Chitosan inactivates spoilage yeasts but ... - Wiley Online Library

Journal of Applied Microbiology 2005, 98, 872–880

doi:10.1111/j.1365-2672.2004.02527.x

Chitosan inactivates spoilage yeasts but enhances survival of Escherichia coli O157:H7 in apple juice G. Kisko´1, R. Sharp2 and S. Roller2 1

Szent Istva´n University, Budapest, Hungary, and 2Microbiology Research Group, Faculty of Health and Human Sciences, Thames Valley University, London, UK

2004/0330: received 24 March 2004, revised 18 June 2004 and accepted 4 September 2004

ABSTRACT ´ , R . S H A R P A N D S . R O L L E R . 2005. G. KISKO

Aims: To develop new measures for controlling both spoilage and pathogenic micro-organisms in unpasteurized apple juice using chitosan. Methods and Results: Micro-organisms were isolated and identified from apple juice treated or untreated with chitosan using enrichment, selective media, microscopy, substrate assimilation patterns and ribosomal DNA profiling. Chitosan (0Æ05–0Æ1%) delayed spoilage by yeasts at 25C for up to 12 days but the effect was species specific: Kloeckera apiculata and Metschnikowia pulcherrima were inactivated but Saccharomyces cerevisiae and Pichia spp. multiplied slowly. In challenge experiments at 25C, total yeast counts were 3–5 log CFU ml)1 lower in chitosan-treated juices than in the controls for 4 days but the survival of Escherichia coli O157:H7 was extended from 1 to 2 days; at 4C, chitosan reduced the yeast counts by 2–3 log CFU ml)1 for up to 10 days but survival of the pathogen was prolonged from 3 to 5 days. The survival of Salmonella enterica serovar Typhimurium was unaffected by chitosan at either temperature. Conclusions: The addition of chitosan to apple juice delayed spoilage by yeasts but enhanced the survival of E. coli O157:H7. Significance and Impact of the Study: The results suggest that the use of chitosan in the treatment of fruit juices may potentially lead to an increased risk of food poisoning from E. coli O157:H7. Keywords: apple juice, chitosan, Escherichia coli O157:H7, Salmonella Typhimurium, yeast.

INTRODUCTION Although the predominant microflora of fruit juices consists of harmless yeasts and lactic acid bacteria, contamination with enteric pathogens through contact with faecal material or improper handling and manufacturing is possible. Cases of food poisoning associated with the consumption of unpasteurized fruit juices have been reported. In the USA, several children have died after drinking unpasteurized apple juice contaminated with enterohaemorrhagic Escherichia coli O157:H7 (CDC 1997; Correspondence to: Sibel Roller, Faculty of Health and Human Sciences, Thames Valley University, 32–38 Uxbridge Road, London W5 2BS, UK (e-mail: [email protected]).

Uljas and Ingham 1999). Outbreaks of salmonellosis involving apple juice have also been reported (CDC 1975). Warning labels are now required in the USA for all unprocessed fruit juices unless a 5-log pathogen reduction treatment has been applied to the product (FDA 1998; Giese 2002). In conventional beverage processing, fruit juices are heated to inactivate all nonspore-forming micro-organisms. Any remaining bacterial spores are generally unable to germinate because of the acidic nature (pH < 4Æ6) of the juices (Splittstoesser et al. 1994). However, heat treatment causes vitamin losses and changes in flavour of the juices. Many consumers regard heat-treated, shelf-stable products as low in quality and are demanding foods that are minimally processed. Increasing consumer pressure for the exclusion of ª 2005 The Society for Applied Microbiology

CHITOSAN IN APPLE JUICE

many chemically synthesized preservatives from foods and beverages has challenged food scientists to find alternative and more natural methods of preserving foods (Dillon and Board 1994; Naidu 2000; Roller 2003a). The inhibitory and biocidal action of chitosan (a polymer of 1,4-linked 2-amino-2-deoxy-b-D-glucose) has been reported widely mainly on the basis of in vitro trials against pure cultures of micro-organisms (Shahidi et al. 1999). Minimum inhibitory concentrations ranging from 0Æ0004 to 1% of chitosan against yeasts, moulds and bacteria in laboratory media have been reported, depending on the type of chitosan used, the conditions of testing (pH, temperature, medium) and the target organism (reviewed in Roller 2003b). Several reports in the literature suggest that chitosan inhibits growth of some food-borne pathogens at concentrations as low as 0Æ01%. In general, concentrations of chitosan required to inactivate or inhibit growth of bacteria appear to be at least one order of magnitude higher than those needed to show an effect on yeasts and moulds. The antimicrobial activity of chitosan is more pronounced at pH values below its pKa of 6Æ3 and at temperatures above 20C (Roller and Covill 1999, 2000; Rhoades and Roller 2000; Helander et al. 2001; Strand et al. 2001; Roller 2003b). As a result of these studies, it has been suggested that chitosan could be used as a novel, natural food preservative. The aim of this study was to evaluate the addition of chitosan as a novel measure for controlling both spoilage and pathogenic micro-organisms in unpasteurized apple juice with due regard to the possible implications of the selective action of chitosan on the safety of the product. Unpasteurized apple juice is often referred to as cider in the USA. This should not be confused with UK cider, which is a fermented alcoholic beverage. In this paper, the term apple juice refers strictly to the raw, unprocessed, unfermented juice of the apple.

