A

eoaibBpciipo©

K

1

(spgfecw

0d

Available online at www.sciencedirect.com

Postharvest Biology and Technology 49 (2008) 121–128

Metschnikowia pulcherrima strain MACH1 outcompetesBotrytis cinerea, Alternaria alternata and Penicillium expansum

in apples through iron depletion

Duraisamy Saravanakumar a, Annalisa Ciavorella a, Davide Spadaro b,∗,Angelo Garibaldi a, M. Lodovica Gullino a

a Centre of Competence for the Innovation in the Agro-environmental Sector, Universita degli Studi di Torino,via L. da Vinci 44, I-10095 Grugliasco (TO), Italy

b DiVaPRA – Plant Pathology, Universita degli Studi di Torino, via L. da Vinci 44, I-10095 Grugliasco (TO), Italy

Received 20 July 2007; accepted 16 November 2007

bstract

A new strain of Metschnikowia pulcherrima (MACH1) was studied for its efficacy as a biocontrol agent against Botrytis cinerea, Penicilliumxpansum and Alternaria alternata on apples stored for 8 months at 1 ◦C. The results of two semi-commercial trials demonstrated the efficacyf the biocontrol strain MACH1. In order to understand the mechanism of action involved, the yeast strain was investigated for its competitivebility for iron against postharvest pathogens of apple. M. pulcherrima strain MACH1 was cultivated on PDA with different concentrations ofron (supplemented as FeCl3) against A. alternata and B. cinerea. The yeast strain MACH1 produced a wider pigmented inhibition zone againstoth pathogens in low iron amendments while less inhibition was measured with increased iron concentrations. At the coloured inhibition zone,. cinerea and A. alternata conidia did not germinate and mycelial degeneration was observed. In addition, a high reduction in infection by bothathogens was recorded in apples treated with M. pulcherrima strain MACH1 supplemented with low iron amendments compared to higher irononcentrations. The same experiments were carried out in vivo and in vitro against P. expansum. M. pulcherrima strain MACH1 amended with low

−1

ron concentration (5 �g mL FeCl3), showing modest lesion diameter reduction and no effect on P. expansum under increased iron and withoutron amendments. This study illustrated that iron depletion by the yeast strain MACH1 under low iron conditions could reduce the growth of someostharvest pathogens in vitro and in vivo. Although, iron depletion seems to be a primary mode of action against the postharvest pathogens studied,ther mechanisms of action cannot be excluded in the biocontrol employed by M. pulcherrima strain MACH1.

2007 Elsevier B.V. All rights reserved.

disea

tpthc

ao

eywords: Biocontrol; Competition; Iron depletion; Pulcherrimin; Postharvest

. Introduction

Apple postharvest rots, caused by Penicillium expansum Linkblue mould), Botrytis cinerea Pers. (grey mould) and Alternariap. (Fr.) Keissl. (Alternaria rot) are particularly severe even inroduction areas where the most advanced storage technolo-ies are available (Snowdon, 2003). When permitted, syntheticungicides are the primary means to control postharvest dis-

ases (Eckert and Ogawa, 1985). However, the growing publiconcern over the health and environmental hazards associatedith pesticide use in orchards (Wilson and Wisniewski, 1994),

∗ Corresponding author. Tel.: +39 011 6708942; fax: +39 011 6709307.E-mail address: davide.spadaro@unito.it (D. Spadaro).

otapean

925-5214/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.postharvbio.2007.11.006

ses; Yeast

he development of fungicide resistant strains of postharvestathogens (Spotts and Cervantes, 1986) and the deregistra-ion of some of the most effective fungicides (Ragsdale, 2000),ave generated interest in the development of alternative non-hemical methods.

Biological control using microbial antagonists (Janisiewicznd Korsten, 2002; Spadaro and Gullino, 2004) has emerged asne of the most promising alternatives, either alone or as partf integrated pest management to reduce pesticide use. Duringhe past 30 years, several biocontrol agents have been exploitednd widely investigated against different postharvest fungal

athogens (Lima et al., 1997; Zahavi et al., 2000; Vivekananthant al., 2004; Calvo et al., 2007). Among the different biocontrolgents, yeasts deserve particular attention, as their activity doesot generally depend on the production of toxic metabolites,

1 Biolog

wiaaaw2

aseoaptnwscftMasa

natop

2

2

ivTC8

aAaitps5wwteRt

ap

2

VcAittw3p(aa6drSo1siusimim

2

sct1om

ywwo(ws2

22 D. Saravanakumar et al. / Postharvest

hich could have a negative environmental or toxicologicalmpact. Recently, Metschnikowia pulcherrima has been reporteds an effective biocontrol agent against postharvest decay ofpple, table grape, grapefruit and cherry tomato (Schena etl., 2000; Janisiewicz et al., 2001; Spadaro et al., 2002) asell as against some food-borne pathogens (Leverentz et al.,006).

