SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Discussion
  5. Acknowledgements
  6. References

Experimental designs using Response Surface Methodology (RSM) were used to determine effects and interactions of Nisin (0–200 IU ml−1), pH values (5·4–6·6), incubation time (0–36 h or 0–144 h) and the lactoperoxidase-thiocyanate-hydrogen peroxide system (LPS) on Listeria monocytogenes CIP 82110 in skim milk, at 25 °C. The LPS varied from level 0–2; LPS at level 1 consisted of lactoperoxidase (35 mg l−1), thiocyanate (25 mg l−1) and H2O2, which was supplied exogenously by glucose-oxidase (1 mg l−1) and glucose (0·2 g l−1); LPS activity was dependent on LPS level and incubation time. In the presence of LPS at level 1, a bacteriostatic phase was followed by growth, whereas at a higher level, a bactericidic phase was observed. Nisin response was time- and pH-dependent. Nisin was bactericidic at acidic pH values and for a short incubation time (12 h) only; then, a re-growth phase was observed. Nisin and LPS in combination gave an original response which lacked the transitory bactericidal effect of Nisin and had a continuously bactericidal affect, leading to 10 cfu ml−1 of L. monocytogenes at 144 h; the response was greatly affected by incubation time. Predicted values were in good agreement with experimental values. Response Surface Methodology is a useful experimental approach for rapid testing of the effects of inhibitors.

INTRODUCTION

Listeria monocytogenes, which is ubiquitous in nature, occurs in soil, vegetation and water; it is frequently carried by healthy humans and animals and is often found in food-processing environments. As a result of its wide distribution, due to its ability to survive for long periods under adverse conditions and grow rapidly at refrigeration temperatures, L. monocytogenes is recognized as an important food-borne pathogen (Carosella 1990). It is always detected in a variety of food, and has caused a number of large-scale outbreaks of listeriosis in the USA, Canada and Europe (Farber & Peterkin 1991).  The antimicrobial lactoperoxidase system (LPS), present in raw milk, consists of lactoperoxidase (LPO), thiocyanate (SCN) and hydrogen peroxide (H2O2). Lactoperoxidase catalyses the oxidation of thiocyanate by hydrogen peroxide (Reiter 1985); the major antibacterial action is due mainly to the hypothiocyanate ion (OSCN), an intermediate reaction product (Aune & Thomas 1977; Reiter 1985). The LPS is bactericidal for Gram-negative bacteria and bacteriostatic for several Gram-positive micro-organisms (Reiter & Harnulv 1984; Pruitt & Reiter 1985). In developing countries, it is used to increase the storage times of raw milk at ambient temperature (Wolfson & Sumner 1993). The activity of the LPS in cow's milk, dairy products or synthetic media has been established against a variety of milk-borne spoilage and pathogenic bacteria (Björck 1988). An anti-L. monocytogenes activity was observed in Trypticase Soy Broth and sterile milk (Denis & Ramet 1989; Earnshaw & Banks 1989; Siragusa & Johnson 1989; Bibi & Bachmann 1990; El-Shenawy et al. 1990; Kamau et al. 1990; Gaya et al. 1991; Zapico et al. 1993).

In various food test systems, the inhibitory effect of bacteriocins from lactic acid bacteria suggests that these peptides have potential as biopreservatives for controlling food-borne pathogens and spoilage micro-organisms. Their ability to inhibit L. monocytogenes has been demonstrated (Harris et al. 1989; Muriana 1996). Nisin, produced by some strains of Lactococcus lactis subsp. lactis, has a broad spectrum of activity. This lanthionin-containing bacteriocin is used as a natural preservative in various food processes. Its main commercial application is in hard cheese in which it inhibits the outgrowth of Clostridium sp. spores (Delves-Broughton 1990). Sensitivity of L. monocytogenes to Nisin has been demonstrated extensively (Benkerroum & Sandine 1988; Harris et al. 1989). In the production of Ricotta-type cheese made without starter culture, the addition of bacteriocins to milk could control contamination by L. monocytogenes over long periods of storage (70 d) at 6–8 °C, in the presence of acetic acid and sorbate (Davies et al. 1997). The simultaneous use of Nisin and pediocin AcH, a bacteriocin produced by Pediococcus acidilactici H, provided a greater antibacterial activity (Hanlin et al. 1993). This illustrates the use of Nisin in the ‘hurdle concept’ of food safety (Muriana 1996), and the combination of bacteriocins with other preservation substances has been proposed as a means of preventing contamination by food-borne pathogens.

Several approaches have been used in the development of predictive equations to describe the effects of various cultural factors on the behaviour of selected food-borne pathogens. Response Surface Methodology (RSM) appears to be particularly promising in microbiology. This empirical method has been used successfully for modelling the effects and interactions of several factors on the growth of L. monocytogenesBuchanan & Phillips 1990; Cole et al. 1990), Bacillus licheniformis (Mansour et al. 1998) and spoilage yeasts (Deak & Beuchat 1993).

The aims of this study were to determine the effects and interactions of LPS, Nisin, pH, and their combinations, on the behaviour of L. monocytogenes in order to obtain synergistic combinations of inhibitors which could prevent growth of L. monocytogenes in food systems. Experiments were carried out in skim milk at 25 °C. Activation of the LPS was made with exogenous LPO and SCN H2O2 was generated through an enzymatic reaction between glucose and glucose-oxidase (GOD). A Doehlert experimental design was used (Doehlert 1970) and validation was performed by classical microbial methods.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Discussion
  5. Acknowledgements
  6. References

Bacterial cultures

Listeria monocytogenes CIP 82110, obtained from the Institut Pasteur Collection (CIP, Paris, France), was used as test organism. The stock culture was maintained at 4 °C in slants on Trypcase-Soy Agar (TSA) (BioMerieux) supplemented with 0·6% yeast extract (YE) and 1·2% agar (Biokar Diagnostics) (TSA-YE). Listeria monocytogenes was transferred from stock cultures into a TS-YE broth and incubated at 37 °C for 24 h. A second transfer was made into reconstituted skim milk (10% w/v, Oxoid) and was similarly incubated.

Preparation of preservatives

A stock solution of Nisin (Sigma) consisted of 10 mg ml−1 (1 × 104 International Units (IU) ml−1) in 0·02 mol l−1 HCl. After heating at 70 °C for 10 min, the pH value was adjusted to 6·0 with 10 mol l−1 NaOH. Appropriate amounts of Nisin solution were immediately added to inoculated skim milk to give final concentrations of 0, 50, 100, 150 or 200 IU ml−1. The LPS was composed of LPO (BioSerae), KSCN (Prolabo); GOD (BioSerae) and glucose (Prolabo). Each component was sterilized by filtration through 0·22 μm filters (Millipore) and added to milk 5 min before inoculation. Final concentrations of LPS components in milk (level 1) were (in mg l−1): LPO, 35; KSCN, 25; GOD, 1 and glucose, 200; LPS was also used at level 0, 0·5, 1·5 or 2.

Microbial analysis

Reconstituted skim milk (Oxoid) at 10% (w/v) was sterilized (110 °C, 10 min) and acidified with lactic acid (99%, Prolabo) at pH 5·4, 6·0 or 6·6. An overnight culture of L. monocytogenes was 10-fold serial diluted in Tryptone-Salt broth (Biokar Diagnostics) and appropriate dilutions were inoculated into 100 ml skim milk to obtain 104 cfu ml−1. Nisin, LPS or a combination of both was then added at 0 h at appropriate concentrations. Cultures were incubated at 25 °C. For determination of the viable L. monocytogenes count, 1 ml of a given dilution of the sample in Tryptone-Salt broth was plated, in duplicate, into TSA-YE and incubated at 37 °C for 48 h. In order to inactivate the inhibitory effect of Nisin, a protease suspension (type XIV from Streptomyces griseus; Sigma) was added to a final concentration of 1 mg ml−1 and incubated at 37 °C for 20 min before plating.

Experimental design and statistics

Preliminary studies were carried out to determine specific levels of each parameter, and a Doehlert experimental design was selected (Doehlert 1970). This experimental matrix displayed a uniform distribution of the points within the experimental domain and allowed a number of distinct levels for each variable. Maximal number levels were the most important factors. The variables investigated were inhibitor concentrations (X1, five levels; LPS levels or Nisin concentrations), incubation time (X2, seven levels) and pH values (X3, three levels). Two distinct models were established, one for a short incubation time (0–36 h) and the other for a long incubation time (0–144 h) (Table 1a,b). To study the effects of LPS and Nisin combined, the variable X1 was the Nisin concentration and the LPS was fixed at level 1. The total number (N) of experiments for three factors (k) was 13 (N = k2 + k + 1), but 16 were carried out (Table 2); the 13th assay, performed at the centre of the experimental domain, was repeated three times (14, 15, 16) in order to estimate residual variance value. A quadratic model containing 10 coefficients, including interaction terms, was assumed to describe relationships between the response, which was L. monocytogenes cell number logarithm (Y, log10 N), and the experimental factors (Xi, Xj):

Table 1.  Experimental domain and level distribution of the variables used for the evaluation of the inhibitory activities of LPS, Nisin, and their combinations on the survey of Listeria monocytogenes as a function of inhibitor concentrations (X1; LPS level or Nisin concentration), time (X2) and pH values (X3). (a) Short incubation time (0–36 h); (b) Long incubation time (0–144 h)
(a)
ModelFactorsLevelsExperimental values
 X1: LPS*50–0·5–1–1·5–2
LPS aloneX2: Time (h)70–6–12–18–24–30–36
 X3: pH35·4–6–6·6
 X1: Nisin (IU ml−1)50–50–100–150–200
Nisin aloneX2: Time (h)70–6–12–18–24–30–36
 X3: pH35·4–6–6·6
 X1: Nisin (IU ml−1)50–50–100–150–200
LPS-NisinX2: Time (h)70–6–12–18–24–30–36
(LPS fixed at level 1)*X3: pH35·4–6–6·6
(b)
ModelFactorsLevelsExperimental values
  • *

    See Materials and Methods for corresponding concentrations.

 X1: LPS*50–0·5–1–1·5–2
LPS aloneX2: Time (h)70–24–48–72–96–120–144
 X3: pH35·4–6–6·6
 X1: Nisin (IU ml−1)50–50–100–150–200
Nisin aloneX2: Time (h)70–24–48–72–96–120–144
 X3: pH35·4–6–6·6
 X1: Nisin (IU ml−1)50–50–100–150–200
LPS-NisinX2: Time (h)70–24–48–72–96–120–144
(LPS fixed at level 1)*X3: pH35·4–6–6·6
Table 2.  Experimental design for the three variables, X1 inhibitor concentration (LPS level or Nisin concentration), X2 time and X3 pH values, including three repetitions at the centre of the domain (14, 15, 16) according to Doehlert (1970) uniform shell design
Variables
X1X2X3
ExperimentLPS level*Nisin (UI ml−1)Time (0–36 h)Time (0–144 h)pH
  • *

    See Materials and Methods for corresponding concentrations.

1220018726
20018726
31·5150361446
40·550006
51·5150006
60·550361446
71·515024966·6
80·55012485·4
91·515012485·4
101100301205·4
110·55024966·6
1211006246·6
13110018726
14110018726
15110018726
16110018726
  • image
  • Responses were recorded on graphs as isopopulation curves predicted by the quadratic model, as functions of two experimental factors while the third was held constant at the centre of the domain.

Data analysis

Data analysis, anova, and multilinear regression were performed using the NEMROD® software (LPRAI, Marseille, France; Mathieu & Phan-Tan-Luu 1997).

Validation of the model

Predicted values were compared with the experimental values obtained with LPS at level 1 in the LPS and LPS-Nisin models, and/or Nisin at 100 IU ml−1 in the Nisin and LPS-Nisin models. In these experiments, the pH was constant at 6·4 and the incubation time varied from 0 to 144 h.

Results

Skim milk had an initial pH value fixed at 5·4, 6·0 or 6·6, which remained constant throughout incubation. Effects and interactions of LPS, Nisin and pH values on the behaviour of L. monocytogenes were studied at 25 °C, in skim milk, using the Doehlert experimental design. Initial population levels were 3·5 × 104 cfu ml−1 and 7 × 103 cfu ml−1 for the LPS or Nisin models, and the LPS-Nisin model, respectively. Two distinct models were established, one for a short incubation time (36 h) and another for a long incubation time (144 h). The response (Log N), analysed using Response Surface Methodology (RSM), was given by a polynomial equation (Tables 3, 4 and 5). The R2 values were significant, indicating that the derived model fitted the experimental data. Response surface graphs showed the effects of LPS (Fig. 1a,b), Nisin (Fig. 2a,b) and the LPS-Nisin combination (Fig. 3a,b) on L. monocytogenes cell counts.

Table 3.  Coefficient estimations of the different quadratic models LPS for the three variables, inhibitor concentration X1 (LPS level), time X2 and pH values X3, for short incubation time (0–36 h) and for long incubation time (0–144 h)
LPS modelPolynomial equation
  1. Response was Listeria monocytogenes cell count (N). aα ≤ 1£b1£ ≤ α < 1% c1% < α ≤ 5% R2A, adjusted R2.