MATERIALS AND METHODS Materials Freshly pressed, unclarified, raw juice from a mixture of Bramley and Cox apples was obtained directly from a manufacturer in Suffolk, UK. The pH of juice prepared from stored apples varied between 3Æ5 and 3Æ7 whereas the pH of juice prepared from freshly harvested apples was 3Æ2 on arrival at the laboratory. All microbiological media and diluents were from Oxoid (Basingstoke, UK) and all chemicals from Sigma Chemicals Co. Ltd (Poole, Dorset, UK) unless otherwise indicated. Chitosan glutamate was from Pronova (Drammen, Norway). It contained 42% glutamate and had a deacetylation range of 75–85% (manufacturer’s data).

873

Micro-organisms and their cultivation Bacteria, yeasts and moulds isolated from the apple juice were routinely cultivated on plate count agar (PCA) and malt extract agar (MEA). PCA plates were incubated at 30C for 2–3 days while MEA plates were incubated at 25C for 3–5 days. Salmonella Typhimurium NCTC74, an attenuated strain of E. coli O157:H7 (Dr P. McClure, Unilever Research, Colworth House, Bedford, UK) and Listeria monocytogenes NCTC11994 were maintained in brain heart infusion (BHI) broth and agar and incubated overnight at 37C. For challenge experiments, fresh overnight cultures of E. coli O157:H7 and Salm. Typhimurium in BHI broth were washed three times and re-suspended in sterile saline solution. Viable counts of the washed cell suspensions were determined by spread-plating 0Æ1 ml of the suspensions (in duplicate) on tryptone soya agar (TSA) and incubating the plates at 37C for 24 h. Isolation and identification of the microflora in apple juice Containers (200 ml capacity) of apple juice (100 ml) were prepared in duplicate for each treatment and stored at 4 and 25C. Duplicate samples (10 ml) were taken periodically from each container for microbiological analysis: on days 0, 1, 2, 4, 7 and 12 from containers stored at 25C and on days 0, 3, 5, 8, 12, 16 and 20 from containers stored at 4C. Serial (1 : 10) dilutions were prepared in sterile maximal recovery diluent. Viable numbers were determined by pour-plating (1Æ0 ml) on PCA for total counts, orange serum agar (OSA) and deMan, Rogosa and Sharpe agar (MRSA) for lactic acid bacteria and ethanol bicarbonate agar (EBA) for acetic acid bacteria. MRSA was supplemented with 0Æ001% w/v actidion (cycloheximide) or 0Æ5% chitosan in some experiments in order to suppress overgrowth of lactic acid bacteria by yeasts. PCA and MRSA plates were incubated at 30C for 2–3 days while EBA plates were incubated at 25C for 3–5 days. Violet red bile glucose agar was used for the enumeration of Enterobacteriaceae and was incubated at 37C for 24 h. Yeasts and moulds were enumerated by spread-plating (0Æ1 ml) on three chloramphenicol-supplemented media: dichloranglycerol (DG 18) agar, tryptone glucose yeast extract agar (TGYEA) and oxytetracycline glucose yeast extract agar. For the determination of heat-resistant spore-forming bacteria, 5 ml samples of the juice were heated at 70C for 20 min, cooled immediately in ice, and 0Æ1 ml was spreadplated (in duplicate) on PCA and incubated at 30C for 2– 3 days. For the enumeration of heat resistant moulds, two methods were used (Pitt and Hocking 1997). In the first method, the juice (5 ml) was heat-treated at 70C for 30 min, cooled, spread-plated (0Æ1 ml, in duplicate) on TGYEA and incubated at 25C for 5–7 days. In the second

ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology, 98, 872–880, doi:10.1111/j.1365-2672.2004.02527.x

