Determination of Nisin Concentration to Control Listeria monocytogenes in Cooked Ham

Abstract

Listeria monocytogenes contamination may occur after thermal processing of cooked ham. This study aimed to determine the effective nisin concentration to inhibit the growth of L. monocytogenes. Two batches of ham were prepared by adding 12.5 mg nisin/kg ham at brine injection (assay 1) or tumbling stage (assay 2). Another piece of cooked ham was prepared by adding 32 mg nisin/kg ham at brine injection (assay 3). Afterward, samples of cooked ham were contaminated with L. monocytogenes cocktail (4 log CFU/mL). The results of assays 1 and 2 revealed that this nisin concentration was ineffective in inhibiting L. monocytogenes in ham. The minimum bactericidal concentration test results indicated that each L. monocytogenes strain had a different sensitivity profile and the cocktail had greater resistance. High concentrations of nisin (32 mg/mL) were efficient in controlling the most resistant strains. Statistical analyses were performed using ANOVA and Tukey’s test. This higher nisin concentration was effective in controlling the L. monocytogenes cocktail in cooked ham over 10 days. The findings highlight the potential of high nisin concentrations to effectively reduce L. monocytogenes contamination in cooked ham.

Introduction

Ham is highly susceptible to microbial contamination due to its high water activity (A_w > 0.9) and neutral pH. Additionally, procedures performed in retail stores (cutting, slicing, and packaging) increase the risk of cross-contamination due to improper handling or inefficient decontamination of equipment [1, 2]. Listeria monocytogenes is the most dangerous pathogen found in ham, causing up to 30% mortality rate in infected individuals [2]. Reportedly, there were 3 outbreaks of L. monocytogenes related to deli meat resulting in 3 deaths and 26 hospitalizations in the US between 2006 and 2022 [3]. Generally, the prevalence of L. monocytogenes in deli meats is 3%; however, it increases to 30% in deli meat from retail stores [1].

Nisin is produced by Lactococcus lactis and is permitted for addition to food items in more than 50 countries due to its inhibitory action against Gram-positive microorganisms [4]. Nisin causes pore formation in the cell membrane and inhibits enzyme activity and cell wall biosynthesis [5]. Nisin has “generally recognized as safe” (GRAS) status, with a maximum allowed level of 25 mg nisin/kg food [6] in cooked meat, poultry products, and frankfurters casings [7]. Similar to the US legislation, in Brazil, nisin can be added as a preservative in meat products at a maximum concentration of 25 mg nisin/kg food [8].

L. monocytogenes is frequently found in food handling environments due to its ability to form biofilms on surfaces and equipment [1]. It is difficult to prevent L. monocytogenes biofilm formation; therefore, addition of an antimicrobial compound directly on the food is an effective strategy to control this microorganism in deli meats. There is growing consumer demand for natural rather than synthetic additives in foods [9]. Nisin is a good choice of a natural antimicrobial for ham and deli meats because it is effective against L. monocytogenes. However, to date, no study has been performed to evaluate the antilisterial efficacy of nisin when integrated into ham at various processing stages. It is important to determine the optimal stage for nisin incorporation during ham production, as this antimicrobial compound may lose its effectiveness on interacting with components of the matrix, such as brine and meat. Additionally, it is essential to establish the minimum bactericidal concentration (MBC) of nisin for various strains of L. monocytogenes, both individually and as a cocktail. This information will enable application of appropriate concentrations and mitigating the risk of bacterial resistance.

Therefore, the objectives of this study were to evaluate the antilisterial activity of commercial nisin (Nisaplin®) to determine a nisin concentration able to inhibit L. monocytogenes in vitro and during ham processing.

Materials and Methods

Bacterial Cultures

Five L. monocytogenes strains: L. monocytogenes ATCC 7644, L. monocytogenes J11 (isolated from meat), L. monocytogenes 55 and L. monocytogenes 47 (both isolated from stainless steel surface), and L. monocytogenes 7459 (isolated from dairy products), were used. Each strain was grown individually in soy tryptone broth (Himedia, India) supplemented with 0.6% of yeast extract at 37 °C for 18 h. The growth of each strain was tested by using spread plate in Chromocult® agar and was found to be similar before preparing the cocktail assembly (8 log CFU/mL) (data not shown). To prepare the L. monocytogenes cocktails, 2 mL of each strain was added to a Falcon tube and centrifuged (HITACHI, model CR 21GIII, Ibaraki, Japan) at 10,000 × g for 10 min. The pellet was suspended in 10 mL of 0.1% peptone water (w/v) and the process was repeated twice. The concentration of suspension in the Falcon tube was adjusted to 8 log CFU/mL by adding 0.1% peptone water to a 0.5 (OD_630nm) optical density using an Ultrospec™ 3100 Pro spectrophotometer (GE Healthcare, England).Serial dilution with 0.1% peptone water was performed to obtain a final concentration of 4 log CFU/mL.

