Evaluation of Penicillium expansum isolates for aggressiveness, growth and patulin accumulation in usual and less common fruit hosts

International Journal of Food Microbiology 143 (2010) 109–117

Contents lists available at ScienceDirect

International Journal of Food Microbiology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j f o o d m i c r o

Evaluation of Penicillium expansum isolates for aggressiveness, growth and patulin accumulation in usual and less common fruit hosts Fiorella Neri ⁎, Irene Donati, Francesca Veronesi, David Mazzoni, Marta Mari Criof, Department of Protection and Improvement of Agricultural Food Products, Alma Mater Studiorum, University of Bologna, via Gandolfi, 19 Cadriano Bologna, Italy

a r t i c l e

i n f o

Article history: Received 10 March 2010 Received in revised form 27 July 2010 Accepted 3 August 2010 Keywords: Apple Apricot Mycelial growth Aggressiveness in fruit hosts Patulin Penicillium expansum Peach

a b s t r a c t Experiments were carried out in vivo and in vitro with four isolates of Penicillium expansum (I 1, E 11, C 28 and I 12) to evaluate their aggressiveness, growth and patulin accumulation in both usual (pears and apples) and less common hosts (apricots, peaches, strawberries and kiwifruits) of the pathogen. The 75% of isolates showed the ability to cause blue mould in all tested hosts. In particular, C 28 and I 1 were the most and the least aggressive isolates, respectively (52.9 and 10.6% infection and 20.7 and 15.4 mm lesion diameters). ‘Candonga’ strawberries and ‘Pinkcot’ apricots showed the largest lesion diameters (29.8 and 25.3 mm), followed by ‘Conference’ pears, ‘Spring Crest’ peaches and ‘Abate Fetel’ pears. With the exception of ‘Candonga’ strawberries, the formation of colonies and mycelial growth of P. expansum isolates on fruit puree agar media (PAMs) was stimulated in comparison to a standard growth medium (malt extract agar, MEA). Two of the most aggressive isolates in our assays (I 12 and C 28) showed the greatest accumulation of patulin both in vitro and in vivo, while the least aggressive isolate (I 1) produced patulin only in a few growth media and cvs. Patulin concentration on fruit PAMs was higher than patulin detected in infected fruit tissues. Apple PAMs were the more favorable substrates for patulin accumulation in vitro (maximum concentration 173.1 and 74.1 μg/mL in ‘Pink Lady and ‘Golden Delicious’ PAMs, respectively) and ‘Pink Lady’ apples inoculated with the isolate E 11 showed the greatest accumulation of patulin in the whole in vivo assay (33.9 μg/mL). However, infected tissue of cv Golden Delicious showed lower average accumulation of patulin (1.7 μg/mL) than that of cv Pink Lady (19.1 μg/mL), and no significant differences in patulin concentrations were found among ‘Golden Delicious’ apples and tested cvs of pears, kiwifruits and strawberries. Peaches were highly susceptible to patulin accumulation, showing average concentrations of 27.4 and 18.6 μg/mL in vitro and in vivo, respectively. Apricots were also consistently positive for patulin accumulation, both in vitro (average values of 20.1 μg/mL) and in vivo (average values of 9.4 μg/mL). Our study showed the potential of some less common hosts of P. expansum (in particular peaches and apricots) to support patulin production, indicating that a steady monitoring of patulin contamination should be carried out in fruit substrates other than apples and pears. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Penicillium expansum Link (agent of blue mould disease) is one of the main causes of spoilage of pears and apples after harvest and is frequently isolated from a wide range of other fruit, including stone fruit, soft fruit and berry fruit (Snowdon, 1990; Sommer et al., 1974). The pathogen penetrates typically through wounds or injuries produced during harvest and handling. Infection may also occur through stem end, open calyx tube and lenticels in pome fruits or it may gain entry through infection sites of other primary fruit pathogens. Over-mature or long-stored fruit are more susceptible to P. expansum infection. Blue mould develops even at low temperatures ⁎ Corresponding author. Criof, Department of Protection and Improvement of Agricultural Food Products, University of Bologna, via Gandolfi, 19, 40057 Cadriano di Granarolo Emilia, Bologna, Italy. Tel.: + 39 051 766563; fax: + 39 051 765049. E-mail address: fi[email protected] (F. Neri). 0168-1605/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.ijfoodmicro.2010.08.002

used for fruit storage (−1°–0 °C), although its development is favored by warm environment at retail and consumer sites (Mari et al., 2009). Besides the economic impact caused by fruit infection, current interest for P. expansum is the health hazard caused by ability of the pathogen to produce patulin (Moake et al., 2005), a mutagenic and embryo toxic substance produced by most isolates of the pathogen (Sommer et al., 1974; Andersen et al., 2004; Morales et al., 2008a). For this mycotoxin the limits of 50, 25 and 10 μg/kg have been set in Europe for fruit juices and fruit nectar, solid apple products and apple based products for infants and young children, respectively (Anon., 2003), and 50 μg/L is the norm for patulin regulation of apple juice and cider in many countries (Anon., 2004). Most studies on patulin are focused on apples and their products (Sant'Ana et al., 2008). This is justified by the following reasons: apples are the most susceptible fruit to P. expansum infection in many producer countries, contamination by patulin frequently occurs in apple industry and limits for patulin has been established for apple-

110

F. Neri et al. / International Journal of Food Microbiology 143 (2010) 109–117

based products. The presence of patulin in fruit other than apples has been less investigated (Laidou et al., 2001; Anon., 2002; Morales et al., 2008b; Spadaro et al., 2008), although P. expansum has a broad host range and early studies on patulin reported the occurrence of this mycotoxin in a variety of fruit (Buchanan et al., 1974; Frank et al., 1977; Scott et al., 1977). In the study by Buchanan et al. (1974), only plums were, for example, a poor substrate for patulin accumulation, while pears, peaches, apricots and cherries infected by P. expansum accumulated concentrations of patulin similar to those reported for apples. The role of patulin in plant pathogenesis is still unknown. In general, secondary metabolites of fungi are not required for their growth under normal conditions, but can presumably confer some selective advantage in certain situations (Bhatnagar et al., 2002). Whether mycotoxins act in pathogenesis as factors of pathogenicity or virulence is a controversial issue. Different results are found in the literature depending on fungal species and type of toxin (Xu and Berrie, 2005; Hof, 2007), and contrasting results have also been reported with regard to the relationship between levels of patulin produced and aggressiveness of P. expansum isolates (Sommer et al., 1974; McCallum et al., 2002; Baert et al., 2007). The aim of this study was to evaluate the aggressiveness of some isolates of Penicillium expansum to different fruit hosts and to investigate the influence exhibited by hosts on the isolates' growth and patulin accumulation in in vivo and in vitro assays. 2. Materials and methods 2.1. Pathogen Four isolates of P. expansum (I 1, E 11, C 28 and I 12) obtained from blue mould decayed pears and belonging to the CRIOF collection were used. A monoconidial culture of each isolate was grown at 20 °C on malt extract agar (MEA; Oxoid, UK) until use. Conidial suspensions of each P. expansum isolate were prepared by washing the pathogen colonies with sterile distilled water containing 0.05% v/v Tween 80. Spore concentrations were adjusted to 103 conidia/mL, by means of a haemocytometer. In the trials of mycelial growth, P. expansum was grown for 2 days on Czapek-dox agar (Oxoid, UK).

