts, 1990; Gullino et al., 1994; Filonow et al., 1996; Piano et al., 1997; El-Ghaouth et al., 1998; Janisiewicz, 1998). At present, three products containing Pseudomonas syringae Van Hall, active against Botrytis, Penicillium, Mucor and Geotrichum spp., named Bio-Save 100, Bio-Save 110 and Bio-Save 1000 and commercialized by EcoScience Corp. (Janisiewicz and Jeffers, 1997), and a product containing Candida oleophila Montrocher, effective against Botrytis and Penicillium spp., named Aspire and commercialized by Ecogen Inc. (Hofstein and Fridlender, 1994), are available on the market for postharvest protection. Information on the mechanisms of action of most of the antagonists is still incomplete, but it is essential to develop appropriate formulation and methods of application, to obtain registration and to select new effective microorganisms. The main mode of action of yeast biocontrol 26


agents is believed to be competition for nutrients and space (Droby and Chalutz, 1994). Moreover, yeasts are able to colonize the carposphere for a long period in low humidity conditions, grow rapidly and are generally poorly sensitive to fungicides (Janisiewicz, 1991). Yeasts deserve particular attention, as their activity does not generally depend on the production of toxic metabolites, which could have a negative environmental

or

animal

toxicological

impact

(Smilanick,

1994).

Additional modes of action such as mycoparasitism, induced resistance and the production of lytic enzymes such as β-1,3-glucanase have been suggested (Wisniewski et al., 1991; El Ghaouth et al., 1998; Jijakli and Lepoivre, 1998). It is likely that there are multiple interactions between antagonist, fruit, pathogen, and other components of the natural epicarpic microflora (Droby and Chalutz, 1994). Recently we have isolated and selected four strains of the yeast Metschnikowia pulcherrima, named BIO126, GS 88, GA102 and GS37, which proved to be effective in containing Botrytis and Penicillium spp. rots in apple. Other strains of M. pulcherrima, coded 2.33 and 4.4, had already proved to be highly effective in the control of Botrytis rot of apple (Gullino et al., 1991; 1994; Migheli et al., 1997; Piano et al., 1997). The aim of this work was to determine the mechanism of action of the antagonistic isolates and to evaluate their efficacy under controlled and semi-commercial conditions with low temperature and controlled atmosphere. 2. Materials and methods 2.1. Microorganisms and culture conditions Metschnikowia pulcherrima (Pitt) M. W. Miller isolates BIO126, GS88, GA102, e GS37 were isolated from the carposphere of apple cv Golden delicious, harvested in unsprayed orchards located in Piedmont and Aosta Valley (GA102), Northern Italy. Cultures were stored at –20°C in cell suspension with 65% V/V of glycerol and 35% V/V of a solution MgSO4 100 mM and Tris ( pH 8.0) 25mM. Yeasts were grown on Yeast Peptone Dextrose (YPD: 10 g l-1 of Extract of Yeast Granulated Merck; 27

Chapter 2 ________________________________________________________________________

20 g l-1 of Triptone-Peptone of Casein Difco; 20 g l-1 of D(+)-Glucose Monohydrate Merck). Inocula of the antagonists for all the experiments were prepared by subculturing in 250 ml Erlenmeyer flasks containing 75 ml of Yeast Peptone Dextrose (YPD) and incubating on a rotary shaker (100 rpm) at 25°C for 48 hours. Yeast cells were collected by centrifugation at 2500 x g for 5 minutes, washed and resuspended in sterilised Ringer solution

(pH 6.9+0.1; Merck), and brought to a

standard concentration of 108 cells ml-1, unless otherwise stated, by direct counting with a haemacytometer. Three strains of Alternaria sp., isolated from rotted apples belonging to the cvs Golden delicious and Red delicious and selected for their virulence by inoculation in artificially wounded apples, were used as a mixture throughout this work, to ensure a higher level of disease. The same operations were accomplished for three strains of Botrytis cinerea, three of Monilia sp., and three of Penicillium expansum. Each strain was stored in tube with Potato Dextrose Agar (PDA; Merck) and 50 mg l-1 of streptomycin Merck at 4°C and was routinely inoculated and re-isolated from apple to maintain pathogenicity. Spore suspensions were prepared by growing the pathogens on Petri dishes for two weeks with Potato Dextrose Agar (PDA; Merck) and 50 mg l-1 of streptomycin Merck (strains of Alternaria sp. and P. expansum) or with Potato Glucose Malt (35 g l-1 of Potato Dextrose Agar Merck, 7 g l-1 of D(+)-Glucose Monohydrate Merck and 3 g l-1 of Malt Extract Merck) and 50 mg l-1 of streptomycin Merck (strains of B. cinerea and Monilia sp.). After two weeks incubation at 25°C, spores from the three strains of each pathogen were collected and suspended in sterile Ringer’s solution (Merck). After filtering through 8 layers of sterile cheese-cloth, spores were counted and brought to a final concentration of 105 ml-1. 2.2. Antagonism in apple artificial wounds Apples (Malus domestica Borkh, cv Golden delicious), disinfected in sodium hypochloride (NaClO, 1.0 % as chlorine) and rinsed under tap water, when dry were punctured with a sterile needle at the equatorial region (3 mm depth; 3 wounds per fruit). An antagonistic yeast cell 28


suspension (30 µl) was pipetted into wound. Autoclaved cells of M. pulcherrima and culture filtrates, prepared by centrifuging cultures of the antagonists and then filtering the supernatant through a 0.2 µm nitro-cellulose filter, were applied into wounds in order to evaluate their efficacy in reducing the incidence of the pathogens on apple fruit. Inoculated control fruits were pipetted, before pathogen inoculation, with 30 µl of Yeast Peptone Dextrose. After 3 hours, 30 µl of the spore suspension of the pathogen strains were pipetted in the wound. When dry, apples from different treatments were randomly packed in commercial plastic trays and either stored at 23°C for 6 days (B. cinerea and P. expansum), 12 days (Monilia sp.) or 18 days (Alternaria sp.) or kept at 4°C for 21 days (B.cinerea) or 28 days (P. expansum). Three fruits per treatment were used (9 inoculation sites) and each experiment was repeated three times. 2.3. Antagonism in vitro The growth rate of the pathogens was tested in different solid synthetic mediums: Potato Dextrose Agar (PDA, Merck), NYDA (as in Droby et al., 1989), Yeast Potato Dextrose-Agar (YPD with 20 g l-1 of Agar-agar Merck), CZAPEK-Agar (1 g l-1 of Potassium Phosphate Merck, 2 g l-1 of Sodium Nitrate Merck, 0.5 g l-1 of Magnesium Sulphate Merck, 0.5 g l-1 Potassium Chloride Merck, 0.01 g l-1 of Ferrous Sulphate Merck, 30 g of Saccharose Merck e 15 g of Agar-Agar Merck), Malt Extract Agar (Merck) and Apple-Agar (80 % v/v of apple homogenised filtered through Whatman no. 1 filter paper and 20 g l-1 of Agar-Agar Merck; pH: 5.5). A drop of the yeast cell suspension was striped on the substrate in 90 mm diameter Petri dishes, 20 mm from the border. A 6 mm mycelium disk of the pathogen was put 32 mm from the border and 32 mm from the strip of the antagonist. The radial growth of the mycelium towards the yeast strip was measured when the pathogen reached 32 mm of diameter towards the dish border, after storage at 23°C and in the dark. If the mycelium did not reach the border in 28 days, the substrate was considered not optimal for that pathogen growth. Three Petri plates per treatment were used and the experiment was repeated twice. 29

Chapter 2 ________________________________________________________________________

2.4. Effect on B. cinerea spores germination The effect of the isolates of M. pulcherrima on spore germination of B. cinerea isolate Gao1 was assessed in potato dextrose broth (PDB, Difco). Aliquots (100 µl) of spore suspension (5x106 spores ml-1) of the pathogen in Ringer’s solution were transferred to 10 ml plastic tubes containing 5 ml PDB. Living cells of each antagonistic yeast (100 µl of a suspension containing 5x107, 5x108, or 5x109 cells ml-1) or cells killed by autoclaving (100 µl of a suspension containing 5x108 cells ml-1) were added to each tube. As a control, the pathogen was added to 5 ml of a mixture (1:1) of PDB and of culture filtrates obtained as described from 48 hours old cultures of the 4 isolates of M. pulcherrima in PDB. After 12 h incubation of the 45° sloping tubes at 25°C on a rotary shaker (100 rpm), 100 spores per replicate were observed microscopically and their germination was evaluated. The treatments were replicated three times and the experiment repeated twice. 2.5.

Experimental

trials

under

semi-commercial

conditions:

storage in controlled atmosphere An experimental trial was carried out in Aosta (Aosta Valley, Northern Italy) during the period November 1998 – July 1999 in cooperation with the Institut Agricole Regional on artificially infected apples of the cv Golden delicious. Apples were harvested in orchards conducted with integrated pest management. Four boxes were used in each treatment (100 fruits per box). Ten apples per box, to reproduce the most probable conditions after harvesting, were artificially wounded at the equatorial region (3 mm depth; 3 wounds per fruit). All fruits were artificially inoculated by dipping for 60 seconds in 100 l tanks containing a conidial suspension (105 spores ml-1 per pathogen) of B. cinerea, P. expansum, and Alternaria sp.. After 3 hours, biocontrol isolates were applied at 107 cells ml-1 completely dipping the boxes of fruits for 60 sec in 100 l tanks containing cell suspensions prepared as described. The treatments included the four isolates of M. pulcherrima (BIO126, GS37, GS88 and GA102) and a chemical control (thiabendazole, Tecto 20 S, Elf 30


Atochem Agri Italy, 19,7 % a.i., 30 g a.i. 100 l-1). The inoculated, with the pathogens suspension, and uninoculated controls were represented by four boxes with 100 fruits per box, ten of them artificially wounded. When dry, apples were incubated at 23°C for 24 h and then stored at 1°C for 8 months in controlled atmosphere (2% O2 and 3% CO2). After 4 and 8 months storage, the rot incidence was evaluated and the relative importance of the different postharvest pathogens determined visually or through isolation on potato dextrose agar (PDA, Merck). 3. Results 3.1. Antagonism in apple artificial wounds M. pulcherrima BIO126, GS88, GA102, and GS37 cell suspension, applied at 108 cells ml-1, on apple artificial wounds stored at 23°C generally reduced the lesion diameter of Alternaria rot but the results were not homogeneous (Table 1). The major reduction to 0.9 % was due to the application of the isolate GS37. Culture filtrates and autoclaved cell suspensions also caused reduction of the lesions but, even in the same treatment, there were differences among single fruits. Culture filtrates and autoclaved cell suspensions of the four isolates were ineffective against Monilia sp. at 23°C, B. cinerea and P. expansum at 23° and 4°C (Tables 1 and 2). Cell suspension (108 cells ml-1) of the four isolates significantly reduced Botrytis rot on apples, but the best results were obtained storing the fruits at 4°C (Table 2). BIO126 and GS37 cell suspensions permitted to obtain a remarkable reduction of the lesions also at 23°C, respectively to 25.7 and 26.8 %. Addition of the cell suspensions of any of the four antagonistic strains completely inhibited the growth of Monilia sp. after 12 days of storage at 23°C (Table 1). The cell suspensions of the four strains applied on apple wounds resulted highly effective against P. expansum at 4°C after 28 days of storage (Table 2). Reduction of lesions was lower, ranging between 35.1 and 60.9 %, but anyway significantly different from the control, storing fruits at 23°C (Table 1).

31

Chapter 2 ________________________________________________________________________

Table 1 - Effect of cell suspension, cell-free culture filtrate and autoclaved cells of M. pulcherrima isolates BIO126, GS88, GA102, and GS37, applied in artificial wounds of “Golden delicious” apple, on the growth of different post-harvest rots. Storage at 23°C for 6 (Botrytis and Penicillium rots), 12 (Monilia rot), and 18 days (Alternaria rot).

Treatment

Percentage of control * Alternaria B.cinerea Monilia P.expansum

Uninoculated control Inoculated control 8

BIO 126 10 cells/ml

0.0

a

0.0 a

0.0 a

0.0 a

100.0

f

100.0 d

100.0 b

100.0 ef

25.7 b

0.7 a

35.1 b

5.7

BIO 126 culture filtrate 96.4

ab

113.2 de 110.5 b

101.7 ef

BIO 126 autoclaved

52.1 cde

f

115.8 de

107.6 f

8

GS 88 10 cells/ml

51.8 cde

60.7 c

GS 88 culture filtrate

69.9 def

GS 88 autoclaved

71.6 b 0.0 a

52.4 bc

103.6 de

75.1 b

81.0 de

34.5 abcd 122.1 de

31.8 b

97.5 ef

0.0 a

60.9 cd

GA 102 culture filtrate

42.3 bcde 113.9 de 105.2 b

102.5 ef

GA 102 autoclaved

81.8

ef

110.2 f

0.9

ab

GA 102 108 cells/ml

8

GS 37 10 cells/ml

0.0

a

102.9 de 125.5 e

114.4 b

26.8 b

0.0 a

40.2 bc


28.0 abc

103.8 de 103.7 b

92.9 ef

GS 37 autoclaved

17.9 abc

109.4 de

97.4 b

107.4 f

*Calculated on the lesion diameter. Values in the same column followed by the same letter are not statistically different by Duncan’s Multiple Range Test (P < 0,05).

In this experiment the effect of all the yeast isolated was studied on disease severity (lesion diameter) and not incidence (percent infection), evaluated in the trials under semi-commercial conditions. 3.2. Antagonism in vitro The co-culture on different solid substrate antagonists and pathogens, permitted to study antagonism in vitro. As shown in Table 3-A and 3-C, on some media, even after 28 days, Alternaria and Monilia spp. did not grow sufficiently to be influenced by the presence of the biocontrol agent, so that it was not possible to measure the inhibition. Alternaria sp. mycelium growth (Table 3-A) was significantly inhibited by 32


Table 2 - Effect of cell suspension, cell-free culture filtrate and autoclaved cells of M. pulcherrima isolates BIO126, GS88, GA102, and GS37, applied in artificial wounds of “Golden delicious” apple, on the growth of different post-harvest rots. Storage at 4°C for 21 (Botrytis rot) and 28 days (Penicillium rot).

Treatment

Percentage of control* B.cinerea

Uninoculated Control

P.expansum

0.0

a

0.0

a

100.0

cd

100.0

cd

57.1

b

0.0

a

BIO 126 culture filtrate 116.3

de

89.7

b

Inoculated control 8

BIO 126 10 cells/ml BIO 126 autoclaved 8

GS 88 10 cells/ml

121.3

e

101.2

d

5.3

a

0.0

a


103.6 cde

98.5

cd

GS 88 autoclaved

113.3 cde

98.2

cd

GA 102 108 cells/ml

9.5

a

0.0

a

95.6

c

85.5

b

100.0

cd

93.1

bcd

0.0

a

0.0

a


102.7

cd

88.5

b

GS 37 autoclaved

103.6 cde

92.4

bc

GA 102 culture filtrate GA 102 autoclaved 8

GS 37 10 cells/ml

*See table 1.

the presence of the antagonist strip on APPLE (Figure 1) and CZAPEK. B. cinerea (Table 3-B) growth was reduced by the four strains on YPD and NYDA and by GA102 on PDA. Monilia sp. (Table 3-C) was partially inhibited by all the potential antagonists on APPLE. P. expansum (Table 3-D) radial growth was significantly inhibited by the four isolates on NYDA and partially on YPD and APPLE. 3.3. Effect on B. cinerea spore germination By co-culturing on potato dextrose broth (PDB, Difco), the effect of M. pulcherrima BIO126, GS88, GA102, and GS37 on spore germination of B. cinerea was evaluated (Table 4).

33

Chapter 2 ________________________________________________________________________

Table 3 - Inhibition of the mycelium growth of Alternaria sp. (3-A), Botrytis cinerea (3-B), Monilia sp. (3-C), Penicillium expansum (3-D) by M. pulcherrima isolates BIO126, GS88, GA102, and GS37 in dual culture on different media at room temperature in the dark. 3-A

Days of

Mean inhibition (%) *

Substrates co-culturing** BIO 126 GS88 GA102 a

0

a

0

a

GS37

PDA

12

0

0

a

YPD

28

No growth on control plates

NYDA

28


CZAPEK

12

29

d

6

ab 10

bc 18

c

MALT

15

0

a

0

a

0

a

0

a

APPLE

12

24

b

25 b

24

b

31

b

Substrates

Days of

3-B

co-culturing**

Mean inhibition (%) * BIO 126 GS88 GA102 b

GS37

PDA

5

0

a

0

a

13

0

a

YPD

5

10

c

7

b

6.3 b

NYDA

5

19

c

13 b

14

bc 15 bc

CZAPEK

5

0

a

0

a

0

a

0

a

MALT

5

0

a

0

a

0

a

0

a

APPLE

5

0

a

0

a

0

a

0

a

8.3 bc

3-C

Substrates

Days of


co-culturing** BIO 126 GS88 GA102

34

b

4

a

PDA

20

12

YPD

28


NYDA

28


CZAPEK

28


MALT

15

0

a

0

APPLE

10

6.3

a

21 b

a

4.2 a

GS37 1

a

0

a

0

a

16

b

16

b


3-D

Substrates

Days of


co-culturing** BIO 126 GS88 GA102 PDA

5

15

b

13 b

GS37

7.3 ab 10

b

YPD

5

18

b

20 b

17

b

19

b

NYDA

5

13

b

21 b

17

b

19

b

CZAPEK

5

0

a

0

a

0

a

0

a

MALT

5

4.2

a

0

a

0

a

0

a

APPLE

5

18

b

19 b

29

c

18

b

*Values in the same row followed by the same letter are not statistically different by Duncan’s Multiple Range Test (P < 0,05). Control, always indicated with an a, is implied. **Days needed by control mycelium to reach a radius of 32 mm.

A complete inhibition of the spore germination emerged in presence of 108 cells ml-1 of the four strains of M. pulcherrima. With 107 cells ml-1 there was a partial inhibition: the percentage of conidia germinated compared with the control varied from 25.7 % of BIO126 to 73.0 % of GA102. In the presence of 106 cells ml-1 of antagonist only a negligible and not significant reduction in the germination was observed. A culture filtrate and killed cell suspension permitted the full germination of the spores. During the experiment, a strong attitude by the antagonist living cells to concentrate and adhere to non germinated spores of B. cinerea was observed. Adhesion was not observed with autoclaved cells. 3.4.

Experimental

trials

under

semi-commercial

conditions:

storage in controlled atmosphere Trial carried out dipping boxes of apples cv Golden delicious in a cell suspension of the antagonist was followed by an 8 month storage in controlled atmosphere at 1°C. First survey, after 4 months, showed a reduction in the incidence of rotted apples for all biological treatments (Table 5). Treatments with cell suspensions differ significantly from the chemical control, while they did not one from the other. BIO126 cell suspension 35

Chapter 2 ________________________________________________________________________

offered a control (9.0 % of rotted apples) similar to thiabendazole (8,7 %). Analyzing pathogens separately, it was possible to point out in all theses a major incidence of B. cinerea rots. Alternaria sp. rots were absent or at a very low level (1.4 %). Table 4 - Effect of cell suspension, cell-free culture filtrate and autoclaved cells of M. pulcherrima isolates BIO126, GS88, GA102, and GS37, on spore germination of Botrytis cinerea, by co-culturing in PDB at 25°C for 12 hours.

B. cinerea spore germination (%)* Control BIO126

99

e

culture filtrate

99

e

autoclaved

99

e

8

-1

0

a

7

-1

26

b

6

-1

99

e

culture filtrate

98

e

autoclaved

10 cells ml 10 cells ml 10 cells ml GS88

99

e

8

-1

0

a

7

-1

48

c

6

-1

99

e

culture filtrate

98

e

autoclaved

10 cells ml 10 cells ml 10 cells ml GA102

99

e

8

-1

0

a

7

-1

73

d

6

-1

97

e

99

e

10 cells ml 10 cells ml 10 cells ml

culture filtrate GS37

autoclaved

100

e

8

-1

0

a

7

-1

52

c

6

-1

98

e

10 cells ml 10 cells ml 10 cells ml

* Values in the same column followed by the same letter are not statistically different by Duncan’s Multiple Range Test (P < 0,05).

Biological treatments showed a control of P. expansum (average 36


incidence of 2.9 %) similar to thiabendazole (3.2 %), but a lower efficacy towards B. cinerea (average incidence of 9.7 %) in comparison with the chemical control (4.1 %). After 8 months’ storage (Table 5), biological treatments offered a control statistically not different from thiabendazole, but significantly different from the inoculated control, with a reduction of the incidence of rotted fruits, compared to first survey. Alternaria rot incidence increased prolonging storage, but remained a minority of the total rotted fruits. In comparison with the first survey, the incidence of B. cinerea was lower and that of P. expansum was similar. Summing up the percentages of rotted apples in the two surveys, the best result was offered by BIO126 (17.1 %) and the worst by GA102 (23.1 %). Compared with the chemical product (13.7 %), biocontrol agents showed an efficacy slightly lower, but the incidence of rotted fruits was significantly reduced with respect to the inoculated control (50.4 %). 4. Discussion 4.1. Mechanisms of action Generally the activity of antagonist yeasts is not based on the production of antibiotics or other secondary toxic metabolites (Droby and Chalutz, 1994); the results of this study show that M. pulcherrima BIO126, GS88, GA102, and GS37 principally act for the competition for space and / or nutrients. From the antagonism in apple artificial wounds, a substantial incapability emerged to antagonize all pathogens tested either by culture filtrate (without yeast cells) or by autoclaved cell suspension (killed cells). Living cells of the antagonists are necessary to guarantee the fungal control. The nutritional environment of the apple wound could be favourable to M. pulcherrima, that would colonize fruit tissues rapidly competing with pathogens for nutrients.