Nisin Solution Preparation

For preparing 1 mg/mL nisin solution, 1 g of de Nisaplin® was dissolved in 10 mL of 0.1 N HCl. Later, 4 mL of this solution was diluted in 10 mL of phosphate buffer (0.1 M).

Minimum Bactericidal Concentration

In vitro antimicrobial activity of nisin was tested against L. monocytogenes strains combined in two different cocktails (Cocktail 1: L. monocytogenes strains ATCC 7644, J11, 55, 47, and 7459. Cocktail 2: L. monocytogenes strains ATCC 7644, J11, 55, and 47) and against each strain separately [10]. A 100 µL volume of Mueller Hinton broth (MH, Sigma-Aldrich Saint Louis) was added to each well before 100 µL of nisin solution (1 mg/mL: “Nisin solution preparation” section) was added to the first well. After homogenization, sequential dilutions were performed as follows: 100 µL from the first well was transferred to the second well, then 100 µL from the second well to the third well, and so on. After the sequential dilutions, 100 µL of bacterial cocktail (4 log CFU/mL) was added to each well. This technique was repeated for all the five L. monocytogenes strains separately. The microdilution plates were incubated at 37 °C for 24 h and after incubation, 20 µL from each well was transferred to Chromocult® agar plates. After 24-h incubation, each plate was checked for growth of specific colonies and the lowest nisin concentration that inactivated the microorganism (i.e., MBC) was determined.

To test for interaction between nisin and brine compounds, MBC was determined with the same methods, using brine (Brine 1 from Table 1) instead of MH broth. This technique was repeated for L. monocytogenes cocktails and for the five L. monocytogenes strains separately.

Table 1.

Brine formulations

Brine ingredients	Brine 1 (g)	Brine 2 (g)	Brine 3 (g)
Isolated soy protein SUPRO® 548	20.6	20.6	20.6
Nisaplin®	0.5	–	1.28
Sodium chloride	15.0	15.0	15.0
Morex® glucose syrup 1940	10.1	10.1	10.1
Morex® Maltodextrin 1920	10.1	–	10.1
Carrageenan Metaline® 1240	5.1	5.1	5.1
Sodium tripolyphosphate	3.8	3.8	3.8
California aroma	5.6	5.6	5.6
Monosodium glutamate	2.4	2.4	2.4
Sodium pyrophosphate acid	1.5	1.5	1.5
Curing salt	2.1	2.1	2.1
Sodium erythorbate	1.0	1.0	1.0
Cochineal extract	0.15	0.15	0.15
Cold water	297.5	297.5	297.5

Ham Preparation and Processing

Three different ham formulations (assay 1, 2, and 3) were prepared in a small-scale processing laboratory in southern Brazil. Nisaplin® was acquired from Danisco Brazil Ltda. In assay 1 and 2, 12.5 mg nisin/kg ham was added in the brine and at tumbling stage, respectively (Fig. 1). In assay 3, 32 mg nisin/kg ham was added to the brine, according to the concentration determined in the MBC test (“Nisin solution preparation” section). Ham prepared by applying the brine 2 formulation, described in Table 1, was kept as control.

Fig. 1.

Flowchart of the production process of ham and steps proposed for nisin addition

Deboned meat samples (5% fat content) were injected with the different brine solutions (Table 1) reaching 36% brine concentration in ham. Later, the deboned meat was ground (disk 5) and tumbled at 12 rpm at 5 °C for 12 h. The samples were cured at 4 °C for 24 h, stored in a cook-in plastic, vacuum packed and cooked in water bath (72 °C), cooled (8 °C for 40 min), sliced, packed, and stored at 8 °C) Fig. 1 demonstrates production process of cooked ham, highlighting the differences between the assays.