of each fruit, using a Chatillon digital penetrometer fitted with an 8 mm probe (11 mm for apples). Soluble solids content (%) was determined using a digital refractometer (Atago Co., Tokyo, Japan) in a portion of filtrate obtained by blending each fruit. The values of pH and titratable acidity (mequiv. 10/mL of pure juice) were determined using an automatic titrator (Crison Instruments, Modena, Italy) by titrating fruit juice (obtained by diluting homogenized flesh with distilled water in a ratio 1:5 and filtering the solution in a vacuum) with 0.1 N NaOH to pH 8.10. Fruit were stored at 0 °C until testing (a maximum of 3 days for stone fruits and strawberries and 1 month for pome fruits and kiwifruits). 2.3. Aggressiveness of P. expansum isolates Batches of fruit were wounded with a sterile nail (one wound per fruit in the equatorial zone; 2 × 2 × 2 mm) and dipped for 1 min in a conidial suspension (103 conidia/mL) of each P. expansum isolate. Three replicates of 20 fruit were used for each isolate and cv. The percentage of infected wounds and the diameter of the lesions (mm) were recorded after 7 days of incubation at 20 °C. Fruit with no infection were not counted for lesion size measurements. 2.4. Measurements of pH in fruit wounded site and in Czapek-dox liquid medium The pH of mesocarp of each fruit used for aggressiveness evaluation was measured by placing the pH electrode InLab 427 (Mettler Toledo) connected to a SG2-SevenGo pH meter (Mettler Toledo) at approximately 15 mm depth through the wound site. The pH of healthy tissue was measured in non-inoculated fruit (controls) kept for 7 days at 20 °C. Three replicates of 20 fruit were used for each isolate, cv and treatment. Conidial suspensions of each isolate were inoculated in 3 tubes containing 10 mL of Czapek-dox liquid medium (CLM, Oxoid, UK) to achieve the final concentration of 103 conidia/mL and incubated at 20 °C for 14 days. The pH of the medium was measured daily starting after 3 days of incubation using the instrument described above. 2.5. Effect of fruit based media on P. expansum isolates development

2.2. Fruit Both typical and less common fruit hosts of P. expansum were tested. Pears (cv Conference and Abate Fetel), apples (cv Golden Delicious and Pink Lady), apricots (cv Pinkcot), peaches (cv Spring Crest), strawberries (cv Candonga) and kiwifruits (cv Hayward), were purchased from the Emilia-Romagna Apofruit packinghouse. Only undamaged and disease-free fruits were used in the experiments. Physical–chemical characteristics were analyzed on 20 fruit of each cv before inoculation (Table 1). Firmness (Newtons), not determined in strawberries, was measured after removing the skin, on opposite sides Table 1 Physical–chemical characteristics of fruit before inoculation (means ± standard deviations). Fruit

Firmness (N)

SSC (%)

pH

‘Abate Fetel’ pear ‘Conference’ pear ‘Golden Delicious’ apple ‘Pink Lady’ apple ‘Pinkcot’ apricot ‘Spring Crest’ peach ‘Candonga’ strawberry ‘Hayward’ kiwifruit

49.71 ± 6.03 67.57 ± 7.93 64.97 ± 6.83

14.97 ± 0.06 13.50 ± 0.06 16.83 ± 0.06

4.40 ± 0.02 4.55 ± 0.02 3.36 ± 0.08

2.90 ± 0.15 2.00 ± 0.06 7.94 ± 0.00

76.43 ± 5.99 10.23 ± 6.36 47.56 ± 7.41 n.d.

14.43 ± 0.06 10.77 ± 0.06 10.53 ± 0.09 9.63 ± 0.25

3.45 ± 0.09 2.76 ± 0.02 2.96 ± 0.01 3.43 ± 0.05

7.61 ± 0.26 33.92 ± 0.00 15.79 ± 0.01 12.31 ± 0.05

6.5 ± 0.00

3.36 ± 0.02

19.48 ± 0.05

n.d. = not determined.

70.78 ± 13.13

Total acidity (meq/100 mL)

To assess only the effect of constitutive characteristics of the host, conidial suspensions and mycelial disks of each P. expansum isolate were cultured in vitro on growth media derived from boiled fruit purees. Within a few days of harvesting, fruit samples of the same batches used for in vivo experiments (with the peel but without the core or the stone) were mixed until a fine puree was obtained, and stored at −24 °C until needed. Puree agar media (PAMs) from each cv were obtained by adding aliquots of 300 mL sterile agar solution (9 g agar technical, Oxoid, in distilled water) to 600 g of fruit purees previously boiled for 45 min into 1 L bottle (Baert et al., 2007). Colony-forming units (CFU) and mycelial growth of P. expansum isolates cultured on fruit PAMs were compared with growth on MEA (control). Six replicate dish cultures were used for each isolate, treatment and assay. Aliquots of 100 μL of conidial suspension (103 conidia/mL) of each P. expansum isolate were spread onto Petri dishes containing 20 mL of each fruit PAM and onto dishes of MEA. The CFU were counted 3 days after incubation at 20 °C. To test rates of mycelial growth, a mycelial disc (6 mm diameter) was taken from the periphery of an actively growing agar culture and placed at the centre of a Petri dish containing MEA or PAM. After 7 days of incubation at 20 °C, the diameter of the colonies was recorded. 2.6. Determination of patulin in vitro and in vivo In vitro and in vivo experiments were carried out to determine the effect of different growth media and fruit hosts on patulin accumulation.

F. Neri et al. / International Journal of Food Microbiology 143 (2010) 109–117

Concentration of patulin accumulated by each P. expansum isolate was determined in agar and liquid media and in infected tissues of fruit, using the method of extraction and analysis in HPLC-UV described by Andersen et al. (2004). Mycelial plugs of each P. expansum isolates were cultured on MEA and fruit PAMs as described above, and conidial suspensions of each isolate were inoculated in tubes containing CLM or modified Pratt's medium broth (Gregori et al., 2008) to a final concentration of 103 conidia/mL. After 7 days of incubation at 20 °C, five 6 mm agar plugs of mycelia cut in different sites of the colonies and 2 mL of liquid cultures were separately used for patulin extraction. Extraction was ultrasonically performed by 2 mL of ethyl acetate/formic acid (200:1, v/v) for 60 min. Each extract was evaporated to dryness in a rotary vacuum concentrator. The dried residue was redissolved ultrasonically in 1 mL of methanol for 30 min and filtered through a 0.45 μm filter before HPLC analyses. Three repetitions for each sample were used. Determination of patulin in vivo was carried out on infected parts of fruits used for the evaluation of aggressiveness. After 7 days of incubation at 20 °C, visibly infected tissues of each cv were removed from the surrounding sound area of fruit and blended with a mixer. A sample of 2 g was extracted as described above. After the addition of 6 mL ethyl acetate/formic acid (100:1, v/v), samples were evaporated to dryness, redissolved in 3.5 mL of methanol by sonication for 30 min and filtered through a 0.45 μm filter. Analysis of patulin was performed on an HP-1100 high-performance liquid chromatograph (Agilent, Waldbronn, Germany) equipped with a diode array detector (DAD) (Agilent). Separations were done on a 125 × 2 mm i.d., 3 μm Hypersil BDS-C18 cartridge column with a 10 × 2 mm i.d., Superspher 100 RP-18 guard column (Agilent). A linear gradient started at 85% water and 15% acetonitrile, changing to 100% acetonitrile in 40 min and maintaining these conditions for 5 min. The solvent composition returned to starting conditions in 8 min followed by 5 min of equilibration. A flow rate of 0.3 mL/min was used. The temperature of the column was 40 °C, and the injection volume was 5.0 μL. The eluate was monitored at 276 nm wavelength. To construct the calibration curve, standard solutions of patulin (Sigma-Aldrich) with concentrations of 0.025, 0.05, 0.1, 0.2, 0.4 and 0.5 μg/mL were injected in triplicate. Standard curve obtained by linear regression of concentrations against peak areas had a correlation coefficient r2 N 0.99. Recovery was determined on a blank apple puree spiked at 0.05, 0.1, 0.2 and 0.5 μg/mL of patulin in triplicate. The mean recovery values were 99.5, 96.8, 98.0 and 98.9%, respectively. The relative standard deviations were 2.26, 11.8, 2.84 and 0.8%, respectively. All in vivo and in vivo experiments were performed twice. 2.7. Statistical analysis The data were subjected to ANOVA. Means were separated using LSD test at P ≤ 0.05. Linear regressions between physical–chemical characteristics of fruits and aggressiveness of P. expansum isolates and between physical–chemical characteristics of fruits and patulin