37

Chapter 2 ________________________________________________________________________

Table 5 - Efficacy of M. pulcherrima isolates BIO126, GS88, GA102, and GS37 against P. expansum, Alternaria sp., and B. cinerea, evaluated by dipping boxes of “Golden delicious” apples in a cell suspension of the antagonist and storing in controlled atmosphere at 1°C for 4 (1st survey) and 8 months (2nd survey).

First Survey

Rotted apples (%)

Treatments

P. expansum B. cinerea Alternaria sp.


Total

2.3

4.9

0.0

7.8

15.2

0.3

23 b

3.2

4.1

1.4

8.7 a

Yeast BIO 126

0.9

8.1

0.0

9a

Yeast GS 88

4.5

9.1

0.0

14 a

Yeast GA 102

1.7

12.2

0.0

14 a

Yeast GS 37

4.6

9.4

0.0

14 a

Inoculated control -2

Thiabendazole (3*10

-1

g l )*

Second Survey

7.2 a**

Rotted apples (%)

Treatments

P. expansum B. cinerea Alternaria sp.


Total

2.4

5.1

1.1

3.1

21.7

2.3

27.0 b

1.5

1.2

2.3

5.0 a

Yeast BIO 126

2.1

5.7

0.3

8.1 a

Yeast GS 88

3.1

4.1

0.7

7.9 a

Yeast GA 102

4.7

2.4

2.0

9.2 a

Inoculated control -2

Thiabendazole (3*10

-1

g l )*

8.7 a**

All isolates were applied at 107cells ml-1. *Apples were treated with 150 ml hl-1 of Tecto 20S (thiabendazole: 19,7 %). **See Table 4.

Also in in vitro experiments on B. cinerea spore germination, neither the culture filtrate nor autoclaved cells of the four isolates had any effect on the germination. The antagonistic activity of M. pulcherrima was dependent on the concentration of the antagonist: when applied at 106 cells ml-1 no yeast provided a satisfactory level of control. During the study a tenacious adhesion of living yeast cells to B. cinerea spores and hyphae was observed, in a manner similar to that described by Wisniewski et al. (1991). Attachment to pathogen conidia was not

38


observed after incubation of autoclaved antagonistic cells. This permits to suppose a direct interaction between antagonist and pathogen. Fig. 1. Dual culture of Alternaria sp. and M. pulcherrima BIO126 (left dish) and GS37 (right dish) on APPLE substrate in Petri dish for 12 days at room temperature in the dark. It is possible to notice the inhibition of the pathogen mycelium growth on the direction of the yeast strip side.

Often microbial antagonists provide different results in in vitro or in vivo conditions (Gullino, 1994). Co-culture experiments of antagonists and pathogens on different solid substrates bring to suppose that, at least in in vitro conditions, antagonistic yeasts could produce some metabolites toxic for the pathogens, differently from what results from the application of culture filtrate in vivo. Because inhibition of mycelial growth is only present on some substrates, it is probable that the nutritional

environment

influences

the

production

of

secondary

metabolites. The CZAPEK substrate, for instance, is poor in simple and complex sugars, therefore it could not favor the radial growth of the pathogen; furthermore it is rich in nitrates, that remarkably reduce antagonistic capability of the isolates, as already observed by Piano et al., 1997. APPLE substrate could easily simulate the nutritional 39

Chapter 2 ________________________________________________________________________

conditions of the wound; for many of the pathogens tested it was possible to point out some inhibition. On MALT, PDA, and NYDA, around the yeast strip, a pink halo was also visible, indicative of the metabolism of some compounds present on the substrate. 4.2. Efficacy To be commercially acceptable, antagonists must be effective under semi-commercial conditions. Since previous experiments, carried out measuring naturally developed rots, were ineffective because of the low incidence of the disease in the control, the trial of storage at 1°C in controlled atmosphere was carried out on partially wounded apples and inoculated with spore suspensions of Alternaria sp., Botrytis cinerea, and Penicillium expansum. This step permitted to obtain, after 8 months of storage, 50.4 % of rotted apples in the inoculated control, greatly increasing

the

probability

to

observe

significant

differences.

P.

expansum and B. cinerea were the pathogens more frequently isolated from the fruits. The low incidence of Alternaria rots gave no significant results. All biocontrol agents were effective in the reduction of the total rots. The origin of M. pulcherrima BIO126, GS88, GA102, and GS37, isolated from apple surface of “Golden delicious”, could have influenced positively the experiments since the isolates are naturally able to colonize the carposphere of the fruit. In trials of antagonism in apple wounds, the isolates offered a satisfactory biocontrol efficacy against B. cinerea, Monilia, sp. and P. expansum. Control of Botrytis and Penicillium rots by antagonist cell suspension was more homogeneous in the repetition, while efficacy towards Alternaria and Monilia rots resulted highly dependent on the fruit. An element of interference in the experiment was the difficulty to find fruits with the same degree of ripening; a high concentration in sugars, in fact, associated with senescence of the fruit, is a factor promoting the attack of certain pathogens, such as P. expansum and B. cinerea (Roberts, 1991). Comparing trials of antagonism carried out at 4 and at 23 °C, antagonists showed a major efficacy at 4°C and, in some cases, pathogens were completely inhibited. 40


Considering the results of the efficacy trials, a sharp difference in the biocontrol capability of the four antagonists does not exist: anyway BIO126 offered a higher control in semi-commercial conditions. 5. Conclusions M. pulcherrima BIO126, GS88, GA102, and GS37 were tolerant to benomyl and thiabendazole (benzimidazoles), and to vinclozolin and procymidone (dicarboximides), all of them registered for postharvest use (data not shown), therefore it could be possible to employ the biocontrol agents together with reduced dosages of these fungicides, in an integrated control perspective. The antagonists were tolerant to calcium chloride (data not shown). Between the strategies experimented during the last years in fruit protection, the association of biological agents with calcium chloride infiltration is to remember. The addition of this salt greatly enhances the action of the antagonist yeast (McLaughlin et al., 1990; Piano et al., 1998). The biocontrol capability of the yeasts, during the experiment in semicommercial conditions, was not affected by the low temperature of storage

(1°C)

of

the

fruits

and

by

the

controlled

atmosphere.

Antagonists are compatible with normal storage methods and with chemical products employed in post-harvest. From growth at different temperatures (data not shown), it resulted that the tested isolates do not grow at 37°C, which is important from a toxicological point of view. Another point favourable for a future commercialization of the studied yeasts is the lack of production of antibiotics active against the tested pathogens in vivo. The main mechanism of action used by the biocontrol agents is competition with pathogens for space and nutrients, but a secondary mechanisms of action with a synergistic effect could be also a direct interaction, such as parasitism, and some production of toxic metabolites in particular nutritional conditions, not investigated in this work. In future studies, to increase the knowledge on the mechanisms of action, it could be useful to purify and characterize the substances 41

Chapter 2 ________________________________________________________________________

released in the culture substrates. Enzymes surely involved in the process of antagonism, such as glucanases or chitinases, specific for the cell wall of the fungi, have already been isolated from the media where yeast antagonists were grown (Jijakli and Lepoivre, 1998; Wilson et al., 1994). It could also be useful supply the microorganisms with different nutrients, sources of carbon or nitrogen, to understand which are involved in the mechanism of competition (Janisiewicz et al., 1992; Piano et al., 1998). Future studies will also concentrate on the potential of resistance induced in the host tissue (Arras, 1996; Wilson and ElGhaouth, 1993). It would be interesting to evaluate the curve of the antagonist population on the carposphere, but it was not possible to mark the biocontrol agents for the sensitivity to antibiotics to differentiate them from the indigenous population. Trials of sensitivity to seven antibiotics resulted in a similar level of

tolerance, between the four isolates and

the population of yeasts present on the fruit (data not shown). The work is now continuing towards a molecular characterization of the isolates, either to evaluate the survival and dynamic of the population after application on the fruit or to study the environmental impact of a possible release in the open field. RAPD-PCR (Random Amplified Polymorphic DNA) and AP-PCR (Arbitrary Primed Polymerase Chain Reaction) techniques permit to distinguish efficiently also strictly correlated strains (Droby et al., 1999; Schena et al., 2000). Finally, the formulation should constitute a fundamental field of the studies, to permit the commercialization of the product (Fravel et al.,1999). Acknowledgements This research was carried out with grants from the Institut Agricole Regional of Aosta, Italy. Particular thanks to R. Pramotton and C. Duverney, who cooperated in part of the experiments, and to the anonymous reviewers for their useful suggestions.

42


References

Arras, G., 1996. Mode of action of an

Rev. Phytopathol. 26, 433-469.

isolate of Candida famata in biological

El-Ghaouth,

control of Penicillium digitatum in orange

Wisniewski,

fruits. Postharvest Biol. Technol. 8, 191-

and

198.

biological control of Botrytis cinerea by

Droby, S., Chalutz, E., 1994. Mode of action

of

biocontrol

agents

of

A., M.,

Wilson, 1998.

cytochemical

Candida

aspects

saitoana

C.

L.,

Ultrastructural

in

of

apple

the fruit.



Filonow, A.B., Vishniac, H.S., Anderson,

Wisniewski,

Biological

J.A., Janisiewicz, W.J., 1996. Biological

Control of Postharvest Diseases - Theory

control of Botrytis cinerea in apple by

and Practice. CRC Press, Boca Raton, pp.

yeasts from various habitats and their

63-75.

putative

M.E.

(Eds.),


mechanisms

of

antagonism.

Biol. Control 7, 212-220.

Wisniewski, M.E., 1989. Characterization

Fravel, D.H., Rhodes, D.J., Larkin, R.P.,

of

of

1999. Production and commercialization

Debaryomices hansenii in the control of

of biocontrol products. In: Albajes, R.,


Gullino, M.L., Van Lenteren, J.C., Elad,

J. Microbiol. 35, 794-800.

Y. (Eds.), Integrated Pest and Disease

the

biocontrol

activity

Droby, S., Cohen, L., Daus, A., Weiss, B., Horev, B., Chalutz, E., Katz, H., Keren-Tzur,

M.,

Shachnai,

A.,

1998.

Management

in

Greenhouse

Crops.

Kluwer Academic Publishers, Dordrecht, pp. 365-376.

Commercial testing of Aspire: a yeast

Gullino, M.L., 1994. Lotta biologica a

preparation for the biological control of

funghi agenti di marciumi della frutta in

postharvest decay of citrus. Biol. Control

post-raccolta. Informatore Fitopatologico

12, 97-101.

44 (9), 5-13.

Droby, S., Lischinski, S., Cohen, L.,

Gullino, M.L., Aloi, C., Palitto, M., Benzi,

Wiess, B., Daus, A., Chand-Goyal, T.,

D., Garibaldi, A., 1991. Attempts at

Eckert,

biological control of postharvest diseases

J.W.,

Manulis,

S.,

1999.

Characterization of an epiphytic yeast

of

population

Rijksuniv. Gent 56, 195-202.

of

grapefruit

capable

of

suppression of green mold decay caused by Penicillium digitatum. Biol. Control 16, 27-34.

apple.

Meded.

Fac.

Landbouww.

Gullino, M.L., Bonino, M., Piano, S., Testoni, A., Salimei, A., Duverney, C., 1994. Biological control of postharvest

Eckert, J. W., Ogawa, J. M., 1988. The

rots

chemical

control

Fourth International Conference on Plant

diseases:

deciduous

of

postharvest

fruits,

berries,

vegetables and root/tuber crops. Ann.

of

apples.

Proceedings

ANPP

-

Diseases, Bordeaux, France, December 6-8, 1994, pp. 333-340.

43

Chapter 2 ________________________________________________________________________


of

Social

solutions

and

political

implications

of

inoculum

concentration

and

biological

control

on

salt of


postharvest

fungicides

Candida sp. Phytopathology 80, 456-

in

Europe.

Ann.

Rev.

Phytopathol. 32, 559-579. Hofstein,

R.,

Friedlender,

diseases

of

apple

with

461. B.,

1994.

Migheli, Q., Gullino, M.L., Piano, S.,

Development of production, formulation

Galliano,

and delivery systems. Proceedings of the

Biocontrol capability of Metschnikowia

Brighton

pulcherrima and Pseudomonas syringae

Crop

Brighton,

Protection

England,

Conference,

November

21-24,

1994, vol. 3, pp. 1273-1280.

Nutritional

biocontrol

of

enhancement

blue

mold

on

of

apples.

Phytopathology 82, 1364-1370. Janisiewicz,

W.J.,

1991.

Duverney,

C.,

1997.

against postharvest rots of apple under semi-commercial condition. Meded. Fac.

Janisiewicz, W.J., Usall, J., Bors, B., 1992.

A.,

Landbouww. Univ. Gent 62, 1065-1070. Piano,

S.,

Cerchio,

sostanze Biological

control of postharvest fruit disease. In: Arora D.K., Rai B., Knudsen G.R. (Eds.), Handbook of Applied Micology, vol. 1:

F.,

Migheli,

Q.,

Gullino, M.L., 1998. Effetto di diverse sull’attività

dell’antagonista

biologica Metschnikowia

pulcherrima 4.4 contro Botrytis cinerea e sulla sua sopravvivenza su mele. Atti Giornate Fitopatologiche, pp. 495-500.

Soil and Plants. Marcel Dekker, Inc.,

Piano,

New York, pp. 301-326.

Gullino, M.L., 1997. Characterization of

Janisiewicz, W.J., 1998. Biocontrol of postharvest diseases of temperate fruits –

Challenges

and

opportunities.

In:

Boland, G.J., Kuykendall, L.D. (Eds.),

S.,

the

Neyrotti,

V.,

biocontrol


Migheli,

Q.,

capability

pulcherrima

postharvest

rot

of against

of

apple.

Postharvest Biol. Technol.11, 131-140.

Plant-Microbe Interactions and Biological

Roberts,

Control, Marcel Dekker, Inc., New York,

biological control of gray mold of apple

pp. 171-198.

by


R.G.,

1990.

Postharvest

Cryptococcus

laurentii.



Roberts, R.G., 1991. Characterization of


postharvest


deciduous fruit diseases by Cryptococcus


spp. In: Biological Control of Postharvest

Jijakli,

M.H.,

Lepoivre,

Characterization

of

an

P.,

1998.

exo-β-1,3-

biological

control

of

Disease of Fruit and Vegetables. ARS, USDA, ARS-92, pp. 32-35.

glucanase produced by Pichia anomala

Romano, M.L., Gullino, M.L., Garibaldi,

strain K, antagonist of Botrytis cinerea

A., 1983. Evaluation of the sensitivity to

on apples. Phytopathology 88, 335-343.

several

McLaughlin,

R.J.,

Wisniewski,

M.E.,


44

fungicides

pathogens

in

of

post-harvest

North-western

Italy.

Meded. Fac. Landbouww. Univ. Gent 48, 591-602.

Mechanism of Action and Efficacy

_________________________________________________________________________ Schena, L., Ippolito, A., Zahavi, T.,

J.L.

Cohen, L., Droby, S., 2000. Molecular

Biologically

approaches to assist the screening and

American

monitoring

Washington, pp. 181-185.

yeasts.

of

postharvest

European

biocontrol

Journal

of

Plant

Pathology 106, 681-691.

(Eds.),

management: Technologies,

Chemical

Society

Press,

Wilson, C.L., El-Ghaouth, A., Chalutz, E., Droby, S., Stevens, C., Lu, J., Khan, V.,

Smilanick, J.L., 1994. Strategies for the

Arul,

isolation

resistance

and

Pest Based

testing

of

biocontrol

J.,

1994. to

Potential

of

control

postharvest

induced

agents. In: Wilson, C.L., Wisniewski,

diseases of fruits and vegetables. Plant

M.E.

Dis. 78, 837-844.

(Eds.),

Postharvest

Biological

Diseases

-

Control Theory

of and

Practice. CRC Press, Boca Raton, pp. 2542. Spotts,

1994. Biological Control of Postharvest Diseases.

R.A.,

Cervantes,

L.A.,

1986.

Population, pathogenicity, and benomyl resistance of Botrytis spp., Penicillium spp.,

Wilson, C.L., Wisniewski, M.E. (Eds.),

and

Mucor

piriformis

in

packinghouses. Plant Dis. 70, 106-108.

Theory

and

Practice.

CRC

Press, Boca Raton, pp. 182. Wisniewski, M.E., Biles, C., Droby, S., McLaughlin, R., Wilson, C.L., Chalutz, E., 1991. Mode of action of the postharvest biocontrol yeast, Pichia guilliermondii. I.

Trevisan, D., Duverney, C., Petitjacques,

Characterization

U., Lantelme, C., Pramotton, R., 1992.

Botrytis cinerea. Physiol. Mol. Pl. Pathol.

L’infestation

39, 245-258.Wisniewski, M.E., Wilson,

par

la

moniliose

des

pommiers de la vallée centrale du Val

C.L.,

d’Aoste.

postharvest

Annales

Institut

Agricole

Regional 1, 195-205. Wilson,

C.L.,

Multifaceted postharvest

vegetables:

El-Ghaouth, biological diseases

1992.

A.,

1993.

control of

fruits

of

attachment

Biological diseases recent

of

control fruits

to

of and

advances.

HortScience 27, 94-98.

of and

vegetables. In: Lumsden, R.D., Vaughn,

45

CHAPTER 3

Control of Penicillium expansum and Botrytis cinerea on apple combining a biocontrol agent with hot water dipping and acibenzolar-S-methyl, baking soda, or ethanol application

D. Spadaro, A. Garibaldi and M. L. Gullino, Postharvest Biology and Technology (2004), accepted.

Integrated Control ________________________________________________________________

Abstract The

application

of

a

cell

suspension

of

the

BIO126

strain

of

Metschnikowia pulcherrima proved to be highly effective in the control of blue and grey mould, two of the most severe postharvest diseases on apple fruit. The possibilities to integrate the application of the antagonist with chemicals, such as acibenzolar-S-methyl (ASM), ethanol, or sodium bicarbonate, and a heat treatment were investigated in this work. The fruits were stored at 23°C for 5 days and at 4°C for 20 days. The antagonist, applied at 108 cells ml-1, proved to be the key element for the control of both pathogens, resulting more efficient after cold storage, with a reduction of 56.6% and 97.2% of the lesion diameter of blue and grey mould. Ethanol and acibenzolar-S-methyl permitted a partial reduction of grey mould severity. Heat treatment and sodium bicarbonate significantly improved the efficacy of the biocontrol agent against blue mould with storage at 23°C. To study the possibility of a single application of the biocontrol agent (107 cells ml-1) with reduced dosages of sodium bicarbonate or ethanol, the viability of the microorganism with these chemicals was studied and a new set of experiments was established. Against both pathogens, the higher reduction of the lesion diameter was obtained treating simply with the biocontrol agent. Significant results on blue mould were provided by the application of 20% ethanol or 5% sodium bicarbonate before the biocontrol agent and by the application of BIO126 in 0.1% sodium bicarbonate. The application of the cell suspension of BIO126 M. pulcherrima, preceded or not by a pre-treatment with sodium bicarbonate or ethanol, could become a successful alternative to fungicide usage in postharvest disease control of pome fruit. The fungistatic effect of ethanol or sodium bicarbonate could be associated to the effect of the biocontrol agent, persistent on the fruit for long periods. Keywords: benzothiadiazole, biocontrol agent, blue mould, ethanol, grey mould, heat treatment, sodium bicarbonate.

49

Chapter 3 ________________________________________________________________

1. Introduction Fungal pathogens are the main cause of postharvest losses of apples. These losses may reach as much as 50% during the shelf life of the fruits (Eckert and Ogawa, 1988). Blue mould, caused by Penicillium expansum Link, and grey mould, caused by Botrytis cinerea Pers.:Fr., are severe diseases worldwide on pome fruit, even in production areas where the most advanced storage technologies are available, such as Northern Italy (Romano et al., 1983). Currently, the most used and effective management strategy is the application of synthetic fungicides, but alternative methods are needed because of growing public concerns over the human health and environmental risks, the development of fungicide resistant strains of both pathogens (Spotts and Cervantes, 1986), and the problems encountered in the reregistration process by some of the most effective fungicides, especially for minor uses (Gullino and Kuijpers, 1994). Biological control with microbial antagonists has emerged as a promising alternative, with lower environmental impact, either alone or as part of an integrated pest management to reduce synthetic fungicides usage (Wilson and Wisniewski, 1994). At present, a class of products containing Pseudomonas syringae Van Hall (Janisiewicz and Jeffers, 1997), a product containing Candida oleophila Montrocher (Hofstein and Fridlender, 1994), and a biofungicide based on Cryptococcus albidus (De Koch, 1998), active against Botrytis spp., Penicillium spp., and other fungal pathogens, are available for postharvest protection in a few countries, but other biological products are under development (Spadaro and Gullino, 2004). Recently, different strains of the yeast Metschnikowia pulcherrima were isolated in our Department and studied for their efficacy and mode of action (Piano et al., 1997; Spadaro et al., 2002). All of them are effective against P. expansum and B. cinerea on apples with a varying degree of control and act through competition for nutrients and/or space, without producing toxic metabolites in vivo. The strain BIO126 was chosen for further studies. Different physical and chemical control

50


methods could be used together with the application of antagonists to obtain more consistent results. Pre-storage hot water dips of fruits at temperatures superior to 40°C are effective in controlling storage decays, not only by reducing the pathogen inoculum but also by enhancing the resistance of the fruit tissue, influencing host metabolism and ripening (Barkai-Golan and Philips, 1991). Postharvest dips are applied for a few minutes at high temperatures, because fungal spores and latent infections are either on the surface or in the first few cell layers under the peel of the fruit (Lurie, 1998). Hot water treatment may eliminate incipient infections, by removing spores from wounds and acting directly on their viability, and induce fruit defence mechanisms in the outer layers of epicarp which inhibit pathogen growth (Schirra et al., 2000). Moreover, generally regarded as safe (GRAS) compounds have been applied in hot water to improve the efficiency of their antifungal action (Smilanick et al., 1995). The chemical products chosen for studying the possibility of integration with the biocontrol agent are two natural compounds, ethanol and sodium bicarbonate, and an elicitor of systemic acquired resistance in the host tissue, acibenzolar-S-methyl (ASM). Ethanol occurs naturally in fruit and many other food products and the toxic effects of the alcohol on spores of fungal pathogens have been reported (Eckert and Ogawa, 1988). The alcohol has been tested for control of brown rot, also associated with hot water treatment (Margosan et al., 1997), with varying degrees of success: the flesh of the fruit treated with the alcohol was significantly firmer and injury to the fruit did not occur. Baking soda (NaHCO3), a carbonic acid salt, is a common food additive for pH-adjustment, taste, texture modification and spoilage control, permitted unrestrictedly for many applications by European and North American regulations. Moreover, it is listed as an approved ingredient on organic products (Mazzini, 2002). Sodium bicarbonate showed an antimicrobial activity against Penicillium digitatum on citrus fruit (Smilanick et al., 1999). Sodium bicarbonate is a poor eradicant that does not kill spores and its inhibitory action is not very persistent. Biocontrol agents, which can persist for long periods, may provide 51

Chapter 3 ________________________________________________________________

protection of the fruit from reinfection after treatment (Teixidó et al., 2001). Acibenzolar-s-methyl (or benzothiadiazole) is a chemical plant activator of the systemic acquired resistance (SAR) for crop protection (Kessmann et al., 1996). It is commercialised in some countries, including Italy, where it can be used on tomatoes, tobacco, cucurbits, pear, and hazelnut trees (Friedrich et al., 1996, Benhamou and Belanger, 1998). Acibenzolar-S-methyl has been tested on strawberry against B. cinerea: sprayed several times it delayed the development of grey mould on harvested fruits by about two days, increasing their shelf-life (Terry and Joyce, 2000). One preharvest spray of the plant activator on melon leaves decreased the incidence and extent of postharvest diseases (Huang et al., 2000). The aim of this study was to determine if the attacks of blue mould and grey rot on apple were reduced by a combination of the biocontrol agent M. pulcherrima strain BIO126 with a chemical elicitor of resistance, sodium

bicarbonate,

or

ethanol

and

hot

water

treatment.