874 G . K I S K O´ ET AL.

method, the juice (5 ml) was heat-treated at 80C for 30 min in flat 100 ml medical flasks, cooled and incubated in a horizontal position (allowing maximum exposure of the surface area to air) at 25C for 5–7 days. The appearance of mould colonies was assessed by visual inspection. Alicyclobacillus acidoterrestris was determined according to the method of Pettipher et al. (1997) by spread-plating diluted juice on OSA and incubating at 44C for 48 h. Samples (5 ml) of juice were also heated at 70C for 20 min (Eiroa et al. 1999) to activate any spores present and then spread-plated (0Æ1 ml, in duplicate) on OSA and incubated at 44C for 48 h. All colonies were counted as presumptive A. acidoterrestris. Micro-organisms isolated from the apple juice were inspected by light microscopy and the bacteria were Gram-stained (Roberts et al. 1995). Phenotypic profiling of the Gram-negative bacteria was undertaken using API 20E (20100) strips (Biomerieux, Marseille, France). Pre-enrichment and enrichment were used to detect the presence or absence of pathogenic bacteria in the apple juice. For Escherichia coli O157:H7, 25 ml of juice in 225 ml buffered peptone water (BPW) (or 2Æ5 ml juice in 22Æ5 ml BPW for the challenge experiments) was incubated at 37C for 24 h. An aliquot (1 ml) was dispensed into 10 ml EC broth (reduced bile salts supplemented with novobiocin) and incubated at 37C for 24 h. A loopful of the EC broth was streaked onto sorbitol MacConkey (SMAC) agar or onto CHROMagar (M-Tech Diagnostic Ltd, EE-220-Trial; both agars supplemented with cefixime and potassium-tellurite) and incubated at 37C for 24 h. For Salmonella spp., 25 ml of juice in 225 BPW (or 2Æ5 ml of juice in 22Æ5 ml BPW for the challenge experiments) was incubated at 37C for 6–18 h (pre-enrichment). Aliquots (1 ml) were dispensed into 10 ml of Rappaport Vassiliadis (RV) broth and incubated at 42C for 18–24 h (enrichment). A loopful of the RV broth was streaked onto brillant green agar (BGA) supplemented with sulphamandalate or onto xylose lysine desoxycholate (XLD) agar and incubated at 37C for 24–48 h. For L. monocytogenes, 25 ml of the juice was dispensed into 225 ml BPW and incubated at 37C for 24 h. An aliquot (25 ml) was added into 225 ml of Listeria enrichment broth containing Listeria Selective Enrichment Supplement and incubated at 37C for 24 h. A loopful was streaked onto Oxford agar containing Listeria Selective Supplement and incubated at 37C for 24 h. Positive controls were prepared for all three pathogenic bacteria. Injured E. coli O157:H7 and Salmonella spp. were enumerated using the thin agar layer method (Kang and Fung 2000). In the E. coli O157:H7 challenge experiments, the organisms were spread-plated on SMAC agar or CHROMagar and then overlaid with TSA agar. In the Salmonella challenge experiments, organisms were spreadplated on XLD agar and then overlaid with TSA agar. The

number of injured cells was calculated by subtracting the viable count on selective agar alone from the viable count obtained on the overlaid agar. The yeasts isolated from apple juice were identified on the basis of colony morphology, microscopic appearance, carbon source utilization patterns (API 20C AUX 20210 kit; BioMerieux) and ribosomal DNA (S18) analysis (see below). Molecular identification of yeasts The DNA was extracted from overnight cultures of yeasts (grown on MEA slopes at 30C) suspended in sterile distilled water (1 ml) and centrifuged at 15 000 g for 5 min. Sterile glass beads (1 g, diameter 6–7 mm), 200 ll Tris– EDTA buffer (10 mM Tris–HCl and 1 mM EDTA), 200 ll breaking buffer (100 mM NaCl, 10 mM Tris–HCl, 1 mM EDTA, pH 8) 2 ll SDS (1% w/v), 4 ll Triton-X (2% v/v) and 200 ll phenol : chloroform : isopropanol were added to the pellet. This mixture was homogenized in a MiniBeadBeaterTM (Biospec Products, Bartlesville, OK, USA) at 4200 rev min)1 for 1 min. The homogenized suspension was centrifuged at 15 000 g for 5 min. Nucleic acid was precipitated from 200 ll of the clear upper phase by adding it to 500 ll ethanol (96%) and frozen at )20C for at least 10 min. Following centrifugation at 15 000 g for 5 min, the alcohol was discarded and the DNA was re-dissolved in 50 ll Tris–EDTA buffer and stored at )20C. The NS1/ITS2 primer pair with internal transcribed spacer was used to amplify the 18S rDNA according to the method of White et al. (1990). The yeast DNA (1 ll) was amplified in a 50 ll reaction volume containing: 25 ll Taq DNA polymerase (Ready MixTM Taq PCR Reaction Mix, Sigma P4600), 5 ll of each primer (Sigma-Genosys, Haverhill, UK) and 14 ll water. Amplification was carried out in a Techne Progene PCR thermocycler (Techne, Duxford, UK). The programme consisted of the following steps: initial denaturation at 93C for 1 min, 30 cycles of denaturation at 93C for 1 min, annealing at 55C for 1 min, extension at 72C for 2 min and a final extension at 72C for 5 min. The PCR products (2Æ5 ll) were digested in 10 ll final volume with 0Æ5 ll (10 units) of each of four restriction enzymes (HaeIII, MspI, AluI and RsaI), 1 ll (10X) buffer and 6 ll water at 37C for 4 h. Fragments were separated in 1Æ5% agarose gel with a PCR marker (8 bands, 50–2000 bp, Sigma P-9577) as a molecular size standard. Gels were stained with ethidium bromide, photographed, scanned and used to identify yeast species against the database of Dlauchy et al. (1999). Treatment of apple juice with chitosan Apple juice (100 ml) was dispensed into 200 ml containers. Chitosan powder was added directly to the juice to give a