The strains of the yeast species most investigated gener-lly act by consuming nutrients present on fruit and vegetablekins that normally allow rot-causing fungi to develop (Pianot al., 1997; Janisiewicz et al., 2001). A good understandingf the mode of action is essential before developing appropri-te commercial formulations and application methods. Severalossible biocontrol mechanisms have been suggested to be effec-ive against postharvest rots on fruit including competition forutrients and space, antibiosis, parasitism or direct interactionith the pathogens and induction of resistance in the host tis-

ue (Smilanick, 1994). Competition for nutrients and space isonsidered to be a primary mode of action against postharvestungal pathogens. In particular, competition for iron is believedo play a significant role in biocontrol interactions (Raaska and

attila-Sandholm, 1995). M. pulcherrima was found to producered pigment, pulcherrimin, that accumulates in cells and is alsoecreted around the colonies (Miller and Phaff, 1998; Kurtzmannd Droby, 2001).

The aim of this research was to assess the efficacy of aew strain of M. pulcherrima against B. cinerea, P. expansumnd Alternaria alternata on apples in semi-commercial condi-ions. A second objective was to study the competitive abilityf M. pulcherrima strain MACH1 for iron sequestration againstostharvest pathogens of apples.

. Materials and methods

.1. Antagonist and pathogens preparation

M. pulcherrima (Pitt) M. W. Miller isolate MACH1 wassolated from the carposphere of apple cv Golden Delicious, har-ested in organic orchards located in Piedmont, Northern Italy.he strain was deposited within the American Type Cultureollection on 19 June, 2007 with deposit designation PTA-487.

Yeast strain MACH1 was grown and cells were prepareds described in Spadaro et al. (2002). Two strains each of. alternata, B. cinerea and P. expansum were isolated frompples with rots and selected for their virulence by inoculationn artificially wounded apples. They were used as a mixturehroughout this work, to ensure a high level of disease and theresence of rots caused by multiple pathogens. Each strain wastored in slants on Potato Dextrose Agar (PDA; Merck) with0 mg L−1 of streptomycin Merck at 4 ◦C. Spore suspensionsere prepared by growing the pathogens on Petri dishes for twoeeks on PDA added with 50 mg L−1 of streptomycin. After

wo weeks of incubation at 25 ◦C, spores from the two strains ofach pathogen species were collected and suspended in sterileinger’s solution (Merck, Darmstadt, Germany). After filtering

hrough eight layers of sterile cheese-cloth, spores were counted

wTao

y and Technology 49 (2008) 121–128

nd brought to a final concentration of 105 spores mL−1 per eachathogen.

.2. Efficacy trials in semi-commercial conditions

Two experimental trials were carried out in Aosta (Aostaalley, Northern Italy) in cooperation with the Institut Agri-ole Regional on artificially infected ‘Golden Delicious’ apples.pples were harvested in orchards conducted by following

ntegrated pest management. Four boxes were used for eachreatment (100 fruit per box). Ten apples per box, to reproducehe most probable conditions after harvesting, were artificiallyounded at the equatorial region (3 mm diameter; 6 mm depth;wounds per fruit). All fruit were artificially inoculated by dip-ing for 60 s in 100-L tanks containing a conidial suspension105 spores mL−1 per pathogen) of B. cinerea, P. expansum,nd A. alternata. After 3 h, biocontrol isolates were appliedt 107 cells mL−1 by completely dipping the boxes of fruit for0 s in 100-L tanks containing the cell suspensions prepared asescribed. The treatments were a cell suspension of M. pulcher-ima MACH1 and a chemical control (thiabendazole, Tecto 20, Elf Atochem Agri Italy, 19,7% a.i., 30 g a.i. 100 L−1). More-ver, an inoculated control was represented by four boxes with00 fruit per box, 10 of them artificially wounded, dipped in thepore suspension of the three pathogens. When dry, apples werencubated at 23 ◦C for 24 h, then stored at 1 ◦C for 8 monthsnder controlled atmosphere (2% O2 and 3% CO2) and theame experiment was duplicated in another cold chamber stor-ng the fruit at 1 ◦C in 1% O2 and 2% CO2. After 5 and 8

onths of storage, the total rot incidence and the relative rotncidence (grey mould, blue mould, Alternaria rot) were deter-

ined.

.3. Effect of iron on pigment production and antagonism

Cells of M. pulcherrima strain MACH1 (107 mL−1) weretreaked onto YPD and PDA plates (10 plates for each con-entration of iron) with different concentrations of iron to testhe pigment production. Both media were amended with 5, 10,5 and 20 �g mL−1 of FeCl3 before autoclaving. The widthsf the reddish halos developing around the yeast colonies wereeasured after 5 days of incubation at 25 ◦C.The antagonistic activity of the pigment produced by the

east strain MACH1 was studied on PDA plates supplementedith different concentrations of FeCl3. The plates were floodedith 20 �L of a conidial suspension (3 × 105 spores mL−1)f the tested pathogens followed by yeast strain MACH1107 cells mL−1) streaked onto the centre of each plate. Theidths of the inhibition zones were measured after 3 (P. expan-

um), 4 (B. cinerea) and 10 days (A. alternata) of incubation at5 ◦C.