The

experiments were devoted to develop an integrated strategy to control postharvest decay on apple fruit caused by B. cinerea and P. expansum, as effective as the traditional chemical control. A specific objective was the evaluation of positive or negative interactions between the three alternative methods of disease control. The possibility of a single application of the biocontrol agent together with a chemical compound – ethanol or sodium bicarbonate – was also considered. 2. Materials and methods 2.1. Antagonist and pathogens preparation Metschnikowia pulcherrima (Pitt) M.W. Miller strain BIO126, was isolated from the carposphere of an apple cv Golden delicious harvested in an unsprayed orchard located in Piedmont, Northern Italy, and it was studied for its efficacy and mechanism of action (Spadaro et al., 2002). The microorganism culture was stored at -20°C in cell suspension with 65 % v/v of glycerol and 35 % v/v of a solution MgSO4 100 mM and Tris 52


(pH 8.0) 25 mM. The strain was grown in Nutrient Yeast Dextrose Broth (NYDB), as described by Droby et al., 1989. Inocula of the antagonist for all experiments were prepared by subculturing in 250 ml Erlenmeyer flasks containing 75 ml of NYDB and incubating on a rotary shaker (100 rpm) at 25°C for 48 h. Yeast cells were collected by centrifugation at 1500 x g for 10 min, washed and resuspended in sterilized Ringer solution (pH 6.9+0.1; Merck), and brought to a standard concentration of 108 cells ml-1 by direct counting with a haemacytometer. Two isolates of B. cinerea and two isolates of P. expansum, obtained from rotted apples cv Golden delicious and selected for their virulence, were used as a mixture during the experiments to ensure a high level of disease. Each strain was stored in tubes with Potato Dextrose Agar (PDA; Merck) and 50 mg l-1 of Streptomycin Merck at 4°C. Spore suspensions used for fruit inoculation were prepared as described in Spadaro et al. (2002). 2.2. Efficacy of the combination of biological, chemical and heat treatments Apples (Malus domestica, cv Golden delicious), harvested in an Italian orchard conducted according to integrated pest management practices, were disinfected in sodium hypochloride (NaClO, 1.0 % as chlorine) and rinsed under tap water, dried at room temperature and punctured with a sterile needle at the equatorial region (3 mm depth; 3-4 mm wide; 3 wounds per fruit). Heat-treated fruit were dipped in deionised water at 50°C for 3 and 10 minutes, and left to dry for 1 hour. Benzo-(1,2,3)thiadiazole-7-carbothioic

acid

S-methyl

ester,

kindly

provided

by

Syngenta (50 % a. i. in wettable powder; Bion; Syngenta), was applied at 1.0 mg ml-1, commercial-grade ethanol at 10.0% wt vol-1, and sodium bicarbonate (pH 8.3 to 8.6; Sigma-Aldrich) at 3.0 % wt vol-1. Also a standard chemical (thiabendazole, Tecto 20S, Elf Atochem Agri Italy, 19.7% a. i.) was employed at 0.3 mg a. i. ml-1. The chemical compounds tested were applied diluted in sterile distilled water. Apples were dipped for 10 seconds in beakers containing 500 ml of the chemical suspension. 53

Chapter 3 ________________________________________________________________

After 3 hours, fruits exposed to treatments alternative to fungicides were treated with 30 µl of the cell suspension (108 ml-1) of M. pulcherrima strain BIO126 per wound. After 24 hours at room temperature, 30 µl of the spore suspension of B. cinerea or P.expansum (105 ml-1) were pipetted into the apple wounds. When dry, fruits were randomly packed in commercial plastic trays and stored at 23°C for 5 days and at 4°C for 20 days. Three fruits per treatment were used (9 inoculation sites). The severity of the diseases was determined measuring the mean lesion diameter of the rotted apples. The experiments were carried out three times. 2.3. Antagonist survival in co-culture with ethanol and sodium bicarbonate Erlenmeyer flasks containing 30 ml of NYDB were prepared and ethanol or sodium bicarbonate added at different concentrations. Commercialgrade ethanol was employed at the final concentration of 20%, 10%, 5%, and 2% wt vol-1 in the first trial and 5%, 4%, 3%, 2%, and 1% wt vol-1 in the second one. Sodium bicarbonate (pH 8.3 to 8.6; SigmaAldrich) was used at the final concentration of 5%, 3%, 1%, and 0.5% wt vol-1 in the first experiment and 0.5%, 0.1%, 0.05%, and 0.01% wt vol-1 in the second one. After subculturing the antagonist in NYDB 48 hours and counting with the haemacytometer as previously described, 30 µl of BIO126 cell suspension (108 or 107 cells ml-1) were added to the Erlenmeyer flasks containing the different suspensions of the two chemicals (final concentration: respectively 105 or 104 cells ml-1). The flasks were incubated on a rotary shaker (100 rpm) at 25°C for 36 hours. The viability of the cells was evaluated through direct observation and the cell suspension was counted by haemacytometer. 2.4. Efficacy of different combinations of the biological and chemical treatments Apples cv Golden delicious, harvested in an Italian orchard conducted by following integrated pest management, were disinfected, dried and 54


punctured as previously described. Some fruits were double-treated, by immersion in an ethanol or sodium bicarbonate suspension and successive inoculation with the antagonist. Other fruits were exposed to one single treatment, by immersion in a combination of the antagonist and lower concentrations of ethanol or sodium bicarbonate. The fruits treated twice were dipped for 60 seconds in a commercial-grade ethanol suspension (10 or 20% wt vol-1) or in a sodium bicarbonate (pH 8.3 to 8.6; Sigma-Aldrich) suspension (3 or 5% wt vol-1), left to dry for 3 hours and then treated with 30 µl of BIO126 cell (107 ml-1) suspension. Fruits exposed to single treatment were dipped for 60 seconds in a water suspension containing 107 cells ml-1 of the antagonist and commercial-grade ethanol (1 or 2% wt vol-1) or sodium bicarbonate (0.1% wt vol-1). After 24 hours at room temperature, 30 µl of the spore suspension of B. cinerea or P. expansum (105 ml-1) were pipetted in the wounds of each fruit. When dry, the apples treated differently were randomly packed in commercial plastic trays and stored at 23°C for 5 days. Five fruits per treatment were used (15 inoculation sites). The severity of the diseases was determined by the mean lesion diameter in mm of the rotted apples. The experiments were carried out twice. Table 1 - Summary of the significant effects (indicated as P>F) of antagonist (Ant), heat treatment (Heat), and chemical compounds (Chem) on the lesion size of grey and blue mould on apples cv Golden delicious after storage at 23°C for 5 days and at 4°C for 20 days.

Ant x Chem x Heat Ant x Chem Ant x Heat Chem x Heat

Grey mould (23°C) 0.20 0.01 0.00 0.13

Grey mould (4°C) 0.00 0.00 0.00 0.00

Blue mould (23°C) 0.02 0.04 0.05 0.09

Blue mould (4°C) 0.00 0.00 0.00 0.00

2.5. Statistical analysis Data of the single experiments of the three combined treatments were analysed

through

a

three-way

variance

analysis.

The

significant

interactions were chosen and successively analysed through a Duncan’s

55

Chapter 3 ________________________________________________________________

Multiple Range Test. The same test was employed for the analysis of the single experiments of antagonist survival and efficacy of different combinations of biological and chemical treatments. The program SPSSWIN was used. 3. Results 3.1. Combinations of biological, chemical and heat treatments The experiments against B. cinerea and P. expansum were carried out three times and a three-way variance analysis was executed on each repetition (Table 1). In the trials against B. cinerea and storage at 23°C, the interaction between the three variables was not significant (P=0.20) as the one between chemical and heat treatments (P=0.13), while the interactions between biological and chemical treatments (Table 2) and between biological and heat treatments (Table 3) were. In the experiment at 4°C against grey mould the analysis of variance gave in the four cases a P of

0.00.

Although

the

interactions

between

chemical

and

heat

treatments were significant, they were not considered because the study was focused on the identification of possible synergisms between the application of the biocontrol agent BIO126 and physical or chemical treatments alternative to the traditional fungicide. In the experiments against P. expansum, after 5 days of storage at 23°C, the interaction between the three variables was significant, as were the interactions between biological and chemical treatments (Table 2) and between biological and heat treatments (Table 3). In the trial at 4°C against blue mould the analysis of variance gave in the four cases a P of 0.00. 3.2. Efficacy of the combination of biological and chemical treatments The application of the cell suspension of BIO126 was effective against B. cinerea, providing an almost complete control of the disease, either 56


alone or in combination with the chemical products, storing the apples at 23°C or 4°C (Table 2). Without biological treatment, the chemicals employed were less effective. After storage at 23°C, only ethanol and acibenzolar-S-methyl

reduced

the

pathogen

growth

significantly

(respectively 29.7% and 14.8% of the lesion diameter). Storing at 4°C, ethanol and acibenzolar-S-methyl resulted in a higher protection of the fruit from grey mould, although not significantly different from the control, with a reduction of the pathogen severity of 26.8% and 21.5%. Thiabendazole, the chemical product

commercially

used,

resulted

completely ineffective against the strains of grey mould used. Table 2 - Effect of a cell suspension of M. pulcherrima strain BIO126, combined or not with acibenzolar-S-methyl, ethanol, and sodium bicarbonate, on B. cinerea and P. expansum growth on apples cv Golden delicious. Storage at room temperature (23°C) for 5 days and at 4°C for 20 days.

Grey mould severity (mm)a

Treatment

23°C storage

23°C storage

4°C storage

37.6

d

30.2

bc

28.7

d

25.8

d

36.2

d

29.4

bc

20.1

bc

16.8

bc

32.0

c

23.7

b

23.4

cd

20.1

c

26.4

b

22.1

b

25.5

d

17.4

bc

NaHCO3e

35.3

d

34.0

c

20.5

bc

19.0

c

BIO126f

1.5

a

0.8

a

20.8

bc

11.2

a

1.7

a

1.1

a

15.9

ab

7.9

a

1.8

a

1.1

a

16.7

ab

11.0

a

+ BIO126

1.5

a

0.3

a

18.5

abc

12.3

a

NaHCO3e + BIO126

1.1

a

1.9

a

12.1

a

11.3

a

Control Thiabendazole

b

Acibenzolar-S-methyl Ethanol

Thiabendazole

b

Ethanol

d

c

d

+ BIO126

Acibenzolar-S-methyl

a

4°C storage

Blue mould severity (mm)

c

+ BIO126

Values in the same column followed by the same letter are not statistically different by

Duncan’s Multiple Range Test (P < 0,05). b

300 µg a.i. ml-1: used as chemical control; c 500 µg a.i. ml-1; d 10.0% wt vol-1; e 3.0% wt

vol-1; f 108 cells ml-1.

57

a

Chapter 3 ________________________________________________________________

From the analysis of the interactions between the biological and chemical treatments in the control of blue mould, after storage at room temperature (Table 2), the most consistent result was offered by the combined application of the antagonist and sodium bicarbonate (57.7% of control). The biocontrol agent employed alone was less effective (27.6%), than applied together with other chemical products, such as ethanol and acibenzolar-S-methyl (35.5% and 41.9%), although the difference was not significant. Only sodium bicarbonate improved significantly the efficacy of the application of BIO126 (56.7%). Also without biological treatment, sodium bicarbonate was effective (28.4%). Thiabendazole, applied alone, provided the more consistent efficacy with respect to the other chemicals (29.8%). In the trial of efficacy against blue mould and storage at 4°C (Table 2), BIO126 acted significantly either alone (56.6% of control) or combined with acibenzolar-S-methyl (57.4%), sodium bicarbonate (56.2%) and ethanol (52.3%). In comparison with the experiment carried out at 23°C, the antagonist was much more efficient at the low temperatures of storage. The three chemicals and thiabendazole reduced significantly the pathogen attack with respect to the control, but the presence of the yeast resulted in a more consistent efficacy. Ethanol alone showed a disease severity of P. expansum (67.5%) similar to thiabendazole (65.1%). 3.3. Efficacy of the combination of biological and heat treatments In Table 3 the effectiveness against B. cinerea of the application of two different

hot

water

treatments

with

BIO126

is

reported.

The

microorganism was effective in every treatment, especially after 20 days of storage of the fruit at 4°C, when the control was complete. When apples were stored at room temperature, the best results were shown by the biocontrol agent applied alone (5.2% of pathogen severity with respect to the control) or with heat treatment (4.8% and 4.5%). Ten minutes of hot water treatment led to a significant reduction of the lesion diameter (12.9%) in apples stored at 23°C, but the same treatment followed by storage at 4°C was ineffective. Three minutes of 58


hot water immersion brought a significant result only in the case of apples stored at 4°C (12.9% of reduction of the lesion diameter). Combining biological and hot water treatments against P. expansum, the strain BIO126 of M. pulcherrima provided a good control of the pathogen at 23°C (29.2% of reduction) and 4°C (38.2%). The heat treatment improved the efficacy of the antagonist against blue mould after storage at 23°C and 4°C, but the difference was statistically significant only at room temperature. Immersion in hot water alone resulted in a significant control, with more effective result for the longer treatment (41.3% of reduction), in the trial carried out at 23°C and in an inconsistent control in the experiment conducted at 4°C. Table 3 - Effect of a cell suspension of M. pulcherrima strain BIO126, combined or not with two hot water treatments (3’ and 10’ at 50°C), on B. cinerea and P. expansum growth on apples cv Golden delicious. Storage at room temperature (23°C) for 5 days and at 4°C for 20 days.

Treatment

Grey mould severity (mm)a Blue mould severity (mm) 23°C storage 4°C storage

23°C storage 4°C storage

Control

35.4

c

27.3

c

32.2

d

23.9

b

3' 50°C

34.7

c

23.8

b

25.2

c

21.2

b

10' 50°C

30.8

b

29.1

c

18.9

ab

23.3

b

108 cells/ml BIO126

1.8

a

0.0

a

22.8

bc

14.8

a

108 cells/ml BIO126 + 3' at 50°C

1.7

a

0.0

a

15.1

a

14.5

a

108 cells/ml BIO126 + 10' at 50°C 1.6

a

0.0

a

15.6

a

12.9

a

a

a

See Table 2.

3.4. Antagonist survival in co-culture with ethanol and sodium bicarbonate When M. pulcherrima strain BIO126 was cultivated for 48 hours in NYDB with different concentrations of ethanol (20%, 10%, 5%, and 2%) no cell growth and no viability was detectable in all alcohol concentrations except at 2%. The experiment was repeated with co-culture in 5%, 4%, 59

Chapter 3 ________________________________________________________________

3%, 2%, and 1% ethanol in the synthetic broth. At 1% and 2% ethanol, the total number of cells and the viability were similar to the control. At 3% ethanol the total number of antagonistic cells were 0.5% compared to the control: the microorganism had a slowed growth but the cells were alive. At 4 and 5% ethanol all yeast cells were not viable and did not multiply. Sodium bicarbonate was applied in the NYDB liquid substrate at 5%, 3%, 1%, and 0.5% in the first experiment. At all concentrations the antagonist could not survive and grow: the yeast cells were not viable. In the second trial the effects of 0.5%, 0.1%, 0.05%, and 0.01% sodium bicarbonate were tested on the viability of the strain BIO126. The antagonist had a slowed growth at 0.1% salt concentration and the growth was similar to the control at 0.05% and 0.01% sodium bicarbonate. 3.5. Efficacy of different combinations of the biological and chemical treatments All treatments were significantly different from the control in the trial against B. cinerea (Table 4). Treatments where the biocontrol agent was applied alone (lesion diameter reduced to 5.7%) or after the application of 10% ethanol (3.9%), 20% ethanol (5.2), 3% sodium bicarbonate (6.9%) or 5% sodium bicarbonate (9.6%)were particularly effective. Fruits treated by immersion in a combination of the antagonist at 107 ml-1 and lower concentrations of ethanol or sodium bicarbonate, still significantly different from the control, were more susceptible to B. cinerea: the BIO126 cell suspension in 2% ethanol reduced the pathogen lesions by 55.4%, in 1% ethanol by 43.3% and in 0.1% sodium bicarbonate by 37.3%. In the experiments carried out against P. expansum all treatments caused a significant reduction of the lesion diameter compared to the control. The application of the cell suspension of BIO126 offered the higher level of control of the pathogen (14.7% of disease severity). When the application of 20% ethanol or 5% sodium bicarbonate preceded the biological treatment, the lesion diameter resulted greatly 60


reduced (27.8 and 22.7%). Lower control resulted from the application of 10% ethanol or 3% sodium bicarbonate before the yeast cell suspension (42.0 and 50.3%). A single application of the antagonist cell suspension in 1% and 2% ethanol resulted in a disease severity of 52.7 and 43.4%. A consistent efficacy was also showed by the application of the BIO126 cell suspension in 0.1% sodium bicarbonate). All fruits treated with 3% or 5% sodium bicarbonate, needed a final brushing or washing to eliminate the residues of the salt. No sign of phytotoxicity was observed. Table 4 - Effect of a cell suspension of M. pulcherrima strain BIO126, applied alone, after or together with different concentrations of ethanol or sodium bicarbonate, on B. cinerea and P. expansum growth on apples cv Golden delicious. Storage at room temperature (23°C) for 5 days.

Treatment

Disease severity (mm) Botrytis cinerea Penicillium expansum 35.2

c

23.5

d

7

2.0

a

3.5

a

7

10 cells/ml BIO126 in 1% ethanol

20.0

b

12.4

c

107 cells/ml BIO126 in 2% ethanol

15.7

b

10.2

bc

1.4

bc

Control 10 cells/ml BIO126

a

9.9

20% ethanol and 10 cells/ml BIO126

1.8

a

6.6

ab

107 cells/ml BIO126 in 0.1% NaHCO3

21.9

b

6.5

ab

2.4

a

11.8

c

a

5.3

ab

10% ethanol and 107 cells/ml BIO126 7

3% sodium bicarbonate and 107 cells/ml BIO126 7

5% sodium bicarbonate and 10 cells/ml BIO126 a

3.4

See Table 2.

4. Discussion The strain BIO126 of Metschnikowia pulcherrima proved its antagonistic potential in controlled and semi-commercial trials reducing blue and grey mould on apples. When apples cv Golden delicious were dipped in an antagonist cell suspension and stored at 1°C for 8 months, BIO126 showed postharvest rot control similar to benzimidazoles (Spadaro et al., 2002). The main mode of action involved in the biocontrol is

61

Chapter 3 ________________________________________________________________

competition for nutrients or space although a direct interaction can not be excluded (Spadaro et al., 2002). The biocontrol agent is very effective against B. cinerea but shows less consistent results towards P. expansum, and it is not as effective towards latent infections or previously established pathogens. Since alternatives to chemical control do not possess generally a broad spectrum of activity and they are not as effective as fungicides, a combination of alternative methods could be more effective and consistent than one alternative alone. Hot water treatment, sodium bicarbonate and ethanol are non-curative treatments whose effects in vivo are primarily fungistatic and not very persistent. Acibenzolar-Smethyl is an elicitor of systemic acquired resistance in the host tissue, that could help in the defence of the fruit from the pathogens. For the experiments carried out, two temperatures of storage were chosen. Room temperature (23°C) normally favours the growth of the pathogens (Snowdon, 1990), whereas 4°C is one of the temperatures for commercial fruit storage and favours the yeast antagonist fitness (Spadaro et al., 2002). Heat treatment is effective in sanitizing the fruit and enhancing the wound curing process. It has the added benefit of improving fruit colour but does not lead to softening, since it inhibits the synthesis of cell wall hydrolytic enzymes in the apple fruit, and reduces ethylene production (Lurie, 1998). Heat treatment could also damage the tissue of the fruit and, for this reason, some preliminary trials were carried out to assess the optimum time-temperature regime (data not published). The lowest times (1’ or 2’ at 50°C) were totally ineffective in controlling blue mould and grey rot and the highest ones (30’ at 50°C) caused damages to the apples, such as peel browning, as already noted by Klein and Lurie (1992). Pasteurisation with hot water at 50°C showed an ET50 of 1.5 minutes for B. cinerea spore germination and an ET50 of 0.9 minutes for the germ tube elongation of the same pathogen (Fallik et al., 1996). Combining heat treatment with an antagonist, in some cases, could complement the sanitary effect of the heat treatment with the residual protection of the biocontrol agent (Conway et al., 1999).

