258

Journal of Food Protection, Vol. 73, No. 2, 2010, Pages 258–265 Copyright G, International Association for Food Protection

Effect of Organic Acids and Marination Ingredients on the Survival of Campylobacter jejuni on

TINA BIRK,1,2 ANNE CHRISTINE GRØNLUND,1 BJARKE BAK CHRISTENSEN,1 SUSANNE KNØCHEL,2 KRISTIN LOHSE,1 AND HANNE ROSENQUIST1*

1Department of Microbiology and Risk Assessment, National Food Institute, Technical University of Denmark, Mørkhøj Bygade 19, 2860 Søborg, Denmark; and 2Department of Food Science, Faculty of Life Sciences, Copenhagen University, Rolighedsvej 30, 1958 Frederiksberg C, Denmark

MS 09-281: Received 29 June 2009/Accepted 12 October 2009 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/73/2/258/1679931/0362-028x-73_2_258.pdf by guest on 24 September 2021

ABSTRACT The aim of this study was to determine whether marination of chicken meat in different food ingredients can be used to reduce populations of Campylobacter jejuni. C. jejuni strains were exposed to different organic acids (tartaric, acetic, lactic, malic, and citric acids) and food marinating ingredients at 4uC in broth and on chicken meat. The organic acids (0.5%) reduced populations of C. jejuni in broth (chicken juice and brain heart broth) by 4 to 6 log units (after 24 h); tartaric acid was the most efficient treatment. Large strain variation was observed among 14 C. jejuni isolates inoculated in brain heart infusion broth containing 0.3% tartaric acid. On chicken meat medallions, reductions of C. jejuni were 0.5 to 2 log units when tartaric acid solutions (2, 4, 6, and 10%) were spread onto the meat. Analysis of acidic food ingredient (e.g., , lemon juice, syrup, and soya sauce) revealed that such ingredients reduced counts of C. jejuni by at least 0.8 log units on meat medallions. Three low pH marinades (pH , 3) based on pomegranate syrup, lemon juice, and vinegar were prepared. When applied to whole filets, these marinades resulted in a of approximately 1.2 log units after 3 days of storage. Taste evaluations of chicken meat that had been marinated and then fried were graded positively for flavor and texture. Thus, success was achieved in creating a marinade with an acceptable taste that reduced the counts of C. jejuni.

Campylobacter species, in particular Campylobacter of C. jejuni, respectively (20). Thus, the effects of organic jejuni, are some of the leading causes of zoonotic enteric acids on Campylobacter may differ depending on whether infections worldwide. In the European Union, Campylo- testing is performed in broth or in a food matrix. For bacter has been the most commonly reported gastrointes- example, broiler meat may have a protective effect due to tinal bacterial pathogen in humans since 2004 (7). In the buffering capacity of the meat (11, 16) and/or the Denmark, despite several national initiatives to reduce the decreased accessibility of the . The limited effect of occurrence of this organism in the farm-to-fork chain (17), lactic acid on broiler meat has resulted in a rejection of the the number of campylobacteriosis cases was 71 per 100,000 use of lactic acid as a decontamination compound during inhabitants in 2007 (1). Although there may be several processing by the European Authority (6). sources of Campylobacter, campylobacteriosis is mainly Demand for marinated broiler meat is increasing in believed to be associated with the consumption and/or Denmark and many other countries. Although flavor handling of meat, especially fresh broiler meat (10). development, tenderization, and consumer convenience Thus, new intervention strategies for broilers and broiler may be the main drivers behind this demand, marination meat are needed in all parts of the food chain. Although also may be an antimicrobial treatment (5) and an additional most emphasis in Denmark has been on biosecurity and method of reducing populations of Campylobacter in meat good hygiene practices during rearing and processing of from Campylobacter-positive broiler flocks. However, a broilers, complete absence at the farm level cannot be study employing inoculated chicken drumsticks marinated ensured (9), and much poultry meat may be imported from in a sauce adjusted to pH 4.5 with lactic and acetic acid sources of uncertain status. Thus, reduction of Campylo- resulted in no effect (15), whereas submerging meat bacter in the steps just prior to consumer handling also is medallions inoculated with C. jejuni in red wine at 4uC important. reduced C. jejuni by approximately 3.5 log units (3). Inactivation of C. jejuni in acidified foods has not been The overall purpose of this study was to identify strains well studied. Although counts of C. jejuni were reduced by with different sensitivities to organic acids and to examine 2.1 log units after exposure to 1% lactic acid for 5 min in the effect of different marination ingredients, including the broth at low temperature (20), chemical decontamination of effect of three low pH marinades on the C. jejuni population broilers dipped in 0.50% acetic or lactic acid for 10 min at at refrigeration temperature. The antibacterial effect of 5uC had little effect, with only a 0.07- and 0.08-log decrease different organic acids against C. jejuni NCTC 11168 was determined in brain heart infusion (BHI) broth and chicken * Author for correspondence. Tel: z45 35 88 70 80; E-mail: juice. The strain variation among 14 C. jejuni isolates was [email protected]. then investigated based on exposure to mild tartaric acid. J. Food Prot., Vol. 73, No. 2 REDUCTION OF CAMPYLOBACTER BY MARINATION INGREDIENTS 259