The sensitivity of the test pathogens to iron depletion

as tested with tropolone (2-hydroxycyclohepta-2,4,6-trienone;89702; Sigma–Aldrich Co.), a chelating agent with a strongffinity for ferric ions (Diouf et al., 2002). An aqueous solutionf the compound (50 �L; 0.5 mg tropolone), was placed into

Biolog

wwt

2d

dgcscItAtmwe

2

fl1(f3ap2

taAiipebas9fw

2

ywc

3

3p

e

TEs(

S

F

S

PAV

D. Saravanakumar et al. / Postharvest

ells (diameter, 5 mm) cut into PDA plates previously floodedith 20 �L of a conidial suspension (3 × 105 spores mL−1) of

he tested pathogens. The experiments were repeated three times.

.4. Microscopic observations of the pathogen myceliumevelopment

To examine the effect of the pigmented inhibition zone pro-uced by MACH1 on the conidial germination and hyphalrowth of the tested pathogens, microscopic observations werearried out. Conidial suspensions (3 × 105 spores mL−1) weremeared (20 �L) onto PDA plates with and without differentoncentrations of ferric chloride (5, 10, 15 and 20 �g mL−1).mmediately after, MACH1 cell suspension was streaked ontohe centre of each plate and the plates were incubated at 25 ◦C.fter 36 h, 10 microscopic fields were randomly selected at

he pigmented zone and they were counted for conidial ger-ination. After 48 h, hyphal germination in the coloured zoneas observed under light microscope (Nikon Eclipse 55i). The

xperiments were repeated twice.

.5. Study of iron competition in apples

The strain MACH1 was inoculated into 250 mL Erlenmeyeraks containing 100 mL of YPD broth amended with 0, 5, 10,5, 20 and 30 �g mL−1 FeCl3 and incubated on a rotary shaker100 rpm) at 25 ◦C for 48 h. ‘Golden Delicious’ apples, disin-ected in sodium hypochloride (NaOCl, 1.0% as chlorine) for

0 s and rinsed under tap water, when dry were punctured withsterile needle at the equatorial region (3 mm depth, 3 woundser fruit). The broth containing the antagonistic yeast (30 �L;× 108 cells mL−1) was pipetted into wounds. Untreated con-

pbsm

able 1fficacy of M. pulcherrima strain MACH1 against P. expansum, Alternaria sp., anduspension of the antagonist and storing in two controlled atmosphere (2% O2 and 3second survey)

torage and treatments Rotted apples (%

B. cinerea

irst survey2%O2/3%CO2, inoculated control 11.0 b2%O2/3%CO2, thiabendazole (3 × 10−2 g L−1)a 6.5 a2%O2/3%CO2, M. pulcherrima MACH1 8.4 a1%O2/2%CO2, inoculated control 11.0 b1%O2/2%CO2, thiabendazole (3 × 10−2 g L−1)a 9.0 a1%O2/2%CO2, M. pulcherrima MACH1 9.7 ab

econd survey2%O2/3%CO2, inoculated control 12.3 c2%O2/3%CO2, thiabendazole (3 × 10−2 g L−1)a 6.5 a2%O2/3%CO2, M. pulcherrima MACH1 9.7 b1%O2/2%CO2, inoculated control 11.0 bc1%O2/2%CO2, thiabendazole (3 × 10−2 g L−1)a 9.0 b1%O2/2%CO2, M. pulcherrima MACH1 9.7 b

ercent rotted apples = (no. of infected apples/total no. of apples) × 100.ll isolates were applied at 107 cells mL−1.alues followed by the same letter are not statistically different by Duncan’s multiplea Apples were treated with 150 mL hl−1 of Tecto 20S (thiabendazole: 19.7%).

y and Technology 49 (2008) 121–128 123

rols were inoculated with YPD broth and a chemical controlpplied at 30 �L per wound (thiabendazole, Tecto 20 S, Elftochem Agri Italy, 19.7% a.i., 200 �g a.i. mL−1) was included

n the experiment to provide comparative studies with the newlysolated yeast strain MACH1. After 24 h, 30 �L of conidial sus-ension (2 × 105 spores mL−1) of B. cinerea, A. alternata and P.xpansum were inoculated into wounds and apples were incu-ated at 25 ◦C. The diameter of the lesions, the weight of thepple pulp with rots and the percent infected wounds were mea-ured for each treatment. The measurements were taken afterdays for B. cinerea, 13 days for P. expansum and 21 days

or A. alternata. Eight apples were used for each treatment (24ounds) and the experiment was repeated twice.

.6. Statistical analysis

The program SPSS-WIN was used for the statistical anal-sis to process the data and the Duncan’s multiple range testas employed at P < 0.05 to separate the values in the semi-

ommercial and controlled condition experiments.

. Results

.1. Efficacy of M. pulcherrima strain MACH1 againstostharvest pathogens of apple at 1 ◦C

The semi-commercial trials were conducted to test thefficacy of the strain MACH1 against different postharvest

athogens of apple at 1 ◦C, to assess the performance of theiocontrol agent under storage conditions. The results (Table 1)howed that M. pulcherrima strain MACH1 after 5 and 8onths storage under 2%O2/3%CO2 significantly reduced the

B. cinerea, evaluated by dipping boxes of “Golden Delicious” apples in a cell% CO2; 1% O2 and 2% CO2) at 1 ◦C for 5 months (first survey) and 8 months

)

P. expansum A. alternata Total

8.4 b 0.0 19.4 b7.1 b 0.0 13.5 a3.2 a 0.0 11.6 a3.9 a 0.0 14.8 a

11.6 c 0.0 20.6 b3.2 a 0.0 12.9 a

29.0 b 0.0 a 41.3 c32.9 c 1.3 b 40.6 c15.5 a 0.0 a 25.2 a23.9 b 0.0 a 34.8 b29.7 b 1.3 b 40.0 c18.1 a 0.0 a 29.0 ab

range test (P < 0.05).