In assays 1 and 3, nisin were added at the brine stage. In assay 2, nisin was added at the tumbling stage (Table 1). Nisaplin® (500 mg) and Morex® Maltodextrin 1920 (10.1 g) were mixed and added to meat emulsion after a 12-h tumbling, remaining in Tumbler (12 rpm) for 1 h. The meat emulsion was followed by the next steps of the production process (Fig. 1).

Microbiological Analysis

Prior to the assays, tests were carried out to confirm the absence of coagulase-positive Staphylococcus, thermotolerant coliforms, Salmonella spp. [11], and L. monocytogenes [12] in ham samples.

Contamination of Ham Slices

Sliced ham (25 g) was placed into individual sterile plastic bags, surfaces were artificially contaminated with 1 mL of the L. monocytogenes cocktail (4 log CFU/mL) and manually massaged for 2 min. For control sample, 25 g of sliced ham was produced without nisin. All samples were stored at 8 °C during the assays. This temperature is the same as the one used in domestic storage (7–8 °C) [13].

L. monocytogenes Count

In assay 1, counts of L. monocytogenes in ham slices were performed just after the inoculation (0 h), 24 h, 48 h, 10 d, 20 d, 30 d, and 40 d. In assays 2 and 3, counts were carried out at 0 h, 24 h, 48 h, 4 d, 6 d, 8 d, 9 d, and 10 d. Each sample (25 g) was placed in a sterile bag containing 225 mL of peptone water (0.1%) and homogenized in a Stomacher (model 542,765, Merck, DE) for 1 min. Serial dilutions were made with 0.1% peptone water and aliquots inoculated by spread plating on Chromocult agar and incubated at 37 °C for 48 h.

Statistical Analyses

Analysis of Variance (ANOVA) was used to test for significant differences in mean MBC values and L. monocytogenes counts among strains and cocktails. When ANOVA picked significant differences, the Tukey’s test was used for multiple comparisons. A significance level of 0.05 was used. All statistical analyses were performed in SPSS version 21. All experiments were performed in triplicates.

Results and Discussion

Minimum Bactericidal Concentration Against L. monocytogenes Cocktail and Isolated Strains

The MBC test performed using MH broth as diluent demonstrated different results depending on the bacterial strain. For L. monocytogenes strains 55 and 7459, the MBC was 31.25 µg/mL whereas for strains J11, 47, and ATCC 7644, the MBC was 15.63 µg/mL (p < 0.05). The MBC value for the two L. monocytogenes cocktails (cocktail 1 and 2) was 62.5 µg/mL, which was significantly higher than the MBC values for strains separately (Table 2). This effect may be due to genetic variability among strains and genetic exchanges [14, 15]. Cocktail 2 was prepared without strain 7459, which exhibited greater resistance to nisin. This test evaluated whether the resistance of cocktail 2 would be different from that of the cocktail containing all strains (cocktail 1). Both cocktails presented the same MBC value (62.5 µg/mL); indicating that presence or absence of this strain did not change the sensitivity of the cocktails to nisin.

Table 2.

Minimum Bactericidal Concentration (MBC) of nisin against L. monocytogenes isolated strains and strains grouped in cocktails

MBC (μg/mL)	L. monocytogenes
Strain	55	J11	47	7459	ATCC 7644	Cocktail 1	Cocktail 2a
Nisin	31.25b	15.63c	15.63c	31.25b	15.63c	62.5a	62.5a
Nisin and Brine	125c	31.25d	250b	250b	7.81e	500a	500a

Mean ± SD. Means values followed of different letter, in the same line, are significantly different (Tukey, p < 0.05)

^aCocktail without the wild isolate L. monocytogenes 74,593

The sensitivity profiles of the L. monocytogenes strains to nisin were different when the MBC test was performed using brine as diluent. Brine was tested to evaluate the synergistic or antagonistic effect of curing salts with nisin. The most resistant L. monocytogenes strains were strains 47 and 7459 (MBC: 250 µg/mL), followed by strain 55 (MBC: 125 µg/mL), strain J11 (MBC: 31.25 µg/mL), and strain ATCC 7644 (MBC: 7.81 µg/mL) (Table 2).

The MBC value for cocktails 1 and 2 was 500 µg/mL, which was higher than the MBC value of the resistant strains 47 and 7459 (MBC: 250 µg/mL). Except ATCC 7644, all other strains showed significantly higher resistance when brine was used instead of MH broth (p < 0.05) (Table 2), showing that brine exerted an antagonistic effect thus, reducing nisin antibacterial activity. Brine used in meat products is composed of various ingredients and compounds such as mono and diglycerid emulsifiers, ester glycerid acids, and phospholipids usually show antagonistic action against nisin [16].