Short incubation timeLog N = 4·3400a − 1·1375a LPS + 0·4994a time + 0·1633c pH + 1·6300a LPS2 + 0·0800 time2 − 0·1325 pH2 − 0·3683 LPS time + 0·0245 LPS pH − 0·3323c time pHR2 = 0·8750; R2= 0·6876
Long incubation timeLog N = 5·8100 − 2·6313a LPS + 0·1321b time + 0·6062a pH + 0·0900 LPS2 − 1·7867a time2 − 1·6333 pH2 − 3·8278a LPS time − 0·1041 LPS pH + 0·8073a time pHR2 = 0·9797; R2= 0·9492
Table 4.  Coefficient estimations of the different quadratic models Nisin for the three variables, inhibitor concentration X1 (Nisin concentration), time X2 and pH values X3, for short incubation time (0–36 h) and for long incubation time (0–144 h)
Nisin modelPolynomial equation
  1. Response was Listeria monocytogenes cell count (N). aα ≤ 1£b1£ ≤ α < 1% c1% < α ≤ 5% R2A, adjusted R2.

Short incubation timeLog N = 2·2200a − 2·1288a Nisin + 0·8956a time + 2·4474a pH + 2·1450a Nisin2 + 2·7717a time2 − 0·2017c pH2 − 0·8776a Nisin time − 0·6389b Nisin pH + 1·3989c time pHR2 = 0·9358; R2= 0·8396
Long incubation timeLog N = 8·4900 − 0·3712a Nisin + 2·6969a time + 1·1002a pH − 0·0600 Nisin2 − 2·8667a time2 − 2·6808a pH2 + 0·0115 Nisin time + 1·7044a Nisin pH − 2·0636a time pHR2 = 0·9370; R2= 0·8424
Table 5.  Coefficient estimations of the different quadratic models LPS-Nisin combination for the three variables, inhibitor concentration X1 (Nisin concentration), time X2 and pH values X3, for short incubation time (0–36 h) and for long incubation time (0–144 h)
LPS-Nisin modelPolynomial equation
  1. Response was Listeria monocytogenes cell count (N). aα ≤ 1£b1£ ≤ α < 1% c1% < α ≤ 5% R2A, adjusted R2.

Short incubation timeLog N = 3·800a − 0·1550a Nisin − 0·0866b time − 0·3184a pH − 0·0100 Nisin2 + 0·0500 time2 − 0·3700a pH2 − 0·0808c Nisin time − 0·1306c Nisin pH + 0·0424 time pHR2 = 0·9671; R2= 0·9167
Long incubation timeLog N = 3·200a − 0·3688b Nisin − 1·5870a time − 0·5818b pH − 0·1200 Nisin2 − 0·7267b time2 − 0·6058b pH2 − 0·3118 Nisin time − 0·3246 Nisin pH − 0·0978 time pHR2 = 0·9892; R2= 0·9731
image

Figure 1. The LPS effect on Listeria monocytogenes behaviour in skim milk at 25 °C using the Doehlert experimental design. (a) Short incubation time (0–36 h): response surface contours show LPS level and pH value effects on cell counts. Observations were made at 18 h, the value at the centre of the domain. (b) Long incubation time (0–144 h): response surface contours show LPS level and time effects on cell counts. pH was fixed at 6·0, the value at the centre of the domain

Download figure to PowerPoint

image

Figure 2. Nisin effects on Listeria monocytogenes behaviour in skim milk at 25 °C using the Doehlert experimental design. (a) Short incubation time (0–36 h): response surface contours show Nisin concentration and pH effects on cell counts. Observations were made at 18 h, the value at the centre of the domain. (b) Long incubation time (0–144 h): response surface contours show Nisin concentration and time effects on cell counts. The pH was fixed at 6·0, the value at the centre of the domain

Download figure to PowerPoint

image

Figure 3. The effects of LPS-Nisin combination on Listeria monocytogenes behaviour in skim milk at 25 °C using the Doehlert experimental design. (a) Short incubation time (0–36 h): response surface contours show Nisin concentration and pH effects on cell counts. The LPS was fixed at level 1. Observations were made at 18 h, the value at the centre of the domain. (b) Long incubation time (0–144 h). Response surface contours show the effects of pH and time on cell counts. The LPS was fixed at level 1 and Nisin concentration at 100 IU ml−1, the values at the centre of the domain

Download figure to PowerPoint

In the presence of LPS, when the incubation time was 36 h, the coefficient −1·1375 indicated that the major inhibitory factor was the LPS level; time and pH were less important (Table 3). After 18 h of incubation, at pH 6·0, the response surface graph showed that L. monocytogenes cell counts were essentially a function of LPS levels (Fig. 1a). For LPS levels higher than 0·75, a bacteriostatic effect was observed, without any significant differences in cell counts between line 1 (N0) and 2. With LPS levels below 0·75, growth always occurred and was enhanced as LPS levels decreased; pH values did not affect the response. In the absence of inhibitor, the L. monocytogenes population increased to 8·8 × 106 cfu ml−1.

For the long incubation time (144 h), coefficient values showed that the major factors which strongly inhibited growth were LPS levels and LPS-time interaction (Table 3). At pH 6·0, the combined effects of LPS level and time gave L. monocytogenes isopopulation curves varying from 29 cfu ml−1 to 4·1 × 107 cfu ml−1 (Fig. 1b). For LPS levels higher than 1, after a 48 h bacteriostatic phase with isopopulation curves close to level N0, a bactericidal phase occurred. The highest LPS level and longest time period induced the most inhibitory effect. Maximum inhibition was obtained after 100 h for LPS level 2, or after 144 h for LPS level 1·5. With an LPS level below 1, a shorter bacteriostatic phase of 24 h was observed. Growth then occurred in relation to LPS levels; highest population counts were obtained with lower LPS levels. An LPS level below 0·5 did not inhibit growth of L. monocytogenes.

In the presence of Nisin, for a short incubation time (36 h), a low constant coefficient value (2·2200) was correlated with a strong inhibitory effect. The coefficient value for Nisin concentrations or pH values indicated that these factors interacted with response, and that they acted antagonistically (Table 4); a high pH value increased the L. monocytogenes cell count, whereas a high Nisin concentration reduced the L. monocytogenes population. Interaction factors were important, particularly for time-pH. Listeria monocytogenes isopopulation curves show the combined effects of pH value and Nisin concentration (Fig. 2a). The bactericidal effect was more important at acidic pH values combined with high Nisin concentrations. No growth of L. monocytogenes was obtained at pH 5·4 with 100 IU ml−1 Nisin. Growth of L. monocytogenes occurred (population levels higher than the initial inoculum N0) at pH 5·6 without Nisin, or at pH 6·4 for Nisin concentrations lower than 100 UI ml−1.

For the long incubation time (144 h), the polynomial equation showed that the high constant coefficient value was correlated with the absence of inhibition, and that growth was dependent on incubation time and pH values (Table 4). pH interacted with time and Nisin concentration. At pH 6·0, Nisin concentration had no significant effect on L. monocytogenes behaviour (Fig. 2b) and growth was only time-dependant, with maximum population levels of 9·8 × 107 cfu ml−1 at 60 h.

In the presence of the LPS-Nisin combination, for the short incubation time (36 h), the low constant coefficient value was related to a bacteriostatic effect (Table 5). At 18 h (Fig. 3a), L. monocytogenes isopopulation curves between 1·2 × 103 cfu ml−1 and 7·2 × 103 cfu ml−1 indicated a bacteriostatic effect.

For the long incubation time (144 h), the constant coefficient value (3·200) was lower than those obtained for LPS or Nisin used alone (Table 5). All factors inhibited growth, for which time was the predominant factor. With 100 IU ml−1 Nisin, L. monocytogenes isopopulation curves ranging from 4·7 × 103 (N0) to 10 cfu ml−1 (Fig. 3b) were related to a bactericidal effect. However, an early and short time bacteriostatic phase was noted which was longer at acidic pH values (12 h at pH 6·3 or 48 h at pH 6 or 5·5).

Comparison of results derived from polynomial equations and experimental values obtained from kinetic growth measurements (Fig. 4a,b,c) showed that equations were valid within the experimental domain. Equations obtained during the short incubation time (36 h) described initial effects of LPS (Fig. 4a), Nisin (Fig. 4b) and LPS-Nisin combination (Fig. 4c); equations obtained for the long incubation time correlated with L. monocytogenes behaviour after the 36 h initial period.

image

Figure 4. Comparison of experimental values and those predicted by polynomial equations for (a) LPS (b) Nisin and (c) LPS-Nisin combination. Results were obtained in skim milk at pH 6·4 and 25 °C. For LPS and LPS-Nisin models, LPS was fixed at level 1; for Nisin and LPS-Nisin models, Nisin concentration was fixed at 100 IU ml−1. (•) Experimental Listeria monocytogenes cell counts; (▵) L. monocytogenes cell counts derived from equations obtained during short incubation time (36 h); (□) L. monocytogenes cell counts derived from equations obtained during long incubation time (144 h)

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Discussion
  5. Acknowledgements
  6. References

‘Hurdle technology’ advocates the combination of preservation techniques in order to establish a series of preservative factors (Leistner & Gorris 1995). Some micro-organisms can overcome a number of hurdles but none can ‘jump’ them all. The site of action of many bacteriocins of lactic acid bacteria is the cell membrane. Nisin molecules may form poration complexes in the cell membrane, resulting in leakage of cellular constituents and the loss of membrane potential (Driessen et al. 1995; Muriana 1996).

However, bacteriocins have limitations which reduce their effectiveness as food biopreservatives. They are not active against Gram-negative bacteria, their host range can be either wide or narrow, and some insensitive cells can multiply in the presence of bacteriocins (Harris et al. 1991; Hanlin et al. 1993). The LPS exerts an antimicrobial effect by the OSCN-mediated oxidation of essential protein and enzyme sulphydryl groups, and many cellular systems (i.e. outer membrane, cell wall, cytoplasmic membrane, transport system, glycolytic enzymes and nucleic acids) can be altered (Wolfson & Sumner 1993). The LPS has a large inhibitory spectrum among Gram-positive and Gram-negative strains. Therefore, LPS and Nisin could act synergistically to inhibit L. monocytogenes.

The effectiveness of LPS for inhibiting growth of L. monocytogenes was a function of LPS concentration and incubation time. The LPS at level 1 induced a 24 h bacteriostatic phase followed by an active growing phase; LPS levels higher than 1 were bactericidal against L. monocytogenes whereas LPS levels below 0·5 did not affect growth. The LPS can induce either a bacteriostatic or a bactericidal phase, depending on environmental conditions, target strains, temperature, substrate and mode of H2O2 generation (Reiter & Harnulv 1984; Wolfson & Sumner 1993). Hydrogen peroxide can be produced by indigenous micro-organisms such as lactic acid bacteria, or it can be added exogenously, or it can be generated through an enzymatic reaction between glucose and glucose-oxidase, or hypoxanthine and xanthine-oxidase. The bacteriostatic effect of LPS against L. monocytogenes inoculated in milk has been described by many authors at 30 °C (Siragusa & Johnson 1989; Bibi & Bachmann 1990) or at 35 °C (El-Shenawy et al. 1990; Kamau et al. 1990). These authors also observed a re-growth at a specific growth rate lower than the control. At refrigeration temperature, this bacteriostatic period was longer (Earnshaw & Bank 1989; El-Shenawy et al. 1990; Kamau et al. 1990). The LPS exhibited a bactericidal activity against L. monocytogenes in raw milk at 4 and 8 °C (Gaya et al. 1991). Listeria monocytogenes populations were reduced by 2 log10 cycles in UHT milk maintained at 10 °C (Earnshaw & Banks 1989), or remained constant for different times; complete inhibition occurred after 30 d at 4 °C, or 21 d at 15 °C, with an initial inoculum of 104 cfu ml−1 (Denis & Ramet 1989). In these studies, concentrations of the different LPS components, and mode of H2O2 generation, were not the same, which could explain differences observed in the effects of LPS.

Nisin had a transitory bactericidal effect on L. monocytogenes for a short incubation time; for a long incubation time, no inhibition of growth was observed. Many studies have indicated an immediate bactericidal effect, with reduction of a population by 1–3 log10 cycles on addition of bacteriocin, and with little or no affect upon further incubation (Harris et al. 1991; Mathieu et al. 1994; Muriana 1996). The re-growth of Listeria cells was probably due to survivors which could multiply in the presence of Nisin (Harris et al. 1991; Hanlin et al. 1993; Rekhif et al. 1994).

The LPS-Nisin combination gave a synergistic and long-time inhibitory effect; a 48 h bacteriostatic phase was followed by a bactericidal phase, which led to the reduction of L. monocytogenes counts by 3 log10 cycles in 140 h with a final population of 10 cfu ml−1. The observed effects were original because the early and transitory bactericidal effects of Nisin were suppressed, and the LPS bacteriostatic effect was enhanced and followed by a bactericidal phase, which did not exist in assays with LPS at level 1 or with Nisin alone. The inhibitory effect of the LPS-Nisin combination was not affected by pH values between 5·4 and 6·6. In UHT milk at 30 °C, with 30% H2O2 added exogenously in aqueous solution, the combined effect of LPS-Nisin reduced L. monocytogenes Scott A counts by 0·5 log10 cycle during the first 3 h of incubation and was then bacteriostatic during a 24 h period (Zapico et al. 1998). When Nisin, produced by lactic acid bacteria, was used in combination with LPS in raw milk at 4 or 8 °C, no significant reduction of L. monocytogenes counts was observed (Rodriguez et al. 1997).

The RSM methodology was an empirical approach based on 13 different experiments. Results showed effects and interactions of different parameters such as pH value, inhibitor concentration and incubation time on L. monocytogenes behaviour. The LPS level and incubation time were the major factors influencing LPS antibacterial activity. The Nisin response was time- and pH-dependent. The LPS-Nisin combination inhibited growth but the response was enhanced by time, which was the major factor. The model obtained by RSM methodology fits the data and was valid in the experimental domain. Depending on the properties of the inhibitor, it was necessary to use two distinct models to predict L. monocytogenes behaviour. The short incubation time model predicted the transitory bactericidal effect of Nisin, and the bacteriostatic effects of LPS or the LPS-Nisin combination. The long incubation time model clearly demonstrated the efficiency of the LPS-Nisin combination for inhibiting an initial L. monocytogenes level of 104 cfu ml−1 in skim milk at 25 °C, when compared with results obtained with Nisin or LPS alone. From a practical point of view, the association LPS × Nisin could considerably increase the margin of safety of raw milk with respect to L. monocytogenes. Furthermore, it is effective for a long time. The combined effect of LPS and Nisin merits further investigation.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Discussion
  5. Acknowledgements
  6. References

The authors thank Delphine Maigné for technical assistance.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Discussion
  5. Acknowledgements
  6. References
  • Aune, T.M. & Thomas, E.L. 1977 Accumulation of hypothiocyanite ion during peroxidase catalysed oxidation of thiocyanate ion. European Journal of Biochemistry, 80, 209214.
  • Benkerroum, N. & Sandine, W.E. 1988 Inhibitory action of Nisin against Listeria monocytogenes. Journal of Dairy Science, 71, 32373245.
  • Bibi, W. & Bachmann, M.R. 1990 Antibacterial effect of the lactoperoxidase-thiocyanate-hydrogen peroxide system on the growth of Listeria spp. in skim milk. Milchwissenschaft, 45, 2628.
  • Björck, L. 1988 Code of practise for the preservation of raw milk by the lactoperoxidase system. International Dairy Federation Bulletin, 234, 115.
  • Buchanan, R.L. & Phillips, J.G.1990 Response surface model for predicting the effects of temperature, pH, sodium chloride content, sodium nitrite concentration and atmosphere on the growth of Listeria monocytogenes. Journal of Food Protection, 53, 370376.
  • Carosella, J.M. 1990 Occurrence of Listeria monocytogenes in meat and poultry. In: Foodborne Listeriosis (eds Miller, A.J. Smith, J.L. & Somkuti, G.A), pp. 165173. New York: Elsevier Science.
  • Cole, M.B., Jones, M.V. & Holyoak, C. 1990 The effects of pH, salt concentration and temperature on the survival and growth of Listeria monocytogenes. Journal of Applied Bacteriology, 69, 6372.
  • Davies, E.A., Bevis, H.E. & Delves-Broughton, J. 1997 The use of the bacteriocin, Nisin, as a preservative in ricotta-type cheeses to control the food-borne pathogen Listeria monocytogenes. Letters in Applied Microbiology, 24, 343346.
  • Deak, T. & Beuchat, L.R. 1993 Use of indirect conductimetry for predicting growth of food spoilage yeasts under various environmental conditions. Journal of Industrial Microbiology, 12, 301308.
  • Delves-Broughton, J. 1990 Nisin and its use as food preservative. Food Technology, 44 (100), 102, 104, 106, 108, 111–112, 117.
  • Denis, F. & Ramet, J.P. 1989 Antibacterial activity of the lactoperoxidase system on Listeria monocytogenes in trypticase soy broth, UHT milk and French soft cheese. Journal of Food Protection, 52, 706711.
  • Doehlert, D.H. 1970 Uniform shell designs. Applied Statistics, 19, 231239.
  • Driessen, A.J.M., Van Den Hooven, H.W., Kuiper, W et al. 1995 Mechanistic studies of lantibiotic-induced permeabilization of phospholipid vesicules. Biochemistry, 34, 16061614.
  • Earnshaw, R.G. & Banks, J.G. 1989 A note on the inhibition of Listeria monocytogenes NTCC 11994 in milk by an activated lactoperoxidase system. Letters in Applied Microbiology, 8, 203205.
  • El-Shenawy, M.A., Garcia, H.S. & Marth, E.H. 1990 Inhibition and inactivation of Listeria monocytogenes by the lactoperoxidase system in raw milk, buffer or a semi-synthetic medium. Milchwissenchaft, 45, 638641.
  • Farber, J.M. & Peterkin, P.I. 1991 Listeria monocytogenes, a food-borne pathogen. Microbial Reviews, 55, 476511.
  • Gaya, P., Medina, M. & Nunez, M. 1991 Effect of the lactoperoxidase system on Listeria monocytogenes behavior in raw milk at refrigeration temperature. Applied and Environmental Microbiology, 57, 33553360.
  • Hanlin, M.B., Kalchayanand, N., Ray, P. & Ray, B. 1993 Bacteriocins of lactic acid bacteria in combination have greater antibacterial activity. Journal of Food Protection, 56, 255255.
  • Harris, J.L., Daeschel, M.A., Stiles, M.E. & Klaenhammer, T.R. 1989 Antimicrobial activity of lactic acid bacteria against Listeria monocytogenes. Journal of Food Protection, 52, 384387.
  • Harris, L.J., Fleming, H.P. & Klaenhammer, T.R. 1991 Sensitivity and resistance of Listeria monocytogenes ATCC 19115, Scott A, and UAL 500 to nisin. Journal of Food Protection, 54, 836840.
  • Kamau, D.N., Doores, S. & Pruitt, K.M. 1990 Antibacterial activity of the lactoperoxidase system against Listeria monocytogenes and Staphylococcus aureus in milk. Journal of Food Protection, 53, 10101014.
  • Leistner, L. & Gorris, L.G.M. 1995 Food preservation by hurdle technology. Trends in Food Science and Technology, 6, 4146.
  • Mansour, M., Linder, M., Millière, J.B. & Lefebvre, G.1998 Combined effect of Nisin, lactic acid and potassium sorbate on Bacillus licheniformis spores in milk. Le Lait, 78, 117128.
  • Mathieu, F., Michel, M., Lebrihi, A. & Lefebvre, G. 1994 Effect of the bacteriocin CP5 and of the producing strain Carnobacterium piscicola CP5 on the viability of Listeria monocytogenes ATCC 15313 in salt solution, broth and skimmed milk, at various incubation temperatures. International Journal of Food Microbiology, 22, 155172.
  • Mathieu, D. & Phan-Tan-Luu, R. 1997 LPRAI Université d’Aix-Marseille. Nemrod®: New efficient methodology for research using optical design.
  • Muriana, P.M. 1996 Bacteriocins for control of Listeria spp. In Food. Journal of Food Protection , Suppl 5463.
  • Pruitt, K.M. & Reiter, B. 1985 Biochemistry of peroxidase system: antimicrobial effects. in the Lactoperoxidase System: Chemistry and Biological Significance (eds Pruitt, K.M. & Tenovuo, J.O), pp. 143178. New York. Marcel Dekker.
  • Reiter, B. 1985 Lactoperoxidase system of bovine milk in the Lactoperoxidase System: Chemistry and Biological Significance (eds Pruitt, K.M. & Tenovuo, J.O), pp. 123–141New York. Marcel Dekker.
  • Reiter, B. & Harnulv, G. 1984 Lactoperoxidase antibacterial system: natural occurrence, biological functions and practical applications. Journal of Food Protection, 47, 724732.
  • Rekhif, N., Atrih, A. & Lefebvre, G. 1994 Selection and properties of spontaneous mutants of Listeria monocytogenes ATCC 15313 resistant to different bacteriocins produced by lactic acid bacteria strains. Current Microbiology, 28, 237241.
  • Rodriguez, E., Tomillo, J., Nunez, M. & Medina, M. 1997 Combined effect of bacteriocin-producing lactic acid bacteria and lactoperoxidase system activation on Listeria monocytogenes in refrigerated raw milk. Journal of Applied Microbiology, 83, 389395.
  • Siragusa, G.R. & Johnson, M.G. 1989 Inhibition of Listeria monocytogenes growth by the lactoperoxidase-thiocyanate-H2O2 antimicrobial system. Applied and Environmental Microbiology, 55, 28022805.
  • Wolfson, L.M. & Sumner, S.S. 1993 Antibacterial activity of the lactoperoxidase system: a review. Journal of Food Protection, 56, 887892.
  • Zapico, P., Gaya, P., Nunez, M. & Medina, M. 1993 Goats’ milk lactoperoxidase system against Listeria monocytogenes. Journal of Food Protection, 56, 988990.
  • Zapico, P., Medina, M., Gaya, P. & Nunez, M. 1998 Synergistic effect of Nisin and the lactoperoxidase system on Listeria monocytogenes in skim milk. International Journal of Food Microbiology, 40, 3542.