1 Biology and Technology 49 (2008) 121–128

p2w(t

(ad(mtcrt

aow

ra

3M

rcwmh(ms

tww5acCde

Fig. 1. (a) Pigmented inhibition zones produced by strain MACH1 in PDA withdifferent concentrations of iron against B. cinerea. M, M. pulcherrima strainMibd

3

TA

T

M5112

V

24 D. Saravanakumar et al. / Postharvest

ercentage of fruit attacked by postharvest pathogens (11.6%;5.2%) compared to the control (19.4%; 41.3%). The treatmentith thiabendazole was significantly different from the control

13.5%) after 5 months storage, but ineffective by the end ofhe trial (40.6%).

In the second trial with storage under 1%O2/2%CO2Table 1), carried out to assess the effectiveness of the biocontrolgent under different semi-commercial conditions, the inci-ence of rotten fruit between the treatment with strain MACH112.9%) and the control (14.8%) was not significant after 5onths storage. Also after 8 months storage, the biological

reatment (29.0%) and the control (34.8%) were not signifi-antly different. After both surveys, an increase in the number ofotten fruit compared to the control resulted after treatment withhiabendazole.

Analyzing pathogens separately, it was possible to observemajor incidence of B. cinerea rots after 5 months storage andf P. expansum rots after 8 months storage. Alternaria sp. rotsere absent or at a very low level.Comparing the two trials, the treatment that more effectively

educed the number of fruit with rots was the strain MACH1pplication followed by storage at 2%O2/3%CO2.

.2. Pigmented inhibition zone by M. pulcherrima strainACH1 against postharvest pathogen growth

M. pulcherrima strain MACH1 produced pale pink to darked colonies under different iron conditions. The intensity of theolour increased with increasing iron concentrations (Table 2)hile the strain produced white colonies without iron amend-ent in PDA. The antagonistic yeast produced wider pigmented

alos in PDA without (7.5 mm) or with 5 �g mL−1 of FeCl35.0 mm). An increase in FeCl3 concentration decreased the pig-ented halo width and the narrowest halo was observed with a

upplement of 20 �g mL−1 of FeCl3.The strain showed an inhibition zone of 5.1 mm when cul-

ured with B. cinerea, of 2.5 mm with A. alternata and of 1.2 mmith P. expansum in PDA without iron amendment. Similarly,hen MACH1 was streaked onto the medium amended with�g mL−1 FeCl3, it produced wider pigmented inhibition zonesgainst the tested pathogens when compared to increased con-

entrations (10, 15 and 20 �g mL−1) of iron salt (Fig. 1a).ompared to the effect on other pathogens, M. pulcherrimaid not show an appreciable inhibition zone in the case of P.xpansum, with or without iron amendment.

po

able 2pigmented inhibition zone produced by M. pulcherrima strain MACH1 in iron ame

reatment Colony colour Halo zone (mm)

ACH1 White 7.5 d�g mL−1 FeCl3 + MACH1 Light red 5.0 c0 �g mL−1 FeCl3 + MACH1 Medium red 2.7 b5 �g mL−1 FeCl3 + MACH1 Red 1.2 a0 �g mL−1 FeCl3 + MACH1 Dark red 0.5 a

alues followed by the same letter are not statistically different by Duncan’s multiple

ACH1; B, B. cinerea. (b) Tropolone activity against B. cinerea. T, Tropolonenhibition zone 5 mm; **sporulation by B. cinerea at the point of entry onto inhi-ition zone; Bc, B. cinerea; E, stimulated growth of B. cinerea in the tropoloneiffused zone.

.3. Tropolone activity

Tropolone was used to study the sensitivity of postharvestathogens to iron chelation. A lack of germination of conidiaf B. cinerea and A. alternata was observed in the zone near to

nded medium against postharvest pathogens in vitro

Inhibition zone produced by strain MACH1 (mm)

B. cinerea A. alternata P. expansum

5.1 c 2.5 c 1.2 b4.2 c 2.2 bc 1.0 b3.1 b 1.5 b 0.5 a1.0 a 0.7 a 0.0 a0.4 a 0.5 a 0.0 a

range test (P < 0.05).

D. Saravanakumar et al. / Postharvest Biology and Technology 49 (2008) 121–128 125

Table 3Conidial germination of postharvest pathogens in the pigmented inhibition zoneproduced by M. pulcherrima strain MACH1 in vitro

Treatment Conidial germination (%)

B. cinerea A. alternata P. expansum

MACH1 16 b 35 b 65 b5 �g mL−1 FeCl3 + MACH1 4 a 16 a 45 a10 �g mL−1 FeCl3 + MACH1 18 b 38 b –15 �g mL−1 FeCl3 + MACH1 24 c 52 c –20 �g mL−1 FeCl3 + MACH1 – – –

Values followed by the same letter are not statistically different by Duncan’smultiple range test (P < 0.05).IdT

w(ntst

3

cmairaei5toa

Fbc

wotgdroiMc

3B

oc

Ft

n some of concentrations of iron amendment, the inhibition zone was not pro-uced by yeast strain MACH1 against postharvest pathogens as indicated inable 1.

ells treated with tropolone aqueous solution in PDA mediumFig. 1b). Also, the mycelial growth of B. cinerea and A. alter-ata was restricted in the tropolone zone, whereas the fungi tendo grow freely in the medium outside the tropolone zone. Con-picuously, P. expansum did not show restricted germination inhe zone near the wells treated with tropolone aqueous solution.

.4. Conidial and mycelial inhibition

Microscopic observations in the inhibition zone by M. pul-herrima in the 5 �g mL−1 iron amendment showed interestingorphological deviations in the growth of B. cinerea and A.

lternata. A count on the conidial germination in the colourednhibition zone produced by strain MACH1 against B. cinereaevealed the lowest (4%) conidial germination in the mediummended with 5 �g mL−1 FeCl3. Similarly, A. alternata and P.xpansum recorded 16% and 45% conidial germination in thenhibition zone produced by M. pulcherrima strain MACH1 in

�g mL−1 FeCl3. When the medium had increased concentra-

ions of iron (10 and 20 �g mL−1 FeCl3), the highest percentf conidial germination was observed with B. cinerea and A.lternata (Table 3).

t(tF

ig. 3. Efficacy of yeast strain MACH1 with different iron concentrations on appleshe mean. The same letters over the bars represent treatments that are not significantl

ig. 2. B. cinerea mycelial disruption at the pigmented inhibition zone producedy MACH1 CIZ, coloured inhibition zone; MD, mycelial distruption; BC, B.inerea.

The conidia in the inhibition zone did not germinate and evenhen germinating, malformation of the germination tube wasbserved. B. cinerea growing at the tip of the coloured inhibi-ion zone exhibited mycelial degeneration (Fig. 2) and furtherrowth was arrested. Similar to B. cinerea, A. alternata mycelialegeneration at the tip of the coloured zone was observed. Withegard to P. expansum, conidial and mycelial inhibition was notbserved in contrast to that of the other two pathogens usedn this study. However, to some extend, M. pulcherrima strain

ACH1 in 5 �g mL−1 iron concentration caused a reduction inonidial germination of P. expansum (Table 3).

.5. Iron competition in artificial wounds of apple against. cinerea

Apples treated with the biocontrol agent amended withr without FeCl3 showed lower B. cinerea disease incidenceompared to the untreated control (Fig. 3). Further, apples

reated with M. pulcherrima strain MACH1 recorded a lower6.4%) percentage of attack by B. cinerea. Similarly, applesreated with the strain MACH1 supplemented with 5 �g mL−1

eCl3 recorded the lowest (3.57%) percent infection by B.

against B. cinerea infection in vivo. Vertical bars indicate the standard error ofy different from each other.

126 D. Saravanakumar et al. / Postharvest Biology and Technology 49 (2008) 121–128

F ppleso nifica

cmr35fh(

3A

(MItr

clpwi5

3P

(ai

Fo

ig. 4. Efficacy of yeast strain MACH1 with different iron concentrations on af the mean. The same letters over the bars represent treatments that are not sig

inerea compared with all other treatments. The measure-ent of diameter of the rot area and weight of the fruit with

ots showed the lowest levels for both parameters (11.5 mm,.0 g) in apples treated with the strain of M. pulcherrima and�g mL−1 FeCl3. This was comparable with that of chemical

ungicide. Apples inoculated with B. cinerea alone recorded theighest rotted area (71.3 mm) and rotted fruit weight (54.0 g)Fig. 3).

.6. Iron competition in artificial wounds of apple against. alternata

In the experiments against A. alternata in artificial woundsFig. 4), 11.3% infection was observed in M. pulcherrima strain

ACH1 treatment compared to 34.3% in the untreated control.nterestingly, M. pulcherrima strain MACH1 supplemented withhe lowest concentrations (5 and 10 �g mL−1 of FeCl3) of ironeduced the number of fruit with rots more than the highest

hFco

ig. 5. Efficacy of yeast strain MACH1 with different iron concentrations on applesf the mean. The same letters over the bars represent treatments that are not significa

against A. Alternata infection in vivo. Vertical bars indicate the standard errorntly different from each other.

oncentrations (20 and 30 �g mL−1 of FeCl3). Similarly, theesion diameter (16.0 mm) and the rotted fruit weight (5.9 g)rovided by M. pulcherrima strain MACH1 + 5 �g mL−1 ironere significantly lower than for other treatments with increased

ron amendments and the untreated control (lesion diameter:2.7 mm; rotted fruit weight: 27.5 g).

.7. Iron competition in artificial wounds of apple against

. expansum

The results of the in vivo experiments against P. expansumFig. 5) reproduced the same situation of the in vitro studiesnd M. pulcherrima strain MACH1 alone was not very effectiven reducing P. expansum infections in apple fruit. On the other

and, M. pulcherrima strain MACH1 amended with 5 �g mL−1

eCl3 significantly increased the efficacy of the strain MACH1ompared to the other concentrations of iron. Also the diameterf the rot area (45.5 mm) and the rotted fruit weight (23.9 g) were

against P. expansum infection in vivo. Vertical bars indicate the standard errorntly different from each other.

Biolog

lwct

4

iatcccaputtictueg

oie2r

aibutpPfifbf1

prbltrastm

eSs

vsbDdIavti

rgagddhMiIpm

aewetiaooor

tiomaaTcmfouf

D. Saravanakumar et al. / Postharvest

ower in the treatment supplemented with 5 �g mL−1 FeCl3,hile high concentrations of iron were less effective than the

ontrol. Among all the treatments, the untreated control showedhe highest infection rate by P. expansum.

. Discussion

In the present study, M. pulcherrima strain MACH1, recentlysolated from the carposphere of ‘Golden Delicious’ applesnd previously selected for its efficacy under controlled condi-ions, was partially effective in controlling postharvest diseasesaused by B. cinerea, A. alternata and P. expansum under semi-ommercial conditions. The experiments in semi-commercialonditions were carried out to test the efficacy of the biocontrolgent under cold storage and controlled atmospheres used byackinghouses. Actually, some biocontrol strains perform wellnder controlled conditions in the laboratory, where normallyhe fruit used have the same level of maturity, but may not dohe same under controlled atmosphere cold storage in the pack-nghouses, where it is not possible to control the physiologicalonditions of the fruit tested. Zheng et al. (2005) reported thathe efficacy of the biocontrol strains at room temperature andnder cold storage conditions was determined by the nutritionalnvironment at the wound site which may or may not favour therowth and colonization.

The only postharvest fungicide admitted in Italy and in manyther countries is benzimidazole thiabendazole, whose efficacys poor due to the development of resistant populations of P.xpansum and B. cinerea (Baraldi et al., 2003; Bertetti et al.,003). Also in our study, thiabendazole was not working in theeduction of both postharvest pathogens on apple.

After testing the efficacy, we tried to elucidate the mech-nism of action deployed by M. pulcherrima strain MACH1n the control of postharvest pathogens on apples. As reportedy Boekhout and Robert (2002), M. pulcherrima occurs nat-rally on fruit, buds and floral parts of apple trees, is ableo ferment glucose, assimilate various carbon compounds androduce the pigment pulcherrimin (Miller and Phaff, 1998).revious studies demonstrated the production of pulcherriminormed non-enzymatically from pulcherriminic acid and ferricons (Cook and Slater, 1956; MacDonald, 1965). Iron is essentialor the fungal growth and pathogenesis, and iron sequestrationy non-pathogenic microbes could be exploited in novel systemsor biological control of postharvest pathogens (Calvente et al.,999; Zhang et al., 2007).

The antagonistic strain MACH1 produced the red pigmentulcherrimin in presence of iron, indicating the uptake of fer-ic ions from the surrounding substrate. Higher inhibition halosy the antagonistic strain against B. cinerea and A. alternata inower iron amendments indicated the depletion of the micronu-rient by the yeast strain under low iron conditions. In contrast, aeduced halo formed by MACH1 against B. cinerea, A. alternata

nd P. expansum in increased concentrations of iron demon-trated the availability of a sufficient amount of iron closer tohe yeast cells. Iron depletion by the M. pulcherrima strain in the

edium inhibited the growth of B. cinerea, A. alternata and P.

s

bt

y and Technology 49 (2008) 121–128 127

xpansum. Our results are consistent with previous findings ofipiczki (2006) who demonstrated the iron competence betweentrains of M. pulcherrima and pathogenic fungi in vitro.

In order to assess the sensitivity of the pathogens to iron depri-ation, tropolone was used in the current study. Tropolone hastrong affinity for ferric ions and is able to inhibit their reductiony catecholates, lowering the redox potential of the iron couple.iouf et al. (2002) reported that tropolone inhibited wood degra-ation by Poria placenta by chelating the iron present in wood.n the current study, tropolone inhibited the growth of B. cinereand A. alternata, demonstrating their sensitivity to iron depri-ation. On the other hand, the overgrowing of P. expansum ontohe zone produced by tropolone indicated its low sensitivity toron chelation.

Similarly to the tropolone activity, the pigmented zone sur-ounding the streaks of the strain MACH1 blocked conidialermination and caused mycelial degeneration of B. cinereand A. alternata. As conidia require a large intake of iron forermination (Charlang et al., 1981; Calvente et al., 1999), ironepletion by the M. pulcherrima strain delays or reduces coni-ial germination. Moreover, the microscopic observation thatyphae crack when entering the pigmented zones around the. pulcherrima streaks demonstrates that iron starvation elic-

ts complex physiological changes in the fungal cells. Similarly,ppolito et al. (2005) reported that the ability of Aureobasidiumullulans to out-compete the pathogens for nutrients and spaceight weaken the pathogen cells.In the study of iron competition in artificial wounds on apple,

reduced disease incidence by B. cinerea, A. alternata and P.xpansum was observed in apples treated with the strain MACH1hen supplemented with lower concentrations of iron. The

xternal supplementation of high concentrations of iron reducedhe efficacy of the biocontrol yeast compared to non-iron and lowron concentrations. Our findings are similar to those of He etl. (2003), who supposed the involvement of iron in the modef action of antagonistic yeasts against postharvest pathogensf apple. The same authors reported that the biocontrol activityf the microbial antagonists was concentration dependent andeversed by the addition of exogenous nutrients.

Some precursors from M. pulcherrima move further fromhe yeast colony before they are immobilized by iron, result-ng in increased efficacy of biocontrol. At higher concentrationsf iron, the infection rate was higher because the precursorolecules did not diffuse far, because they bound sufficient

mount of iron closer to the yeast cells, and the pathogens hadn increased availability of free ferric irons for their growth.he pulcherrimic acid-ferric ion complex formed near the yeastells depletes the iron in the substrate and creates an environ-ent unsuitable for fungal microbes requiring the micronutrients

or their growth (Sanson et al., 2005; Sipiczki, 2006). More-ver, the increased efficacy by M. pulcherrima strain MACH1nder low iron conditions indicated the movement of precursorsurther from the yeast to find sufficient amounts of iron in the

ubstrate.

The higher reduction in the incidence of fruit with rots causedy B. cinerea, A. alternata and P. expansum, when apples werereated with MACH1 + 5 �g mL−1, indicated that small amounts

1 Biolog

od

iMt

accsp

A

ilIg

R

B

B

B

C

C

C

C

D

E

H

I

J

J

K

L

L

M

M

P

R

R

S

S

S

S

S

S

S

S

V

W

Z

Zhang, H., Zheng, X., Yu, T., 2007. Biological control of postharvest diseases

28 D. Saravanakumar et al. / Postharvest

f iron could elicit the production of pulcherrimic acid thatepleted iron necessary for the pathogen growth.

From the above results, this study clearly demonstrated thatron depletion from the substrate, used by M. pulcherrima strain

ACH1 for the production of pulcherrimin, ultimately inhibitedhe growth of postharvest pathogens.

The measurement of the iron content in the fruit could becomen interesting parameter to establish in advance the potentialontrol of postharvest pathogens on fruit, before their appli-ation, and could possibly be used to establish if a low ironupplementation is needed to guarantee an effective level ofathogen control during postharvest.

cknowledgements

The authors gratefully acknowledge Dr. Sandro Frati for help-ng in the experimental trials and Dr. Jeanne Griffin for theinguistic advice. Duraisamy Saravanakumar acknowledges thetalian Ministry for University and Research for the PostDocrant.

eferences

araldi, E., Mari, M., Chierici, E., Pondrelli, M., Bertolini, P., Pratella, G.,2003. Studies on thiabendazole resistance of Penicillium expansum of pears:pathogenic fitness and genetic characterization. Plant Pathol. 52, 362–370.

ertetti, D., Keji, S., Garibaldi, A., Gullino, M.L., 2003. Valutazione della sensi-bilita di alcuni agenti di marciume su pomacee in post-raccolta nei confrontidi diversi fungicidi. Inf. Tore Fitopat. 53, 57–60.

oekhout, T., Robert, V., 2002. Yeasts in Food—Beneficial and DetrimentalAspects. Behr’s Verlag, Hamburg.

alvente, V., Benuzzi, D., de Tosetti, M.I.S., 1999. Antagonistic action ofsiderophores from Rhodotorula glutinis upon the postharvest pathogen Peni-cillium expansum. Int. Biodeterior. Biodegrad. 43, 167–172.

alvo, J., Calvente, V., de Orellano, M.E., Benuzzi, D., Tosetti, M.I.S., 2007.Biological control of postharvest spoilage caused by Penicillium expansumand Botrytis cinerea in apple by using the bacterium Rahnella aquatilis. Int.J. Food Microbiol. 113, 251–257.

harlang, G., Bradford, N., Horowitz, N., Horowitz, R., 1981. Cellular and extra-cellular siderophores of Aspergillus nidulans and Penicillium chrysogenum.Mol. Cellular Biol. 1, 94–100.

ook, A.H., Slater, C.A., 1956. The structure of pulcherrimin. J. Chem. Soc.,4133–4135.

iouf, P.N., Delbarre, N., Perrin, D., Gerardin, P., Rapin, C., Jacquot, J.P., Gel-haye, E., 2002. Influence of tropolone on Poria placenta wood degradation.Appl. Environ. Microbiol. 68, 4377–4382.

ckert, J.W., Ogawa, J.M., 1985. The chemical control of postharvest diseases:subtropical and tropical fruits. Annu. Rev. Phytopathol. 23, 421–454.

e, D., Zheng, X., Yin, Y., Sun, P., Zhang, H., 2003. Yeast application for con-trolling apple postharvest diseases associated with Penicillium expansum.Bot. Bull. Acad. Sin. 44, 211–216.

ppolito, A., Schena, L., Pentimone, I., Nigro, F., 2005. Control of postharvestrots of sweet cherries by pre and postharvest applications of Aureobasid-ium pullulans in combination with calcium chloride or sodium bicarbonate.Postharvest Biol. Technol. 36, 245–252.

anisiewicz, W.J., Korsten, L., 2002. Biological control of postharvest diseasesof fruits. Annu. Rev. Phytopathol. 40, 411–441.

anisiewicz, W.J., Tworkoski, T.J., Kurtzman, C.P., 2001. Biocontrol potential ofMetchnikowia pulcherrima strains against blue mold of apple. Phytopathol-ogy 91, 1098–1108.

Z

y and Technology 49 (2008) 121–128

urtzman, C.P., Droby, S., 2001. Metschnikowia fructicola, a new ascosporicyeast with potential for biocontrol of postharvest fruit rots. Syst. Appl.Microbiol. 24, 395–399.

everentz, B., Conway, W.S., Janisiewicz, W., Abadias, M., Kurtzman, C.P.,Camp, M.J., 2006. Biocontrol of the food-borne pathogens Listeria mono-cytogenes and Salmonella enterica serovar Poona on fresh-cut apples withnaturally occurring bacterial and yeast antagonists. Appl. Environ. Micro-biol. 72, 1135–1140.

ima, G., Ippolito, A., Nigro, F., Salerno, M., 1997. Effectiveness of Aureoba-sidium pullulans and Candida oleophila against postharvest strawberry rots.Postharvest Biol. Technol. 10, 169–178.

acDonald, J.C., 1965. Biosynthesis of pulcherriminic acid. Biochem. J. 96,533–538.

iller, M.W., Phaff, H.J., 1998. Metschnikowia Kamienski. In: Kurtzman, C.P.,Fell, J.P. (Eds.), The Yeasts. Elsevier, Amsterdam, pp. 256–267.

iano, S., Neyrotti, V., Migheli, Q., Gullino, M.L., 1997. Character-ization of the biocontrol capability of Metschnikowia pulcherrimaagainst Botrytis postharvest rot of apple. Postharvest Biol. Technol. 11,131–140.

aaska, L., Mattila-Sandholm, T., 1995. Effects of iron level on the antagonis-tic action of siderophores from non-pathogenic Staphylococcus spp. J. Ind.Microbiol. Biotechnol. 15, 480–485.

agsdale, N.N., 2000. The Impact of the Food Quality Protection Act on theFuture of Plant Disease Management. Annu. Rev. Phytopathol. 38, 577–596.

anson, G., Rezza, I., Calvente, V., Benuzzi, D., de Tosetti, M.I.S., 2005. Controlof Botrytis cinerea strains resistant to iprodione in apple with rhodotorulicacid and yeasts. Postharvest Biol. Technol. 35, 245–251.

chena, L., Ippolito, A., Zahavi, T., Cohen, L., Droby, S., 2000. Molecularapproaches to assist the screening and monitoring of postharvest biologicalyeasts. Eur. J. Plant Pathol. 106, 681–691.

ipiczki, M., 2006. Metschnikowia strains isolated from botrytized grapesantagonize fungal and bacterial growth by iron depletion. Appl. Environ.Microbiol. 72, 6716–6724.

milanick, J.L., 1994. Strategies for the isolation and testing of biocontrol agents.In: Wilson, C.L., Wisniewski, M.E. (Eds.), Biological Control of PostharvestDiseases—Theory and Practice. CRC Press, Boca Raton, pp. 25–42.

nowdon, A.L., 2003. A Colour Atlas of Postharvest Diseases and Disorders ofFruits and Vegetables. Vol. 1: Fruits, Manson Publishing, London.

padaro, D., Gullino, M.L., 2004. State of art and future perspectives of bio-logical control of postharvest fruit diseases. Int. J. Food Microbiol. 91,185–194.

padaro, D., Vola, R., Piano, S., Gullino, M.L., 2002. Mechanisms of action,efficacy and possibility of integration with chemicals of four isolates of theyeast Metschnikowia pulcherrima active against postharvest pathogens onapples. Postharvest Biol. Technol. 24, 123–134.

potts, R.A., Cervantes, L.A., 1986. Population, pathogenicity, and benomylresistance of Botrytis spp., Penicillium spp., and Mucor piriformis in pack-inghouses. Plant Dis. 70, 106–108.

ivekananthan, R., Ravi, M., Saravanakumar, D., Kumar, N., Prakasam,V., Samiyappan, R., 2004. Microbially induced defense related proteinsagainst post harvest anthracnose infection in mango. Crop Prot. 23, 1061–1067.

ilson, C.L., Wisniewski, M.E. (Eds.), 1994. Biological Control of PostharvestDiseases. Theory and Practice. CRC Press, Boca Raton.

ahavi, T., Cohen, L., Weiss, B., Schena, L., Daus, A., Kaplunov, Tl, Zutkhi, J.,Ben-Arie, R., Droby, S., 2000. Biological control of Botrytis, Aspergillus andRhizopus rots on table and wine grapes in Israel. Postharvest Biol. Technol.20, 115–124.

of peach with Cryptococcus laurentii. Food control 18, 287–291.heng, X.D., Zhanga, H.Y., Sun, P., 2005. Biological control of postharvest green

mold decay of oranges by Rhodotorula glutinins. Eur. Food Res. Technol.220, 353–357.