RESULTS

10 25°C 8 6 4 Viable count (log CFU ml–1)

final concentration of 0Æ05% and 0Æ1%. For challenge testing, Salm. Typhimurium or E. coli O157:H7 were inoculated at a level of 104 CFU ml)1. All treatments were prepared in duplicate. The juices were stored at 25 and 4C for 6–12 days and 20 days respectively. Juices containing no added chitosan were also inoculated with the bacteria as controls. Absolute controls containing no chitosan and no added pathogens were also prepared. Single samples were removed from each duplicate batch of juice periodically during incubation for viable counting and plating out in triplicate; therefore, mean counts for each time point were calculated from six replicate determinations.

875

2 0 4°C 10 8

The microflora of unprocessed apple juice Micro-organisms were isolated from three separate batches of unprocessed juice immediately upon arrival at the laboratory from the manufacturer (this was always within 24 h of pressing). Of the 542 isolates, 454 were identified as yeasts while the remaining colonies were identified as bacteria and moulds. The 28 moulds included species of Penicillium (39% of all isolates), Cladosporium, Mucor, Rhizopus, Botrytis and Alternaria. The majority (52) of the 60 bacterial strains were Gram-negative and included Serratia and Acinetobacter species. Of the 17 different groups of yeasts isolated from juice, the three most frequently isolated groups were identified, on the basis of colony morphology and biochemical profiling (API strips), as Saccharomyces cerevisiae (white, shiny colonies, peaked in centre), Kloeckera spp. (yellowish-white, shiny, round, flat colonies) and Candida spp. (matt white, conical, wrinkled colonies surrounded by ring of different coloured agar). Subsequent molecular work confirmed the identity of S. cerevisiae. Furthermore, Kloeckera spp. was identified as Kloeckera apiculata and Candida spp. as Metschnikowia pulcherrima, a teleomorph of Candida pulcherrima. A fourth group containing Pichia spp. was also often isolated and molecular analysis confirmed that this group contained at least two species: Pichia fermentans and Pichia kluyveri. Figure 1 illustrates the development of the total microflora (on PCA and OSA), total yeasts and moulds (on TGYEA and DG18), lactic acid bacteria (on MRSA), acetic acid bacteria (on EBA) and two yeast species (S. cerevisiae and K. apiculata, identified on the basis of their morphological features and S18 rDNA) in apple juice (pH 3Æ5) stored at 25 and 4C over a period of 6 and 20 days, respectively. The results show that at 25C, total viable counts increased in the first 2–3 days to a level of ca 8 log CFU ml)1, where they remained up to day 6. The yeast counts reflected the total counts. Saccharomyces cerevisiae and K. apiculata grew at a similar rate to the total

6 4 2 0 0

5

10

15

20

Time (days)

Fig. 1 Survival and/or growth of bacteria, yeasts and moulds in apple juice (pH 3Æ5) at 25 and 4C. Data points represent means of six replicate counts ±0Æ25 log CFU ml)1 determined from duplicate batches of juice: total counts on PCA (d), total counts on OSA (s), yeast counts on TGYEA (.), yeast counts on DG18 (,), moulds on TGYEA (j), moulds on DG18 ((), EBA count (¤), MRSA count ()), Saccharomyces cerevisiae (- - s - -) and Kloeckera apiculata (- - , - -). Sampling was discontinued on day 6 at 25C as the juices had visibly spoiled

yeast flora but absolute numbers were generally ca 0Æ5– 1Æ0 log CFU ml)1 lower. Metschnikowia pulcherrima was detected at a level of 2Æ3 log CFU ml)1 on day 0 but was not detectable thereafter at either 25 or 4C. Mould counts were at ca 3 log CFU ml)1 in the first 2 days, decreasing to ca 2 log CFU ml)1 for the remainder of the storage period. Lactic acid bacteria and acetic acid bacteria were detected at a level of ca 3 log CFU ml)1 on day 0 but numbers decreased to ca 2 log CFU ml)1 after 2 days at 25C. After 3 days at 25C, both MRSA and EBA plates were overgrown with yeasts thereby rendering the lactic acid and acetic acid bacteria uncountable. Growth and survival of the same organisms at 4C were similar relative to each other but maximum numbers were reached more slowly, i.e. after 17–18 days, compared with 2–3 days at 25C (Fig. 1). Although a few colonies were isolated on XLD, Oxford and MacConkey agars following enrichment of the apple



juice on day 0 of storage, their appearance was atypical for Salmonella spp., Listeria spp. and E. coli O157:H7 respectively. Furthermore, the results of substrate assimilation tests (API) suggested that these colonies were E. coli, Enterobacter amnigenus and Enterobacter agglomerans. No colonies were detected on the selective agars after storage of the apple juice at either temperature. These results showed that the few Enterobacteriaceae, which may have been present at the start of storage, died off rapidly. No heat-resistant fungi or spore-forming bacteria (including A. acidoterrestris) were detected. The selective action of chitosan on yeasts As shown in Fig. 2, S. cerevisiae and K. apiculata dominated the yeast flora during spoilage of apple juice (pH 3Æ2) at 25C, reaching levels of 7–8 log CFU ml)1 within 4 days; M. pulcherrima and Pichia spp. grew at a slower rate and

8 Control 6

Viable count (log CFU ml–1)

4

2

0 Chitosan (0·05%)

8

6

4

2

0 0

2

4

6

8

10

12

Time (days)

Fig. 2 Survival and/or growth of Saccharomyces cerevisiae (¤), Kloeckera apiculata ((), Metschnikowia pulcherrima (m) and Pichia spp. (s) in apple juice (pH 3Æ2) untreated or treated with 0Æ05% chitosan and stored at 25C. Data points represent means of six replicate counts ±0Æ25 log CFU ml)1 determined from duplicate batches of juice. Sampling of the control batch was discontinued on day 7 as the juices had visibly spoiled

reached a level of ca 5 log CFU ml)1 in 7 days. In the presence of 0Æ05% chitosan, M. pulcherrima and K. apiculata were inactivated to levels near or below the detection limit of the assay (1Æ0 log CFU ml)1) for the duration of the experiment (12 days). However, following an initial drop in counts of ca 1 log CFU ml)1, S. cerevisiae and Pichia spp. multiplied slowly in the presence of chitosan, reaching the same numbers in 7 days as those present in the controls within 2 days (Fig. 2). Survival of enteric pathogens in apple juice treated with chitosan As shown in Fig. 3, the total viable counts in juices treated with 0Æ05 and 0Æ1% chitosan were similar to those in the untreated controls at both 25 and 4C for 12 and 20 days respectively. However, in the presence of chitosan, yeast counts were 3–5 log CFU ml)1 lower than in the controls for 4 days at 25C and ca 2–3 log CFU ml)1 lower for up to 10 days at 4C. The difference between the total and yeast counts in chitosan-treated juices was accounted for by a predominance of lactic acid bacteria. However, it was not possible to enumerate the lactic acid bacteria because of overgrowth of MRSA plates by yeasts. The initial effect of chitosan on yeasts was biocidal with numbers reduced by 1–2 log CFU ml)1 immediately after exposure, followed by slow growth. The number of yeasts remained below the levels in the control at both storage temperatures up to 7 days at 25C (Fig. 3a) and 15 days at 4C. Therefore, chitosan delayed the spoilage of apple juice by yeasts at both ambient and chill storage temperatures (Fig. 3). When inoculated into raw apple juice, E. coli O157:H7 survived for up to 1 day at 25C and 3 days at 4C (Fig. 3a,b; Table 1). In the presence of chitosan, E. coli O157:H7 was detectable up to 2 and 5 days at 25 and 4C respectively (Table 1). Although there was no growth on directly plated selective agar (SMAC), cells could be recovered using enrichment, indicating that injured organisms were present in chitosan-treated juices. By knocking out a portion of the yeast population selectively, chitosan may have enabled the injured Gram-negatives to survive for longer than they would otherwise have done. When inoculated into raw apple juice, Salm. Typhimurium survived for up to 1 day at 25C and 8 days at 4C (Fig. 4; Table 1). Although Enterobacteriaceae and salmonellae on BGA died off more rapidly in the presence of chitosan than in the untreated controls (Fig. 4), Salmonella was detectable for the same length of time as in the unsupplemented controls (with the sole exception of the sample treated with 0Æ1% chitosan and stored at 25C for 2 days; Table 1), suggesting that the survival of Salm. Typhimurium was not substantially affected by the


877


(a)

Total and mould counts

10

6

Enterobacteriaceae

5

8

4 6 Viable count (log CFU ml–1)

3 4

2

2

1 0

0 10

Yeast counts

E. coli O157:H7

6

8

5

6

4 3

4

2 2

1

0

0 0

4

6

8

10

12

0

2

4

6

Time (days) (b) 10

Total and mould count

s

6

Enterobacteriaceae

5

8

4 6 3 Viable count (log CFU ml–1)