Regardless of the diluent (MH broth or brine), MBC values for cocktails were higher than those for separate strains, demonstrating that bacterial strain behavior can be different when analyzed separately or in combination [17]. Cocktail 1 showed higher resistance to nisin than the isolated strains, therefore, it was used for artificial contamination of ham in subsequent experiments.

Nisin Activity Against L. monocytogenes in Cooked Ham

The nisin concentration used in this study was based on a previous work of the group, where ham was produced using the same formulation and methodology proposed here. In the previous study, 12.5 mg nisin/kg ham applied at the tumbling stage was effective in controlling lactic acid bacteria naturally occurring in ham, maintaining counts equal to the initial levels (1 log CFU/g) over 8 d after slicing [18].

In this study, the same nisin concentration (12.5 mg nisin/kg ham) was used to evaluate its effectiveness in inhibiting the L. monocytogenes cocktails. The ham formulation was identical for all samples. The differences between the samples were in nisin concentration and the stage at which nisin was added (brine injection or tumbling). All ham samples were tested before the assays to ensure they were free of contamination by Staphylococcus, thermotolerant coliforms, Salmonella spp., and L. monocytogenes (data not shown).

The counts of L. monocytogenes in cooked ham for assay 1 (12.5 mg nisin/kg ham at brine injection) are shown in Fig. 2. No growth of L. monocytogenes was seen until day 2 for nisin containing ham sample and control (p > 0.05). On day 10, nisin-containing ham sample had a L. monocytogenes level of 6.14 log CFU/g and the control sample had a level of 7.18 log CFU/g, with no significant difference (p > 0.05). After 20 days, L. monocytogenes counts reached 9 log CFU/g (p > 0.05). These results showed that 12.5 mg nisin/kg ham added in the brine was not effective for controlling L. monocytogenes growth in ham.

Fig. 2.

Viable cell counts of Listeria monocytogenes in cooked ham tested with 12.5 mg nisin/kg ham added at the brine injection stage (assay 1) and control sample (without nisin) incubated at 8 °C for 40 d

In assay 2, 12.5 mg nisin/kg ham nisin was added at the tumbling stage. The L. monocytogenes counts are shown in Fig. 3a. There was difference in L. monocytogenes counts between the nisin-containing sample (3.67 log CFU/g) and the control sample (4.74 log CFU/g) (p < 0.05) on the first day. Until day 6, L. monocytogenes count in the nisin-containing sample was 1 log CFU/g lower than in the control sample (p < 0.05). From day 8 until the end of assay, both samples presented L. monocytogenes counts close to 6 log CFU/g, with no significant difference between them (p > 0.05). These results were similar to those of assay 1 on day 10, demonstrating that regardless of method of nisin application (brine injection or during tumbling) this concentration of nisin (12.5 mg nisin/kg ham) is not effective in controlling L. monocytogenes in cooked ham. Additionally, the addition of nisin in the tumbling stage led to an adverse effect on the appearance of the ham (data not shown).

Fig. 3.

Viable cell counts of Listeria monocytogenes in cooked ham tested with: a 12.5 mg nisin/kg ham added at the tumbling stage (assay 2) and control sample (without nisin) incubated at 8 °C for 10 d. b 32 mg nisin/kg ham added at brine injection stage (assay 3) and control sample (without nisin) incubated at 8 °C for 10 d

Ham samples were evaluated over 10 d because it was observed that L. monocytogenes counts were constant since day 8, i.e., L. monocytogenes was in the stationary growth phase (p > 0.05).

Antibacterial activity of nisin depends on intrinsic and extrinsic factors of food, such as pH, acidity, A_w, salt and fat content, additives, enzymes, and inherent microbiota [19]. Nisin shows better performance than other antibiotics because of its stability at low or high temperatures (during short periods) and tolerance to acidity (low pH values) [20].

Nisin exhibits higher antimicrobial activity in liquid than in solid media [20]. Due to the complexity of food, higher concentrations of nisin are required to control microorganisms in food than in in vitro tests [21]. The reasons for this behavior might be the binding of nisin with meat proteins, nisin adsorption onto fat or degradation by meat proteases, making nisin unavailable to interact with microorganisms, thereby decreasing its antimicrobial activity [22].