111

accumulation were calculated The data were processed using the statistical package Statistica for Windows (Statsoft Inc.). 3. Results 3.1. Aggressiveness of P. expansum isolates The ability to cause blue mould rot was influenced by fruit hosts (Table 2). In general, I 12 and C 28 caused the highest incidence of infection (from 23.3 to 85% and 18.3 to 95%, respectively), E 11 showed intermediate aggressiveness (from 10 to 73%), while I 1 caused the lowest incidence of decay (ranging from 0 to 55%). However, no significant differences in incidence of infection were found among the P. expansum isolates in ‘Hayward’ kiwifruits, or among E 11, C 28 and I 12 isolates in ‘Pink Lady’ apples, ‘Conference’ pears, ‘Pinkcot apricots’, ‘Springcrest’ peaches and ‘Abate Fetel’ pears. Among the tested hosts, ‘Candonga’ strawberries showed the highest incidence of infection (from 55.0 to 95%), followed by ‘Abate Fetel’ pears (from 11.7 to 83.2%) and ‘Pinkcot' apricots (from 10 to 65.0%). However, no significant differences in incidence of infection were found among ‘Pinkcot’ apricots and ‘Abate Fetel’ pears inoculated with I 1, or among ‘Pinkcot’ apricots, ‘Abate Fetel’ pears and ‘Candonga’ strawberries inoculated with E 11 and among ‘Abate Fetel’ pears and ‘Candonga’ strawberries inoculated with I 12. The cvs Pink Lady, Conference and Golden Delicious showed intermediate susceptibility to P. expansum infection (average values ranging from 26.3 to 34.6%). The lowest incidence of infection was generally observed in ‘Hayward’ kiwifruits (from 10.0 to 33.3%) or ‘Springcrest’ peaches (from 1.7 to 31.7%), although no decay or very low decay (1.7%) developed in cvs Conference, Golden Delicious and Pink Lady inoculated with I 1. In general, inoculation with C 28 caused the largest lesion diameters (from 13.2 to 37.7 mm, depending on the host), while I 1 induced, where virulent, the lowest lesion diameters (from 11.8 to 20.7 mm) (Table 3). However, no significant differences in lesion diameters were found among E 11, I 12 and C 28 for inoculation of cvs Golden Delicious, Pink Lady, Hayward and Conference, while I 12 caused the largest lesion diameters in cv Pinkcot. Among the tested hosts, the largest lesion diameters were observed in ‘Candonga’ strawberries, with values ranging from 20.7 and 37.7 mm, depending on the P. expansum isolate. However, no significant differences in lesion diameter were observed between ‘Pinkcot’ apricots and ‘Candonga’ strawberries inoculated with E 11 or I 12. The smallest lesion diameters were observed in ‘Pink Lady’ and ‘Golden Delicious’ apples, with values ranging from 11.6 to 13.6 mm in three out of the four tested. No significant differences were found in lesion diameters among cv Pink Lady, Golden Delicious, Hayward and Springcrest inoculated with E 11, or among cvs Pink Lady, Golden Delicious and Hayward inoculated with I 12. ‘Abate Fetel’ pears, ‘Springcrest’ peaches and ‘Conference’ pears showed intermediate lesion diameters (16.2, 16.9 and 17.9 mm on average, respectively).

Table 2 Infected wounds (%) in fruits inoculated with isolates (I 1, E 11, C 28 or I 12) of Penicillium expansum. Fruit

I1A

E 11

C 28

I 12

Averagea

‘Abate Fetel’ pear ‘Conference’ pear ‘Golden Delicious’ apple ‘Pink Lady’ apple ‘Pinkcot’ apricot ‘Spring Crest’ peach ‘Candonga’ strawberry ‘Hayward’ kiwifruit Averageb

11.7 ± 7.6 bc A 0.0 ± 0.0 a A 0.0 ± 0.0 a A 1.7 ± 2.9 ab A 10.0 ± 5.0 abc A 1.7 ± 2.9 ab A 55.0 ± 13.2 d A 13.3 ± 10.4 c A 11.7

73.0 ± 14.7 c B 35.7 ± 16.1 b B 36.7 ± 5.8 b B 31.7 ± 7.6 b B 65.0 ± 8.7 c B 26.7 ± 2.9 ab B 66.7 ± 7.6 c B 10.0 ± 13.2 a A 43.2

71.7 ± 12.6 d B 46.4 ± 10.0 bc B 51.7 ± 8.0 bc C 33.3 ± 12.6 ab B 63.3 ± 16.1 cd B 31.7 ± 10.4 ab B 95.0 ± 8.7 e C 18.3 ± 17.6 a A 51.4

83.2 ± 8.5 e B 55.0 ± 10.0 cd B 50.0 ± 5.0 cd C 38.3 ± 11.5 ab B 61.1 ± 5.3 d B 23.3 ± 12.5 a B 85.0 ± 5.0 e C 33.3 ± 20.8 ab A 53.7

59.9 34.3 34.6 26.3 49.9 20.9 75.4 18.7

Fruit were wounded and inoculated with conidial suspensions (103 conidia/mL) of Penicillium expansum. Incidence of infected wounds (%) was assessed after 7 days of incubation at 20 °C. Data represent the mean of three replicates ± standard deviation. Within each column (lower case) and row (upper case), different letters indicate significant differences according to LSD test at P ≤ 0.05. a Mean of row data. b Mean of column data.

112

F. Neri et al. / International Journal of Food Microbiology 143 (2010) 109–117

Table 3 Lesion diameters (mm) in infected fruit inoculated with isolates (I 1, E 11, C 28 or I 12) of Penicillium expansum. Averagea

Fruit

I1A

E 11

C 28

I 12

‘Abate Fetel’ pear ‘Conference’ pear ‘Golden Delicious’ apple ‘Pink Lady’ apple ‘Pinkcot’ apricot ‘Spring Crest’ peach ‘Candonga’ strawberry ‘Hayward’ kiwifruit Averageb

14.2 ± 5.6 a A – – – 14.7 ± 10.2 ab A – 20.7 ± 8.9 b A 11.8 ± 4.8 a A 15.4

15.5 ± 4.6 bc A 17.1 ± 5.3 c A 11.6 ± 4.6 a A 12.8 ± 4.4 ab A 25.6 ± 9.7 d B 13.3 ± 8.7 ab A 27.7 ± 6.6 d B 13.1 ± 7.3 ab A 17.1

19.5 ± 4.9 cd B 18.3 ± 6.6 bc A 13.6 ± 5.2 a A 13.2 ± 4.7 a A 29.4 ± 6.5 e BC 21.5 ± 8.0 d B 37.7 ± 6.4 f D 16.6 ± 4.3 abc A 21.2

15.5 ± 3.8 18.3 ± 7.0 13.3 ± 4.8 12.5 ± 5.8 31.4 ± 7.4 16.0 ± 7.3 33.2 ± 5.9 12.3 ± 6.7 19.1

bA cA ab A aA dC bc AB dC aA

16.2 17.9 12.8 12.8 25.3 16.9 29.8 13.5

Fruit were wounded and inoculated with conidial suspensions (103 conidia/mL) of Penicillium expansum. Lesion diameters were assessed after 7 days of incubation at 20 °C. Data represent the mean of three replicates ± standard deviation. Only infected fruit were counted for the measure. Within each column (lower case) and row (upper case), different letters indicate significant differences according to LSD test at P ≤ 0.05. –isolate did not produce sufficient decay. a Mean of row data. b Mean of column data.