Fig. 3 (a) Survival and/or growth of total microflora, yeasts, moulds, Enterobacteriaceae and Escherichia coli in apple juice untreated and treated with chitosan and stored at 25C for 12 days. Data points represent means of six replicate counts ±0Æ25 log CFU ml)1 determined from duplicate batches of juice: control with no additions (d), control inoculated with E. coli O157:H7 (,), 0Æ05% chitosan (j) and 0Æ1% chitosan ()). The pH of the juice on day 0 was 3Æ7; on day 12, the pH was 3Æ5 in the untreated controls and 3Æ1 in chitosan-treated juices. Sampling of the control batches was discontinued on day 7 as the juices had visibly spoiled. (b) Survival and/or growth of total microflora, yeasts, moulds, Enterobacteriaceae and E. coli in apple juice untreated and treated with chitosan and stored at 4C for 20 days. Data points represent means of six replicate counts ±0Æ25 log CFU ml)1 determined from duplicate batches of juice: control with no additions (d), control inoculated with E. coli O157:H7 (,), 0Æ05% chitosan (j) and 0Æ1% chitosan ()). The pH of the juice on day 0 was 3Æ7; on day 20, the pH was 3Æ5 in the untreated controls and 3Æ1 in chitosan-treated juices

2

4

2

2

1 0

0 Yeast counts

10

E. coli O157:H7

6

8

5

6

4 3

4

2 2

1

0

0 0

presence of chitosan. Although small differences in counts between samples plated directly onto selective agars and those plated using the thin agar layer technique (Kang and Fung 2000) on day 0 were observed, these differences disappeared after 24 h of storage of the apple juice at either temperature.

5

10

15

20

0

2

4

6

8

10 12 14

Time (days)

DISCUSSION As expected, yeasts constituted the predominant microflora of unprocessed apple juice in this study (Fig. 1). The viable counts obtained within 1–2 days of pressing were consistent with levels expected (3–5 log CFU ml)1) from juice prepared



Table 1 Presence/absence of Escherichia coli and Salmonella in apple juice as shown in Figs 3 and 4. Results were determined in duplicate Storage temperature 25C

4C

Sampling days

E. coli Control, no additions Control, E. coli O157:H7 inoculated Chitosan, 0Æ05% Chitosan, 0Æ1% Salmonella Control, no additions Control, E. coli O157:H7 inoculated Chitosan, 0Æ05% Chitosan, 0Æ1%

0

1

2

4

0

3

5

8

12

) +

) +

) )

) )

) +

) +

) )

) )

ND ND

+ +

+ +

+ +

) )

+ +

+ +

+ +

) )

ND ND

) +

) +

) )

) )

) +

) +

) +

) +

) )

+ +

+ +

) +

) )

+ +

+ +

+ +

+ +

) )

ND, not determined.

from healthy fruit (Stratford et al. 2000). Moulds, acetic acid bacteria and lactic acid bacteria were detected at a level of ca 3– 4 log CFU ml)1. Very few representatives of the Enterobac-

25°C

4°C 6

6

Enterobacteriaceae

Enterobacteriaceae

Viable count (log CFU ml–1)

teriaceae were isolated following enrichment and these tended to die-off rapidly. Salmonella spp., E. coli O157:H7 and Listeria spp. or the spore-forming spoilage organism A. acidoterrestris were not detected. These results suggest that good processing and hygiene practices had been used during pressing (Deak and Beuchat 1996; Stratford et al. 2000). The pH of apple juice can vary between pH 3Æ2 and 4Æ4 (Stratford et al. 2000). The pH of the juice used in this study was at the lower end of this range (3Æ2–3Æ7). Nevertheless, even at low pH, both E. coli O157:H7 and Salm. Typhimurium had the capability of surviving for several days, especially when stored at refrigeration temperatures. These findings agree with those reported elsewhere (Zhao et al. 1993; Miller and Kaspar 1994; Leyer et al. 1995; Roering et al. 1999; McClure and Hall 2000; Roller and Covill 2000; Roller et al. 2002; Sagoo et al. 2002). The acidic nature of apple juice does not ensure its safety as some pathogens may survive for extended periods of time and cause disease (CDC 1975, 1997). The biocidal action of chitosan, tested against individual yeast and bacterial cultures in vitro, has been reported previously (Roller and Covill 1999; Shahidi et al. 1999; Roller 2003b). The action of chitosan on the mixed microflora of some foods such as ground beef, sausage and mayonnaise has also been reported (Roller and Covill 2000; Roller et al. 2002; Sagoo et al. 2002). However, this is the

5

5

4

4

3

3

2

2

1

1

0

0 Salmonella

6

Salmonella 6

(BGA)

5

5

4

4

3

3

2

2

1

1

0

(BGA)

0 0

2

4

6

8

10

12

14

16

0

Time (days)

1

2

3

4

Fig. 4 Survival of Enterobacteriaceae and Salmonella (on BGA) in apple juice (starting pH 3Æ5) untreated and treated with chitosan at 25C for 4 days and 4C for 16 days. Data points represent means of six replicate counts ±0Æ25 log CFU ml)1 determined from duplicate batches of juice: control with no additions (d), control inoculated with Salm. Typhimurium (,), 0Æ05% chitosan (j) and 0Æ1% chitosan ())



first report of the selective action of chitosan on four species of yeasts occurring naturally in the mixed microflora of apple juice. The results (Fig. 2) demonstrate that the presence of as little as 0Æ05% chitosan inactivated M. pulcherrima and K. apiculata and prevented their re-growth for at least 12 days at 25C. However, the growth of S. cerevisiae and Pichia spp. was merely delayed in the presence of chitosan. Therefore, the action of chitosan on yeasts was highly species specific. While the partial inactivation of the yeast microflora could be advantageous to the fruit juice processing industry in preventing spoilage and extending the shelf life of apple juice, it was important to investigate the implications of reduced yeast numbers for the survival of pathogenic bacteria. The results in Fig. 3 and Table 1 suggest that the presence of chitosan in apple juice containing a natural, mixed microflora extended the ability of E. coli O157:H7 to survive thereby potentially compromising the safety of the product. Furthermore, it is known that direct plating on selective media following exposure to physical or chemical stresses (including exposure to chitosan) can lead to gross underestimation of viable counts by as much as 3–4 log CFU ml)1 (Brashears et al. 2001; Helander et al. 2001). Although a thin agar layer method (Kang and Fung 2000) and enrichment techniques (Roberts et al. 1995) were used in this study to allow some resuscitation of the inoculated pathogens, it is possible that the results shown in Fig. 3 represent an underestimate of numbers. Furthermore, it is possible that a larger sample volume (25 ml instead of 2Æ5 ml) in the pre-enrichment step would have allowed the resuscitation and recovery of an even greater number of E. coli O157:H7. Although the survival of Salm. Typhimurium (Fig. 4 and Table 1) appeared generally unaffected by chitosan, it is notable that the starting pH of the juice was slightly lower (pH 3Æ5) than that used in the E. coli challenge experiment (pH 3Æ7). All too often in the literature, scientists focus on the behaviour of a pure culture of a selected micro-organism in a laboratory medium or food that has been sterilized to remove the background flora (Fleet 1999). Although the results obtained from such studies are valuable in providing basic physiological information about the specific strain, they give no indication of how that organism might behave in the mixed flora normally present in foods. Yet, such information is essential if improvements in our ability to predict and control micro-organisms in foods are to be made in the future. The study presented here is a case in point: the elimination of a harmless group of micro-organisms (spoilage yeasts) has created a niche for and enhanced the survival of a pathogenic bacterium. This, in turn, could have profound implications for the safety of the product. In this study, we set out to develop new measures for controlling spoilage and pathogenic micro-organisms in

879

unpasteurized apple juice using the natural antimicrobial agent chitosan. While better control over the proliferation of spoilage yeasts was achieved, chitosan treatment resulted in the opposite effect with respect to the survival of E. coli O157:H7. As shown in this study, food scientists must exercise caution when applying novel agents such as chitosan to foods containing a mixed microflora to ensure that their selective action does not create a new public health hazard. ACKNOWLEDGEMENTS We would like to thank D. Dlauchy for assistance and advice in the molecular identification of yeasts. The financial support of the European Commission in the form of a Marie Curie Individual Fellowship awarded to author Kisko´ (QLK1-CT-2000–51126) to undertake this work in author Roller’s laboratory in London is also acknowledged. REFERENCES Brashears, M.M., Amezquita, A. and Stratton, J. (2001) Validation of methods used to recover Escherichia coli O157:H7 and Salmonella spp. subjected to stress conditions. Journal of Food Protection 64, 1466–1471. CDC (1975) Salmonella Typhimurium outbreak traced to commercial apple cider – New Jersey. Morbidity and Mortality Weekly Report 24, 87–88. CDC (1997) Outbreaks of Escherichia coli O157:H7 infection and criptosporidiosis associated with drinking unpasteurized apple cider – Connecticut and New York, October 1996. Journal of the American Medical Association 277, 781–787. Deak, T. and Beuchat, L.R. (1996) Handbook of Food Spoilage Yeasts. 210 pp. Boca Raton, FL: CRC Press. Dillon, V.M. and Board, R.G. (1994) Natural Antimicrobial Systems and Food Preservation. 328 pp. Wallingford, UK: CAB International. Dlauchy, D., Tornai-Lehoczki, J. and Peter, G. (1999) Restriction enzyme analysis of PCR amplified rDNA as a taxonomic tool in yeast identification. Systematic and Applied Microbiology 22, 445– 453. Eiroa, M.N.U., Junqueira, V.C.A. and Schmidt, F.L. (1999) Alicyclobacillus in orange juice: occurrence and heat resistance of spores. Journal of Food Protection 62, 883–886. Fleet, G.H. (1999) Microorganisms in food ecosystems. International Journal of Food Microbiology 50, 101–117. Food and Drug Administration (FDA) (1998) Food labelling: warning and notice statements; labelling of juice products. Federal Register 63, 20486–20493. Giese, J. (2002) FDA issues juice guidelines. Food Technology 56, 24. Helander, I.M., Nurmiaho-Lassila, E.-L., Ahvenainen, R., Rhoades, J. and Roller, S. (2001) Chitosan disrupts the barrier properties of the outer membrane of Gram-negative bacteria. International Journal of Food Microbiology 71, 235–244. Kang, D.H. and Fung, D.Y.C. (2000) Application of thin agar layer method for recovery of injured Salmonella typhimurium. International Journal of Food Microbiology 54, 127–132.



Leyer, G.J., Wang, L.L. and Johnson, E.A. (1995) Acid adaptation of Escherichia coli O157:H7 increases survival in acidic foods. Applied and Environmental Microbiology 61, 3752–3755. McClure, P.J. and Hall, S. (2000) Survival of Escherichia coli in foods. Journal of Applied Microbiology Supplement 88, 61S–70S. Miller, L.G. and Kaspar, C.W. (1994) Escherichia coli O157:H7 acid tolerance and survival in apple cider. Journal of Food Protection 57, 460–464. Naidu, A.S., ed. (2000) Natural Food Antimicrobial Systems. 818 pp. Boca Raton, FL: CRC Press. Pettipher, G.L., Osmundson, M.E. and Murphy, J.M. (1997) Methods for the detection and enumeration of Alicyclobacillus acidoterrestris and investigation of growth and production of taint in fruit and fruit juice-containing drinks. Letters in Applied Microbiology 24, 185–189. Pitt, J.I. and Hocking, A.D. (1997) Methods for isolation, enumeration and identification. In Fungi and Food Spoilage, 2nd edn. pp. 35–85. Gaithersburg, MD: Aspen Publishers. Rhoades, J. and Roller, S. (2000) Antimicrobial actions of degraded and native chitosan against spoilage organisms in laboratory media and foods. Applied and Environmental Microbiology 66, 80–86. Roberts, D. Hopper, W. and Greenwood, M. (1995) Practical Food Microbiology. Forum, Somerset, UK: Public Health Laboratory Service (PHLS). Roering, A.M., Luchansky, J.B., Ihnot, A.M., Ansay, S.E., Kaspar, C. and Ingham, S.C. (1999) Comparative survival of Salmonella typhimurium DT104, Listeria monocytogenes, and Escherichia coli 0157:H7 in preservative-free apple cider and simulated gastric juice. International Journal of Food Microbiology 46, 263–269. Roller, S., ed. (2003a) Natural Antimicrobials for the Minimal Processing of Foods. 256 pp. Cambridge, UK: Woodhead Publishing Ltd. Roller, S. (2003b) Chitosan: new food preservative or laboratory curiosity? In Natural Antimicrobials for the Minimal Processing of Foods ed. Roller, S., pp. 158–175. Cambridge, UK: Woodhead Publishing Ltd. Roller, S. and Covill, N. (1999) The antifungal properties of chitosan in laboratory media and apple juice. International Journal of Food Microbiology 47, 67–77.

Roller, S. and Covill, N. (2000) The antimicrobial properties of chitosan in mayonnaise and mayonnaise-based shrimp salads. Journal of Food Protection 63, 202–209. Roller, S., Sagoo, S., Board, R., O’Mohony, T., Caplice, E., Fitzgerald, G., Fogden, M., Owen, M. et al. (2002) Novel combinations of chitosan, carnocin and sulphite for the preservation of chilled pork sausages. Meat Science 62, 165–177. Sagoo, S., Board, R. and Roller, S. (2002) Chitosan inhibits growth of spoilage microorganisms in chilled pork products. Food Microbiology 19, 175–182. Shahidi, F., Arachchi, J.K.M. and Jeon, Y. (1999) Food applications of chitin and chitosans. Trends in Food Science and Technology 10, 37–51. Splittstoesser, D.F., Churey, J.J. and Lee, C.Y. (1994) Growth characteristics of aciduric sporeforming bacilli isolated from fruit juices. Journal of Food Protection 57, 1080–1083. Strand, S.P., Vandvik, M.S., Va˚rum, K.M. and Østgaard, K. (2001) Screening of chitosans and conditions for bacterial flocculation. Biomacromolecules 2, 126–133. Stratford, M., Hofman, P.D. and Cole, M.B. (2000) Fruit juices, fruit drinks and soft drinks. In The Microbiological Safety and Quality of Food, Vol. I ed. Lund, B.M., Baird-Parker, T.C. and Gould, G.W. pp. 836–869. Gaithersburg: Aspen Publishers Inc. Uljas, H.E. and Ingham, S.C. (1999) Combinations of intervention treatments resulting in 5-log unit reductions in numbers of Escherichia coli 0157:H7 and Salmonella typhimurium DT104 organisms in apple cider. Applied and Environmental Microbiology 65, 1924–1929. White, T.J., Bruns, T. Lee, S. and Taylor, J. (1990) Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In PCR Protocols – Guide to Methods and Applications ed. Innis, M.A., pp. 315–322. New York: Academic Press Inc. Zhao, T., Doyle, M.P. and Besser, R.E. (1993) Fate of enterohemorrhagic Escherichia coli 0157:H7 in apple cider with and without preservatives. Applied Environmental Microbiology 59, 2526– 2530.