Glutathione (GSH) is an important agent for cellular defense found in raw meat (beef, poultry, and pork). Several studies revealed that nisin can bind to GSH in fresh meats, making nisin unavailable to microorganisms. However, on cooking the meat, GSH levels decrease which subsequently reduces the amount of nisin bound to GSH thus, enhancing its antimicrobial activity [22].

Nisin acts against microorganisms by interacting with anionic lipids of the bacterial cytoplasmic membrane, which results in formation of pores. The pore formation causes leakage of ATP, amino acids, and other essential cellular components, leading to bacterial death [20, 23]. However, when nisin is applied to high lipid content foods, such as meat products, nisin might bind to the lipid content of meat instead of the cytoplasmic membrane lipids of microorganisms, thereby decreasing its antimicrobial activity [24, 25].

The curing of ham involves activating the proteolytic enzymes present in meat, producing substances responsible for the particular organoleptic characteristics of cured ham, such as texture, aroma, and flavor [26, 27]. However, the same proteolytic enzymes responsible for curing ham can inactivate nisin due to their ability to break down the peptide chain of nisin [20].

Besides the interaction with food components, the reduced antimicrobial activity of nisin might be caused by the presence of food additives, such as sodium metabisulfite, titanium dioxide [5], emulsifiers, ester glycerid acids, and phospholipids present in brine [16]. Other variables that might decrease antimicrobial activity of nisin in the food matrix include temperature of the assays and differences in the biology of L. monocytogenes, which grows planktonically in broths, but forms biofilms in foods [28].

As the nisin concentration of 12.5 mg nisin/kg ham was ineffective in controlling L. monocytogenes, the results of MBC test were used to determine the nisin concentration in assay 3. Based on Table 2, a concentration of 32 mg nisin/kg ham was chosen, because this concentration was efficient in controlling the most resistant strains, 55 and 7459. The concentration of nisin (62.5 mg/kg of ham) effective in inhibiting the cocktail could not be added to ham, because a large amount of Nisaplin® (powder form containing 2.5% pure nisin) would be needed.

Figure 3b shows that during assay 3, L. monocytogenes counts were equal or significantly lower (p < 0.05) on nisin-containing ham than in the control sample. Until the eighth day, the bacterial count was similar (p < 0.05) to that obtained after bacterial inoculation at the beginning of the test (4 log CFU/g). These results demonstrated that the concentration of nisin determined by the MBC test (32 mg nisin/kg ham) was effective in controlling the L. monocytogenes cocktail in ham.

A comparison of Fig. 3a, b reveals that until day 4, both samples exhibited L. monocytogenes counts close to the initial count (4 log CFU/g), probably due to the bacteria being in the lag phase of growth. From day 6 to day 10, samples with the highest nisin concentration (32 mg nisin/kg ham) had lower counts (p < 0.05) of L. monocytogenes than samples with the lowest concentration of nisin (12.5 mg nisin/kg ham). These findings are in agreement with those of Ruiz et al. [29] who demonstrated that higher nisin concentration greatly reduce the bacterial count. These results suggest that the nisin concentration of 32 mg nisin/kg ham is effective for controlling high levels of L. monocytogenes cocktail in ham over 10 days. These findings are promising as several previous studies reported consistent antimicrobial activity of nisin in in vitro tests, but a reduction in antibacterial activity in real meat products due to the complexity of the food matrix [30].

Conclusion

The MBC test demonstrated that brine had an antagonistic effect, which diminishes the antilisterial activity of nisin. Additionally, the test revealed that a higher concentration of nisin was required to inhibit the bacterial cocktail than for individual strains. The application of 12.5 mg nisin/kg of cooked ham, both at the brine injection and tumbling stages, was ineffective in controlling the growth of L. monocytogenes. Effective growth control of L. monocytogenes for 10 days was achieved by incorporating 32 mg of nisin/kg of ham into the brine. These findings suggest that a high concentration of nisin is effective in controlling L. monocytogenes in ham. Further studies are required to assess the effectiveness of applying nisin in combination with other food preservation methods, and the nanoencapsulation of this antimicrobial compound to protect it from interactions with ham constituents.

Author Contributions

Conceptualization: DCL, PdaSM, and ECT; Methodology: DCL, PdaSM, and ECT; Formal analysis and investigation: DCL; Writing–original draft preparation: DCL; Writing–review and editing: FAC and LTC; Supervision: PdaSM and ECT.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest.