The combined effect of nisin, moderate heating and high hydrostatic pressure on the inactivation of Bacillus sporothermodurans spores


  • C. Aouadhi,

    Corresponding author
    1. LUNAM Université, Oniris, GEPEA, Nantes, France
    2. CNRS, Nantes, France
    3. Laboratory of Animal Resources and Food, National Institute of Agronomy, University Carthage, Tunis, Tunisia
    • Laboratory of Epidemiology and Veterinary Microbiology, Bacteriology and Biotechnology Development Groups, Pasteur Institute of Tunisia (IPT), University Manar, Tunis, Tunisia
    Search for more papers by this author
  • H. Simonin,

    1. LUNAM Université, Oniris, GEPEA, Nantes, France
    2. CNRS, Nantes, France
    Search for more papers by this author
  • S. Mejri,

    1. Laboratory of Animal Resources and Food, National Institute of Agronomy, University Carthage, Tunis, Tunisia
    Search for more papers by this author
  • A. Maaroufi

    1. Laboratory of Epidemiology and Veterinary Microbiology, Bacteriology and Biotechnology Development Groups, Pasteur Institute of Tunisia (IPT), University Manar, Tunis, Tunisia
    Search for more papers by this author


Chedia Aouadhi, Laboratoire d'Epidémiologie et Microbiologie Vétérinaire, Groupes de Bactériologie et Développement Biotechnologique. Institut Pasteur de Tunis, BP 74, 13 place Pasteur, Belvédère, 1002 Tunis, Tunisia. E-mail:



To investigate the combined effect of hydrostatic pressure (HP), moderate temperature and nisin on the inactivation of Bacillus sporothermodurans spores which are known to be contaminant of dairy products and to be extremely heat-resistant.

Methods and Results

A central composite experimental design with three factors, using response surface methodology, was used. By analysing the response surfaces and their corresponding contour plots, an interesting interaction with the three factors was observed. The inactivation observed was shown to be well fitted with values predicted by the quadratic equation, since the adjusted determination coefficient (Radj2) was 0·979. The optimum process parameters for a 5-log spores ml−1 reduction of B. sporothermodurans spores were obtained, 472 MPa/53°C for 5 min in presence of 121 UI ml−1 of nisin.


Nisin and temperature treatments improve the effectiveness of pressure in the inactivation of highly heat-resistant spores of B. sporothermodurans.

Significance and Impact of the Study

This study shows the potential of using high HP for a short time (5 min) in combination with moderate temperature and nisin to inactivate B. sporothermodurans spores in milk. Such treatments could be applied by the dairy industry to ensure the commercial sterility of UHT milk.


Bacillus sporothermodurans is a nonpathogenic Gram-positive bacterium firstly described by Pettersson et al. (1996). It is responsible of nonsterility problems in UHT milk because it produces highly heat-resistant endospores which survive ultra-high temperature (UHT) treatment or industrial sterilization (Pettersson et al. 1996; Scheldeman et al. 2006). Moreover, these spores can germinate and grow to 105 CFU ml−1 in stored UHT milk and therefore, cause its spoilage and reduce its shelf-life (Klijn et al. 1997). This problem was spread in different countries in and outside Europe (Hammer et al. 1995; Guillaume-Gentil et al. 2002).

The use of hydrostatic pressure (HP) for the inactivation of bacterial endospores in foods has received particular attention in recent years (Raso and Barbosa-Canovas 2003; Sasagawa et al. 2006). HP has been shown to be efficient for the destruction of vegetative bacteria, viruses and yeasts (Nakayama et al. 1996). Typically, a 10-min exposure to HP of 250–300 MPa at low or moderate temperature (lower than 50°C) results in what could be called ‘cold pasteurization’ (Hoover 1997); however, bacterial spores are not significantly inactivated by such treatment. Basset and Macheboeuf (1932) have shown that some Bacillus spores could even survive after a treatment at 1750 MPa for 45 min at ambient temperature. Nevertheless, other possibilities exist to inactivate spore using the property of HP to initiate the first stage of spore germination at 100–600 MPa and temperature below 50°C (Opstal et al. 2004). It is then possible to inactivate germinated spores by other treatments. However, such treatment is not usually 100% effective for inactivation of spores due to spore population heterogeneity in germination efficiency.

Consequently, the inactivation of spores can occur by combining HP with temperature (Minh et al. 2010). The combination of nisin with HP has also been proposed as an interesting approach for the inactivation of bacterial spores in dairy products (Black et al. 2008; Sobrino-Lopez and Martin-Belloso 2008). Nisin is the only bacteriocin that has been approved as a food additive by the Food and Drug Administration (FDA). It was first introduced commercially as a food preservative in the UK approximately 30 years ago. First established use was as a preservative in processed cheese products and since then numerous other applications in foods and beverages have been identified. It is currently recognized as a safe food preservative in approximately 50 countries. It is used to inhibit the outgrowth of Clostridium and Bacillus spores in the production of processed cheeses, in canned vegetables, in various pasteurized dairy products (Delves-Broughton et al. 1996). Nisin inhibits the step between spore germination and pre-emergent swelling by modifying membrane sulfhydryl groups (Liu and Hansen 1993). Recent studies suggest that nisin–lipid II interactions inhibit the transition between a germinated spore and a vegetative cell membrane through the formation of pores in the germinated spore. Membrane disruption by nisin prevents the establishment of a membrane potential and oxidative metabolism, thus inhibiting outgrowth of germinated spores (Gut et al. 2011).

Previous studies stated that a combination of pressure and nisin is effective to inactivate pathogenic bacteria and spores of Bacillus and Clostridium species (Black et al. 2005, 2008; Gao et al. 2011). For example, the inactivation of Clostridium perfringens spores was shown to be higher (6·8-log spores ml−1) after a treatment at 600 MPa and 65°C for 12 min in presence of nisin (496 UI ml−1) than after the same treatment without nisin (2·5-log spores ml−1) (Gao et al. 2011). Black et al. (2008) showed that the viability of Bacillus subtilis spores in milk and buffer was reduced by 2·5-log spores ml−1 by cycled HP (500 MPa/40°C), while the addition of nisin (500 IU ml−1) to HP treatment resulted in 5·7- and 5·9-log spores ml−1 reductions in phosphate-buffered saline and milk, respectively.

In a previous work, B. sporothermodurans spores were treated in distilled water at 600 MPa/20°C for 5 min, and it was found that the level of inactivation did not exceed 0·1-log spores ml−1 (Aouadhi et al. 2012). When spores were pressurized at 200 MPa/50°C for 5 min, the inactivation of spores was 0·6-log spores ml−1. These results suggest that the level of inactivation of B. sporothermodurans spores is poorly improved by combining high pressure with moderate heating. In the current work, we investigate the combined effect of pressure (for 5 min), temperature and nisin on the inactivation of B. sporothermodurans spores. For this purpose, response surface methodology (RSM) was used to predict optimized processing conditions to inactivate B. sporothermodurans spores.

Materials and methods

Bacterial strain and spore preparation

Bacillus sporothermodurans strain LTIS27 described by Aouadhi et al. (2012) was used in this study. Cells were grown on brain–heart infusion agar supplemented with 1 mg l−1 vitamin B12 (Sigma-Aldrich, St Louis, MO, USA) (BHI-B12) at 37°C. Twenty-four hours before use, cultures were held in BHI-B12 broth on a rotary shaker at 37°C to obtain a working culture. Spores were prepared according to the modified method of Ireland and Hanna (2002). A working culture (18 h, 37°C under agitation) was diluted 1 : 10 into sporulation medium (25 g l−1 nutrient broth, 1 mg l−1 vit B12, 8 mg l−1 MnS04 H20 and 1 g l−1 CaC12 H20) then incubated for 7 days at 37°C. This sporulation medium was then centrifuged at 6000 g for 10 min. The resulting pellet, containing spores and nonsporulated vegetative cells, was washed three times with sterile distilled water, resuspended in 5 ml of sterile distilled water and treated at 100°C for 30 min to inactivate the vegetative cells. The endospore suspension was centrifuged and washed 4 times as described above. The cleaned endospore preparation was stored at a final concentration of approximately 107–10CFU ml−1 in distilled water at 4°C.

Effect of nisin on bacterial spore

The effect of nisin on B. sporothermodurans LTIS27 spore (5 × 107 spores ml−1) was tested. Standard stock solution of nisin (Sigma-Aldrich) containing 1 × 104 IU ml−1 was prepared by dissolving 10 mg of nisin in 1 ml sterile 0·02 mol l−1 HCl. Spores, after preparation, were suspended in sterile distilled water containing different concentrations of nisin (0, 50, 100, 500, 1000 and 5000 UI ml−1). After 24 h of incubation at 37°C, the spore population was determined. Positive controls consisted of inoculated sterile distilled water without nisin. Negative controls consisted of uninoculated sterile distilled water in order to determine sterility.

High hydrostatic-pressure treatment

High-pressure processing was carried out in a 3-l vessel (ACB Pressure Systems, Nantes, France) equipped with temperature and pressure regulatory devices, capable of achieving a maximum pressure of 600 MPa and a temperature of up to 60°C. The samples were diluted 1 : 2 in sterile distilled water (5 × 107 spores ml−1), enclosed in polyethylene pouches, and subjected to combined pressure, initial temperature in the vessel and nisin treatments protocols given in Table 1. The compression rate was 34 Pa s−1, and the pressure release was almost instantaneous. The temperature of the water in the vessel was maintained at its initial value by circulating water in the double wall of the device from a water bath. The temperature increase due to the adiabatic heating effect was equivalent to approximately 1 to 3°C per 100 MPa depending in the initial temperature of water into the vessel. Due to the double wall-circulating water, temperature gradually decreased to its initial value during the treatment. After treatment, the pouches were immediately placed on ice and analysed within 10 min after the treatment.

Table 1. Experimental design and results for the central composite design
Trial no.FactorsY = log N0/N
X1 Pressure (MPa)X2 Temperature (°C)X3 Nisin (UI ml−1)ObservedPredicted

Enumeration of surviving spores

Serial dilutions of treated sample were prepared in peptone water (0·01%), and the viable counts were determined by plating on BHI-B12 agar. The plates were incubated at 37°C for 24 h, and then the colonies were enumerated. Inactivation was expressed as a logarithmic viability reduction log (N0/N) with N and N0, respectively the colony counts after a treatment and in the untreated sample.

Experimental design

A rotatable central composite design with three independent factors (α = 1·68) was performed in order to study the inactivation of B. sporothermodurans LTIS27 spores. The factors investigated were pressure (300–500 MPa), initial temperature into the vessel (30–50°C) and nisin concentration (50–150 UI ml−1). Each variable at five coded levels (indicated as −1·68, −1, 0, +1, +1·68 as shown in Table 2) were designed as experimental runs using the software STATGRAPHICS Centurion XV version 15.2.06 (Sigma Plus, Levallois-Perret, France). The real values of the independent variables (X) were coded according to eqn (1):

display math(1)
Table 2. Code and level of variables chosen for the trials
  1. a

    x 1 = (X 1 − 400)/100; x 2 = (X 2 − 40)/10; x 3 = (X 3 − 100)/50.

Pressure (MPa) x 1 X 1 231·8300400500568·2
Temperature (°C) x 2 X 2 23·230405056·8
Nisin concentration (UI ml−1) x 3 X 3 1550100150184

where X is the real value of the independent variable; xi is the coded value of the independent variable given by the experimental matrix; X0 is the real value of the central point and ΔX is the step change. X and x values are shown in Table 2. The experimental response was the number of log-cycle reduction of B. sporothermodurans LTIS27 spores [log (N0/N)]. It was estimated, taking into account the influence of the experimental factors. Three experiments were performed at the centre of the experimental domain in order to estimate residual variance value. An analysis of variance and estimation of response surface by multiple linear regressions were performed using the software STATGRAPHICS Centurion XV version 15.2.06.


Effect of nisin on the inactivation of spores

The effect of different concentrations of nisin on the inactivation of B. sporothermodurans LTIS27 spores is shown in Fig. 1. The spore inactivation was concentration-dependent and increased from 0·4-log spores ml−1 at 50 UI ml−1 to 4-log spores ml−1 at 5 × 103 UI ml−1.

Figure 1.

Effect of nisin concentrations on inactivation of Bacillus sporothermodurans LTIS27 spores (5 × 107 spores ml−1). Values given are means (error bars represent standard deviations) of three independent experiments.

Effect of moderate heat and nisin on the inactivation of spore by pressure

Predictive response model

The combined effects of pressure, moderate heat and nisin on the inactivation of B. sporothermodurans LTIS27 spores were studied, using RSM. The response (Y) was measured in terms of log-cycle reduction (log N0/N). The design matrix of the variables in coded units is summarized in Table 2 with the experimental results. The response was represented by the following second-order polynomial equation, containing 10 estimated coefficients, where x1 is pressure, x2 is temperature and x3 is nisin concentration.

display math(2)

The analysis of variance (anova) showed that the value was 0·9792 indicating a high degree of correlation between the observed and predicted values. The coefficient values of eqn (2) were calculated and tested for their significance. The P-values are used as a tool to check the significance of each coefficient which in turn may indicate the pattern of interactions between variables. The linear coefficients (x1, x2 and x3), the quadratic term coefficient (x12, x22 and x32) and the cross coefficients (x1x2 and x1x3) were significant with small P-values (P < 0·05). The other term coefficient (x2x3) was not significant (P < 0·05).

The qualification of the polynomial equation (eqn (2)) was validated with ten verification experiments carried out under different pressure–temperature–nisin combinations of process parameters (Table 3). The (0·99) and the statistical significance level of P < 0·0001 indicated that the derived model fitted the experimental data, and that the model was satisfactory and accurate. Therefore, the model was quite reliable method for predicting the inactivation of B. sporothermodurans LTIS27 spores.

Table 3. Experimental and predicted results of the ten check-points of inactivation of Bacillus sporothermodurans LTIS27 spores
Trial no.Factors= log N0/N
X1 Pressure (MPa)X2 Temperature (°C)X3 Nisin (UI ml−1)ObservedPredicted

Localization of optimum conditions

Bacillus sporothermodurans is responsible for nonsterility problems in UHT milk because of the high resistant of their spores to heat treatment (Pettersson et al. 1996). These bacteria grow in stored milk and their load can be higher than or equal to 10CFU ml−1 after 5 days of storage at 30°C (Hammer et al. 1995; Klijn et al. 1997; Montanari et al. 2004). Therefore, a five-log-cycle reduction of B. sporothermodurans LTIS27 spores was used as a target level of inactivation by combined pressure, moderate heat and nisin concentrations in this study. The process parameters for a five-log-cycle reduction of B. sporothermodurans spores were optimized. The response surfaces and their corresponding contours plots obtained by solving the regression equation (eqn (2)) are presented in Fig. 2. These graphical representations permit the visualization of the relationship between the response and experimental levels of each variable. In each figure, one factor is maintained constant at its optimal value determined by Statgraphics. Figure 2(a) summarizes the effect of pressure and temperature by keeping nisin concentration at its optimal value (121 UI ml−1). The reduction for B. sporothermodurans spores increases with increasing pressure and temperature, reaching a 5-log spores ml−1 reduction at pressures of 460–500 MPa and temperatures of 48–55°C. The interaction of pressure and nisin concentrations at a constant temperature (53°C) is shown in Fig. 2b. In the range of pressure 440–500 MPa and 90–150 UI ml−1 of nisin, a reduction of 5-log spores ml−1 of B. sporothermodurans spores was calculated. The reduction of B. sporothermodurans spores as a function of temperature and nisin concentration at a fixed pressure of 472 MPa is given by the response surface and contour plots in Fig. 2c. In the range of temperatures of 48 to 55°C and nisin concentrations of 110 to 150 UI ml−1, the reduction of B. sporothermodurans spores corresponds to 5-log spores ml−1.

Figure 2.

Response surface and contour plots showing the effects of pressure and temperature (a), pressure and nisin concentration (b), temperature and nisin concentration (c) on the inactivation of Bacillus sporothermodurans LTIS27 spores. The data on the contour maps are log N0/N. N0 represents the plate count of the untreated spore suspension (5 × 107 spores ml−1) and N represents the plate count after pressure treatment.


The result of the effect of different concentrations of nisin on the inactivation of spores shows that this bacteriocin, at high concentration, has sporicidal effect on B. sporothermodurans LTIS27 spores. This is not always the case for other Bacillus and Clostridium species. For example, Mansour et al. (1998) reported that nisin at 40 UI ml−1 had nonsporocidal effect on Bacillus licheniformis spores. Mazzotta et al. (1997) demonstrated that Clostridium botulium 56A spores were significantly resistant to 104 UI ml−1 of nisin, and sporocidal effect of nisin has been observed by De Vuyst and Vandamme (1994) on Clostridium sporogenes PA3679 spores previously damaged by heat treatment (3 min at 121°C). Given that nisin is able to reduce the initial count of B. sporothermodurans LTIS27 spores of 4-log cycle, we can hypothesize that these spores are particularly sensitive to nisin.

Generally, nisin exhibits antimicrobial activity towards a wide range of Gram-positive bacteria. It is able to prevent the growth of vegetative cells and outgrowth of spore derived from some species of Bacillus and Clostridium (Black et al. 2008; Cruz and Montville 2008; Gut et al. 2008). Its action against spores is far less understood than for vegetative cells. It is predominantly sporostatic rather than sporocidal (Delves-Broughton 2005). Morris et al. (1984) demonstrated that the nisin action against spores is caused by binding to sulfhydryl groups of protein residues. It was observed that spores became more sensitive to nisin the more heat damaged they are, and it is an important factor in the use of nisin as a food preservative in heat processed foods (Delves-Broughton et al. 1996).

The inactivation of B. sporothermodurans spores by combined HP, heat treatment and nisin was also investigated using design experiments. By analysing the pressure–temperature–nisin plots, it can be noticed that the spore inactivation increased with increasing pressure, temperature and nisin concentration. Furthermore, a significant synergistic effect was observed between pressure and temperature and between pressure and nisin, as confirmed by the significant interaction between these factors in the model. In fact, the 5-log spores ml−1 reduction of B. sporothermodurans LTIS27 spores was obtained under different combinations of pressure, temperature and nisin concentration. And the same inactivation can be obtained at lower pressure treatment by increasing temperature or nisin concentration. The optimal inactivation (5-log spores ml−1 reduction) could be obtained after a treatment at 472 MPa/53°C for 5 min in presence of 121 UI ml−1 of nisin according to the model optimization. The optimal conditions were tested in skim milk (data not shown). The inactivation found in milk was not significantly different from the inactivation found in distilled water. This remark is different from that reported in previous studies, which demonstrated that food components may act on treatment efficiency either by decreasing or increasing the inactivation. For example, Aouadhi et al. (2012) showed that pressure-induced germination was enhanced in milk. Whereas, milk had a protective effect against spores inactivation by combined high-pressure and thermal treatments (Aouadhi et al. 2013).

When nisin was applied at 100 UI ml−1 with no further treatment, an inactivation of 0·9-log spores ml−1 was observed. Besides, a HP treatment applied alone (600 MPa/5 min/20°C) has no significant effect on Bsporothermodurans spore inactivation (Aouadhi et al. 2012). Furthermore, B. sporothermodurans is resistant to high temperature treatments (141°C, 5 s) (Klijn et al. 1997). However, a synergistic effect was obvious when combining the three treatments. The antimicrobial properties of nisin were amplified by pressure and moderate heating. Sobrino-Lopez and Martin-Belloso (2008) reported that combining nisin with thermal and nonthermal processing techniques can act synergistically to reduce the population of different micro-organisms, including bacterial spores. In fact, on the one hand, HP treatment destabilizes membrane structure and renders cells more susceptible to antimicrobial action (Kalchayanand et al. 1998; Masschalck et al. 2001; Ray 2001). On the other hand, treatment with cell wall-weakening agents sensitizes pressure-resistant bacteria to HP (Earnshaw et al. 1995). Lopez-Pedemonte et al. (2003) showed that HP treatment may also improve the efficacy of nisin on spore inactivation by increasing the permeability of the spore coat. Black et al. (2008) studied the extent of physical damages after HP treatments on B. subtilis spores by using scanning electron micrographs. They demonstrated that, untreated spores have a typical smooth morphology. HP-treated spores (500 MPa/5 min) do not display much damage to their exterior structure; however, those treated at 500 MPa for 5 min cycled twice in the presence of 500 UI ml−1 nisin displayed extensive surface injuries, giving them a sunken and empty appearance. Through the high levels of inactivation of B. sporothermodurans spores LTIS27 seen with combined pressure and temperature treatments with nisin, we can hypothesize that HP and moderate heat causes some physical damage to coats of B. sporothermodurans spores, allowing entry of nisin molecules to their site of action.

The effect of nisin on inactivation of B. sporothermodurans LTIS27 spores by pressure increased with increasing temperature. This result is in accordance with the study of Stewart et al. (2000) on B. subtilis and Clostridium spores. They show that the inactivation of spores by combined HP (404 MPa, 15 min) and nisin (10 UI ml−1) treatments was improved by increasing temperature from 25 to 45°C and 70°C. Spore germination and inactivation under HP treatment generally increase with a moderate heating at around 50°C (Oh and Moon 2003; Opstal et al. 2004), meaning that modifications occurring in the spore coat under HP are favored under moderate heating. It can explain synergy between nisin and temperature observed in the present study.

A 5-log spores ml−1 reduction of B. sporothermodurans spores could be obtained by combining HP with nisin and moderate heating with the treatments conditions 472 MP/53°C and 121 UI ml−1 of nisin. Black et al. (2008) also demonstrated a 5- to 6-log spores ml−1 reduction of Bsubtilis spores after processing at 500 MPa/40°C for 5 min in presence of nisin (500 UI ml−1). According to Stewart et al. (2000), 5-log spores ml−1 reduction of B. subtilis spores was obtained with a treatment at 404 MPa, 45°C for 30 min at pH 5·0 in presence of 0·6 UI ml−1 of nisin. At least a 6-log spores ml−1 reduction of B. coagulans spores was generally achieved with pressurization at 400 MPa in pH 4·0 buffer at 70°C for 30 min in presence of nisin 0·8 IU ml−1 (Roberts and Hoove 1991). The optimum process parameters for a 6-log spores ml−1 reduction of B. coagulans AS 1.2009 spores were 550 MPa, 41°C, for 12 min and 120 IU ml−1 of nisin (Gao and Ju 2011). It is clear that there is a marked advantage of using the combination of three factors at sublethal injury level to inactivate Bacillus spores. The present study shows that such combination is even efficient on highly heat-resistant spores such as B. sporothermodurans.

However, it must be taken into account that the inactivation of spores under HP is highly variable and depends on the species, the processing conditions (pressure level, treatment time and temperature), the physicochemical parameters of the environment (pH of the suspension medium) and the presence or absence of antimicrobials agents. Masschalck et al. (2001) reported that the sensitization of bacteria by high pressure to antimicrobial compounds varies even among strains belonging to the same species. Black et al. (2008) showed the presence of significant variation in the levels of inactivation between species. For example, the application of a HP treatment (500 MPa/for 5 min) in the presence of nisin (500 UI ml−1) to four strains of B. cereus suspended in skimmed milk proved to be less effective than for B. subtilis.

It is thus necessary to further improve the mechanistic knowledge on spore inactivation under HP and the interaction with additional hurdles as the use of antimicrobial to design efficient and implementable treatments.


RSM involving experimental design and regression analysis was used to reveal the influence of pressure, temperature and nisin concentrations on the inactivation of Bsporothermodurans. Our results suggest that combining the treatments, HP-heating with nisin addition, results in synergistic inactivation of B. sporothermodurans, as has been shown for other spore species. The optimum process parameters obtained for a 5-log spores ml−1 reduction of B. sporothermodurans spores were 472 MPa/53°C for 5 min in presence of nisin (121 UI ml−1). This study shows the potential of using HP in combination with moderate temperature and nisin to inactivate B. sporothermodurans spores. It is particularly interesting because the initial contamination density (N0) of HRS spores in raw milk is generally lower than 5-log. Finally, this work illustrates the efficacy of applying such treatment to milk.


The authors gratefully acknowledge the financial support provided by the Tunisian Ministry of Higher Education and Scientific Research. A special thank to Dr. Marie de Lambellerie (LUNAM Université, Oniris, UMR 6144, GEPEA, BP 82225, Nantes, CNRS, Nantes, F-44322, France) and Dr. Hervé Prévost (LUNAM Université, Oniris, UMR 1014, SECALIM, BP 82225, Nantes, INRA, Nantes, F-44322, France).