No significant correlations were found between firmness, SSC, pH or acidity of fruit measured before inoculation and incidence of infected wounds after incubation, nor between SSC, pH or acidity and lesion diameters (data not reported). An inverse correlation was found instead between firmness and lesion diameters in fruit infected with C 18, I 12 or E 11 P. expansum isolate (coefficient of correlations of −0.95, −0.90 and −0.85 respectively, p ≤ 0.01). 3.2. pH in fruit wounded sites and in Czapek-dox liquid medium In ‘Candonga’ strawberries and ‘Hayward’ kiwifruits, the pH of inoculated tissues was lower than the pH of non-infected controls (delta of pH ranging from 0.31 to 0.45 units and 0.67 to 1.05 units, respectively), regardless of the P. expansum isolate used for inoculation (Table 4). A decrease of pH in comparison to controls was also observed in tissues of ‘Pinkcot’ apricots (about 0.75 unit of pH), ‘Conference’ and ‘Abate Fetel’ pears (about 0.4 unit of pH) inoculated with E 11, C 28 or ISCI I2 isolate. In these hosts wound tissues of non-infected fruits had pH≥ 4.0. Very little changes or no significant difference in pH in comparison with controls were found in P. expansum inoculated tissues of ‘Golden Delicious’ and ‘Pink Lady’ apples (pH in control fruit of 3.83 and 3.82). A slight increase of pH to values of about 3.7 was found in inoculated tissues of ‘Springcrest’ peaches (pH in control fruit 3.51). The pH of CLM inoculated with each P. expansum isolate decreased during the period of incubation at 20 °C, from the initial value of 6.5 to values ranging from 3.7 and 5.4, depending on the isolate, after 14 days of incubation (Fig. 1). The decrease of pH began after 3 days of incubation for cultures of C 28 or E 11, while after 4 days for culture of other isolates. I 1 retained the highest pH, reaching values of about 5 after 10 days of incubation, followed by a little increase of pH after further days of incubation. C 28 and E 11 isolates showed the greatest decrease of pH during the first 7 days of incubation. After 14 days of incubation, no significant differences in pH were found among cultures of I 12, E 11 and C 28.

3.3. Colony-forming units After 3 days of incubation, the conidial suspensions of P. expansum isolates inoculated on MEA formed 50.2 colonies on average, without significant differences among the isolates (Table 5). The inoculation on most fruit PAMs led to more colonies than those on MEA. The greatest formation of colonies was observed on ‘Golden Delicious’ apple PAM (97.2 colonies on average), showing 48.1% average stimulation in comparison with MEA. On the contrary, conidial suspensions of P. expansum inoculated on strawberry PAM formed colonies only after 4 days of incubation. Moreover, no significant increase or low increase in the number of CFU were found for growth on strawberry PAM in comparison with MEA. The colonies grown on kiwifruit PAM showed the largest size, while the colonies grown on strawberry PAM were much smaller than those on MEA (data not reported). 3.4. Mycelial growth In most growth media, isolates I 1 and C 28 showed the weakest and the highest mycelial growth respectively (average values of 38.9 and 52.7 mm) (Table 6). E 11, on the other hand, showed the fastest growth in strawberry PAM, while no significant differences were found between E 11 and C 28 for mycelial growth on MEA, nor among I 12, E 11 and C 28 for growth on ‘Conference’ and ‘Hayward’ PAMs. After 7 days of incubation, mycelial growth on MEA ranged from 33.1 to 38.8 mm, depending on the isolate. Mycelial growth on fruit PAMs was faster than that on MEA (from 14 to 40% average stimulation, depending on the cv), with the exception of strawberry PAM where mycelial growth rate was always reduced (38% average reduction). Major mycelial growth was observed on ‘Hayward’ PAMs (from 50.3 to 68.6 mm), although no significant differences in mycelial growth were found between kiwifruit and peach PAMs in C 28, or among ‘Hayward’ and ‘Conference’ PAMs in E 11. Cultures on pear and peach PAMs showed faster mycelial growth (from 40.4 to 65.4 mm) than culture on

Table 4 pH in wounded tissues of healthy and Penicillium expansum isolates (I 1, E 11, C 28 or I 12) inoculated fruit. Fruit

Control

‘Abate Fetel’ pear ‘Conference’ pear ‘Golden Delicious’ apple ‘Pink Lady’ apple ‘Pinkcot’ apricot ‘Spring Crest’ peach ‘Candonga’ strawberry ‘Hayward’ kiwifruit

4.52 ± 0.21 4.57 ± 0.24 3.83 ± 0.17 3.82 ± 0.17 4.65 ± 0.59 3.51 ± 0.26 4.04 ± 0.39 5.02 ± 0.29

I1A C B A B B A C C

4.53 ± 0.33 4.62 ± 0.35 3.89 ± 0.22 3.80 ± 0.17 4.67 ± 0.53 3.75 ± 0.32 3.73 ± 0.40 4.30 ± 0.41

E 11 C B AB B B B B B

4.22 ± 0.30 4.26 ± 0.39 3.95 ± 0.23 3.70 ± 0.20 3.88 ± 0.79 3.73 ± 0.26 3.64 ± 0.27 4.35 ± 0.43

C 28 B A B A A B AB B

4.10 ± 0.31 4.30 ± 0.40 3.87 ± 0.20 3.72 ± 0.21 3.96 ± 0.77 3.66 ± 0.27 3.59 ± 0.27 4.23 ± 0.40

I 12 A A AB A A B A B

4.01 ± 0.29 4.13 ± 0.47 3.93 ± 0.36 3.81 ± 0.28 3.85 ± 0.78 3.75 ± 0.17 3.68 ± 0.25 4.22 ± 0.38

A A B B A B AB B

Fruit were wounded and inoculated with conidial suspensions (103 conidia/mL) of Penicillium expansum and incubated 7 days at 20 °C. Control fruit were only wounded and kept 7 days at 20 °C. Data represent the mean of three replicates ± standard deviation. Within each cv, different letters indicate significant differences according to LSD test at P ≤ 0.05.

F. Neri et al. / International Journal of Food Microbiology 143 (2010) 109–117

Fig. 1. pH of culture of Penicillium expansum isolates (I 1, E 11, C 28 or I 12) in Czapekdox liquid medium.

‘Pink Lady’ apple PAM (from 38.0 to 46.8 mm). Cultures on ‘Pinkcot’ and ‘Golden Delicious’ PAMs showed intermediate mycelial growth (average values of 49.8 and 50.9 mm, respectively). 3.5. Patulin accumulation in infected fruit tissues P. expansum isolates showed different capacity to accumulate patulin (Table 7). The isolate I 1 accumulated patulin only in a few hosts (‘Abate Fetel’ pears, ‘Pinkcot’ apricots and ‘Spring Crest’ peaches), E 11 accumulated patulin in most hosts (no accumulation in cvs Abate Fetel and Golden Delicious), while I 12 and C 28 accumulated patulin in all tested hosts. In general, I 1 accumulated also the lowest concentrations of patulin (0.7 μg/mL on average), while I 12 and or C 28 accumulated the greatest concentrations (10.5 and 9.1 μg/mL on average, respectively). The greatest accumulation of patulin was observed only in ‘Pink Lady’ apples in tissues inoculated with E 11 (33.9 μg/mL). Among the tested hosts, the lowest concentrations of patulin were detected in infected tissues of ‘Candonga’ strawberries, ‘Abate Fetel’ pears, ‘Golden Delicious’ apples, ‘Hayward’ kiwifruits and ‘Conference’ pears (average values increasing from 1.4 and 2.5 μg/mL). Irrespective of the isolate used for inoculation, patulin was always detected in ‘Pinkcot’ apricots, although with intermediate concentrations (average values of 9.4 μg/mL). In most isolates, ‘Pink Lady’ apples or Spring Crest’ peaches accumulated the highest average concentration of patulin (18.9 and 18.6 μg/mL, respectively). 3.6. Patulin accumulation in vitro After 7 days of incubation, the isolates E 11, C 28 and I 12, accumulated patulin in all agar media (MEA and fruit PAMs), while I 1

113

produced patulin only in MEA and apricot PAM (Table 8). The isolate I 1 also showed the lowest accumulation of patulin (0.7 μg/mL on average), while I 12 or C 28 showed the highest patulin accumulation (42.6 and 39.4 μg/mL, on average). However, E 11 showed the highest patulin accumulation in Golden Delicious PAM. No significant differences in patulin concentrations were found among E 11, C 28 and I 12 grown on MEA or in ‘Hayward’ PAMs. Mycelia grown on pear PAMs accumulated less patulin than mycelia grown on apple PAMs. In particular, ‘Abate Fetel’ PAM showed the lowest average accumulation of patulin (from 0 to 15.5 μg/mL), while mycelia grown on ‘Pink Lady’ apple PAMs showed the greatest accumulation of the mycotoxin (from 0 to 173.1 μg/mL). All P. expansum isolates accumulated patulin in apricot PAM, although with intermediate concentrations (from 6.0 to 38.0 μg/mL). The greatest patulin accumulation of all the assays was found by I 12 mycelia grown on ‘Pink Lady’ PAM (173.1 μg/mL). No P. expansum isolate produced patulin in cultures of CLM or Pratt's medium (data not reported). Patulin concentrations detected in vitro (mycelia of P. expansum isolates grown on fruit PAMs) was 1.5–28.8 fold higher than that accumulated in vivo (infected tissues of fruit inoculated with P. expansum). The highest differences were found for ‘Golden Delicious’ apples, ‘Hayward’ kiwifruits and ‘Candonga’ strawberries, where patulin concentration was, respectively, 28.8, 22.0 and 13.6 fold higher in vitro than in vivo. No significant correlations were found between physical–chemical characteristics (firmness, SSC, pH or acidity) of fruit before inoculation and the amount of patulin accumulated in vitro or in vivo (data not reported). 4. Discussion One aim of our study was to evaluate the aggressiveness of some isolates of P. expansum in different fruit hosts. In vivo tests showed that most P. expansum isolates tested were capable of causing blue mould in both usual (pears and apples) and less common hosts of P. expansum (apricots, peaches, strawberries and kiwifruits). In particular, isolate C 28 showed a consistently high incidence of infection (53.7% on average) and the largest lesion diameters after 7 days of incubation (21.2 mm on average), while I 1 showed the lowest incidence of infection (11.7% on average) and the smallest lesion diameters (15.4 mm, where pathogenic). Different factors were probably involved in the aggressiveness of isolates. The most aggressive isolate C 28 showed the longest germ tube elongation after 24 h of incubation in CLM, while the less aggressive isolate I 1 showed the lowest germ tube elongation in this culture (data not reported). The more rapid conidial germination observed for C 28

Table 5 Colony-forming units of Penicillium expansum isolates (I 1, E 11, C 28 or I 12) grown on MEA and in puree agar media (PAMs) of different fruit. Growth medium

I1A

E 11

C 28

I 12

Averagea

MEA ‘Abate Fetel’ pear PAM ‘Conference’ pear PAM ‘Golden Delicious’ apple PAM ‘Pink Lady’ apple PAM ‘Pinkcot’ apricot PAM ‘Spring Crest’ peach PAM ‘Candonga’ strawberry PAM ‘Hayward’ kiwifruit PAM Averageb

50.8 ± 7.6 a A 73.0 ± 9.3 bc AB 72.1 ± 8.5 bc BC 102.0 ± 12.0 d A 84.8 ± 16.4 c B 59.8 ± 4.3 ab A 58.3 ± 9.0 ab A 64.8 ± 12.2 b B 66.3 ± 11.0 b A 70.2

49.9 ± 5.4 a A 75.9 ± 10.1 c AB 56.1 ± 7.9 ab A 104.8 ± 13.7 d A 65.3 ± 8.9 b A 58.7 ± 5.6 ab A 61.5 ± 7.2 b AB 55.3 ± 8.0 ab AB 81.3 ± 9.5 c C 67.6

47.4 ± 5.7 a A 81.0 ± 9.2 e B 66.1 ± 8.3 cd AB 91.5 ± 10.3 f A 60.8 ± 4.2 bc A 60.3 ± 5.5 bc A 54.8 ± 8.0 ab A 52.8 ± 4.4 ab A 69.5 ± 7.7 d AB 64.9

52.8 ± 7.6 a A 68.3 ± 8.4 bc A 77.2 ± 8.6 b C 90.5 ± 11.3 d A 65.3 ± 9.2 b A 69.1 ±7.1 bc B 69.0 ± 8.2 bc B 52.0 ± 8.6 a A 77.5 ± 6.9 b BC 69.1

50.2 74.5 67.9 97.2 69.1 62.0 60.9 56.2 73.7

Colonies were counted after 4 days of incubation at 20 °C for inoculation on strawberry PAM and after 3 days of incubation for inoculation on other growth media. Data represent the mean of six dishes for each growth medium ± standard deviation. Within each column (lower case) and row (upper case), different letters indicate significant differences according to LSD test at P ≤ 0.05. a Mean of row data. b Mean of column data.

114

F. Neri et al. / International Journal of Food Microbiology 143 (2010) 109–117

Table 6 Mycelial growth (mm) of Penicillium expansum isolates (I 1, E 11, C 28 or I 12) cultured on MEA and in puree agar media (PAMs) of different fruit. Growth medium

I1A

E 11

C 28

MEA ‘Abate Fetel’ pear PAM ‘Conference’ pear PAM ‘Golden Delicious’ apple PAM ‘Pink Lady’ apple PAM ‘Pinkcot’ apricot PAM ‘Spring Crest’ peach PAM ‘Candonga’ strawberry PAM ‘Hayward’ kiwifruit PAM Averageb

34.7 ± 1.9 b A 40.4 ± 1.2 d A 40.8 ± 2.9 d A 42.8 ± 1.7 d A 38.0 ± 1.9 c A 42.3 1.2 ± d A 40.8 ± 2.9 d A 20.3 ± 1.4 a A 50.3 ± 2.3 e A 38.9

40.1 ± 2.1 b B 55.9 ± 1.5 de B 61.7 ± 3.3 ef B 51.2 ± 2.5 cd B 42.4 ± 3.5 b B 50.5 ± 3.2 c B 58.6 ± 6.1 de B 28.2 ± 1.2 a C 64.0 ± 8.8 f B 50.3

38.8 ± 1.8 61.1 ± 4.3 62.9 ± 3.7 58.4 ± 5.9 46.8 ± 1.6 56.0 ± 4.0 65.4 ± 1.2 19.3 ± 1.5 65.3 ± 5.6 52.7

bB ef C fg B de C cC dC gC aA gB

I 12

Averagea

33.1 ± 1.2 b A 58.8 ± 4.3 e BC 62.3 ± 1.5 f B 51.1 ± 1.1 d B 43.8 ± 0.5 c B 50.2 ± 4.0 d B 59.2 ± 3.4 ef B 22.7 ± 2.8 a B 68.6 ± 5.8 g B 50.0

36.7 54.1 56.9 50.9 42.8 49.8 56.0 22.6 62.1

Colony diameters were measured after 7 days of incubation at 20 °C. Data represent the mean of six dishes for each growth medium ± standard deviation. Within each column (lower case) and row (upper case) different letters indicate significant differences according to LSD test at P ≤ 0.05. a Mean of row data. b Mean of column data.

could have favoured the colonization of wounds in vivo, increasing the competition of this isolates against other microorganisms residing in the fruit. Beside this, the low capacity of acidification observed in I 1 could have negatively influenced the progress of infection of this less aggressive isolate. It is known that Penicillium spp. utilize the decrease of pH to support pathogen attack and that change in pH around the host infection sites can modulate the expression of factors of pathogenicity (Hadas et al., 2007; Prusky and Lichter, 2008). In particular, a study by McCallum et al. (2002) indicated that aggressively growing isolates of P. expansum greatly reduced the pH of potato dextrose broth, while slower growing isolates produced a lower reduction of pH of the substrate. In agreement with these results, more aggressive isolates in our assays showed a higher reduction in pH of growth medium than the less aggressive isolate. Moreover, a decrease of pH in tissues around the infected sites was observed in vivo in hosts with pH ≥ 4.0, and in these fruit (kiwifruits, apricots, pears and strawberries) isolate I 1 generally showed a lower decrease of pH or no significant change of pH compared with the controls. Conversely, no reduction of pH was found in our assays in apples with pH of 3.82 and 3.83 and a low increase of pH was found in peaches with pH of 3.51. It is possible that the changes of pH to values of 3.7–4.3 observed in our assays provided better conditions for the expression of genes encoding cell wall-degrading enzymes, while no adjustment of pH was needed in fruit which already had a favorable pH for their expression. As found by Prusky et al. (2004), the endopolygalacturonase-encoding gene (pepg1) transcripts accumulated in fact between pH 3.5 and 5, but most accumulation was observed at pH 4. With regard to patulin, a different tendency to its accumulation was found among the tested P. expansum isolates, with similar trends both in vitro and in vivo. Different capacity of P. expansum isolates to produce patulin have also been reported in other studies (McCallum et al., 2002; Pianzzola et al., 2004; Morales et al., 2008a), and

McCallum et al. (2002) indicated that P. expansum isolates with more aggressive growth appeared to be able to produce the greatest patulin. In agreement with this study, I 12 and C 28, two of the most aggressive isolates in our assays, accumulated the greatest amount of patulin both in vitro and in vivo, while the least aggressive isolate I 1 produced patulin only in a few substrates and cvs. These results seemed to indicate that patulin was consistently produced by P. expansum isolates with a broad pathogenicity spectrum. However, the largest lesion diameters in vivo or the highest mycelial growth in vitro did not necessary lead to the highest accumulation of patulin. In agreement with Sommer et al. (1974), a high patulin concentration does not, therefore, seem a necessary requirement to cause a high level of blue mould, and patulin does not appear to be a primary virulence factor. On other hand, since production of secondary fungal metabolites is presumably costly to maintain in the producing organism, the presence of patulin probably conferred some advantages to the pathogen. The antimicrobial activity of patulin has been known since the 1940s and an involvement of patulin in the process of competition against microorganisms of the same ecologic niche could be possible. Some studies have shown that patulin could act as a chemical ecological defense against several pathogenic fungi (Nicoletti et al., 2004). Moreover, in another pathosystem it was found that the mycotoxins of Fusarium graminearum (trichotecenes) promoted the spreading of pathogen by inhibition of the host defense response (Jansen et al., 2005). Similar mechanisms of activity could also be hypothesized for patulin on fruit. Another aim of our study was to evaluate the effect of possible fruit hosts on P. expansum development and patulin accumulation. As pathogen infection can activate several defense responses in living fruit, we tried to separate this interference from the constitutive characteristics of fruit. Besides fruit inoculation, P. expansum isolates were therefore cultured on fruit puree agar media (PAMs). Results of

Table 7 Patulin accumulation (μg/mL) in fruit tissues infected by Penicillium expansum isolates (I 1, E 11, C 28 or I 12). Fruit

I1A

E 11

C 28

I 12

Averagea

‘Abate Fetel’ pear ‘Conference’ pear ‘Golden Delicious’ apple ‘Pink Lady’ apple ‘Pinkcot’ apricot ‘Spring Crest’ peach ‘Candonga’ strawberry ‘Hayward’ kiwifruit Averageb

1.4 ± 0.3 b A 0.5 ± 0.5 a A 0.0 ± 0.0 a A 0.0 ± 0.0 a A 3.7 ± 0.8 c A 0.2 ± 0.0 a A 0.0 ± 0.0 a A 0.0 ± 0.0 a A 0.7

0.0 ± 0.0 a A 2.1 ± 0.5 a A 0.0 ± 0.0 a A 33.9 ± 6.1 d D 8.1 ± 1.8 b AB 16.2 ± 7.0 c B 0.8 ± 1.8 a A 0.3 ± 0.0 a A 7.7

0.7 ± 0.2 a A 1.5 ± 0.3 a A 2.8 ± 0.8 a B 17.2 ± 2.0 b B 14.1 ± 2.4 b C 30.1 ± 7.8 c C 3.2 ± 1.2 a B 3.5 ± 1.6 a B 9.1

4.4 ± 2.4 ab B 5.7 ± 1.7 b B 3.8 ± 0.7 ab B 25.4 ± 2.3 d C 11.5 ± 1.8 c BC 28.0 ± 3.2 d C 1.6 ± 0.6 a AB 3.2 ± 0.1 ab B 10.5

1.6 2.5 1.7 19.1 9.4 18.6 1.4 1.8

Fruit were wounded and inoculated with conidial suspensions (103 conidia/mL) of Penicillium expansum. Patulin was detected after 7 days of incubation at 20 °C. Data represent the mean of three repetitions ± standard deviation. Within each column (lower case) and row (upper case), different letters indicate significant differences according to LSD test at P ≤ 0.05. a Mean of row data. b Mean of column data.

F. Neri et al. / International Journal of Food Microbiology 143 (2010) 109–117

115

Table 8 Patulin accumulated (μg/mL) by Penicillium expansum isolates (I 1, E 11, C 28 or I 12) cultured on MEA and in puree agar media (PAMs) of different fruit. Growth medium

I1A

E 11

C 28

I 12

Averagea

MEA ‘Abate Fetel’ pear PAM ‘Conference’ pear PAM ‘Golden Delicious’ apple PAM ‘Pink Lady’ apple PAM ‘Pinkcot’ apricot PAM ‘Spring Crest’ peach PAM ‘Candonga’ strawberry PAM ‘Hayward’ kiwifruit PAM Averageb

1.4 ± 2.5 a A 0.0 ± 0.0 a A 0.0 ± 0.0 a A 0.0 ± 0.0 a A 0.0 ± 0.0 a A 6.0 ± 1.8 b A 0.0 ± 0.0 a A 0.0 ± 0.0 a A 0.0 ± 0.0 a A 0.8

31.9 ± 10.0 bc B 8.4 ± 3.3 a B 7.0 ± 1.3 a B 74.1 ± 3.0 e C 39.8 ± 3.5 c B 23.7 ± 15.9 b AB 41.1 ± 1.6 c C 23.5 ± 1.9 b B 53.7 ± 6.2 d B 33.7

36.2 ± 4.3 b B 11.5 ± 3.0 a BC 27.0 ± 7.0 b C 61.3 ± 2.1 c B 30.6 ± 7.8 b B 38.0 ± 13.7 b B 59.3 ± 6.2 c D 38.3 ± 13.2 b C 52.3 ± 4.0 c B 39.4

27.8 ± 9.3 c B 15.5 ± 3.2 abc C 22.1 ± 0.3 bc C 60.2 ± 2.0 d B 173.1 ± 12.3 e C 12.6 ± 7.3 ab A 9.3 ±3.0 a B 14.4 ± 4.5 ab B 48.0 ± 9.9 d B 42.6

24.3 8.9 14.0 48.9 60.9 20.1 27.4 19.1 38.5

Patulin was determined after 7 days of incubation at 20 °C. Data represent the mean of three repetitions ± standard deviation. Within each column (lower case) and row (upper case), different letters indicate significant differences according to LSD test at P ≤ 0.05. a Mean of row data. b Mean of column data.

in vitro assays showed that growth medium influenced both pathogen development and patulin accumulation. Culture of P. expansum isolates on most PAMs stimulated the formation of colonies and mycelial growth in comparison to a standard growth medium (MEA). Major stimulation of formation of colonies and mycelial growth was observed for ‘Golden Delicious’ apples (48.1%) and ‘Hayward’ kiwifruits (40%) PAMs, respectively. Viceversa, strawberry PAM did not stimulate or showed little increase of conidial germination, while it reduced the mycelial growth of all P. expansum isolates. Germination rates lower than 100% in a standard growth medium have also been observed in Monilinia spp. (Casals et al., 2010) and mycostasis (a fungistatic mechanism that protects the propagules from spontaneous germination in the absence of a potentially colonisable substrata or sufficient available nutrients) has been suggested as a possible explanation of this phenomenon (Lockwood, 1988). Higher mycelial growth of P. expansum generally observed on pear PAM than on apple PAM agree with results by Morales et al. (2008b), who found higher dry weight of P. expansum in pear juice than in apple juice. To our knowledge, no other studies have previously compared the conidial germination and mycelial growth of P. expansum in different fruit agar media. Some differences in host susceptibility to P. expansum were found between in vitro and in vivo tests. The highest severity of infection in vivo was in general observed in fruit with softer texture, such as strawberries. Fruit softening is related to degradation of cell wall components that occurs during maturation. Since an increase of depolymerisation of pectin also occurs during P. expansum infection (Miedes and Lorences, 2006), colonization of tissues was probably facilitated in softer fruit where the pathogen could already find favorable conditions for the progress of infection. The delay of P. expansum growth induced by natural constituents of strawberry in vitro seemed, instead, not sufficient to avoid the development of severe blue mould decay in vivo. The opposite was observed for kiwifruit, that was an excellent substrate for P. expansum growth only in vitro. Some factors occurring in the live host probably decreased kiwifruit susceptibility to blue mould decay in vivo. Kiwifruit has, for instance, a natural high concentration of volatile compounds with fungicidal activity against P. expansum such as trans-2-hexenal (Young and Paterson, 1985; Neri et al., 2006) and a hard and thick skin (Hallet and Sutherland, 2005) that could confer an important defensive structure to pathogen penetration. Inhibitors of fungal degrading enzymes (Di Matteo et al., 2006) could also have been involved in the mechanism of defense against P. expansum in vivo. Further studies are needed to elucidate these issues. Regarding patulin, no accumulation of the mycotoxin was found when the pathogen was cultured in Pratt's medium or in CLM (data not reported), while patulin was always detected when isolates were grown on MEA and, in most cases, on fruit PAMs. The ability of P. expansum isolates to produce patulin in MEA was also observed by Andersen et al. (2004), and concentrations of patulin detected in our

study for growth on MEA were similar to those found in culture of yeast extract sucrose by Abramson et al. (2009). The lack of patulin production in CLM agrees with the results of White et al. (2006). Different production levels of patulin in relation to growth medium were also found in other studies on patulin-producing fungi, and various components of growth medium, such as the type of carbohydrate and source of nitrogen, have been reported to affect patulin production (Stott and Bullerman, 1975; Rice et al., 1977; Abrunhosa et al., 2001). Pratt's medium and CLM have a low content of nutrients; they probably did not provide satisfactory nutrition to P. expansum for patulin production. Regarding culture on fruit media, the greatest accumulation of patulin was observed in mycelia grown on apple PAMs (173.1 and 74.1 μg/mL in ‘Pink Lady’ and ‘Golden Delicious’ PAMs, respectively). The ‘Pink Lady’ apples inoculated with isolate E 11 also showed the greatest accumulation of patulin in in vivo assays (33.9 μg/mL). These results confirm that apples are an important source of patulin contamination and, because of their high consumption, are probably the greatest dietary source of patulin in humans. However, in agreement with what was found in other studies (Martins et al., 2002; McCallum et al., 2002; Morales et al., 2008b), some differences were observed in vivo among apple cvs. Infected tissue of ‘Golden Delicious’ apples showed lower patulin accumulation than ‘Pink Lady’ tissue. Furthermore, no significant differences in patulin concentrations were found in our study among infected tissue of ‘Golden Delicious’ apples, ‘Abate Fetel’ pears, ‘Conference’ pears and ‘Candonga’ strawberries. Besides apples and pears, our study showed the potential of some less common hosts of P. expansum to support patulin occurrence and in some cases with values comparable to those on apples, confirming what was previously pointed out by Buchanan et al. (1974). In particular, apricots were consistently positive for patulin accumulation, both in vitro (average values of 20.1 μg/mL) and in vivo (average values of 9.4 μg/mL). Infected tissues of peaches were highly susceptible to patulin accumulation too, showing an average accumulation of 27.4 and 18.6 μg/mL in vitro and in vivo, respectively. Patulin accumulated in infected tissue of peaches was higher than that in ‘Golden Delicious’ apples and, for isolate I 12, also comparable with patulin accumulated in ‘Pink Lady’ apples. On the contrary, strawberries and kiwifruits accumulated intermediate or high values of patulin in vitro (19.1 and 38.5 μg/mL on average, respectively), while much lower amounts were found in vivo (1.4 and 1.8 μg/mL on average, respectively). No significant correlations were found between physical-chemical characteristics of fruit before inoculation and patulin accumulated both in in vitro and in vivo (data not reported). Other components of fruit not analyzed in the study may play a role on patulin synthesis or stability. Patulin has been found to react with thiol and/or amino groups of proteins, as well as with free radicals generated by oxidation of acid ascorbic, and a partial degradation of patulin in fruit juices has been observed during storage by the presence of sulfhydryl groups or

116

F. Neri et al. / International Journal of Food Microbiology 143 (2010) 109–117

after the addition of ascorbic acid (Scott and Sommers, 1968; Fliege and Metzeler, 1999; Drusch et al., 2007). The medium-low amount of patulin detected in our study in strawberries could have been influenced by the high content of both thiols (such as cysteine, glutathione and N-acetylcysteine) and ascorbic acid that naturally occurs in these fruit (Demirkol et al., 2004; Szajdek and Borowska, 2008). The very low concentration of thiols in apples has been associated with the stability of patulin added to apple juice (Scott and Sommers, 1968). Interestingly, the amount of patulin detected in vitro (fruit PAMs) was higher than the mycotoxin accumulation in vivo (infected tissues of fruit). This behavior was also observed for several P. expansum isolates by Sommer et al. (1974) and Morales et al. (2008a), but the reasons for this pattern are not known. An involvement of the host in limitation of patulin synthesis or in detoxification of patulin in response to P. expansum infection could be hypothesized. In conclusion, our study showed that differences in germ tube elongation and decrease of environmental pH may have influenced the aggressiveness of P. expansum isolates in vivo. In contrast, patulin accumulation did not appear to be a primary virulence factor. Although patulin was consistently accumulated by P. expansum isolates pathogenic to a broad spectrum of fruit hosts and no or low amounts of patulin were produced by the lowest virulent P. expansum isolate, the accumulation of high amounts of patulin did not seem a necessary requirement to cause high level of blue mould. In vitro assays also demonstrated that constitutive characteristics of the host can influence the rate of P. expansum conidial germination and mycelial growth, showing ‘Golden Delicious’ apple and ‘Hayward’ kiwifruit as particularly favorable substrates for conidial germination and mycelial growth, respectively. Finally, our results proved that, besides apples, other fruit can support patulin production, indicating that a real threat of patulin contamination exists also in less common hosts of P. expansum, and in particular in apricots and peaches. A wide offer of juices, nectars and purees of pure or mixed species of fruit is currently available on the market, and more frequently consumed by children and the elderly. The kind of fruit products preferred for consumption varies among countries and in Italy apricot, peach and pear nectars are consumed in higher amounts than apple juices. In a recent survey carried out on pear, peach and apricot juices purchased from Italian supermarkets, 34.4% of samples were positive for patulin and 15.2% contained more than 10 μg/Kg of patulin, which is the maximum level permitted for baby food (Spadaro et al., 2008). In another study on juices produced by various Italian and European companies, apple juices has concentrations of patulin less that limits set by regulation, while 40% of the pear juices had concentrations higher than the legal limit (up to 92 μg/L of patulin) and samples with lower market cost had the higher amount of patulin contamination (Bonerba et al., 2010). Results of our study and of recent surveys on fruit juices indicate that a steady monitoring of patulin contamination should be carried out in all fruit substrates to effectively control the contamination of patulin in the human diet. References Abramson, D., Lombaert, G., Clear, R.M., Sholberg, P., Trelka, R., Rosin, E., 2009. Production of patulin and citrinin by Penicillium expansum from British Columbia (Canada) apples. Mycotoxin Research 25, 85–88. Abrunhosa, L., Paterson, R.R.M., Kozakiewicz, Z., Lima, N., Venancio, A., 2001. Mycotoxin production from fungi isolates from grapes. Letters in Applied Microbiology 32, 240–242. Andersen, B., Smedsgaard, J., Frisvad, J., 2004. Penicillium expansum: consistent production of patulin, chaetoglobosins, and other secondary metabolites in culture and their natural occurrence in fruit products. Journal of Agricultural and Food Chemistry 52, 2421–2428. Anon., 2002. Reports on tasks for scientific cooperation, task 3.2.8. Assessment of dietary intake of patulin by the population of EU member States. Brussels: SCOOP Report 2002. Anon., 2003. Commission Regulation (EC) No. 1425/2003. Official Journal of the European Communities L 203, 1–3.

Anon., 2004. U.S. Food and Drug Administration (USFDA). Compliance policy guide. Compliance policy guidance for FDA staff. Sec. 510.150. Apple juice, apple juices concentrates, and apple juice products-adulteration with patulin. Baert, K., Devlieghere, F., Flyps, H., Oosterlinck, M., Ahmed, M.M., Rajkovic, A., Verlinden, B., Nicolai, B., Devebevere, J., De Meulenaer, B., 2007. Influence of storage conditions of apples on growth and patulin production by Penicillium expansum. International Journal of Food Microbiology 119, 170–181. Bhatnagar, D., Yu, J., Ehrlich, K.C., 2002. Toxins of filamentous fungi. Chemical Immunology 81, 167–206. Bonerba, B., Ceci, E., Conte, R., Tantillo, G., 2010. Survey of the presence of patulin in fruit juices. Food Additives and Contaminants 3, 114–119. Buchanan, J.R., Sommer, N.F., Fortlage, R.J., Maxie, E.C., Mitchell, F.G., Hsieh, D.P.H., 1974. Patulin from Penicillium expansum in stone fruits and pears. Journal of the American Society for Horticultural Science 99, 262–265. Casals, C., Vinas, I., Torres, R., Griera, C., Usall, J., 2010. Effect of temperature and water activity on in vitro germination of Monilinia spp. Journal of Applied Microbiology 108, 47–54. Demirkol, O., Adams, C., Ercol, N., 2004. Biologically important thiols in various vegetables and fruits. Journal of Agricultural and Food Chemistry 52, 8151–8154. Di Matteo, A., Bonivento, D., Tsernoglou, D., Federici, L., Cervone, F., 2006. Polygalacturonase-inhibiting protein (PGIP) in plant defence: a structural review. Phytochemistry 67, 528–533. Drusch, S., Kopka, S., Kaeding, J., 2007. Stability of patulin in a juice-like aqueous model system in the presence of ascorbic acid. Food Chemistry 100, 192–197. Fliege, R., Metzeler, M., 1999. The mycotoxin patulin induces intra- and intermolecular protein crosslinks in vitro involving cysteine, lysine and histidine side chains, and α-amino groups. Chemico-Biological Interactions 123, 85–103. Frank, H.K., Orth, R., Figge, A., 1977. Patulin in lebensmitteln pflanzlicher herkunft [Patulin in foods of vegetable origin. 2. Several kinds of fruit and vegetables and fruit and vegetable products]. Zeitschrift fur Lebensmitteluntersuchung undForschlung A 111–114. Gregori, R., Mari, M., Bertolini, P., Sañudo Barajas, J.A., Tian, J.B., Labavitch, J.M., 2008. Reduction of Colletotrichum acutatum infection by a polygalacturonase inhibitor protein extracted from apple. Postharvest Biology and Technology 48, 309–313. Hadas, E., Goldberg, I., Pines, O., Pruscky, D., 2007. Involvement of gluconic acid and glucose oxidase in the pathogenicity of Penicillium expansum of apples. Phytopathology 97, 384–390. Hallet, I.C., Sutherland, P.W., 2005. Structure and development of kiwifruit skins. International Journal of Plant Science 166, 693–704. Hof, H., 2007. Mycotoxins: pathogenicity factors or virulence factors? Mycoses 51, 93–94. Jansen, C., von Wettstein, D., Schafer, W., Kogel, K.-H., Felk, A., Maier, F.J., 2005. Infection patterns in barley and wheat spikes inoculated with wild-type and trichodiene synthase gene disrupted Fusarium graminearum. PNAS 102, 16892–16897. Laidou, I.A., Thanassoulopoulos, C.C., Liakopoulou-Kyriakides, M., 2001. Diffusion of patulin in the flesh of pears inoculated with four post-harvest pathogens. Journal of Phytopathology 149, 457–461. Lockwood, J.L., 1988. Evolution of concepts associated with soilborne plant pathogens. Annual Review of Phytopathology 26, 93–121. Mari, M., Neri, F., Bertolini, P., 2009. Management of important diseases in Mediterranean high value crops. Stewart Postharvest Review 2, 2. Martins, M.L., Gimeno, A., Martins, H.M., Bernardo, F., 2002. Co-occurrence of patulin and citrinin in Portuguese apples with rotten spots. Food Additives and Contaminants 19, 568–574. McCallum, J.L., Tsao, R., Zhou, T., 2002. Factors affecting patulin production by Penicillium expansum. Journal of Food Protection 65, 1917–1942. Miedes, E., Lorences, E.P., 2006. Changes in cell wall pectin and pectinase activity of apple and tomato fruits during Penicillium expansum infection. Journal of the Science of Food and Agriculture 86, 1359–1364. Moake, M.M., Padilla-Zakour, O.L., Worobo, R.W., 2005. Comprehensive review of patulin control methods in food. Comprehensive Reviews in Food Science and Food Safety 1, 8–21. Morales, H., Marin, S., Obea, L., Patino, B., Domenech, M., Ramos, A.J., Sanchis, V., 2008a. Ecophysiological characterization of Penicillium expansum population in Lleida (Spain). Internation Journal of Food Microbiology 122, 243–252. Morales, H., Barros, G., Marin, S., Chulze, S., Ramos, A.J., Sanchis, V., 2008b. Effects of apple and pear varieties and pH on patulin accumulation by Penicillium expansum. Journal of the Science of Food and Agriculture 88, 2738–2743. Neri, F., Mari, M., Brigati, S., 2006. Control of Penicillium expansum by plant volatile compounds. Plant Pathology 55, 100–105. Nicoletti, R., De Stefano, M., De Stefano, S., Tricone, A., Marziano, F., 2004. Antagonism against Rhizoctonia solani and fungitoxic metabolita production by some Penicillium isolates. Mycopathologia 158, 465–474. Pianzzola, M.J., Moscatelli, M., Vero, S., 2004. Characterization of P. expansum isolates associated with Blue Mold on apple in Uruguay. Plant Disease 88, 23–28. Prusky, D., Lichter, A., 2008. Mechanisms modulating fungal attach in post-harvest pathogen interactions and their control. European Journal of Plant Pathology 121, 281–289. Prusky, D., McEvoy, J.L., Saftner, R., Conway, W.S., Jones, R., 2004. Relationship between host acidification and virulence of Penicillium spp. on apple and citrus fruit. Phytopathology 94, 44–51. Rice, S.L., Beuchat, L.R., Worthington, R.E., 1977. Patulin production by Byssochlamys spp. in fruit juices. Applied and Environmental Microbiology 34, 791–796. Sant'Ana, A.d.S., Rosenthal, A., De Massaguer, P.R., 2008. The fate of patulin in apple juice processing: a review. Food Research International 41, 441–453.

F. Neri et al. / International Journal of Food Microbiology 143 (2010) 109–117 Scott, P.M., Sommers, E., 1968. Stability of patulin and penicillic acid in fruit juices and flour. Journal of Agricultural and Food Chemistry 16, 483–485. Scott, P.M., Fuleki, T., Harwig, J., 1977. Patulin content of juice and wine produced from moldy grapes. Journal of Agricultural and Food Chemistry 25, 434–437. Snowdon, A.L., 1990. A colour atlas of post-harvest diseases and disorders of fruits and vegetables. Volume 1: General Introduction and Fruits. Wolfe Scientific Ltd, London, pp. 178–179; 223; 250. Sommer, N.F., Buchanan, J.R., Fortlage, R.J., 1974. Production of patulin by Penicillium expansum. Applied Microbiology 28, 589–593. Spadaro, D., Garibaldi, A., Gullino, M.L., 2008. Occurrence of patulin and its dietary intake through pear, peach and apricot juices in Italy. Food Additives and Contaminants, Part B 1, 134–139.

117

Stott, W.T., Bullerman, L.B., 1975. Influence of carbohydrate and nitrogen source on patulin production by Penicillium patulum. Applied Microbiology 30, 850–854. Szajdek, A., Borowska, E.J., 2008. Bioactive compounds and health-promoting properties of berry fruits: a review. Plant Foods for Human Nutrition 63, 147–156. White, S., O'Callaghan, J., Dobson, A.D.W., 2006. Cloning and molecular characterization of Penicillium expansum genes upregulated under conditions permissive for patulin biosynthesis. FEMS Microbiology Letters 255, 17–26. Xu, X.M., Berrie, A.M., 2005. Epidemiology of mycotoxigenic fungi associated with Fusarium ear blight and apple blue mould: a review. Food Additives and Contaminants 22, 290–301. Young, H., Paterson, V.J., 1985. The effects of harvest maturity, ripeness and storage on kiwifruit aroma. Journal of the Science of Food and Agriculture 36, 352–358.