62


Mainly for its inability to survive at 50°C, the antagonist was applied after hot water treatment. The problem of applying the biocontrol agent before hot water treatment is that the microorganism must be heattolerant (Leverentz et al., 2000), but in this case problems of registration could rise. From growth at different temperatures (data not shown), it resulted that the BIO126 isolate does

not grow at

temperatures of 37°C or more, which is important from a toxicological point of view, especially in the case of contact with immunosuppressed patients (Mohl et al., 1998). Ethanol can be effective in reducing postharvest decay immediately after harvest by disinfecting the fruits. The major target of ethanol stresses is the lipid membrane but it has many other effects, such as denaturation of proteins on fungal cells (Mishra, 1993). A 10% ethanol solution, concentration chosen for the experiments carried out, had previously shown to be effective in controlling Monilinia fructicola and Rhizopus stolonifer on peaches and nectarines (Margosan et al., 1997) and Penicillium digitatum on lemons (Smilanick et al., 1995). Injury to the fruit did not occur, no odours or residues (differently from sodium bicarbonate) from the fruit were detected and an increased firmness of the fruit was a benefit, permitting an extension of the shelf-life (Margosan et al., 1997). On the other side, ethanol vapours can induce concern about manipulation and storage, so that a vapour abatement system should be developed, with increased cost for equipment and energy to operate it and a delay in cooling fruit before storage. The loss of ethanol efficacy after prolonged storage periods is probably an indication that decays developing at this time are the result of latent or secondary infections, rather than of surface wounds infections (Lichter et al., 2002). Sodium bicarbonate is inexpensive, readily available and can be used with a minimal risk of injury to the fruit. The inhibitory activity of sodium bicarbonate depends on the presence of salt residues within the wound infection courts occupied by the fungus and on interactions between this residue and constituents of the peel. In previous trials sodium bicarbonate was applied for control of B. cinerea on apple at 1% but it resulted ineffective (data not published). Oranges dipped for three 63

Chapter 3 ________________________________________________________________

minutes at room temperature in water with 2 to 4 % of sodium bicarbonate reduced decay caused by Penicillium italicum more than 50 % (Palou et al., 2001). For our experiments a concentration of 3% sodium bicarbonate was chosen. A disadvantage of sodium bicarbonate is that heating the solution will cause carbon dioxide evolution into air with a concomitant increase in solution pH but the addition of hypochlorite should permit the heating of the salt solution (Smilanick et al., 1999). Another issue of the treatments with sodium bicarbonate, differently from ethanol, is that the salt residues should be eliminated from the fruit skin before commercialisation. Acibenzolar-S-methyl has been used until now before harvesting for the protection of fruit from postharvest diseases. The chemical has an efficacy inferior to traditional fungicides and it needs a relatively long period of time after its application, before pathogen infection, to provide positive results (Kessmann et al., 1996). Moreover, to show positive results, it needs more than one application. In these experiments, it has been

used

once

in

postharvest

inoculation. In previous trials

48

hours

before

the

pathogen

conducted in our laboratory, also

acetylsalicylic acid was employed but it resulted totally ineffective (data not shown). In the experiments carried out, the strains of B. cinerea and P. expansum used were probably resistant to benzimidazoles, as can be observed from the low efficacy of thiabendazole. This low sensitivity is confirmed by recent evaluations on postharvest pathogens (Bertetti et al., 2003). The yeast antagonist resulted really effective in the control of grey mould on apples stored at room temperature or at 4°C. The experiments of combination with other physical or chemical treatments resulted unnecessary, with no significant increase of the protection from the pathogen. Hot water treatment alone showed inconsistent results against grey mould. Ethanol and acibenzolar-S-methyl permitted a partial reduction of the disease severity of B. cinerea but at a level not commercially acceptable. The effect of the application of BIO126 on apples was less consistent against blue mould, a more harmful disease also involved in the 64


production of mycotoxins. After storage at 23°C, heat treatment significantly improved the efficacy of the biocontrol agent, but not storing at low temperatures, where BIO126 was effective alone. Sodium bicarbonate significantly improved the efficacy of the antagonistic microorganism when apples were stored at 23°C but any chemical did not improve significantly the effect of BIO126 at 4°C. In the experiments carried out to study the possibility of a single application of the biocontrol agent with reduced dosages of sodium bicarbonate or ethanol, the strain of M. pulcherrima resulted compatible with low concentrations of ethanol (1 to 2%), as results also from the fact that this species of yeast is involved in the first step of the fermentation process of apples for cider-making (Beech, 1993). As the ethanol level raises (2 to 4%), these initial fermenters die out and the microbial succession is taken over by Saccharomyces cerevisiae. BIO126 and other biocontrol agents are not in general compatible with high concentrations of sodium bicarbonate, that reduces the growth and the viability of the microorganisms. Other organisms, such as Pantoea agglomerans,

are

tolerant

to

2%

sodium

bicarbonate

at

room

temperature, although the culturability of the bacterium is reduced by more than 1000-fold after 30 minutes in 2% sodium bicarbonate (Teixidó et al., 2001). In

the

new

set

of experiments, BIO126 was applied at lower 7

concentrations (10 cells ml-1 instead of 108 cells ml-1) to assess possible synergistic effects with the two chemicals. Against grey mould, the best results were obtained when the yeast was applied alone. Ethanol and sodium bicarbonate, when applied before, were not necessary to improve the efficacy. When the biocontrol agent was applied in a solution with 1% and 2% ethanol or 0.1% sodium bicarbonate, the control was reduced, probably because the fitness of the microorganism was lower. All the treatments against P. expansum showed a significant reduction of the disease severity. The higher reduction of the lesion diameter was obtained simply treating with the biocontrol agent. Significant results were provided also by the application of 20% ethanol or 5% sodium bicarbonate before the biocontrol agent. Also the application of BIO126 65

Chapter 3 ________________________________________________________________

in 0.1% sodium bicarbonate significantly reduced the lesion diameter of the rots. P. expansum has a behaviour related to the physiology of the fruit: smaller and less ripe apples are more resistant to the attack of the pathogen. During the experiments, great effort was employed in the selection of uniform fruits. In conclusion, it is possible to associate the fungistatic effect of ethanol or sodium bicarbonate to the effect of the biocontrol agent, persistent on the fruit for long periods. It is not useful, even if it could be more practical to apply the two treatments in one single step, because positive effects can not be revealed and the chemicals could inhibit the growth of the antagonist. An evaluation of the antagonist population survival in apple wound could clarify this question. Pre-treatment with sodium bicarbonate or ethanol and successive application of the cell suspension of BIO126 M. pulcherrima could become an alternative to fungicide usage in postharvest disease control of pome fruit, but registration and development studies to obtain a commercial product are necessary. Acknowledgments The authors gratefully acknowledge Dr. Francesca Alloati and Ms. Incoronata Luongo for their help in the realization of the trials and Dr. Jeanne Griffin for the linguistic advice. A special thank to the anonymous Referees, for their useful suggestions. This work was carried out with a grant from the Italian Ministry for the Environment and Territory within the Framework Agreement “Crop Protection with Respect of the Environment”. References

Barkai-Golan, R., Philips, D.J., 1991. Postharvest

heat

treatment

of

fresh

fruits and vegetables for decay control. Plant Dis. 75, 1085-1089. Beech,

66

F.W.,

1993.

Yeasts

in

Cider

Making. In: Rose, A.H., Harrison, J.S. (Eds.),

The

Yeasts

Vol.5

“Yeast

Technology”. Academic Press, London, pp. 169-213. Benhamou,

N.,

Belanger,

R.,

1998.


Benzothiazole-mediated

induced

Meier, B., Dincher, S., Staub, T., Uknes,

resistance to Fusarium oxysporum f. sp.

S., Métraux, J.P., Kessmann, H., Ryals,

radicis-lycopersici

J.,

in

tomato.

Plant

Physiol. 118, 1203-1212.

1996.

Benzothiadiazole

induces

systemic acquired resistance in tobacco.

Bertetti, D., Kejji, S., Garibaldi, A.,

Plant J. 10, 61-70.

Gullino, M.L., 2003. Valutazione della


sensibilità di alcuni agenti di marciume

Social

su


pomacee

in

post-raccolta

nei

and

political

implications

of

confronti di diversi fungicidi. Inf.tore

fungicides

fitopatol. 53 (6), 57-60.


Conway, W.S., Janisiewicz, W.J., Klein,

Hofstein,

J.D., Sams, C.E., 1999. Strategy for


combining

heat

and delivery systems. Proc. Brighton

infiltration,

and

reduce

treatment, biological

postharvest

calcium

control

decay

of

to

“Gala”

Crop

in

R.,

Europe.

Annu.

Friedlender,

Protection

B.,

Conf.,

Rev.

1994.

Brighton,

England, 21-24 November 1994, vol. 3,

apples. HortScience 34, 700-704.

pp. 1273-1280.

De Koch, S., 1998. Control of post-

Huang, Y., Deverall, B.J., Tang, W.H.,

harvest decay of fruit by means of

Wang,

antagonistic

microorganisms.

W.,

Wu,

F.W.,

2000.

Foliar

Ph.D.

application of acibenzolar-S-methyl and

dissertation, University of Stellenbosch,

protection of postharvest rock melons

Capetown, South Africa.

and Hami melons from disease. Eur. J.


Plant Pathol. 106, 651-656.

Wisniewski, M., 1989. Characterization


of

of


Debariomyces hansenii in the control of




J. Microbiol. 35, 794-800.


the

biocontrol

activity

Eckert, J.W., Ogawa, J.M., 1988. The

Kessmann, H., Oostendorp, M., Staub,

chemical

control

T., Goerlach, J., Friedrich, L., Lawton,

diseases:

deciduous

of

postharvest

fruits,

berries,

K., Ryals, J., 1996. CGA 245704: mode

vegetables and root/tuber crops. Annu.

of action of a new plant activator. Proc.

Rev. Phytopathol. 26, 433-469.

Brighton Crop Protection Conf., 18-21

Fallik, E., Grinberg, S., Alkalai, S., Lurie, S.,

1996.

The

effectiveness

of

November 1996, pp. 961-966. Klein, J.D., Lurie, S., 1992. Prestorage

postharvest hot water dipping on the

heating

control of grey and bleck moulds in

postharvest quality: interaction of time

sweet red pepper (Capsicum annuum).

and temperature. HortScience 27, 326-

Plant Pathol. 45, 644-649.

328.

Friedrich, L., Lawton, K., Ruess, W.,

Leverentz, B., Janisiewicz, W.J., Conway

Masner, P., Specker, N., Gut Rella, M.,

W.S., Saftner, R.A., Fuchs, Y., Sams,

of

apple

fruit

for

enhanced

67

Chapter 3 ________________________________________________________________

C.E.,

Camp,

M.J.,

2000.

Combining

yeasts or a bacterial biocontrol agent and

heat

treatment

postharvest decay of

to

reduce

“Gala” apples.


Botrytis

postharvest

rot

of

apple.

Postharvest Biol. Technol. 11, 131-140. Romano, M.L., Gullino, M.L., Garibaldi, A., 1983. Evaluation of the sensitivity to several

fungicides

post-harvest

Lichter, A., Zutkhy, Y., Sonego, L., Dvir,

pathogens

O., Kaplunov, T., Sarig, P., Ben-Arie, R.,

Meded. Fac. Landbouww. Univ. Gent 48,

2002.

591-602.

Ethanol

controls

postharvest

decay of table grapes. Postharvest Biol. Technol. 24, 301-308.

Schirra,

in

of

M.,

North-western

D’Hallewin,

Italy.

G.,

Ben-

Yehoshua, S., Fallik, E., 2000. Hostheat

pathogen interactions modulated by heat

treatments. Postharvest Biol. Technol.

treatment. Postharvest Biol. Technol. 21,

14, 257-269.

71-85.

Lurie,

S.,

Margosan, Simmons,

1998.

Postharvest

D.A., G.F.,

Smilanick,

Henson,

D.J.,

J.L., 1997.

Smilanick, J.L., Margosan, D.A., Henson, D.J.,

1995.

Evaluation

of

heated

Combination of hot water and ethanol to

solutions of sulfur dioxide, ethanol, and

control postharvest decay of peaches

hydrogen

and nectarines. Plant Dis. 81, 1405-

postharvest green mold of lemons. Plant

1409.

Dis. 79, 742-747.

Mazzini, F., 2002. Come cambiano le

Smilanick, J.L., Margosan, D.A., Usall, J.,

norme sui prodotti fitosanitari e loro

Michael, F., 1999. Control of citrus green

coadiuvanti. Inf.tore fitopatol. 52(1), 42-

mold by carbonate and bicarbonate salts

46.

and

Mishra, P., 1993. Tolerance of fungi to ethanol. In: Mycol. Ser., vol. 10. Marcel

W.,

Lerch,

M.M.,

the

influence

to

of

control

commercial

postharvest practices on their efficacy. Plant Dis. 83, 139-145. Snowdon, A.L., 1990. A colour atlas of

Dekker, New York, pp. 189-208. Mohl,

peroxide

Klotz,

M.,

Freidank, H., Zeitz, M., 1998. Infection of an intravenous port system with

post-harvest diseases and disorders of fruits and vegetables. Volume 1. Wolfe Scientific, London, Great Britain, 302 pp.

Metschnikowia pulcherrima Pitt et Miller.

Spadaro, D., Gullino, M.L., 2004. State

Mycoses 41, 425-426.

of

Palou, L., Smilanick, J.L., Usall, J., Viñas, I., 2001. Control of postharvest blue and green molds of oranges by hot water, sodium

carbonate,

and

sodium

S.,

Neyrotti,

V.,

Migheli,

Q.,

Gullino, M.L., 1997. Characterization of the

biocontrol

Metschnikowia

68

capability

pulcherrima

of against

and

future

perspectives

of

diseases. Int. J. Food Microbiol., in press (available online 25 September 2003). Spadaro, D., Vola, R., Piano, S., Gullino, M.L.,

bicarbonate. Plant Dis. 85, 371-376. Piano,

art

biological control of postharvest fruit

2002.

Mechanisms

of

action,

efficacy and possibility of integration with chemicals of four isolates of the yeast Metschnikowia pulcherrima active against

postharvest

pathogens

on



R.A.,

Cervantes,

L.A.,

1986.

Populations, pathogenicity and benomyl resistance of Botrytis spp., Penicillium spp.

and

Mucor

piriformis

in

packinghouses. Plant Dis. 70, 106-108. Teixidó, N., Usall, J., Palou, L., Asensio, A., Nunes, C., Viñas, I., 2001. Improving control of green and blue molds of oranges

by

agglomerans

combining (CPA-2)

Pantoea

and

sodium

bicarbonate. Eur. J. Plant Pathol. 107, 685-694. Terry,

L.A.,

Suppression

Joyce, of

grey

D.C., mould

2000. on

strawberry fruit with the chemical plant activator acibenzolar. Pest Manag. Sci. 56, 989-992. Wilson, C.L., Wisniewski, M.E. (Editors), 1994. Biological Control of Postharvest Diseases.

Theory

and

Practice.

CRC

Press, Boca Raton, FL, USA, 182 pp.

69

CHAPTER 4

Metschnikowia

pulcherrima:

a

promising

species

isolated from different food matrixes for biological control of postharvest diseases in apple

D. Spadaro, W. Sabetta, A. Acquadro, A. Garibaldi and M. L. Gullino, International Journal of Food Microbiology (2004), submitted.

Molecular characterization ________________________________________________________________

Abstract Seven strains of the yeast Metschnikowia pulcherrima, isolated from the carposphere of apples cv. Golden delicious, showed biocontrol capability against B. cinerea and P. expansum. PCR-RFLP of the 18S+ITS rDNA was tested as rapid and easy method for yeast species identification. The efficacy of these strains was compared with that of nineteen M. pulcherrima

strains,

isolated

from

different

matrixes

in

different

geographical regions. The strains were more effective in the control of B. cinerea than of P. expansum, after storage for 28 days at 4°C, with a mean reduction of the pathogen growth respectively to 30.0% and 49.3% of the control. Antagonistic properties could be owned by microorganisms of different origin. Strain 3043 isolated from grape must offered the best control of both diseases. To assess the genetic diversity of M. pulcherrima, the RAPD and AFLP techniques were used. With six RAPD primers 33 polymorphic bands were obtained, while with six AFLP primer pairs 729 polymorphic bands were scored. The genetic distances obtained with AFLP technique were visualized on a dendrogram. Strains isolated in different locations with high genetic diversity could have similar biocontrol potential. One primer pair, such as McaEaa or McgEat, resulted highly informative and sufficient to describe the genetic distance among the strains. Keywords: amplified fragment length polymorphism, biocontrol agent, Botrytis cinerea, molecular characterization, Penicillium expansum, yeast. 1. Introduction Fungal pathogens cause severe losses on apples during postharvest storage and commercialisation. Two of the most aggressive pathogens are Penicillium expansum and Botrytis cinerea (Snowdon, 1990). Currently

postharvest

fungicides

represent

the

main

method

of

controlling wound pathogens. However, alternative methods are needed because of growing public concerns about the human health and 73

Chapter 4

_________________________________________________________________________ environmental risks associated with fungicide residues in fruit. In many industrialised countries, an increasing part of the population prefers to buy organically grown pesticide-free fruit and vegetables (Verhoog et al., 2003). There is concern as well regarding the development of fungicide resistant strains of both pathogens (Spotts and Cervantes, 1986), the prohibitive costs of developing new active ingredients and the problems encountered in the reregistration process by some of the most effective fungicides, especially for minor uses (Gullino and Kuijpers, 1994). Biological control, using microbial antagonists, has emerged as an effective strategy to combat major postharvest decays of fruit (Wilson and Wisniewski, 1994; Janisiewicz and Korsten, 2002; Spadaro and Gullino, 2004). Yeasts deserve particular attention because of their ability to colonize the fruit carposphere for long periods under unfavourable conditions, rapid growth, poor sensitivity to fungicides and lack of production of toxic metabolites (Janisiewicz, 1991). At present, yeast-based biofungicides, active against Botrytis cinerea and Penicillium spp. on apple, pear and citrus fruit, are commercialised in the USA, Israel (a product containing Candida oleophila; Hofstein and Fridlender, 1994) and South Africa (a product based on Cryptococcus albidus; De Koch, 1998). During the last decade several yeast strains have been selected for their antagonistic properties in the control of postharvest pathogens of fruit (Gullino et al., 1991; 1994). Recently three strains of the yeast Metschnikowia pulcherrima, named BIO126, GS 88 and GS37 proved to be effective in containing Botrytis and Penicillium spp. rots in apple, especially at low storage temperatures (Spadaro et al., 2002; 2004). The strain 4.4 of M. pulcherrima already proved to be highly effective in the control of Botrytis rot of apple (Piano et al., 1997). M. pulcherrima is not a new yeast. This species of yeast is involved in the first step of the fermentation process of apples for cider-making (Beech, 1993). As the ethanol level raises (2 to 4%), these initial fermenters begin to die out and the microbial succession is taken over by Saccharomyces cerevisiae. In grape must and during the early phase of fermentation, apiculate yeasts belonging to the species Kloeckera apiculata are dominant and, to a lesser extent, isolates of M. 74


pulcherrima, can also be detected (Fleet and Heard, 1993). This species, present on different food matrixes, naturally occurs on fruits, buds and floral parts of certain apple trees (Boekhout and Robert, 2002) and has been reported to be a yeast species effective as biocontrol agent against postharvest decay of apple, table grape, grapefruit and cherry tomato (Schena et al., 2000; Janisiewicz et al. 2001; Spadaro et al., 2002; 2004). M. pulcherrima normally acts by consuming the nutrients on fruit and vegetable skins that allow rotcausing fungi to thrive (Piano et al., 1997; Janisiewicz et al., 2001). Molecular tools can assist to monitor the genetic and environmental fate of these agents after releasing. Random amplified polymorphic DNA (RAPD) and arbitrarily primed-PCR (AP-PCR) techniques were useful methods to identify and evaluate the survival rate of some fungi (Aureobasidium

pullulans)

and

yeasts

as

agents

for

postharvest

biological control (Schena et al., 1999; 2000). Specific fingerprints using amplified fragment length polymorphism (AFLP) technique have also been developed to monitor the population of Rhodotorula glutinis, Cryptococcus laurentii and Aureobasidium pullulans in both the field and cold room (Lima et al., 2003). One goal of this study was to compare the biocontrol capability against B. cinerea and P. expansum of the seven M. pulcherrima strains, all of them isolated from the carposphere of apples, with the same capability of other strains of the same species coming from different sources. Another goal was the assessment of the genetic diversity of M. pulcherrima strains having a different origin and the evaluation of the relationship between biocontrol capability and genetic distance. To obtain specific fingerprints of M. pulcherrima strains the AFLP technique was preferred to the RAPD technique. AFLP technique has been widely used to study plant genomes but rarely for fungal or yeast studies. 2. Materials and methods 2.1. Microorganisms and cultures Seven yeast strains, isolated in Northern Italy from the carposphere of 75

Chapter 4

_________________________________________________________________________ apples organically grown (BIO114, BIO126, BIO131, GS9, GS37, GS88 and 4.4), showed antagonistic properties in postharvest biological control against Botrytis cinerea and Penicillium expansum on apple cv. Golden delicious. Table 1 - The twenty-six Metschnikowia pulcherrima strains studied in this paper