Different food ingredients and marinades also were tested initial level of C. jejuni NCTC 11168 in chicken juice and BHI for their effects on C. jejuni inoculated onto broiler meat. broth was 7.07 (¡0.07) log CFU/ml. Two replicate experiments were conducted. MATERIALS AND METHODS Strain variation in BHI broth. For each of the 14 strains, Bacterial strains. Fifteen strains of C. jejuni were tested. The 4 ml of inoculum prepared in BHI broth was transferred into sterile sequenced clinical human isolate C. jejuni NCTC 11168 from the tubes, and 60 mlofa20% (wt/vol) stock solution of tartaric acid National Collection of Type Cultures was used in the screening of was subsequently added. The tubes were immediately stored at organic acids. The variation of response to tartaric acid in broth 4uC. After 0, 3, 7.5, 23, 29, and 46 h of storage, samples were was tested on 14 strains: three chicken isolates (992, 1099, and collected to determine populations of C. jejuni. The mean (¡SD) 835-770B) and five turkey isolates (305A, 355A, 4-3, 327, and initial level of all the strains tested was 6.96 (¡0.16) log CFU/ml. 5-1) (Department of Food Science, Copenhagen University, Two replicate experiments were conducted. Frederiksberg) and three matched strain pairs, where each pair includes a chicken isolate and a human isolate of the same Inhibitory effect of tartaric acid and food ingredients on serotype. The matched strains, which have been isolated and meat medallions. The inhibitory effects of pure tartaric acid Downloaded from http://meridian.allenpress.com/jfp/article-pdf/73/2/258/1679931/0362-028x-73_2_258.pdf by guest on 24 September 2021 characterized in a previous study (14), represent the most common solutions alone and in combination with NaCl and ethanol (C. serotypes associated with human infections in Denmark (serotypes jejuni strains 305A and 327) and different low pH food ingredients 2, 1.44, and 4-complex [4C]) (1). For the experiments with meat (Table 1) (C. jejuni strains 4C and 305A) were tested. A 50-ml medallions treated with tartaric acid, strains 305A and 327 were aliquot of inoculum prepared in chicken juice was added to meat used. For testing the effect of other marination ingredients, strains medallions by pipetting. The meat medallions were left at ambient 305A and 4C were used. Strain 305A also was used for room temperature for 10 min, and then 200 ml of a tartaric acid experiments with whole breast filets treated with various food solution or a food ingredient was spread onto the medallions (see ingredients and marinades. Table 1 for details). The samples were immediately stored at 4uC. Two medallions treated with each marinade combination were Preparation of inocula. Strains were maintained at 280uCin collected for analysis on days 0, 1, 2, and 3. Two independent BHI broth (CM225, Oxoid, Basingstoke, UK) containing 15% experiments were performed. The mean (¡SD) initial levels of glycerol. During propagation and growth, all plates were incubated strains 305A, 327, and 4C were 5.85 (¡0.27), 5.97 (¡0.13), and at 37uC in sealed gas jars under microaerobic conditions (6% O , 2 6.21 (¡0.17) log CFU/ml, respectively. 7% H2,7% CO2, and 80% N2). Cultures were prepared by growing the strains for 3 to 4 days on blood agar base no. 2 Inhibitory effect of marinades on breast filets. Three low (CM271, Oxoid) containing 5% horse blood. One loopful of each pH (pH , 3) marinades (Table 1) were prepared: marinades 1 culture was subsequently streaked onto new blood agar base no. 2 through 3 for which the main ingredients were pomegranate syrup, plates, which were incubated for 24 h. The cells were harvested lemon juice, and white wine vinegar, respectively. A 200-ml aliquot with 2 ml of phosphate-buffered saline (BR0014, Oxoid). The of strain 305A inoculum prepared in chicken juice was spread on inoculum was diluted based on the observation that the optical whole breast filets (120 to 140 g) with a sterile spatula, and the density at 600 nm of 0.1 corresponded to approximately 8 log CFU/ml. Subsequently, the inocula were diluted to approximately filets were left at ambient temperature for 30 min to allow 7 log CFU/ml in either chicken juice or BHI broth and were ready diffusion of cells into the meat. Each breast filet was placed in a for use. stomacher bag, and 8 g of marinade per filet was added and massaged into the meat by hand for 5 min. The inoculated breast Preparation of chicken juice. Chicken juice was prepared as filets with marinade were then placed in sterile plastic dishes previously described (2). The chicken juice was obtained from covered with a plastic film and immediately stored at 4uC. frozen retail broilers with giblets. Microbiological analyses and sampling. Counts (CFU) of Preparation of broiler meat. Breast filets from frozen thermotolerant Campylobacter in chicken juice and/or BHI broth Campylobacter-free and Salmonella-free broilers (Lantma¨nnen were determined by placing 100 ml of sample from the respective Danpo A/S, Give, Denmark) were used. Absence of Campylobac- tubes into 900 ml of maximum recovery diluent (MRD; CM733, ter was verified on Abeyta-Hunt-Bark (AHB) agar plates Oxoid). A serial 10-fold dilution was performed, and 10 ml of the containing 1% triphenyl-tetrazoliumchloride (18). To determine appropriate dilutions were subsequently spotted onto three blood the antibacterial effect of three low pH marinades, thawed whole agar base no. 2 plates containing 5% horse blood. All plates were breast filets were used. To screen for the effect of tartaric acid and incubated under microaerobic conditions for 24 to 48 h at 37uC. food ingredients, meat medallions were used because they were Counts of thermotolerant Campylobacter on medallions and relatively easy to handle. Meat medallions of 9.6 cm2 were breast filets were determined by placing two arbitrary meat aseptically cut using a plug center bit from chicken breast filets, medallions or two breast filets in each stomacher bag and adding and each piece of broiler meat was placed on gauze in a petri 10 ml or 50 ml of MRD, respectively. Meat medallions were dish. stomached for 2 min. Breast filets were massaged by hand for 2 min. The homogenized samples were diluted 10-fold in MRD, and Inhibitory effect of organic acids in chicken juice and BHI 10 ml of the dilution was spotted in triplicate onto Campylobacter broth. Four-milliliter aliquots of the inoculum (NCTC 11168) selective AHB agar plates containing 1% triphenyltetrazoliumchlor- prepared in chicken juice or BHI broth were transferred into sterile ide (18). All plates were incubated under microaerobic conditions tubes, and 100 ml from a 20% stock solution of acetic acid for 24 to 48 h at 37uC. The Campylobacter colonies on the plates (vol/vol), lactic acid (vol/vol), citric acid (wt/vol), tartaric acid were counted, and the CFU per meat piece was estimated. (wt/vol), or malic acid (wt/vol) was pipetted into the tubes, resulting in an organic acid concentration of 0.5%. The tubes were pH measurement. pH values of all solutions and food immediately stored at 4uC. The mean (¡standard deviation [SD]) ingredients were measured before these additives were spread onto 260 BIRK ET AL. J. Food Prot., Vol. 73, No. 2

TABLE 1. Characteristics of different types of marinades tested drying out during the process, approximately four times as on chicken meat at 4uC much marinade was used as was used in the microbiological tests. The frying time was approximately 10 min. The meat was Meata Text figure Marinade or ingredients Percentage pH evaluated based on the following categories: acidity-sweetness M 2A Tartaric acid 2 2.7 balance, spiciness, texture, and overall impression. M 2A Tartaric acid 4 2.5 M 2A, 2B Tartaric acid 6 2.3 Statistical methods and data analysis. C. jejuni counts were M 2A Tartaric acid 10 2.2 log transformed to approximate a normal distribution. Statistical M 2B Ethanol 8 4.4 analysis was performed by using SAS statistical software (version M 2B NaCl 10 4.5 9.1, SAS Institute, Cary, NC). Results with P values ,0.05 were M 2B Tartaric acid 6 2.6 considered significant. To test for significant C. jejuni strain Ethanol 8 differences in the reduction of counts when exposed to 0.3% M 2B Tartaric acid 6 1.1 (wt/vol) tartaric acid in chicken juice and the variation among the Ethanol 8 five organic acids in chicken juice versus BHI broth, analyses of Downloaded from http://meridian.allenpress.com/jfp/article-pdf/73/2/258/1679931/0362-028x-73_2_258.pdf by guest on 24 September 2021 NaCl 10 variance were used with the PROC MIXED procedure for repeated M 3 Honey 50 3.7 measurements. To compare the effect of different food ingredients, Water 50 tartaric acid concentrations, ethanol, and NaCl on C. jejuni on M 3 powder 66.7 4.9 broiler meat medallions and the effects of the three marinades on Water 33.3 breast filets, the PROC GLM procedure was applied. M 3 Pomegranate syrup 100 1.7 M 3 Red wine 100 3.7 RESULTS M 3 Soya sauce 100 4.3 Inhibitory effect of organic acids in chicken juice M 3 Lemon juice 100 2.7 and BHI broth. The reductions in the level of C. jejuni M 3 Lemon juice 50 3.0 Red wine 50 NCTC 11168 in chicken juice and BHI broth after exposure M 3 Lemon juice 50 3.3 to acetic acid, citric acid, tartaric acid, lactic acid, and malic Red wine 25 acid are illustrated in Figure 1. Table 2 lists pH values for all Soya sauce 25 solutions used and the percentages of undissociated acid. M 3 White wine vinegar 100 3.4 Initially, the pH of chicken juice (pH 6.6) was lower than for M 3 White wine vinegar 50 3.4 BHI broth (pH 7.1), and this situation remained after organic Red wine 50 acids were added to a final concentration of 0.5% (Table 2). M 3 White wine vinegar 25 3.5 The lowest pH value (3.85) was observed for 0.5% (wt/vol) Red wine 50 tartaric acid in chicken juice and the highest pH value was Soya sauce 25 measured in BHI broth with 0.5 (vol/vol) acetic acid (4.61). WF 4 Marinade 1: 2.45 % Pomegranate syrup 45.5 All the organic acids reduced C. jejuni very efficiently, Water 23 causing a 4- to 6-log reduction in after 24 h of exposure at Honey 17 4uC, although the reduction differed depending on the Mustard powder 11.5 medium and acid used. The most pronounced effect, NaCl 2 approximately 6 log CFU/ml after 6 h of exposure, was Pepper 1 observed with 0.5% (wt/vol) tartaric acid in chicken juice, WF 4 Marinade 2: 2.10 which also had the lowest pH (Fig. 1). When exposed to Lemon juice 73 citric acid, more C. jejuni survived in BHI broth than in Honey 18 chicken juice (P ~ 0.003), whereas the opposite effect was 6 ~ NaCl 2 observed with acetic acid (P 0.003). Pepper 1 WF 4 Marinade 3: 2.87 Strain variation. After testing the reducing effect of White wine vinegar 52 different organic acids on a single strain (C. jejuni NCTC Sugar 24 11168), the response variation among 14 other C. jejuni Onion 18.5 strains to tartaric acid stress was investigated in BHI broth. Estragon 2.5 Tartaric acid was selected for this experiment because it is NaCl 2 one of the main acids in wine and it had a strong reducing Pepper 1 effect in the initial experiments. Table 3 shows the effect of a WF, whole chicken filet; M, chicken medallion. 0.3% (wt/vol) tartaric acid after 29 and 46 h of exposure. The diversity among strains was most pronounced after 29 h of exposure, when counts of the most sensitive strain, C. the meat. After spreading, the surface pH of the meat was measured jejuni 327, was reduced by 4.48 log CFU/ml compared with after 5 to 10 min and after 1, 2, and 3 days of storage. the most robust strain, C. jejuni 305A, which was reduced Sensory evaluation of marinated prepared meat. An by 1.59 log CFU/ml, a difference of nearly 3 log CFU/ml. untrained panel of 12 people was used to evaluate the meat. After 46 h of exposure, all strains were reduced more than Uninoculated chicken breast filet was cut into small pieces and 3.5 log CFU/ml. At this point, strain 835-770B was the most marinated in the refrigerator overnight. To prevent the meat from sensitive and strain 355 was the most robust. J. Food Prot., Vol. 73, No. 2 REDUCTION OF CAMPYLOBACTER BY MARINATION INGREDIENTS 261 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/73/2/258/1679931/0362-028x-73_2_258.pdf by guest on 24 September 2021

FIGURE 1. Effect of 0.5% acetic acid (vol/vol) (A), citric acid (wt/vol) (B), tartaric acid (wt/vol) (C), lactic acid (vol/vol) (D), and malic acid (wt/vol) (E) on the population of C. jejuni NCTC 11168 at 4uC in BHI broth (m) and chicken juice (&). Data points are means ¡ SEM (n ~ 2).

Survival on meat medallions. The inhibitory effect of Thereafter, no remarkable decline was observed. Addition tartaric acid solutions and food marinade ingredients of 8% (vol/vol) ethanol or 8% (vol/vol) ethanol and 10% (Table 1) on C. jejuni inoculated onto meat medallions was (wt/vol) NaCl together with 6% (wt/vol) tartaric acid had tested on the most acid robust C. jejuni strain (305A) and the effects similar to those caused by 6% (wt/vol) tartaric alone most sensitive strain (327) found after 29 h in the broth (Peth ~ 0.73 and Peth/NaCl ~ 0.58). Solutions of 8% experiment. On meat, strain 327 was still more sensitive than (vol/vol) ethanol or 10% (wt/vol) NaCl on their own had no strain 305A, but the differences in reductions were not as inhibitory effect on C. jejuni (Fig. 2B). pronounced as those in BHI broth (data not shown). Marinade ingredients such as honey, mustard, pome- The reductions obtained with tartaric acid (2, 4, 6, and granate syrup, red wine, soya sauce, lemon juice, and wine 10%, wt/vol) for strain 305A are presented in Figure 2A. vinegar alone or in combinations were tested for their Increasing reductions in counts of C. jejuni were observed antibacterial activity against the most common human C. with increasing concentrations of tartaric acid (P , 0.001), jejuni serotype 4C and the robust strain 305A inoculated with a maximum reduction of 2 log CFU/ml for 10% onto meat medallions. All food ingredients tested reduced (wt/vol) tartaric acid. For all concentrations, the largest drop populations of C. jejuni. The reductions obtained with red in C. jejuni counts occurred within the first day (Fig. 2A). wine, honey, pomegranate syrup, and mustard were less 262 BIRK ET AL. J. Food Prot., Vol. 73, No. 2

TABLE 2. pH values and percentage of undissociated acid forms TABLE 3. Relative reduction of C. jejuni strains exposed to 0.3% of organic acids in BHI broth and chicken juice (wt/vol) tartaric acid in BHI at 4uCa

% undissociated Reduction in C. jejuni (log[N/N0]) after pH acida tartaric acid exposure for:

BHI Chicken BHI Chicken 29 h 46 h Organic acid solution broth juice broth juice Mean Mean 0.5% (vol/vol) L-acetic acid 4.61 4.38 58 70 Strain (n ~ 2) SEM (n ~ 2) SEM 0.5% (vol/vol) L-lactic acid 4.26 4.01 25 38 0.5% (wt/vol) L-tartaric acid 4.06 3.85 7 11 Human isolates 0.5% (wt/vol) L-malic acid 4.31 4.08 11 17 Serotype 4C 2.58 BC 0.27 4.05 ABC 0.19 0.5% (wt/vol) L-citric acid 4.51 4.17 4 8 Serotype 2 3.24 AB 0.34 4.62 ABC 0.62 Serotype 1.44 3.15 B 0.11 4.98 ABC 0.54 a 2 The Henderson-Hasselbach equation, pH ~ pKa z log(cA / Downloaded from http://meridian.allenpress.com/jfp/article-pdf/73/2/258/1679931/0362-028x-73_2_258.pdf by guest on 24 September 2021 cHA), was calculated, where cA2 is the dissociated acid form Poultry isolates and cHA is the undissociated acid form. Serotype 4C 2.53 BC 0.13 3.98 BC 0.08 Serotype 2 2.69 BC 0.15 4.24 ABC 0.31 Serotype 1.44 3.31 AB 0.39 4.43 ABC 0.62 327 4.48 A 0.58 4.79 ABC 0.40 than 1 log CFU/ml after 3 days at 4uC (Fig. 3), whereas 5-1 3.68 AB 0.05 4.84 ABC 0.43 wine vinegar (pH 3.4) and wine vinegar with red wine 305A 1.59 C 0.44 4.18 AB 1.22 (pH 3.4) resulted in reductions higher than 2.5 log CFU/ml. 355A 3.30 AB 0.45 3.59 C 0.27 When mixing food ingredients to prepare a marinade, the 4-3 3.72 AB 0.01 4.65 ABC 0.46 most antibacterial ingredient in the mixture is diluted. For 992 3.47 AB 0.77 4.88 ABC 0.50 example, the combination of wine vinegar, red wine, and 1099 2.55 BC 0.57 3.82 BC 0.30 835-770B 3.85 AB 0.58 5.28 A 0.08 soya sauce (pH 3.5) resulted in a 1.5-log reduction compared with wine vinegar alone. However, there was a Within a column, means with the same letter are not significantly no correlation between the lowest pH of the food ingredients different (P . 0.05). and the highest reduction. Lemon juice (pH 2.7) had the lowest pH but wine vinegar (pH 3.4) was the most effective A hedonic taste evaluation with a panel of 12 untrained ingredient for reducing C. jejuni populations. Thus, the tasters was carried out on the marinated prepared breast specific ingredient is a relevant factor. Lemon juice filets. In this test all the marinades were graded positively, primarily contains citric acid, whereas wine vinegar contains but marinade 3 (based on white wine vinegar) had the various organic acids and metabolites from fermentation highest scores (data not shown). that might contribute to the antibacterial effect. DISCUSSION Inhibitory effect of marinades on breast filets. Three Several acidic marinade ingredients alone or in marinades whose main ingredients were pomegranate syrup combination were effective for reducing populations of C. (marinade 1, pH 2.45), lemon juice (marinade 2, pH 2.10), jejuni. In both BHI broth and chicken juice, 0.5% and white wine vinegar (marinade 3, pH 2.87) were tested concentrations of the organic acids were very efficient in for their antibacterial activity against C. jejuni 305A reducing C. jejuni. The effect of the organic acids was inoculated onto whole breast filets. The acid robust strain correlated with the level of undissociated acid. The lower 305A was selected for this experiment because this strain the pH, the higher level of undissociated organic acid that was expected to be most robust to various acid treatments. can passively cross the cell membrane and enter the cell, The results obtained with strain 305A were therefore resulting in a decrease in pH (8). For all the acids examined, considered a worst-case scenario compared with other the pH of the chicken juice was lower than that of BHI broth strains. Lemon juice and white wine vinegar were chosen (Table 2). However, this difference did not result in a because of their inactivation of human strain 4C in the meat greater reduction of C. jejuni in chicken juice than in BHI medallion experiments (Fig. 3). Pomegranate syrup was broth for any of the organic acids tested. Survival in the selected because of its very low pH and because it reduced presence of organic acids seems not to be medium specific, the acid robust strain 305A approximately 0.8 log CFU/ml. as observed for other stresses such as freezing (4). Influence An even more pronounced effect was expected with the of the medium on survival of C. jejuni during acid stress has more sensitive human strain 4C. Other ingredients were been studied by Shaheen et al. (19), who reported that added to the marinades to balance the flavors. Brucella broth appeared to be more protective than tryptic The antibacterial effects of each marinade on C. jejuni soy broth for C. jejuni cells at pH 4.0. Other studies in 305A inoculated onto whole breast filets are presented in which acid tolerance of C. jejuni cells at pH 4.5 was Figure 4. All marinades caused reductions in C. jejuni of compared in three Brucella broths from different manufac- approximately 1.2 log CFU/ml after 3 days of storage. No turers also revealed variation in survival (12, 13). These and significant difference was observed between the three our findings confirm that the composition of the test marinades despite differences in pH and ingredients. medium influences C. jejuni inactivation during acid stress. J. Food Prot., Vol. 73, No. 2 REDUCTION OF CAMPYLOBACTER BY MARINATION INGREDIENTS 263 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/73/2/258/1679931/0362-028x-73_2_258.pdf by guest on 24 September 2021

FIGURE 2. Reducing effects of various concentrations of tartaric acid (TA) and TA in combination with ethanol and NaCl on populations of C. jejuni 305A inoculated onto broiler meat medallions that were then stored at 4uC. (A) 2% (wt/vol) TA (¤), 4% (wt/vol) TA (&), 6% (wt/vol) TA (m), and 10% (wt/vol) TA (’). (B) 8% (vol/vol) ethanol (e), 10% (wt/vol) NaCl (%), 6% (wt/vol) TA (m), 6% (wt/vol) TA plus 8% (vol/vol) ethanol (z), and 6% (wt/vol) TA plus 8% (vol/vol) ethanol plus 10% (wt/vol) NaCl (2). Data points are means ¡ SEM for four meat medallions (n ~ 2).

Some strain variability was observed when 14 C. jejuni after 3 days of storage, although the clinical isolates strains were exposed to 0.3% (wt/vol) tartaric acid in BHI survived slightly better after prolonged storage. broth at 4uC. However, no differences were found within We found that the antibacterial effect of organic acids strains of the same serotype (serotypes 2, 1.44, and 4C) on C. jejuni on broiler meat was less pronounced than the regardless of their origin (human or poultry). In a cold effect in culture broth. This difference was likely due to the tolerance study conducted at 4uC, Chan et al. (5) found no buffering capacity of the meat (16). After lowering pH by difference in survival between clinical and poultry isolates adding organic acids onto the meat surface, we observed a

FIGURE 3. Reducing effect of different food ingredients on populations of C. jejuni inoculated onto meat medallions that were then stored for 3 days at 4uC. Strain 305A was exposed to pomegranate syrup (PS), honey (HO), and mustard powder (MU) diluted with water (WA). Strain 4C human was exposed to red wine (RW), soya (SO), lemon juice (LJ), and wine vinegar (WV). The pH values listed next to the specific food ingredients and next to the columns are the pH of the solution before addition to the meat and pH on the meat after 3 days of storage, respectively. Data are means ¡ SEM for four meat medallions and two independent experiments (n ~ 2). 264 BIRK ET AL. J. Food Prot., Vol. 73, No. 2

Ma¨kela¨ et al. (15) found no difference in populations of C. jejuni between nonmarinated and marinated broiler meat after 10 days in an oil-water emulsion marinade with 5.9% NaCl and pH adjusted to 4.5 with lactic acid; therefore, it seems that low pH is necessary to obtain a rapid effect. Despite the very low pH of the tested marinades, a tasting panel that evaluated the prepared, marinated broiler meat reported positive scores with regard to taste and texture. Of the organic acids tested, tartaric acid reduced C. jejuni most effectively. However, significant variation was observed among the 14 strains exposed to this acid. On meat, low pH food ingredients had the highest inhibitory effect on C. jejuni, and wine vinegar and lemon juice were Downloaded from http://meridian.allenpress.com/jfp/article-pdf/73/2/258/1679931/0362-028x-73_2_258.pdf by guest on 24 September 2021 the most effective ingredients. Mixing acidic food ingredi- ents with other ingredients did not result in an additive reducing effect. FIGURE 4. Reducing effect of different marinades at 4uCon In summary, this study revealed that it is possible to populations of C. jejuni 305A inoculated onto broiler breast filets. compose a marinade that both has an antimicrobial effect on (m) Marinade 1, pH 2.45 (45.5% pomegranate syrup, 23% water, C. jejuni on broiler meat and results in an acceptable taste of 17% honey, 11.5% mustard powder, 2% NaCl, and 1% pepper); the prepared meat. Marination of broiler meat could be used (&) marinade 2, pH 2.10 (73% lemon juice, 18% honey, 2% as an intervention to control Campylobacter in this branch garlic, 2% NaCl, and 1% pepper); (’) marinade 3, pH 2.87 (52% of production. white wine vinegar, 24% sugar, 2.5% estragon, 18.5% onion, 2% In this study organic acids had remarkably less effect on NaCl, and 1% pepper). Each sample point represents six breast filets. Error bars represent the standard deviation. C. jejuni when it was inoculated onto meat than when it was inoculated into broths. Thus, it is important to carry out experiments on meat rather than in growth medium when the antibacterial effect of the components on meat is to be rapid rise in pH within a few minutes, which to some extent evaluated. neutralized the antibacterial effect. Still, organic acids are more promising candidates for antibacterial marinades than ACKNOWLEDGMENTS are low levels of such additives as NaCl and ethanol, which had little or no effect. The authors gratefully acknowledge Lantma¨nnen Danpo A/S for providing Campylobacter-free breast filets and Thomas Alter (Institute of To compose an efficient antibacterial marinade, the Medical Microbiology and Hygiene, University of Ulm, Ulm, Germany) for antibacterial effects of different food ingredients on C. jejuni Campylobacter strains. This work was financially supported by the Danish inoculated onto broiler meat medallions were evaluated. The Industry Agency. most efficient food ingredient was wine vinegar either alone or in combination with red wine and soya. The antibacterial REFERENCES activity of wine vinegar continued during the 3 days of 1. Anonymous. 2009. Annual report on Zoonoses in Denmark 2007. storage, whereas the other ingredients and acids were Technical University of Denmark, Søborg. effective only within the first 24 h of storage. Wine vinegar 2. Birk, T., H. Ingmer, M. T. Andersen, K. Jorgensen, and L. Brondsted. 2004. Chicken juice, a food-based model system suitable is a fermentation product of wine and contains a to study survival of Campylobacter jejuni. Lett. Appl. Microbiol. 38: combination of organic acids (acetic acid, tartaric acid, 66–71. and citric acid) and other fermentation metabolites. Lemon 3. Birk, T., and S. Knøchel. 2009. Fate of food-associated bacteria in juice, which had the lowest pH (2.7), was not as effective as as affected by marinade, temperature, and ultrasound. J. Food wine vinegar (pH 3.4) for reducing the population of C. Prot. 72:549–555. jejuni. 4. Birk, T., H. Rosenquist, L. Brondsted, H. Ingmer, A. Bysted, and B. B. Christensen. 2006. A comparative study of two food model In this screening study, red wine was not very effective systems to test the survival of Campylobacter jejuni at 218uC. J. for reducing C. jejuni, although we previously found that C. Food Prot. 69:2635–2639. jejuni on pork medallions could be reduced by approxi- 5. Chan, K. F., H. L. Tran, R. Y. Kanenaka, and S. Kathariou. 2001. mately 3.5 log units after 3 days at 4uC (3). However, in that Survival of clinical and poultry-derived isolates of Campylobacter jejuni at a low temperature (4uC). Appl. Environ. Microbiol. 67:4186– study, the pork meat medallions were fully submerged in 4191. red wine during storage, thereby constantly stressing the 6. European Food Safety Authority. 2006. The community summary bacteria with red wine, whereas the comparisons in the report on trends and sources of zoonoses, zoonotic agents, present study were made with small amounts of liquid on antimicrobial resistance and foodborne outbreaks in the European the surface of the meat. Union in 2005. EFSA J. 94:1–288. 7. European Food Safety Authority. 2007. The community summary The three low pH (,3) marinades examined reduced report on trends and sources of zoonoses, zoonotic agents and the population of C. jejuni inoculated onto whole breast antimicrobial resistance in the European Union in 2006. EFSA J. 130: filets by more than 1 log unit after 3 days of storage. Perko- 1–351. J. Food Prot., Vol. 73, No. 2 REDUCTION OF CAMPYLOBACTER BY MARINATION INGREDIENTS 265

8. Foster, J. W. 2000. Microbial responses to acid stress, p. 99–115. In infections can be found in retail poultry products and food animals. G. Storz and R. Hengge-Aronis (ed.), Bacterial stress responses. Epidemiol. Infect. 134:758–767. ASM Press, Washington, DC. 15. Perko-Ma¨kela¨, P., M. Koljonen, M. Miettinen, and A.-L. Hanninen. 9. Gibbens, J. C., S. J. S. Pascoe, S. J. Evans, R. H. Davies, and A. R. 2000. Survival of Campylobacter jejuni in marinated and nonmari- Sayers. 2001. A trial of biosecurity as a means to control nated chicken products. J. Food Saf. 20:209–216. Campylobacter infection of broiler chickens. Prev. Vet. Med. 48: 16. Po¨so¨, A. R., and E. Puolanne. 2005. Carbohydrate metabolism in 85–99. meat animals. Meat Sci. 70:423–434. 10. Humphrey, T., S. O’Brien, and M. Madsen. 2007. Campylobacters as 17. Rosenquist, H., L. Boysen, C. Galliano, S. Nordentoft, S. Ethelberg, zoonotic pathogens: a food production perspective. Int. J. Food and B. Borck. 2009. Danish strategies to control Campylobacter in Microbiol. 117:237–257. broilers and broiler meat: facts and effects. Epidemiol. Infect. 137: 11. Kyla-Puhju, M., M. Ruusunen, R. Kivikari, and E. Puolanne. 1742–1750. 2004. The buffering capacity of porcine muscles. Meat Sci. 67: 18. Rosenquist, H., H. M. Sommer, N. L. Nielsen, and B. B. Christensen. 587–593. 2006. The effect of slaughter operations on the contamination of 12. Murphy, C., C. Carroll, and K. N. Jordan. 2003. Induction of an chicken carcasses with thermotolerant Campylobacter. Int. J. Food adaptive tolerance response in the foodborne pathogen, Campylo- Microbiol. 108:226–232.

bacter jejuni. FEMS Microbiol. Lett. 223:89–93. 19. Shaheen, B. W., M. E. Miller, and O. A. Oyarzabal. 2007. In vitro Downloaded from http://meridian.allenpress.com/jfp/article-pdf/73/2/258/1679931/0362-028x-73_2_258.pdf by guest on 24 September 2021 13. Murphy, C., C. Carroll, and K. N. Jordan. 2005. The effect of survival at low pH and acid adaptation response of Campylobacter different media on the survival and induction of stress responses by jejuni and Campylobacter coli. J. Food Saf. 27:326–343. Campylobacter jejuni. J. Microbiol. Methods 62:161–166. 20. Stern, N., P. J. Rothenberg, and J. M. Stone. 1985. Enumeration and 14. Nielsen, E. M., V. Fussing, J. Engberg, N. L. Nielsen, and J. reduction of Campylobacter jejuni in poultry and red . J. Food Neimann. 2006. Most Campylobacter subtypes from sporadic Prot. 48:606–610.