They

Number

Code

Source

Geographical origin

1

BIO114

apple

Northern Italy

2

BIO126

apple

Northern Italy

3

BIO131

apple

Northern Italy

4

GS37

apple

Northern Italy

5

GS88

apple

Northern Italy

6

GS9

apple

Northern Italy

7

4.4

apple

Northern Italy

8

291

pear

Southern Italy

9

311

quince

Southern Italy

10

320

table grape

Southern Italy

11

3008

grape must

Central Italy

12

3041

cherry

Italy

13

3042

sour black cherry

Italy

14

3043

grape must

Central Italy

15

3345

wine

Northern Italy

16

3348

---

Sicily

17

3435

---

Sardinia

18

3527

grape

Spain

19

3835

Helix pomatia (snail)

---

20

3938

grape must

---

21

4064

grape must

Greece

22

4129

grape

Greece

23

4130

grape must

Slovenia

24

4185

grape must

Greece

25

4292

winery surface

Central Italy

26

4354

seawater

Italy

were

identified

by

PCR-RFLP

(restriction

fragment

length

polymorphism) analysis on the ITS region of the 5.8S rRNA gene and the two ribosomal internal transcribed spacers as Metschnikowia pulcherrima (Pitt) M. W. Miller (Esteve-Zarzoso et al., 1999) and the classification was confirmed by identification carried out by the DBVPG,

76


University of Perugia, Italy (www.agr.unipg.it/dbvpg). These isolates were compared for their biocontrol efficacy and genetic fingerprints with nineteen different strains of the same species. Three strains, 311, 291 and 320, were kindly given by the University of Bari (Schena et al., 2000; see table 1). Sixteen strains were purchased from the Industrial Yeast Collection of the Department of Plant Biology, University of Perugia, Italy (www.agr.unipg.it/dbvpg). Cultures were stored at –20°C in cell suspension with 65% V/V of glycerol and 35% V/V of a solution MgSO4 100 mM and Tris (pH 8.0) 25mM. Yeasts were grown on Yeast Peptone Dextrose (YPD: 10 g l-1 of granulated yeast extract Merck, Darmstadt, Germany; 20 g l-1 of triptone-peptone of casein Difco, Detroit, MI, USA; 20 g l-1 of D(+)glucose monohydrate Merck). Antagonists inocula were prepared by subculturing in 100 ml Erlenmeyer flasks containing 30 ml of YPD and incubating at 25°C on a rotary shaker (100 rpm) for 48 h. Yeast cells were collected by centrifugation at 2500 x g for 5 minutes, washed and resuspended in sterilised Ringer solution

(pH 6.9+0.1; Merck), and 8

brought to a standard concentration of 10 cells ml-1 by direct counting with a haemacytometer. Three strains of Botrytis cinerea Pers. : Fr. and Penicillium expansum Link, isolated from apples and selected for their virulence, were used as a mixture, to ensure a consistent level of disease. Spore suspensions were prepared by growing the pathogens on Petri dishes for 10 days with PDA and 50 mg l-1 of streptomycin Merck. Pathogen spores were suspended in sterile Ringer’s solution (Merck), filtered through 8 layers of sterile cheesecloth and brought to a final concentration of 105 ml-1. 2.2. Biocontrol capability trials The experiments of biocontrol efficacy against B. cinerea and P. expansum were carried out in apples (Malus domestica Borkh, cv Golden delicious). The fruits, disinfected in sodium hypochlorite (NaClO, 1.0 % as chlorine) and rinsed under tap water, when dry, were punctured with a sterile needle at the equatorial region (3 mm depth; 3 wounds per fruit). 30 µl of the cell suspension (108 cells ml-1) of the 26 M. 77

Chapter 4

_________________________________________________________________________ pulcherrima strains were pipetted into the wound. Control fruits were inoculated, before pathogen inoculation, with 30 µl of YPD. Also a standard chemical (thiabendazole, Tecto 20S, Elf Atochem Agri Italy, 19.7% a. i.) was employed at 0.3 mg a. i. ml-1. After 3 h, 30 µl of the B. cinerea or P. expansum suspension (105 spores ml-1) were pipetted into the wound. When dry, apples were randomly packed in commercial plastic trays and stored at 4°C for 28 days. Five fruits per treatment were used (15 inoculation sites) and both biocontrol trials were repeated twice. 2.3. DNA extraction Two ml of YPD culture of the yeast isolates were centrifuged at 2500 x g for 3 min. The pellets were suspended in 280 µl of EDTA 50 mM (pH 88.5) with 400 µg of lyticase (Sigma, St Louis, MO, USA) and incubated at 37°C for 45 min. After 3 min centrifugation, the pellets were treated with the Wizard Genomic DNA Purification kit (Promega Corp., Madison, WI, USA). Genomic DNA was controlled by electrophoresis (30 min at 100 V/cm) on 1% SeaKem LE agarose gel (FMC BioProducts, Rockland, ME, USA) in 1X TAE buffer (40 mM Tris, 40 mM acetate, 2 mM EDTA, pH 8.0; Maniatis et al., 1982); the gel was stained with ethidium bromide and visualized through UV light. Gel images were acquired with a Gel Doc 1000 System (Bio-Rad Laboratories, Hercules, CA, USA). A 1 kb DNA ladder (Gibco BRL, Rockville, MD, USA) was used as a molecular weight marker for an approximate quantification of the genomic DNA. A precise quantification in ng/µl was obtained by a BioPhotometer (Eppendorf, Hamburg, Germany). Purified DNA was stored in TE buffer (10 mM Tris-HCl; 0.1 mM EDTA; pH 8) at 4°C. 2.4. RAPD analysis Six 10-mer oligonucleotides were randomly chosen among the OPA, OPB, OPE, OPM and OPT series of Operon Technologies Inc. (Alameda, CA, USA: OPA 10, OPA 18, OPB 17, OPE 19, OPM 05 and OPT 16) and tested as primers for RAPD analysis on all strains. Amplification 78


reactions were performed in a volume of 15 µl containing 10 mM TrisHCl (pH 9.0), 1.5 mM MgCl2, 50 mM KCl, 0.1% Triton X-100, 0.01% (w/v) gelatin, 60 µM each dNTP (Promega), 5 pM each primer, 0.75 U SuperTaq DNA polymerase (HT Biotechnology, Qiagen, Cambridge, UK) and 1 µl of genomic DNA (0.1 ng/µl). PCR reactions were performed in a GeneAmp

PCR

System

9700

(Perkin-Elmer,

Norwalk,

CT,

USA)

programmed for one cycle of 2 min 30 s at 94°C, 45 cycles of 30 s at 94°C, 1 min at 36°C and 2 min at 72°C and one final cycle of 5 min at 72°C.

RAPD

analysis

was

repeated

at

least

twice

per

primer.

Amplification products were size-separated by electrophoresis on 1.5% agarose gel in 1X TAE buffer (1h and 30 min at 80 V/cm). 2.5. AFLP analysis The AFLP protocol was similar to that described by Vos et al. (1995). Each genomic DNA was diluted to 100 ng/µl with TE buffer. DNA digestion was carried out using EcoRI and MseI (BioLabs, Beverly, MA, USA) as restriction enzymes; 2 µl of genomic DNA were added to a reaction mixture containing 1X NEB buffer2, 1 µg/µl BSA, 10 U EcoRI, 10 U MseI and water to a final volume of 20 µl. After 3 h incubation at 37°C, 20 µl containing 100 pmol/µg MseI-adapter and 10 pmol/µg EcoRI-adapter, 4 U T4-DNA-Lygase (BioLabs) and 1X T4-buffer were added to the restriction mixture; the ligation was carried out for 16 h at 16°C. Two primers, Mc and Ea (Table 2), both with one selective base in the 3’ position, were used for the pre-amplification reaction. 5 µl of DNA template 10-fold diluted with TE buffer were amplified in a final volume of 20 µl of a reaction mixture containing 1X PCR buffer (10 mM Tris-HCl; 50 mM KCl; pH 8.3), 1.5 mM MgCl2, 200 µM dNTP, 50 ng each primer and 1 U Taq-polymerase (Promega). The PCR reaction was performed with the following programme: 1 min at 94°C; 35 cycles of 30 s at 94°C, 30 s at 55°C and 1 min at 72°C; 10 min at 72°C. The presence of the pre-amplified products was verified through electrophoresis on a 2% agarose gel in 1X TAE buffer. The selective amplifications were carried out using couples of primers both with a selective extension; the primer combinations indicated in 79

Chapter 4

_________________________________________________________________________ Table 2 were used. Table 2 - Adapter and primer sequences for the AFLP preamplification and selective amplification Restriction

Primer

Sequence

site Adapter

MseI

5’-GACGATGAGTCCTGAG-3’

Universal primer

3’-TACTCAGGACTCAT-5’

Preamplification

Mc

5’-GATGAGTCCTGAGTAAC-3’

1st selective amplification

Mca

5’-GATGAGTCCTGAGTAACA-3’

2nd selective amplification

Mct

5’-GATGAGTCCTGAGTAACT-3’ 5’-GATGAGTCCTGAGTAACA-3’

rd

selective amplification

Mca

4th selective amplification

Mct

5’-GATGAGTCCTGAGTAACT-3’


Mca

5’-GATGAGTCCTGAGTAACA-3’

Mcg

5’-GATGAGTCCTGAGTAACG-3’

3

6th selective amplification Adapter

EcoRI

5’-CTCGTAGACTGCGTACC-3’

Universal primer

3’-CATCTGACGCATGGTTAA-5’

Preamplification

Ea

5’-GACTGCGTACCAATTCA-3’

1st selective amplification

Eaa

5’-GACTGCGTACCAATTCAA-3’

2nd selective amplification

Eac

5’-GACTGCGTACCAATTCAC-3’

3rd selective amplification

Eag

5’-GACTGCGTACCAATTCAG-3’


Eag

5’-GACTGCGTACCAATTCAG-3’


Eat

5’-GACTGCGTACCAATTCAT-3’

Eat

5’-GACTGCGTACCAATTCAT-3’

6

th

selective amplification

Note: Abbreviations of the primers are given. Nucleotide extensions for preamplification and selective amplification are indicated in boldface.

PCR mixture was the same as the pre-amplification reaction, except for the primers (2 ng/µl each). The amplification program was: one cycle at 94°C for 1 min; 13 cycles of 30 s at 94°C, 30 s ramping from 65°C to 56.6°C (-0.7°C per cycle) and 1 min at 72°C; 23 cycles of 30 s at 94°C, 30 s at 56°C and 1 min at 72°C; one cycle at 72°C for 10 min. 4 µl of amplification product were added to 15 µl loading buffer (98% formamide; 10 mM EDTA pH 8.0; 0.01% bromophenol blue; 0.01% xylene cyanol), denaturated at 95°C for 5 min and kept at 0°C before loading in a denaturing (7M urea) 5 % polyacrylamide (19:1) gel. A 5 ng/µl marker mixture, formed by 10 bp DNA ladder (Sigma), 100 bp DNA 80

ladder

(Sigma)

and

formamide

blue

(5:5:40),

was

used.


Electrophoresis was carried out at 80 W for 2.5 h. Polyacrylamide gels were fixed for 30 min in a 10% acetic acid solution. Silver staining was carried out as described by Bassam et al. (1991). Gel profiles were scanned and visualised by a Gel-Documentation System (Quantity One Programme, Bio-Rad Labs). Every AFLP analysis was repeated twice for each couple of primers. 2.6. Data scoring and statistical analysis AFLP amplifications were repeated at least once in order to test their consistency. Polymorphisms were identified by the name of the primers and the size of the DNA amplified products. Each PCR product was assumed to represent a single locus and only reproducible polymorphic bands were scored as present (1) or absent (0). All fragments were given equal weights. A binary matrix of isolates and markers for cluster analysis was compiled using the NTSYS-pc (Numerical Taxonomy and Multivariate Analysis System) version 1.80 package (Rohlf, 1993). Genetic similarity among accessions was calculated according to Dice’s Similarity Index (DSI; Dice, 1945) in all possible pair-wise comparisons, using the SIMQUAL (Similarity of Qualitative Data) routine. DSIs are defined as: DSIxy=2a/(2a+b+c), where a = number of bands shared from individuals x and y, b = number of bands present in x and absent in y, c = number of bands present in y and absent in x; thus, DSIxy=1 indicates identity between x and y, whereas DSIxy=0 indicates complete diversity.

The

similarity

coefficients

were

used

to

construct

a

dendrogram using the UPGMA (unweighted pair-group method with arithmetic

average)

through

the

SHAN

(sequential,

hieratical,

agglomerative and nested clustering) routine and a thousand bootstrap were performed over AFLP loci using PHYLIP software (Felsenstein, 1993;

http://evolution.genetics.washington.edu/phylip.html).

A

co-

phenetic matrix was produced using the hierarchal cluster system, by means of the COPH routine, and correlated with the original distance matrixes for AFLP data, in order to test for association between the cluster in the dendrogram and the DSI matrix. Mantel tests (Mantel, 1967) were performed to check the correlation 81

Chapter 4

_________________________________________________________________________ between the similarity matrixes generated by the single primer combinations and the total similarity matrix. 3. Results 3.1. Classification through PCR-RFLP ITS1 and ITS4 primers were used to amplify the region of the rDNA (ribosomal DNA) repeat unit that includes the 5.8S rRNA gene and the two non-coding regions designated the internal transcribed spacers – ITS1 and ITS2 (White et al., 1990) – of the seven strains isolated in Northern Italy from the carposphere of apple. The sizes of the PCR products and the restriction fragments obtained using the restriction endonucleases CfoI, HaeIII and HinfI were compared with results obtained by Esteve-Zarzoso et al. (1999). Identification was compared with classification carried out by the DBVPG with classical method (Figure 1). Fig. 1. - Microphotography of the seven Metschnikowia pulcherrima strains isolated from the carposphere of apples and selected for their efficacy against Botrytis cinerea and Penicillium expansum.

82


3.2. Biocontrol efficacy trials Twenty-six Metschnikowia pulcherrima strains of different origin have been evaluated for their biocontrol efficacy in artificial wound of apple against Penicillium expansum and Botrytis cinerea (Table 3). After 28 days of storage at 4°C the mean reduction of the lesion diameter of blue mould was 49.3% of the control. Five strains (BIO114, GS9, 3008, 3435 and 4129) did not reduce significantly the growth of P. expansum. The other twenty-one strains reduced the pathogen growth ranging from 15.7 to 68.9% of the control. The most effective strains were BIO131, 3043, 4292 and 4185, one coming from apple surface, two from grape must and one from a winery surface. Against B. cinerea, after storage for 28 days at 4°C, generally all strains were more effective, with a mean reduction of the pathogen growth to 30.0% of the control. Two strains, BIO114 and 4354, did not significantly controlled grey mould on apple. The other twenty-four strains reduced the lesion diameter of the mould ranging from 0.0% to 69.5% of the control. GS37 and 3043 provided a complete control of the disease. Fig. 2 - RAPD patterns, using primer OPB17 of the twenty-six M. pulcherrima strains. 1.5% agarose gel stained with ethidium bromide. The first lane on the left and last one on the right contain a DNA marker (1kb DNA ladder, GibcoBRL). W is water control.

83

Chapter 4

_________________________________________________________________________ Table 3 - Effect of the cell suspensions of the twenty-six M. pulcherrima strains on Penicillium expansum and Botrytis cinerea growth on apples cv Golden delicious. Storage at 4°C for 28 days

Blue mould

a

Treatment

severity (%)

Control

*100.0 g

Grey mould a

severity (%)a **100.0

h

Thiabendazole

66.8 e-f

74.0

f-h

BIO114

84.3 f-g

81.9

g-h

BIO126

32.5 b-d

8.4

a-c

BIO131

15.7 a-b

23.0

a-e

GS37

32.7 b-d

0.0

GS88

33.6 b-d

36.5

a c-e

GS9

81.8 f-g

36.0

c-e

4.4

31.8 b-d

24.3

a-e

291

53.9 d-e

33.5

b-e

311

36.6 b-d

18.6

a-d

320

68.2 e-f

39.0

d-e

3008

79.0 f-g

17.3

a-d

3041

68.9 e-f

7.5

a-c

3042

68.0 e-f

21.7

a-e

3043

16.6 a-b

0.0

3345

35.9 b-d

23.0

a-e

3348

36.9 b-d

69.5

f-g

a

3435

87.3 f-g

38.0

c-e

3527

34.3 b-d

51.4

e-f

3835

32.5 b-d

3.8

a-b

3938

66.6 e-f

30.7

a-e

4064

47.2 c-e

29.2

a-e

4129

88.9 f-g

38.8

d-e

4130

45.2 c-d

10.7

a-d

4185

26.5 b-c

25.0

a-e

4292

25.6 b-c

37.5

c-e

4354

50.0 d-e

74.4

f-h

Calculated on the lesion diameter. Values in the same column followed by the same letter

are not statistically different by Tukey b Test (P < 0,05). b

300 µg a.i. ml-1: used as chemical control.

Real attack in the control: *28.9 mm;**35.4 mm.

84


Among the most effective strains, also 3835 (3.8% of the control), 3041 (7.5%) and BIO126 (8.4%) showed a good capability of controlling the disease. Eight strains were able to limit the pathogen growth to values inferior to 20% of the control: three were isolated from grape must, two from apple, one from quince, one from cherry and one from a snail. 3.3. Molecular characterization Genetic variability among the M. pulcherrima strains was assessed by RAPD and AFLP. The molecular weight of amplicons obtained with RAPD technique ranged from 200 to 3000 bp, and six primer combinations generated 33 polymorphic bands (26% of the total amplified bands). A representative result, obtained with primer OPB17, is given in figure 2. The 33 polymorphic bands obtained were not analysed to generate a dendrogram of similarities. The AFLP technique was preferred to continue the studies on the genetic variability among the twenty-six yeast strains. Preamplification reaction was carried out successfully using primer pair McEa. Six primer combinations of an initial 16 tested combinations were chosen for their ability to generate informative patterns rich of polymorphic bands. A representative result, obtained with primer pair MctEac, is given in figure 3. After scoring the AFLP profiles obtained with the six primer combinations selected, a total of 729 polymorphic bands (39% of the total amplified bands) were scored. The size of AFLP fragments was in a range from 40 bp to 1500 bp, but only fragments between 100 and 500 bp were taken into account to avoid scoring problems due to excess primer peaks near the front of the electrophoresed fragments and a decreasing signal for fragments longer than 500 bp. The dendrogram, generated from the Dice distance matrix, using UPGMA clustering analysis, is shown in Figure 4. The matrix of pair-wise distances

calculated

according

to

Dice

similarity

index

and

the

cophenetic matrix are not shown, but the co-phenetic correlation coefficient (r-value) between the data matrix and the co-phenetic matrix for AFLP data was 0.978, suggesting a very good fit between the 85

Chapter 4

_________________________________________________________________________ dendrogram clusters and the similarity matrixes from which they were derived. Reproducibility of the groupings below each node of the dendrogram were verified by analysing 1000 multiple datsets from bootstrapping. Two major clusters could be distinguished that are supported by high bootstrap value (97): Cluster 1, including the seven strains isolated from the carposphere of apples, coming from orchard of Piedmont, in Northern Italy, and Cluster 2, including all the other strains, except for 311 and 291, that did not cluster to a specific group. Strains belonging to Cluster 1 showed a well-structured grouping. This was characterised by high bootstrap values. The branch lengths in the tree reflect the distribution of genetic variation among the strains, calculated using the DSI. Genetic similarities among the strains of M. pulcherrima ranged from 0.21 (between 291 and 311 and the other strains studied) and 0.94 (between 3435 and 3527). Within Cluster 1, the strains BIO114 and BIO126 were the most similar strains (DSI of 0.91); quite high bootstrap values were obtained (96). BIO131 was the yeast more similar to them with a DSI of 0.80. GS37 was more similar to GS88 (DSI of 0.58) and GS9 to 4.4 (DSI of 0.59). The strains appeared to be less structured in Cluster 2. In fact lower bootstrap values were obtained here indicating a lower reliability of structure. Within Cluster 2, two sub-clusters could be identified: Cluster 2a, comprising strains 3041, 3042, 3043, 3435, 3527 and 3345, which separates from Cluster 2b, including 3835, 4130, 4185, 3938, 4064 and 4292, with quite high bootstrap values (72). Within Cluster 2a, the strains 3435 and 3527 were the most similar strains (DSI of 0.94) and were found in 100% of the bootstrap resamples. Also 3041, 3042 and 3043 were quite similar (DSI of 0.86) and were found in all the resamples. In the Cluster 2b, strains 4130 and 4185 are more similar (DSI of 0.91) and were found in 99% of the resamples. In general, the bootstrap values that unite the strains are among the most highly supported: all bootstrap values are greater than 50 except in the case of cluster between 3348 and 4129, 3008, Cluster 2a and Cluster 2b (bootstrap value of 45). 86


In this study the choice of one of the six primer combinations did not have a large influence on the amount of fragments recovered (Table 4) but did have a strong effect on the correlation between the similarity matrixes generated by the single primer combination and the total similarity matrix, ranging from 0,770 to 0.937. To check this correlation, Mantel tests were performed. In general, anyway, the similarity matrixes generated by every single primer pair showed few differences from the total matrix. Primer combinations McaEaa and McgEat resulted highly informative, because both generated a similarity matrix closely related with the total similarity matrix (r of 0.937 and 0.931) and were able to provide a unique fingerprint for every genotype. All the primer combinations permitted to obtain twenty-six unique electrophoretic patterns, except for primer pair McaEat, that did not distinguish between strain 3041 and 3042, and primer pair MctEac, that was not able to separate strain BIO 126 from BIO 131. Table 4 - Correlation between the similarity matrixes generated by the single primer combinations and the total similarity matrix and number of distinguishable genotypes

Primer combinations

Number of polymorphic

% of the total

r value

bands

Distinguishable genotypes

McaEaa

135

18.5

0.937

26

MctEac

124

17.0

0.770

25

McaEag

119

16.3

0.889

26 26

MctEag

130

17.8

0.885

McaEat

109

15.0

0.883

25

McgEat

112

15.4

0.931

26

4. Discussion 4.1. Classification through PCR-RFLP The seven strains isolated by our Department were identified through a molecular method, already applied by Esteve-Zarzoso et al. (1999) for the identification of wine yeast, also showing an interesting application 87

Chapter 4

_________________________________________________________________________ for fruit surface yeast identification. M. pulcherrima (Pitt) M. W. Miller is an ascomycetous teleomorphic yeast and it is the teleomorph of the anamorph Candida pulcherrima. Yeast identification is normally based on morphological, physiological and biochemical traits, which in some cases can lead to an incorrect classification of species or a false identification of strains. We tested the application of the restriction analysis of the rRNA region spanning the 5.8 rRNA gene and two ITSs (internal transcribed spacers) as a rapid and easy method for yeast species identification. ITS1 and ITS4 primers were used to amplify the region of the rDNA repeat unit that includes the 5.8SrRNA gene and two non-coding regions designated ITS1 (internal transcribed spacers) and ITS2 (White et al., 1990). Comparing the sizes of the PCR products and the restriction fragments obtained using the restriction endonucleases CfoI, HaeIII and HinfI, we obtained the same results shown by EsteveZarzoso et al. (1999) and by Guillamon et al. (1998). The same technique was used also for the classification of other antagonistic yeast species

(Debaryomyces

hansenii,

Kloeckera

lindneri,

Pichia

guilliermondii, Hanseniaspora uvarum and Rhodotorula glutinis) with antagonistic properties for the control of postharvest diseases, not used in this study (data not shown). In all cases classification obtained with this molecular method was confirmed with identification carried out by the

DBVPG,

University

of

Perugia,

Italy,

using

morphological,

physiological and biochemical methods. This method proved to be reproducible and it can be a very useful technique to easily and rapidly identify and classify the majority of the yeast species for biocontrol of postharvest diseases. 4.2. Biocontrol efficacy trials The twenty-six strains studied for their biocontrol activities were coming from different sources: seven strains were isolated from the carposphere of apple, two from the carposphere of other pome fruit (pear and quince), two from the carposphere of stone fruit (cherry), eleven from different steps in the wine production chain (grape, must, wine and winery), two from unusual origins (a snail and seawater) and for two of 88


them the origin is unknown. The strains were tested for their efficacy in the control of Botrytis cinerea and Penicillium expansum, causal agents of grey and blue mould on apple. Some strains already proved to be more effective in the control of these diseases at the low temperatures of storage of the fruits than at room temperature, probably because at 4°C the growth rate of the biocontrol agents is reduced less than the growth rate of the pathogens. The main mode of action involved in the biocontrol is competition for nutrients or space although a direct interaction can not be excluded (Spadaro et al., 2002). In the biocontrol activity experiments carried out, thiabendazole was used as chemical control but the strains of Penicillium expansum and Botrytis cinerea used were resistant to benzimidazoles, as can be observed from the low efficacy of the fungicide. In effect, this low sensitivity is also confirmed by some recent evaluations carried out in Italy on postharvest pathogens of pome fruit (Bertetti et al., 2003). In general the strains were more effective in the control of B. cinerea than of P.expansum, after storage for 28 days at 4°C, with a mean reduction of the pathogen growth respectively to 30.0% and 49.3% of the control. Five strains did not reduce significantly the growth of P. expansum and only two the growth of B. cinerea. Only six strains controlled better blue than grey mould. On an average, M. pulcherrima is a yeast species that possesses good antagonistic characteristics for biological control of postharvest diseases of apple, and it is meanly more effective against B. cinerea than P.expansum. The seven strains isolated from apple carposphere are the result of a selection for biocontrol capabilities against B. cinerea and P. expansum on apple among about 400 strains. The other strains were randomly chosen in the yeast collection of DBVPG, University of Perugia, Italy, or (strains 291, 311 and 320) are the result of a selection based on their biocontrol potential but on different host species. The first group of microorganisms, on an average, controlled better B. cinerea (23.1% compared to the control) and P. expansum (40.3%) than the second group (29.8% and 50.8%). The strains previously selected in our laboratory were among the more effective but not all of them. GS37 and 89

Chapter 4

_________________________________________________________________________ 3043 provided a complete control of Botrytis rot. BIO131 and 3043 were the most effective against Penicillium rot. 3043 offered the best control of both diseases and was randomly chosen among the isolates of DBVPG from grape must. In this study we tried to clarify one question of biological control of postharvest diseases. Is the source of the antagonists so important in the determination of the biocontrol capability of the microorganisms? Normally it is believed that the fruit surface is an excellent source of naturally occurring microorganisms against postharvest rot agents (Wilson and Wisniewski, 1994; Droby et al., 1999; Janisiewicz and Korsten, 2002). The carposphere or the phylloplane have provided the major source for antagonists and in a few cases microorganisms have been isolated from other matrixes: one yeast collection has been screened (Filonow et al., 1996) and, in one example, starter cultures used in the food industry were used as possible sources of biocontrol agents (Wilson and Chalutz, 1991). Anyway, seldom there has been a comparison between the efficacy of microorganisms coming from the carposphere and the biocontrol capability of other microorganisms coming from other sources (Filonow et al., 1996). In this paper we showed that antagonistic properties for biological control in the carposphere can be possessed by microorganims isolated from the same source where they will be applied as antagonists but these characteristics can be owned also by microorganisms of different origin. 4.3. Molecular characterization In the second part of the work, we wanted to assess the genetic diversity of M. pulcherrima strains of different origins and to discover if a relationship between biocontrol activity and genetic distance existed. To obtain specific fingerprints of the M. pulcherrima strains, the RAPD and AFLP techniques were used. Specific fingerprints are useful to integrate the morphology-based monitoring of the genetic and environmental fate of these microbial antagonists. 90


Fig. 3 - AFLP patterns of the twenty-six M. pulcherrima strains, using primer pair MctEac. 5 % polyacrylamide denaturing (7M urea) gel silver stained.

Using molecular markers, it is possible to study the persistence and survival of the biocontrol agents, the effects of the introduced microorganisms on microbial communities (composed by endogenous

91

Chapter 4

_________________________________________________________________________ bacteria, yeasts and fungi) and their genetic stability (Gullino et al., 1995). It is also possible to monitor the biocontrol agent population on the skin of treated fruit during storage (Lima et al., 2003). Moreover, specific molecular markers are necessary to monitor the antagonists released in orchard when a preharvest treatment is realized (Schena et al., 2001). Field application of the biocontrol agents (BCAs) may enable early colonisation of the fruit surfaces, protecting them from these infections (Ippolito and Nigro, 2000). Finally, the availability of molecular fingerprints is an important pre-requisite requested by the companies interested in the development of commercial products based on the biocontrol agents, since it prevents their unauthorized use. Genetic variability was assessed by RAPD and AFLP techniques. RAPD technique is faster, easier and less expensive. Both methods were effective, but AFLP was more informative and permitted to obtain a superior

number

of

polymorphisms.

With

six

RAPD

primers

33

polymorphic bands were obtained, while with six AFLP primer pairs 729 polymorphic bands were scored. AFLP technique was about twenty times more informative than RAPD technique. For this reason, AFLP was preferred for the genetic analysis and to develop an STS marker for a specific antagonistic strain. AFLP technique resulted more reproducible and resolute than RAPD technique, but its analysis is more laborious and expensive (Jones et al., 1997; Blears et al., 1998). AFLP technique has been widely used to study plant genomes but rarely for yeast studies. Some applications in literature are for human pathogens (Forche et al., 2000), wine spoilage yeasts (Barros-Lopes et al., 1999) and recently for postharvest biological control (Lima et al., 2003). The genetic distances among M. pulcherrima strains were visualized on a dendrogram, created from the similarity matrix generated using Dice’s similarity coefficient. The similarity matrix was converted to genetic diversity estimates, in turn used for UPGMA cluster analysis. Co-phenetic correlation

values

showed

that

the

genetic

clusters

accurately

represented the estimates of genetic similarity. Reproducibility of the groupings below each node of the dendrogram were verified by 92


analysing 1000 multiple datsets from bootstrapping. The high bootstrap values at each node indicate that the this tree is robust. Isolates coming from the carposphere of apple were grouped in a well structured cluster (Cluster 1) with a high bootstrap value (97). Within Cluster 1, the genetic distance was supported by the different origin of the strains. BIO114, BIO126 and BIO131 came from organic orchards of Piedmont (Northern Italy); GS9, GS37, GS88 and 4.4 were isolated in Piedmont orchards grown following integrated control strategies. The other two strains isolated from the carposphere of pome fruit in Southern Italy, 311 from quince and 291 from pear clustered together (100) but were quite different from the other strains. Fig. 4 - Dendrogram describing the relationships among the twenty-six isolates of M. pulcherrima, based on the Dice similarity index of AFLP-PCR banding profiles. The analysis of grouping was undertaken by unweighted pair-group method using arithmetic averages (UPGMA). Numbers at the nodes represent the proportion of 1,000 bootstrap samples in which a particular clade was found.

The most effective strains in the control of P. expansum were BIO131, 3043, 4292 and 4185, the first one grouped in Cluster 1, the second in Cluster 2a and the last two in Cluster 2b. Against B. cinerea, after 93

Chapter 4

_________________________________________________________________________ storage for 28 days at 4°C, the strains that provided better control of B. cinerea were BIO126 and GS37 (Cluster 1), 3041 and 3043 (Cluster 2a) and 3835 (Cluster 2b). We showed that there was not a relationship among biocontrol capability and origin of the microorganisms, but also that biocontrol efficacy and genetic distance among the strains were not related. Strains of the same species isolated from the same location (BIO114, BIO126 and BIO131) can be very similar from the genetic point of view but greatly differ for their biocontrol potential. On the contrary, strains isolated in different locations (GS37 and 3043), with a high genetic diversity, can have a similar biocontrol potential. The biocontrol potential is the result of a large quantity of genetic traits that contribute to provide the antagonist a fitness advantage towards the pathogen. This is especially true for microbial antagonists using competition as main mechanism of biocontrol. When other mechanisms are involved, such as antibiosis or mycoparasitism, it is easier to identify key genetic traits, such as genes involved in the antibiotic biosynthetic pathway or genes for lytic enzymes. It is anyway important to point out that it is not known yet to which degree AFLP amplification patterns are representative for the overall genetic similarity among the strains. AFLP fragments can be amplified from single-copy or repetitive-DNA fractions. There are indications that AFLP bands preferably are amplified from repetitive

DNA and therefore are a biased sample of heritable

polymorphism (Potokina et al., 2002). Many of the AFLP fragments only occurred in one or a few strains of M. pulcherrima. This pointed towards a high mutation rate and a low information content of a large fraction of the AFLP bands. Such rare polymorphisms would add random noise when AFLP patterns were analysed for overall similarity, and there is the possibility that this highly variable fraction changes quickly with time. The six primer combinations permitted to obtain twenty-six unique patterns, except for primer pair McaEat and MctEac, that were able to provide twenty-five patterns. Assessing the correlation between the similarity matrixes generated by the single primer combination and the total similarity matrix we found that one primer pair, such as McaEaa or McgEat, resulted highly informative and sufficient to describe the genetic 94


distance among the strains. AFLP patterns, although moderately difficult to develop and use, result highly informative with a few primer combinations, while many RAPD patterns need to be analysed to obtain the same quantity of information (Pasquali et al., 2003). 5. Conclusions AFLP patterns could clearly distinguish the different strains of M. pulcherrima and, within the limits of the restricted samples, we have found some putative specific bands for single tag sequence (STS) conversion. AFLP fingerprinting confirmed in other studies to be a useful method for the identification of specific DNA sequences suitable as a source of template for the production of STS markers (Shan et al., 1999). Research is in progress to develop STS isolate-specific markers to further improve identification and monitoring in the environment of our M. pulcherrima biocontrol agents. One of the major characteristics for an antagonist to be used in biological control is its precise identification and its traceability, to permit to follow its environmental fate in space (dispersion) and in time (survival), after release (Gullino et al., 1995). Molecular tools can assist to monitor the genetic and environmental fate of these agents after releasing. Suitable and reproducible strain authentication methods are necessary in commercial procedures such as filling patents and product licensing. Moreover it is essential to develop a quality control system that allows to monitor the genetic stability over time of the biofungicide as a commercial product (Avis et al., 2001). Specific STS primers might be very useful to fast track the presence of the strains released and, thanks to a future application of the real time PCR, to quantify the residual population of the microorganism on the fruit surface after application will become possible. These data might be applied for disease control management and to establish the threshold level effective in the control of the pathogens. A better knowledge of the ecology of the antagonists applied in postharvest

biological

control

may

lead

to

an

optimisation

of

formulations and mode of applications, with benefits for the level of 95

Chapter 4

_________________________________________________________________________ protection of the fruit achieved. Acknowledgements The authors gratefully acknowledge Dr. Franco Nigro (University of Bari, Italy) and Dr. Ann Vaughan-Martini (University of Perugia, Italy) for having provided some of the strains studied in this work, Dr. Jeanne Griffin for the linguistic advice and Dr. Ezio Portis (University of Torino, Italy) for his help in the AFLP technique and statistical analysis. This research was carried out with a grant from the Italian Ministry for the Environment and Territory within the Framework Agreement “Crop Protection with Respect of the Environment”. References Avis, T.J., Hamelin, R.C., Bélanger, R.R.,

Blears, M.J., De Grandis, S.A., Lee, H.,

2001.

Approaches

molecular

Trevors, J.T., 1998. Amplified fragment

biocontrol

length polymorphism (AFLP): review of

agents: some case studies. Can. J. Plant

the procedure and its applications. J.

Pathol. 23, 8-12.

Ind. Microbiol. Biot. 21, 99-114.

characterization

to

of

fungal

Barros-Lopes, M., Rainieri, S., Henschke,

Boekhout, T., Robert, V., 2002. Yeasts in

P.A.,

AFLP

Food

yeast

Aspects. Behr’s Verlag, Hamburg.

Landridge,

fingerprinting

P.,

for

1999.

analysis

of

–

Beneficial

and

Detrimental

genetic variation. Int. J. Syst. Bacteriol.


49, 915-924.


Bassam,

B.J.,

Caetano-Anolles,

G.,

antagonistic

microorganisms.

Ph.D.

Gresshoff, P.M., 1991. Fast and sensitive

dissertation, University of Stellenbosch,

silver staining of DNA in polyacrilamide

Capetown.

gels. Anal. Biochem. 196, 80-83.

Dice,

Beech,

F.W.,

1993.

Yeasts

in

Cider

L.R.,

amount

1945.

of

Measures

ecological

of

the

association

Making. In: Rose, A.H., Harrison, J.S.

between species. Ecology 26, 297-302.

(Eds.),

“Yeast

Droby, S., Lischinski, S., Cohen, L.,

Technology”. Academic Press, London,

Wiess, B., Daus, A., Chand-Goyal, T.,

pp. 169-213.

Eckert,

Berretti, D., Kejji, S., Garibaldi, A.,

Characterization of an epiphytic yeast

Gullino, M.L., 2003. Valutazione della

population

sensibilità di alcuni agenti di marciume

suppression of green mold decay caused

su

by Penicillium digitatum. Biol. Control

The

pomacee

Yeasts,

in

Vol.

5,

post-raccolta

nei

J.W., of

Manulis, grapefruit

S.,

1999.

capable

of

confronti di diversi fungicidi. Inf.tore

16, 27-34.

Fitopatol. 53 (6), 57-60.

Esteve-Zarzoso, B., Belloch, C., Uruburu,

96


F., Querol, A., 1999. Identification of

Gullino, M.L., Bonino, M., Piano, S.,

yeasts by RFLP analysis of the 5.8S rRNA

Testoni, A., Salimei, A., Duverney, C.,

and

internal

1994. Biological control of postharvest

Syst.

rots of apples. Proceedings of the ANPP

the

two

transcribed

ribosomal

spacers.

Int.

J.

– 4th International Conference on Plant

Bacteriol. 49, 329-337. Felsenstein, Inference

J.,

1993.

Package

Phylogeny

(PHYLIP).

Version

Diseases.

Bordeaux,

France,

6-8

December 1994, 333-340.

3.5. University of Washington, Seattle.

Gullino, M.L., Migheli, Q., Mezzalama,

Filonow, A.B., Vishniac, H.S., Anderson,

M., 1995. Risk analysis in the release of

J.A., Janisiewicz, W.J., 1996. Biological

biological control agents: antagonistic

control of Botrytis cinerea in apple by

Fusarium oxysporum as a case study.

yeasts from various habitats and their

Plant Dis. 79, 1193-1201.

putative

Hofstein,

mechanisms

of

antagonism.

R.,

Friedlender,

B.,

1994.

Biol. Control 7, 212-220.


Fleet, G.H., Heard, G.M., 1993. Yeasts:

and delivery systems. Proceedings of the

growth during fermentation. In: Fleet,

Brighton

G.H.

Brighton, Great Britain, 21-24 November

(Ed.),

Wine

microbiology

and

Crop

Protection

Conference.

biotechnology. Harwood, Chur, pp. 27-

1994, vol. 3, 1273-1280.

54.

Ippolito, A., Nigro, F., 2000. Impact of

Forche, A., Xu, J., Vilgalys, R., Mitchell,

preharvest

T.,

and


characterization of a genetic linkage map


of

19, 610-619.

2000.

Development

Cryptococcus

neoformans length

using

neoformans amplified

polymorphisms

var.

fragment

and

Janisiewicz,

application

W.J.,

of

biological

1991.

Biological

other

control of postharvest fruit disease. In:

markers. Fungal Genet. Biol. 31, 189-

Arora, D.K., Rai, B., Knudsen, G.R.

203.

(Eds.), Handbook of Applied Micology.

Guillamon, J.M., Sabaté, J., Barrio, E.,

Vol. 1: Soil and Plants. Marcel Dekker,

Cano,

New York, pp. 301-326.

J.,

Querol,

A.,

1998.

Rapid

identification of wine yeast species based

Janisiewicz,

on

ribosomal

Biological control of postharvest diseases

internal transcribed spacer (ITS) region.

of fruits. Annu. Rev. Phytopathol. 40,

Arch. Microbiol. 169, 387-392.

411-441.


Janisiewicz,

Social

Kurtzman,

RFLP

analysis

and

of

political

the

implications

of

W.J.,

Korsten, L.,

W.J., C.P.,

2002.

Tworkoski, 2001.

T.J.,

Biocontrol


potential of Metschnikowia pulcherrima

fungicides

strains

in

Europe.

Annu.

Rev.

against

blue

mold

of

apple.



Gullino, M.L., Aloi, C., Palitto, M., Benzi,

Jones, C.J., Edwards, K.J., Castaglione,

D., Garibaldi, A., 1991. Attempts at

S., Winfield, M.O., Sala, F., van de Wiel,

biological control of postharvest diseases

C.,

of apple. Meded. Fac. Landbouww. Univ.

Matthes, M., Daly, A., Brettschneider, R.,

Gent 56, 195-202.

Bettini,

Bredemeijer, P.,

Buiatti,

G., M.,

Vosman, Maestri,

B., E.,

97

Chapter 4

_________________________________________________________________________ Malcevschi, A., Marmiroli, N., Aert, R.,

Cohen, L., Nigro, F., Droby, S., 1999.

Volckaert, G., Rueda, J., Linacero, R.,

Genetic diversity and biocontrol activity

Vasquez,

of

A.,

Karp,

A.,

1997.

Reproducibility testing of RAPD, AFLP

Aureobasidium

against

pullulans

postharvest

rots.

isolates

Postharvest

and SSR markers in plants by a network

Biol. Technol. 17, 189-199.

of European laboratories. Mol. Breeding

Schena, L., Ippolito, A., Zahavi, T.,

3, 381-390.

Cohen, L., Droby, S., 2000. Molecular

Lima, G., De Curtis, F., Castoria, R., De

approaches to assist the screening and

Cicco, V., 2003. Integrated control of

monitoring

apple

yeasts. Eur. J. Plant Pathol. 106, 681-

postharvest

pathogens

and

of

postharvest

biological

survival of biocontrol yeasts in semi-

691.

commercial

Schena, L., Finetti Sialer, M.M., Gallitelli,

conditions.

Eur.

J.

Plant

Pathol. 109, 341-349.

D., 2001. Molecular detection of strain

Maniatis, T.E., Fritsch, E.F., Sambrook,

L47

J.,

biocontrol agent of postharvest disease.

1982.

Molecular

Cloning:

a

of

Aureobasidium

pullulans,

a

Laboratory Manual. Cold Spring Harbor

Plant Dis. 86, 54-60.

Laboratory, Cold Spring Harbor.

Shan, X., Blake, T.K., Talbert, L.E.,

Mantel,

of

1999. Conversion of AFLP markers to

disease clustering and a generalized

N.,

1967.

The

sequence-specific PCR markers in barley

regression approach. Cancer Res. 27,

and wheat. Theor. Appl. Genet., 98,

209-220.

1072-1078.

Pasquali, M., Acquadro, A., Balmas, V.,

Snowdon, A.L., 1990. A colour atlas of

Migheli, Q., Garibaldi, A., Gullino, M.L.,

post-harvest diseases and disorders of

2003. RAPD characterization of Fusarium

fruits and vegetables. Volume 1. Wolfe

oxysporum

Scientific, London.

isolates

Argyranthemum

detection

pathogenic

frutescens

L.

on J.

Phytopathol. 151, 1-6. Piano,

S.,

Neyrotti,

Spadaro, D., Gullino, M.L., 2004. State of

V.,

Migheli,

Q.,

art

and

diseases.

the

press.


capability

pulcherrima

postharvest

rot

of

of

perspectives

of

biological control of postharvest fruit

Gullino, M.L., 1997. Characterization of biocontrol

future

Int.

J.

Food

Microbiol.,

in

against

Spadaro, D., Garibaldi, A., Gullino, M.L.,

apple.

2004. Control of Botrytis cinerea and


Penicillium

Potokina, E., Blattner, F.R., Alexandrova,

combining a biocontrol agent with hot

T., Bachmann, K., 2002. AFLP diversity

water dipping and acibenzolar-S-methyl,

in the ommon vetch (Vicia sativa L.) on

baking

the world scale. Theor. Appl. Genet. 105,

Postharvest Biol. Technol., accepted.

58-67.

Spadaro, D., Vola, R., Piano, S., Gullino,

soda,

expansum

or

on

ethanol

apple

application.

Rohlf, F.J., 1993. NTSYS-pc numerical

M.L.,

taxonomy

analysis

efficacy and possibility of integration

Publ.,

with chemicals of four isolates of the

system.

and Version

multivariate 1.8.

Exeter

2002.

Mechanisms

of

action,

Setauket.

yeast Metschnikowia pulcherrima active

Schena, L., Ippolito, A., Zahavi, T.,

against

98

postharvest

pathogens

on



R.A.,

Cervantes,

L.A.,

1986.

Populations, pathogenicity and benomyl resistance of Botrytis spp., Penicillium spp.

and

Mucor

piriformis

in

packinghouses. Plant Dis. 70, 106-108. Verhoog, H., Matze, M., Lammerts van Bueren, E., Baars, T., 2003. The role of the concept of the natural (naturalness) in organic farming. J. Agr. Environ. Ethics 16, 29-49. Vos, P., Hogers, R., Bleeker, M., Reijans, M., van de Lee, T., Hornes, M., Frijters, A., Pot, J., Peleman, J., Kuiper, M., Zabeau,

M.,

1995.

AFLP:

a

new

technique for DNA fingerprinting. Nucleic Acid Res. 23, 4407-4414. White, T.J., Bruns, T., Lee, S., Taylor, J., 1990. PCR protocols. A guide to methods and

applications.

Amplification

and

Direct Sequencing of Fungal Ribosomal RNA Genes for Phylogenetics, 315-322. Wilson,

C.L.,

Biological Diseases

Chalutz,

Control of

Workshop

Fruits

E.,

1991.

of

Postharvest

and

Vegetables.

Proceedings.

US

GPO,

Washington. Wilson, C.L., Wisniewski, M.E., 1994. Biological

Control

of

Postharvest

Diseases – Theory and Practice. CRC Press, Boca Raton.

99

CHAPTER 5

General discussion

General Discussion ________________________________________________________________

1. Introduction Apple postharvest rots, caused by Penicillium expansum, Botrytis cinerea, Monilia and Alternaria sp., are particularly severe even in production areas where the most advanced storage technologies are available (Snowdon et al., 1990). Over the past 15 years, biological control using microbial antagonists (Wilson and Wisniewski, 1994; Spadaro and Gullino, 2004) has emerged as one of the most promising alternatives, either alone or as part of an integrated pest management, to reduce pesticide use. Yeasts deserve particular attention because of their ability to colonize the fruit carposphere for long periods under unfavourable conditions, rapid growth, poor sensitivity to fungicides and lack of production of toxic metabolites (Janisiewicz, 1991). Cryptococcus laurentii, Pichia guilliermondii, Kloeckera apiculata, Sporobolomyces roseus and Candida oleophila have shown biocontrol effectiveness against grey and blue mould on apple (Roberts, 1990; McLaughlin et al., 1990; Mercier and Wilson, 1994; Janisiewick et al., 1994). At present, yeast-based biofungicides, active against Botrytis cinerea and Penicillium spp. on apple, pear and citrus fruit, are commercialised in the USA, Israel (a product containing Candida oleophila; Hofstein and Fridlender, 1994) and South Africa (a product based on Cryptococcus albidus; De Koch, 1998). 2. Efficacy Several strains of the yeast Metschnikowia pulcherrima effective against B. cinerea, Monilia sp. and P. expansum were isolated, selected and studied in our Department for their efficacy (Gullino et al., 1991; 1994; Piano et al., 1997; Spadaro et al., 2002; 2004a; 2004b). Dipping boxes of apples cv Golden delicious in a suspension of 107 antagonist cells ml-1 and storing for 8 months in controlled atmosphere at 1°C, the isolates showed control capability against B. cinerea and P. expansum similar to thiabendazole. The application of a cell suspension (108 cells ml-1) of the antagonists in artificial wounds of apples permitted to reduce the growth of B. cinerea 103

Chapter 5 ________________________________________________________________

and P. expansum after storage at 23°C. A complete suppression of the pathogen was obtained against Monilia sp., storing at 23°C, and against B. cinerea and P. expansum, storing at 4°C. The results against Alternaria sp. were more variable (Spadaro et al., 2002). The biocontrol agents are very effective against B. cinerea but show less consistent results towards P. expansum. Yeast cells survive on the apple carposphere for long periods of time, competing for nutrients and space with pathogens, agents of new infections, but the antagonist is not so effective towards latent infections or previously established pathogens. 3. Broadening of the spectrum of activity Potential biocontrol agents often have some significant limitations: they have a narrow range of activity, because they act on specific hosts against

well-defined

pathogens

under

particular

environmental

conditions. A method to select antagonists with a broader spectrum of activity, preferably for commercial development, includes efficacy tests for various pathogens and fruit species (Wilson et al., 1993; Lima et al., 1999). Most of the selected yeasts have frequently been tested for antagonistic activity on a few host fruits and against a limited number of postharvest pathogens, while it is essential for the commercial development of a biocontrol agent (BCA) to have a wide range of activity. Some trials of biocontrol of B. cinerea on artificial wounds of kiwifruit and of Penicillium sp. on artificial wounds of orange were carried out using strains BIO126, GA102, GS37 and GS88 (Spadaro, 2000; Vola 2000). A good efficacy towards B. cinerea on kiwifruit was expected from the antagonists effective on apple against the same pathogen, but the results obtained after 14 days of storage did not show a significant control of the pathogen. Artificial wounds were created at the stem-end because this region is preferably attacked by the pathogen. Probably the different nutritional environment and lower pH of the kiwifruit with respect to the apple favour the colonization of the wound by the pathogen. Because some of the biofungicides against postharvest disease products 104


on

the

market

at

the

moment,

Aspire

e

Bio-Save,

containing

respectively Pseudomonas syringae and Candida oleophila, are registred on pome fruit and citrus fruit, the efficacy of BIO126, GA102, GS37 and GS88 was evaluated also against Penicillium digitatum and P. italicum on orange. Control of blue and green mould was significant during the first days, but afterwards a rapid growth of the pathogen on the fruit, not controlled anymore by the antagonists, was observed. The BCAs could be used on citrus fruit only combined with reduced dosages of the fungicides. The efficacy of these strains of M. pulcherrima strains, some of them active against Monilia sp. on apple could also be evaluated against Monilia sp. on stone fruit, such as on peach, nectarine or apricot. Searching to enlarge the potential spectrum of activity of some strains, GS88 and BIO126, have been tried against Plasmopara viticola on grapevine (Perfumo, 2002). These BCAs, selected against postharvest diseases, were found effective also to control an oomycete pathogen on a different environment, the phylloplane. The two strains, together with 7 others, were selected among 231 initial isolates. The work, started with trials of efficacy in vitro, on leaf disks and in planta, is in progress. 4. Mechanisms of action The mechanisms of action and efficacy of four isolates (GS37, GS88, GA 102, and BIO126) of the yeast M. pulcherrima against B. cinerea, P. expansum, Alternaria sp., and Monilia sp., all postharvest pathogens of apple fruits, were studied in vitro and on apples (Spadaro et al. 2002). Applications of culture filtrates and autoclaved cells of the isolates were ineffective in reducing the diameter of the lesions on the fruits, supporting the hypothesis that living cells are necessary for biocontrol. In experiments of antagonism in vitro, on different solid substrates, a reduction of the mycelial growth of the pathogens emerged, so that, at least in vitro, the antagonists could produce some diffusible toxic metabolites. Co-cultivating in vitro on a synthetic medium, B. cinerea spore (105 ml-1) germination was completely inhibited by the presence of 108 cells of the antagonists, while culture filtrates and autoclaved 105

Chapter 5 ________________________________________________________________

suspensions were not able to reduce germination. To improve the knowledge on which kind of competition could be involved, an in vitro system for the cultivation of the antagonist separated from the pathogen by a hydrophilic membrane was used, permitting the free movement of liquids and nutrients. At lower concentrations of apple juice (0.5%) less pathogen conidia germinated. The main mode of action involved in the biocontrol is competition for nutrients (Spadaro et al., 2003a). However a direct interaction can not be excluded.In future studies to increase the knowledge on the mechanisms of action, it could be useful to purify and characterize the substances released in the culture substrates. Enzymes surely involved in the process of antagonism, such as glucanases or chitinases, specific for the cell wall of the fungi, have already been isolated from media where yeast antagonist were grown (Jijakli and Lepoivre, 1998; Wilson et al., 1994). It could also be useful to apply the microorganisms with different nutrients, sources of carbon or nitrogen, to understand which of them are involved in the mechanism of competition (Janisiewicz et al., 1992; Piano et al., 1998). The main aminoacids present in the fruit tissues are asparagine, glutamate and aspartate. Trials already carried out on other strains of M. pulcherrima show that, with slight differences, they use these aminoacids with an increase in the population growth (Janisiewicz et al., 2001). Glutamine was found to influence directly the antagonistic activity,

because,

when

applied

on

the

fruit

together

with

the

antagonist, it determines a significant reduction of grey rot on apple. Other aminoacids, such as arginine, phenilalanine, leucine, lysine and tyrosine contribute to control this pathogen (Piano et al., 1998). Future studies will also concentrate on the potential of resistance induced in the host tissue (Arras, 1996; Wilson and El-Ghaouth, 1993). 5. Sensitivity and compatibility to fungicides Biological means cannot at the moment solve all the problems of postharvest rots during fruit storage and they must be considered instruments to be used in combination with other methods in an 106


integrated vision of disease management. For example, BCAs can be combined with waxes and reduced dosages of fungicides applied not only in post but also in pre-harvest (Pusey, 1994). M. pulcherrima BIO126, GS88, GA102 and GS37 were studied for their sensitivity to fungicides, insecticides and calcium chloride, to clear the possibility of integration with the usual chemical products used in preand post-harvest in pome fruit production. Benomyl and thiabendazole, benzimidazoles widely used in post-harvest on pome fruits, did not show to inhibit the growth of the antagonists, except at 1000 ppm, therefore it could be possible employ the BCAs with reduced dosages of these fungicides. Yeast isolates were also tolerant to vinclozolin and procymidone, dicarboximides registered for post-harvest use. Moreover M. pulcherrima strains were not sensible to triphorine and sulphur, used in orchard. The yeast isolates were sensitive to low concentrations of imazalil, used in post-harvest, and cyproconazole, tebuconazole, triadimefon, bitertanol and penconazole, employed in open field; all of them are inhibitors of the ergosterole biosynthesis (Spadaro et al. 2000; Vola et al., 2000). Insecticides

tested,

normally

used

in

Northern

Italy

during

the

vegetative season on apple trees, were consistent with the growth of the antagonists. The antagonists were tolerant to 1000 ppm of calcium chloride. Between the strategies experimented during the last years in fruit protection, it is to remember the association of biological agents associated with calcium chloride infiltration. The addition of this salt greatly enhance the action of the antagonist yeast (McLaughlin et al., 1990; Piano et al., 1998). The biocontrol capability of the yeasts, during the experiment in semicommercial conditions, was not affected by the low temperature of storage

(1°C)

of

the

fruits

and

by

the

controlled

atmosphere.

Antagonists are compatible with normal methods of storage and with chemical products employed in post-harvest. For an use in open field before harvest, sensitivity to many inhibitors of sterols biosynthesis should be taken into account.

107

Chapter 5 ________________________________________________________________

6. Pre-harvest use One of the major obstacles to the development of postharvest BCAs is their inability to control previously established infections, such as latent infections. Field application of the BCAs may enable early colonisation of the fruit surfaces, protecting them from these infections (Ippolito and Nigro, 2000). First trials to evaluate the possibility to apply the microorganisms in preharvest were carried out on apple cv. Renetta del Canada. In field numerous micro- and macrolesions can already originate and be colonized by postharvest pathogens. Two mixes of antagonists were used to exploit the possible synergies among the strains: one formed by BIO126, GS37, GS88 and GA102 of M. pulcherrima and another with strains 4.4 (M. pulcherrima) and ME134 (Pseudomonas syringae), previously studied (Gullino et al., 1992; Bonino et al., 1994). The two mixes were not able to colonize continuously the fruit surface. Neither application was able to provide significant disease control in orchard. Among the possible reasons for the failure, a scarce adaptability of the antagonists to the different climate and nutritional conditions could be hypothesized. It is important to outline that the season of the experiments was very rainy and that the antagonist was applied without co-formulants, such as adhesive substances, that could have helped the colonization of the fruit surface (Spadaro, 2000). Other experiments are in progress to evaluate the possibility to apply the strain BIO126 of M. pulcherrima in orchard on apple and on peach trees to protect the fruits from the attack of the main postharvest pathogens. 7. Integrated control Since alternatives to chemical control do not generally possess a broad spectrum of activity and they are not as effective as fungicides, a combination of alternative methods could be more effective and consistent than one alternative alone. Hot water treatment, sodium bicarbonate and ethanol are non-curative treatments whose effects in 108


vivo are primarily fungistatic and not as persistent as the BCA. Acibenzolar-S-methyl is an elicitor of systemic acquired resistance in the host tissue, that could help in the defence of the fruit from the pathogens. The fruits were stored at 23°C for 5 days and at 4°C for 20 days. The antagonist, applied at 108 cells ml-1, proved to be the key element for the control of both pathogens, resulting more efficient after cold storage, with a reduction of 97.2% and 56.6% of the lesion diameter of grey

mould

and

blue

mould.

Ethanol

and

acibenzolar-S-methyl

permitted only a partial reduction of grey mould severity. Heat treatment and sodium bicarbonate improved significantly the efficacy of the BCAs against blue mould with storage at 23°C (Spadaro et al., 2004a). To study the possibility of a single application of the BCA (107 cells ml-1) with reduced dosages of sodium bicarbonate or ethanol, the viability of the microorganism with these chemicals was studied and a new set of experiments was established. Against both pathogens, the higher reduction of the lesion diameter was obtained by simply treating with the BCA. Significant results on blue mould were provided by the application of 20% ethanol or 5% sodium bicarbonate before the BCA and by the application of BIO126 in 0.1% sodium bicarbonate (Spadaro et al. 2004a). It is possible to associate the fungistatic effect of ethanol or sodium bicarbonate to the effect of the BCA, persistent on the fruit for longer periods.

Pre-treatment

with

sodium

bicarbonate

or

ethanol

and

successive application of the cell suspension of BIO126 M. pulcherrima could become a successful alternative to fungicide usage in postharvest disease control of pome fruit. 8. Molecular characterization One of the major characteristics for an antagonist to be used in biological control is its precise identification and its traceability, to permit to follow its environmental fate in space (dispersion) and in time (survival), after release (Gullino et al., 1995). Molecular tools can assist 109

Chapter 5 ________________________________________________________________

to monitor the genetic and environmental fate of these agents after releasing. Suitable and reproducible strain authentication methods are necessary in commercial procedures such as filling patents and product licensing. Moreover it is essential to develop a quality control system that allows to monitor the genetic stability over time of the biofungicide as a commercial product (Avis et al., 2001). Random amplified polymorphic DNA (RAPD) and arbitrarily primed-PCR (AP-PCR) techniques were useful methods to identify and evaluate the survival rate of some fungi (Aureobasidium pullulans) and yeasts as agents for postharvest biological control (Schena et al., 1999; 2000). Specific fingerprints using amplified fragment length polymorphism (AFLP) technique have also been developed to monitor the population of Rhodotorula glutinis, Cryptococcus laurentii and A. pullulans in both the field and cold room (Lima et al., 2003). Seven

strains

of

the

yeast

M.

pulcherrima,

isolated

from

the

carposphere of apples cv. Golden delicious, showed biocontrol capability against B. cinerea and P. expansum. PCR-RFLP of the 18S+ITS rDNA was tested as rapid and easy method for yeast species identification. To assess the genetic diversity of M. pulcherrima, the RAPD and AFLP techniques were used (Spadaro et al. 2004b). With six RAPD primers 33 polymorphic bands were obtained, while with six AFLP primer pairs 729 polymorphic bands were scored. The genetic distances obtained with AFLP technique were visualized on a dendrogram. Strains isolated in different locations with high genetic diversity could have similar biocontrol potential. One primer pair, such as McaEaa or McgEat, resulted highly informative and sufficient to describe the genetic distance among the strains. AFLP fingerprints were used to develop STS strain-specific markers to improve identification and monitoring in the environment of the BCAs. 9. Formulation Finally formulation should constitute a fundamental field of further studies, to permit the product access to the market (Fravel et al.,1999). The fruit surface is not generally rich in nutrients. For this reason, to add 110


the formulation substances exclusively of preferentially metabolised by the antagonist could be useful. The putative positive effect of the addiction of aminoacids, such as L-asparagine or L-proline, should be assessed (Janisiewicz et al., 1992; Piano et al., 1998). The addiction of some salts to the culture media can positively or negatively influence the growth and activity of the BCAs. In particular, calcium chloride, applied alone or mixed with other salts (potassium chloride o ferrous sulphate), determines a reduction of apple grey mould (Piano et al., 1998). Among the carbon sources, glucose, saccharose, galactane, cellobiose, mannitol and sorbose significantly increase the biocontrol capability of M. pulcherrima 4.4 (Piano et al., 1997; 1998). Preharvest application of the biofungicides could become feasable if the formulate is able to protect the microorganism from water shortage and ultraviolet light during and after application, to avoid a rapid decrease of the population density. Adhesivants also play a strong role, because they permit a higher survival of the antagonists on the fruit carposphere and a possible preharvest use of the biofungicides, avoiding the washing away of the product in case of rain. Among the already tested compounds, sodium alginate,

sodium

polipectate,

chitosan

and

sodium

carboxymethylcellulose, particularly if mixed with glutamine, mannitol or sorbitol, increase the survival on fruit and the control efficacy of some M. pulcherrima strains (Piano et al., 1998). Finally a correct formulation should tend to extend the shelf-life of the product, facilitating the storage for periods ranging from 6 months to 2 years.

For

example,

Bio-Save

products,

based

on

Pseudomonas

syringae, are formulated to permit storage at –70°C for long periods. Formulation will be the major part of the next studies on the BCAs, and possibly two different formulations should be developed, one for postharvest use and a different one for orchard application. 10. Registration M. pulcherrima (Pitt) M. W. Miller is an ascomycetous teleomorphic yeast and it is the teleomorph of the anamorph Candida pulcherrima. 111

Chapter 5 ________________________________________________________________

From growth at different temperatures (Spadaro, 2000), it resulted that the isolates do not grow at 37°C, which is important from a toxicological point of view. Another point favourable for a future commercialisation is the lack of production of antibiotics active against the tested pathogens in vivo. The main mechanism of action used by the BCAs is competition with pathogens for space and nutrients, but could be secondary mechanisms of action with synergistic effect also a direct interaction, such as parasitism, and some production of toxic metabolites in particular nutritional condition, not investigated in this work. This species of yeast is involved in the first step of the fermentation process of apples for cider-making (Beech, 1993). As the ethanol level raises (2 to 4%), these initial fermenters begin to die out and the microbial succession is taken over by Saccharomyces cerevisiae. In grape must and during the early phase of fermentation, apiculate yeasts belonging to the species Kloeckera apiculata are dominant and, to a lesser extent, isolates of M. pulcherrima, can also be detected (Fleet and Heard, 1993). This species, present on different food matrixes, naturally occurs on fruits, buds and floral parts of certain apple trees (Boekhout and Robert, 2002) and has been reported to be a yeast species as effective as BCA against postharvest decay of apple, table grape, grapefruit and cherry tomato (Schena et al., 2000; Janisiewicz et al. 2001; Spadaro et al., 2002). M. pulcherrima is a yeast species with good antagonistic properties in general against B. cinerea and P. expansum. The efficacy of seven strains of M. pulcherrima, isolated from the carposphere of apples cv. Golden delicious, was compared with that of nineteen M. pulcherrima strains, isolated from different matrixes in different geographical regions. The strains were more effective in the control of B. cinerea than of P. expansum, after storage for 28 days at 4°C, with a mean reduction of the pathogen growth respectively to 30.0% and 49.3% of the control. Antagonistic properties could be owned by microorganisms of different origin.

112


11. Environmental risks Molecular tools can assist to monitor the genetic and environmental fate of these agents after releasing. One of the major characteristics for an antagonist to be used in biological control is its precise identification and its traceability, to permit to follow its environmental fate in space (dispersion) and in time (survival), after release (Gullino et al., 1995). The RAPD-PCR technique was useful to monitor the survival of the antagonists after fruit application and to study its competitive action on the epiphytic microflora. The OPB17 primer was chosen for the identification of the GS88 strain of Metschnikowia

pulcherrima, 7

that

6

was

applied

at

three

different

5

concentrations (10 , 10 and 10 cells/ml) for different times (5 to 600 seconds) on the carposphere of apple fruit Golden delicious. Fungi, yeast and bacteria were isolated on different substrates. A rapid extraction of the yeast genomic DNA was carried out: cells were kept 1h at 37°C in 15 ml lyticase (Sigma) and 10’ in boiling water. After 10 minutes dip in the antagonistic cell suspension, no other yeast strains were detectable. After 5 seconds dip, the number of antagonistic cells found after RAPD-PCR analysis was positively related with the initial concentration of the treatment. The number of fungi and other yeast isolates decreased, while the number of bacteria was higher, with superior concentrations of GS88 (Spadaro et. al, 2003b). It would be interesting to evaluate the curve of the antagonists population on the carposphere. AFLP fingerprints could be used to develop new STS marker specific for the antagonistic strains to improve identification and monitoring in the environment. Concluding, some isolates of M. pulcherrima well studied provided satisfying biocontrol results. They are ready to be used as BCAs. To reach commercialisation, an industrial development of the biofungicides, including fermentation optimisation, formulation studies and large-scale production, should be undertaken, with a close collaboration among university and private partners.

113

Chapter 5 ________________________________________________________________

References

Arras, G., 1996. Mode of action of an isolate of Candida famata in biological control of Penicillium digitatum in orange fruits. Postharvest Biol. Technol. 8, 191198. Avis, T.J., Hamelin, R.C., Bélanger, R.R., 2001.

Approaches

characterization

of

to

molecular

fungal

biocontrol

agents: some case studies. Can. J. Plant Pathol. 23, 8-12. Beech,

F.W.,

1993.

Yeasts

in

Cider

Making. In: Rose, A.H., Harrison, J.S. (Eds.),

The

Yeasts,

Vol.5,

“Yeast

Technology”. Academic Press, London, pp. 169-213.

–

Beneficial

and

Detrimental

Aspects. Behr’s Verlag, Hamburg. Bonino, Gullino,

M.,

Y. (Eds.), Integrated Pest and Disease Management

in

Greenhouse

Crops.

Kluwer Academic Publishers, Dordrecht, pp. 365-376. Gullino, M.L., Aloi, C., Palitto, M., Benzi, D., Garibaldi, A., 1991. Attempts at biological control of postharvest diseases of apple. Meded. Fac. Landbouww. Univ. 56, 195-202. Gullino, M.L., Benzi, D., Ozino-Marletto, I., Avataneo, M., 1992. Caratteristiche di due potenziali mezzi biologici di lotta contro

agenti

di

marciumi

in

post-

raccolta. Atti Giornate Fitopatologiche 2, 381-388.

Boekhout, T., Robert, V., 2002. Yeasts in Food

Gullino, M.L., Van Lenteren, J.C., Elad,

Fioretta,

M.L.,

1994.

C.,

Gullino, M.L., Bonino, M., Piano, S., Testoni, A., Salimei, A., Duverney, C., 1994. Biological control of postharvest

Berio,

T.,

Selezione

di

rots

of

apples.

International

Proc.

ANPP

Conference

on

–

4th

Plant

microrganismi attivi contro il marciume

Diseases, Bordeaux, France, December

bruno

6-8, 1994, pp. 333-340.

su

mele.

Atti

Giornate

Fitopatologiche 3, 351-358.

Gullino, M.L., Migheli, Q., Mezzalama,


M., 1995. Risk analysis in the release of


biological control agents: antagonistic

antagonistic

Fusarium oxysporum as a case study.

microorganisms.

Ph.D.

dissertation, University of Stellenbosch, Capetown.

Plant Dis. 79, 1193-1201 Hofstein,

R.,

Friedlender,

B.,

1994.

Fleet, G.H., Heard, G.M., 1993. Yeasts:


growth during fermentation. In: Fleet,

and delivery systems. Proc. Brighton

G.H.

Crop Protection Conference, Brighton,

(Ed.),

Wine

microbiology

and

biotechnology. Harwood, Chur, pp. 2754.

21-24 November 1994, pp. 1273-1280. Ippolito, A., Nigro, F., 2000. Impact of

Fravel, D.H., Rhodes, D.J., Larkin, R.P.,

preharvest

1999. Production and commercialization


of biocontrol products. In: Albajes, R.,

114

application

of

biological



McLaughlin,

19, 610-619.


Janisiewicz,

W.J.,

1991.

Biological

control of postharvest fruit disease. In: Arora, D.K., Rai, B., Knudsen, G.R. (Eds.), Handbook of Applied Micology. Vol. 1: Soil and Plants. Marcel Dekker, New York, pp. 301-326.

biocontrol of blue mold with nutrient analog 2-deoxy-D-glucose on apples and Appl.

Environ.

Microbiol.

60,

2671-2676. Janisiewicz, Kurtzman,

W.J., C.P.,

Tworkoski, 2001.

T.J.,

Biocontrol

potential of Metschnikowia pulcherrima strains

against

blue

mold

of

apple.


Nutritional

biocontrol

of

enhancement

blue

mold

on

of

apples.

Phytopathology 82, 1364-1370. Jijakli,

M.H.,

Characterization

Lepoivre, of

M.E.,

and

biological

control

diseases

of

salt

apple

of with

Candida sp. Phytopathology 80, 456461. J.,

Wilson,

Colonization naturally

of

C.L.,

apple

occurring

1994.

wounds

by

microflora

and

introduced Candida oleophila and their effect on infection by Botrytis cinerea during storage. Biol. Control 4, 138-144. Perfumo,

C.,

2002.

Selezione

di

microrganismi attivi contro Plasmopara viticola su vite. MS Thesis. University of Turin. S.,

Neyrotti,

V.,

Migheli,

Q.,

Gullino, M.L., 1997. Characterization of the

biocontrol


capability

pulcherrima

postharvest

rot

of against

of

apple.


P.,

an

Wisniewski,

concentration

on

postharvest

Piano,

Janisiewicz, W.J., Usall, J., Bors, B., 1992.

inoculum

solutions

Mercier,

Janisiewicz, W.J., 1994. Enhancement of

pears.

of

R.J.,

1998.

exo-β-1,3-

glucanase produced by Pichia anomala strain K, antagonist of Botrytis cinerea on apples. Phytopathology 88, 335-343.

Piano,

S.,

Cerchio,

F.,

Migheli,

Q.,

Gullino, M.L., 1998. Effetto di diverse sostanze

sull'attività

dell'antagonista

Metschnikowia pulcherrima 4.4 contro Botrytis

cinerea

e

sulla

sua

Lima, G., Arru, S., De Curtis, F., Arras,

sopravvivenza sulle mele. Atti Giornate

G., 1999. Influence of antagonist, host

Fitopatologiche, 3-7 maggio 1998, pp.

fruit and pathogen on the biological

495-500.

control of postharvest fungal diseases by yeasts. J. Ind. Microbiol. Biotechnol. 23, 223-229.

Pusey,

P.L.,

biocontrol

1994.

agents

Enhancement for

of

postharvest

diseases and their integration with other

Lima, G., De Curtis, F., Castoria, R., De

control

Cicco, V., 2003. Integrated control of

Wisniewski,

apple

postharvest

pathogens

strategies. M.E.

In:

Wilson,

(Eds.),

C.L.,

Biological

and

Control of Postharvest Diseases - Theory

survival of biocontrol yeasts in semi-

and Practice. CRC Press, Boca Raton, pp.

commercial

77-88.

conditions.

Pathol. 109, 341-349.

Eur.

J.

Plant

Schena, L., Ippolito, A., Zahavi, T., Cohen, L., Nigro, F., Droby, S., 1999.

115

Chapter 5 ________________________________________________________________

Genetic diversity and biocontrol activity

Spadaro, D., Nania, D., Garibaldi, A.,

of

isolates

Gullino, M.L., 2003b. RAPD approach to

Postharvest

identify and monitor yeast biocontrol

Aureobasidium

against

pullulans

postharvest

rots.

Biol. Technol. 17, 189-199.

agents against postharvest diseases of

Schena, L., Ippolito, A., Zahavi, T., Cohen, L., Droby, S., 2000. Molecular approaches to assist the screening and monitoring

of

postharvest

biological

yeasts. Eur. J. Plant Pathol. 106, 681-

pome

fruit.

Proc.

7th

International

Conference on Public Goods and Public Policy

for

Agricultural

Biotechnology,

Ravello, Italia, 29 giugno – 3 luglio 2003, in press. Spadaro, D., Garibaldi, A., Gullino, M.L.,

691. Snowdon, A.L., 1990. A colour atlas of post-harvest diseases and disorders of fruits and vegetables. Volume 1. Wolfe Scientific, London. Spadaro,

D.,

and Botrytis cinerea on apple combining a biocontrol agent with hot water dipping and acibenzolar-S-methyl, baking soda, or ethanol application. Postharvest Biol.

2000.

Studio

del

meccanismo d’azione e dell’efficacia di lieviti

2004a. Control of Penicillium expansum

antagonisti

nei

confronti

di

patogeni della frutta in post-raccolta. MS Thesis. University of Turin.

Technol., accepted. Spadaro, D., Sabetta, W., Acquadro, A., Garibaldi,

A.,

Gullino,

M.L.

2004b.

Metschnikowia pulcherrima: a promising species

isolated

from

for

biological

different

food

Spadaro, D., Gullino, M.L., 2004. State

matrixes

of

postharvest diseases in apple. Int. J.

art

and

future

perspectives

of

biological control of postharvest fruit diseases. Int. J. Food Microbiol., in press (available online 25 September 2003).

control

of

Food Microbiol., submitted. Vola,

R.,

2000.

Lotta

biologuca

ai

marciumi della frutta in post-raccolta:

Spadaro, D., Vola, R., Piano, S., Gullino,

studio delle caratteristiche biologiche,

M.L,

dei

2002.

Mechanisms

of

action,

meccanismi

di

azione

e

delle

efficacy and possibility of integration

potenzialità di impiego di due lieviti

with chemicals of four isolates of the

antagonisti. MS Thesis. University of

yeast Metschnikowia pulcherrima active

Turin.

against

postharvest

pathogens

on

apples. Postharvest Biol. Technol. 24, 123-134.

Wilson,

C.L.,

El-Ghaouth,

Multifaceted postharvest

biological diseases

A.,

1993.

control of

fruits

of and

Spadaro, D., Alloati, F., Gullino, M.L.,

vegetables. In: Lumsden, R.D., Vaughn,

2003a. Biocontrol of grey mould on

J.L.

apple:

Biologically

a

Metschnikowia

pulcherrima

(Eds.),

Pest

management:

Based

Technologies.

strain competes for nutrients with the

American

pathogen. 23rd International Specialised

Washington, pp. 181-185.

Symposium

on

Yeasts,

Budapest,

Ungheria, 26-29 agosto 2003, Book of Abstracts, 92.

116

Chemical

Society

Press,

Wilson, C.L., Wisniewski, M.E., 1994. Biological

Control

of

Postharvest


Diseases – Theory and Practice. CRC Press, Boca Raton. Wilson, C.L., El-Ghaouth, A., Chalutz, E., Droby, S., Stevens, C., Lu, J., Khan, V., Arul,

J.,

1994.

resistance

to

Potential

of

induced

control

postharvest

diseases of fruits and vegetables. Plant Dis. 78, 837-844. Wilson, C.L., Wisniewski, M.E., Droby, S.,

Chalutz,

E.,

1993.

A

selection

strategy for microbial antagonists to control postharvest disease of fruits and vegetables. Scientia Hort. 53, 183-18.

117

ACKNOWLEDGEMENTS I desire to thank: •

Professor Maria Lodovica Gullino, for her continuous suggestions;

•

Professor Angelo Garibaldi, for his far-seeing wisdom;

•

Davide Nania, Wilma Sabetta, Cristina Perfumo, Francesca Alloati, Valentina Garibaldi, Laura Rostagno, Chiara Beltramo, Alberto Brero and Maria Maddalena Macrì, for accepting my tutoring during their thesis work;

•

Matias Pasquali, Alberto Acquadro and Ezio Portis, for their practical suggestions and moral support during these three years of doctorate;

•

Federico Tinivella, for his cheerfulness and for tolerating every day my moods;

•

Professor Richard Sikora and Professor Sri Sriskandarajah, for his enthusiasm in teaching crop protection and organic farming;

•

Serenella Piano, for introducing me into postharvest biological control;

•

Renata Luongo, for her faultless practical help;

•

Jeanne Griffin, for her useful English revision;

•

Davide Baridon, Domenico Bertetti, Mauro Busso, Andrea Camponogara, Flavia Dematheis, Giorgia Fenocchio, Marina Ferrari, Dario Ghiringhelli, Giovanna Gilardi, Valeria Grasso, Milan Ivic, Said Kejji, Manuela La Rocca, Luisa Marena, Guido Martano, Andrea Minuto, Giovanni Minuto, Massimo Mocioni, Andrea Rettori, Patrizia Titone, Elizabeth Tridico and the others that worked or are working at Agroinnova, for sustaining me and sharing with me happy and difficult moments;

•

Alessandro Catti, for his joy of life, never forgotten;

•

Manuela and Marta, for their love.

Curriculum vitae Davide Spadaro was born in Carmagnola (Turin), Italy, on the 26th of May 1975. In July 1994 he finished the classical studies at the Liceo “G. Baldessano” of Carmagnola. In July 2000 he got his degree in Agricultural Plant Biotechnology with full marks. He won the prize of the Piedmont Region as best MS thesis on Consumer Care with his thesis on “Study of the mechanism of action and efficacy of antagonistic yeasts against postharvest fruit pathogens”. In November 2000 he started his Ph.D. research at the Plant Pathology Sector – Di.Va.P.R.A. / Agrinnova (University of Turin) under the supervision of Prof. Maria Lodovica Gullino. His research activities implied mainly biological and integrated control against postharvest diseases of pome fruit, citrus fruit, kiwifruit, stone fruit, and rose. He worked also on the molecular characterization of biocontrol agents, using RAPD and AFLP markers, and on the biological control of Plasmopara viticola in grapevine. He spent part of the Ph.D. training period at the Institute for Plant Diseases, University of Bonn (Germany), under the supervision of Prof. Richard Sikora. From 2000 to 2001 he was also the editor of the Italian monthly “Informatore fitopatologico”, specialised in crop protection, belonging to IlSole-24Ore Publishing Group. At present (January 2004), he started a Post-Doc position at Agrinnova (University of Turin), focusing on formulation, molecular characterisation using STS markers, and environmental impact studies on the release of biocontrol agents.

Publications

Scientific papers on international journals (ISI)

SPADARO D., VOLA R., PIANO S., GULLINO M.L. (2002) Mechanisms of action, efficacy and possibility of integration with chemicals of four isolates of the yeast Metschnikowia pulcherrima

active against

postharvest pathogens on apples. Postharvest Biology & Technology 24 (3), 123-134. SPADARO

D.,

GARIBALDI

A.,

GULLINO

M.L.

(2003)

Postharvest

biological control of grey and blue mould on apple. Phytopathology 93 (6S), 805 (abstract). SPADARO D., GULLINO M.L. (2004) State of art and future perspectives of biological control of postharvest fruit diseases. International Journal of Food Microbiology, in press (available online 25 September 2003). SPADARO D., GARIBALDI A., GULLINO M.L. (2004) Control of Penicillium expansum and Botrytis cinerea on apple combining a biocontrol agent with hot water dipping and acibenzolar-S-methyl, baking soda, or ethanol application. Postharvest Biology and Technology, accepted. SPADARO D., SABETTA W., ACQUADRO A., GARIBALDI A., GULLINO M.L. (2004) Metschnikowia pulcherrima: a promising species isolated from different food matrixes for biological control of postharvest diseases in apple. International Journal of Food Microbiology, submitted.

Papers published on journals with editorial board

SPADARO D. (2001) Una piccola infestante sta rivoluzionando il mondo vegetale

–

Sequenziato

il

genoma

di

Arabidopsis.

fitopatologico – La difesa delle piante 51 (6), 3-4.

Informatore

DECOIN M., WHIPPS J.M., NICOT P., GULLINO M.L., SPADARO D. (2002) Micro-organismes contre agents pathogènes – Des spécialités trés spéciales! Phytoma – LDV, 549, 32-36. SPADARO D. (2002) La sfida delle piante geneticamente modificate: opportunità o rischio? Informatore fitopatologico – La difesa delle piante, 52 (10), 7-10. SPADARO

D.,

PALLOTTINO

F.,

SMITHSON

A.

(2003)

Alcune

considerazioni etiche sull’applicazione delle biotecnologie nei paesi in via di sviluppo. Informatore fitopatologico – La difesa delle piante, 53 (2), 31-38.

Papers presented to congresses and/or published on proceedings

SPADARO D., PIANO S., VOLA R., DUVERNEY C., GALLIANO A., GULLINO M.L. (2001) Lotta biologica ai marciumi delle pomacee in post-raccolta. Convegno su “La difesa delle colture in agricoltura biologica”. Notiziario sulla Protezione delle Piante 13, 153-156. SPADARO D., PIANO S., GULLINO M.L. (2002) Use of microorganisms, heat treatment and natural compounds against Botrytis rot on apple. Proc. 2nd International Conference on the alternative control methods against plant pests and diseases, Lille, France, 4-7 March 2002, 446453. GULLINO M.L., SPADARO D. (2002) Biological control of postharvest diseases

of

fruit:

present

situation

and

perspectives.

Proc.

2nd

International Conference on the alternative control methods against plant pests and diseases, Lille, France, 4-7 March 2002, 41-48. SPADARO D., ALLOATI F., GARIBALDI A., GULLINO M.L. (2002) BIO126: an effective Metschnikowia pulcherrima strain against grey and blue mould on apple (Abstract). Journal of Plant Pathology, 84, 195. SPADARO

D.,

GARIBALDI

A.,

GULLINO

M.L.

(2003)

Additional

treatments to improve biological control of grey and blue mould on apple (Abstract). Proc. 8th International Congress of Plant Pathology, Christchurch, New Zealand, 2-7 February 2003, Volume 2, 32. PASQUALI M., SPADARO D., GARIBALDI A., GULLINO M.L. (2003) Identification of Fusarium oxysporum strains using molecular tools (Abstract).

Proc.

8th

International

Congress

of

Plant

Pathology,

Christchurch, New Zealand, 2-7 February 2003, Volume 2, 85. SPADARO D., NANIA D., GARIBALDI A., GULLINO M.L. (2003) RAPD approach to identify and monitor yeast biocontrol agents against postharvest diseases of pome fruit. Proc. 7th International Conference on Public Goods and Public Policy for Agricultural Biotechnology, Ravello, Italy, 29 June – 3 July 2003. SPADARO D., ALLOATI F., GULLINO M.L. (2003) Biocontrol of grey mould on apple: a Metschnikowia pulcherrima strain competes for nutrients with the pathogen. 23rd International Specialised Symposium on Yeasts, Budapest, Hungary, 26-29 August 2003, Book of Abstracts, 92. SPADARO D., GARIBALDI A., GULLINO M.L. (2003) Identificazione e monitoraggio di lieviti antagonisti di patogeni post-raccolta delle pomacee tramite impiego di marcatori molecolari. Notiziario sulla Protezione delle Piante, 17, 163. SPADARO

D.,

GARIBALDI

characterization

of

A.,

GULLINO

Metschnikowia

M.L.

pulcherrima

(2004)

Molecular

strains

against

postharvest diseases of pome fruit. Proc. International Workshop: Development

of

biocontrol

agents

of

diseases

for

commercial

applications in food production systems, Sevilla, Spain, 24-27 March 2004, in press. SPADARO D., GARIBALDI A., GULLINO M.L. (2004) Efficacy and molecular characterization of Metschnikowia pulcherrima strains against postharvest diseases of pome fruit. Proc. 5th International Postharvest Symposium, Verona, Italy, 6-11 June 2004, in press.

Lay-out: Davide Spadaro Printing: T-Art Printing support: AGRINNOVA

Biological Control of Postharvest Diseases of Pome Fruit Using Yeast Antagonists

Biological Control of Rice Diseases (Progress in Biological Control, 8)

Postharvest Diseases of Fruits and Vegetables: Development and Control

Handbook of Biological Control: Principles and Applications of Biological Control

Allelochemicals: Biological Control of Plant Pathogens and Diseases

Biological Control of Vertebrate Pests:

Control of Human Parasitic Diseases

Control of Human Parasitic Diseases

Regulation of Biological Control Agents

Biological Control of Microbial Plant Pathogens

Control Of Communicable Diseases Manual 18th Edition

BIOLOGY AND BIOLOGICAL CONTROL KNAPWEED OF

Biological Control of Microbial Plant Pathogens

Dynamic Models and Control of Biological Systems

Evaluating Indirect Ecological Effects of Biological Control

Biological and Biotechnological Control of Insect Pests

Diseases of Tropical and Subtropical Fruit Crops (Cabi Publishing)

Conservation Biological Control

Quality Control and Production of Biological Control Agents

Biological Control: (Cabi International)

Fruit of the month

Potatoes Postharvest

Plant Defence: Biological Control (Progress in Biological Control)

Fruit Quality and Its Biological Basis

Fruit Of Knowledge

Fruit of All Evil

Yeast Protocols

Biological Control: A Global Perspective

Biological Control of Weeds and Plant Diseases: Advances in Applied Allelopathy

Biological Control of Crop Diseases (Books in Soils, Plants & the Environment)

Leptin and Leptin Antagonists

Biological Control of Postharvest Diseases of Pome Fruit Using Yeast Antagonists

Biological Control of Rice Diseases (Progress in Biological Control, 8)

Postharvest Diseases of Fruits and Vegetables: Development and Control

Handbook of Biological Control: Principles and Applications of Biological Control

Allelochemicals: Biological Control of Plant Pathogens and Diseases

Biological Control of Vertebrate Pests:

Control of Human Parasitic Diseases

Control of Human Parasitic Diseases

Regulation of Biological Control Agents

Biological Control of Microbial Plant Pathogens

Control Of Communicable Diseases Manual 18th Edition

BIOLOGY AND BIOLOGICAL CONTROL KNAPWEED OF

Biological Control of Microbial Plant Pathogens

Dynamic Models and Control of Biological Systems

Evaluating Indirect Ecological Effects of Biological Control

Biological and Biotechnological Control of Insect Pests

Diseases of Tropical and Subtropical Fruit Crops (Cabi Publishing)

Conservation Biological Control

Quality Control and Production of Biological Control Agents

Biological Control: (Cabi International)

Fruit of the month

Potatoes Postharvest

Plant Defence: Biological Control (Progress in Biological Control)

Fruit Quality and Its Biological Basis

Fruit Of Knowledge

Fruit of All Evil

Yeast Protocols

Biological Control: A Global Perspective

Biological Control of Weeds and Plant Diseases: Advances in Applied Allelopathy

Biological Control of Crop Diseases (Books in Soils, Plants & the Environment)

Leptin and Leptin Antagonists

Recommend Documents