Previous Article | Next Article ![]()
Applied and Environmental Microbiology, April 2004, p. 2271-2278, Vol. 70, No. 4
0099-2240/04/$08.00+0 DOI: 10.1128/AEM.70.4.2271-2278.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Research Group of Industrial Microbiology, Fermentation Technology and Downstream Processing (IMDO), Department of Applied Biological Sciences, Vrije Universiteit Brussel, B-1050 Brussels, Belgium
Received 29 October 2003/ Accepted 5 January 2004
|
|
|---|
|
|
|---|
At present, consumers are requesting safe, natural food products as well as mild and light foods with a low acid, sugar, salt, or fat content (27). This trend has stimulated research on the additives used in meat processing, such as NaCl and curing agents. For instance, to decrease the high Na+ content of meat, attempts are being made to replace at least part of the NaCl with other salts such as potassium chloride or calcium chloride, without affecting the fermentation process or the quality of the end product (26). The request for more-natural products has also provided an incentive to search for safe, food-grade preservatives of biological origin (16, 60). Furthermore, some food-borne pathogenic bacteria, such as Listeria monocytogenes, can survive in fermented sausage despite the various hurdles they encounter, and they may pose a health risk to consumers (21, 35, 40). The spread of antibiotic resistance genes among listeriae and other pathogens is raising new concerns (61). Moreover, countries such as the United States have adopted a zero-tolerance policy for Listeria contamination of processed meat products (59). This policy has serious consequences, in particular for European exporting companies, since the European Union applies a tolerance of less than 100 cells of L. monocytogenes in fermented meat products. Hence, there is a need for novel and/or extra preventives in meat processing.
Bacteriocins produced by LAB are antibacterial peptides or proteins active against other gram-positive, mainly closely related bacterial species, including some undesirable spoilage bacteria and food-borne pathogens (17). The only legally approved bacteriocin in many countries is nisin, approved for use as a preservative in a limited number of food products (15). Unfortunately, this bacteriocin is not very efficient in a meat environment, because of its low solubility, uneven distribution, and lack of stability (60). An alternative and interesting approach for the use of bacteriocins in fermented sausages is the use of bacteriocinogenic LAB starter cultures adapted to the specific meat environment (43). Several bacteriocins produced by strains isolated from fermented sausages are highly active against listeriae (13, 29, 33, 48, 55, 66). Moreover, several investigators have already shown the possibility of using LAB strains that produce listericidal bacteriocins to reduce Listeria counts in fermented meat products (6, 7, 10, 11, 22, 33, 34, 38, 58, 66). This is one of the reasons why Listeria contamination of fermented sausage is very rare, although Listeria contamination of the raw meat materials does occur (56). Although the meat itself may be contaminated, it is also possible that the contamination occurs in the meat plant. Hence, one of the most important objectives remains the prevention of Listeria contamination through good manufacturing practice, thereby decreasing the hygiene risk. However, the presence of this organism in the food-processing environment seems inevitable. Although the risk of contamination can be reduced, it seems unlikely that it can be eliminated. In addition, in situ production of bacteriocins enhances the competitiveness of the starter organisms toward fortuitous flora and hence ensures a stable and safe end product (31, 66). Therefore, it is of the utmost importance to know how the growth and bacteriocin production of the functional starter cultures used are affected by various sausage ingredients such as salt.
Lactobacillus curvatus LTH 1174, a strain originating in fermented sausage, is the producer of the listericidal bacteriocin curvacin A and has been shown to be a successful starter strain for European sausage fermentations (66). Previously, it was also shown that this strain displays maximum bacteriocin activity levels under the temperature and pH conditions used for these fermentations (47). However, the effects of different sausage ingredients on the behavior of this strain have yet to be determined so as to estimate the bioavailability of bacteriocin in the meat. The present study was undertaken to examine and model the inhibitory effects of different NaCl concentrations in a modified de Man-Rogosa-Sharpe (mMRS) medium on the growth and bacteriocin production of the meat starter culture L. curvatus LTH 1174.
|
|
|---|
mMRS was used as the fermentation medium for L. curvatus LTH 1174. The modification consisted in a doubling of the concentration of the complex nutrient source (bacteriological peptone [Oxoid], Lab Lemco [Oxoid], and yeast extract [VWR International, Darmstadt, Germany]). This was done to avoid a severe growth limitation of L. curvatus strain LTH 1174 due to nutrient depletion in standard MRS medium (44). In addition, calculations of the amino nitrogen content of this medium (9) indicated that this composition more closely simulates the actual sausage environment (14). Furthermore, maximum curvacin A activity was found when the complex nutrient source was doubled, while specific bacteriocin production and the specific apparent bacteriocin inactivation rate were comparable to those with standard MRS (J. Verluyten, W. Messens, F. Leroy, V. Schrijvers, and L. De Vuyst, unpublished data). All media and solutions were sterilized at 121°C for 20 min.
Fermentation experiments.
A series of in vitro fermentations were performed using mMRS supplemented with different concentrations of salt (0, 2, 4, and 6% [wt/vol] NaCl) to investigate their effects on both the growth and bacteriocin production of L. curvatus LTH 1174. The fermentations without added sodium chloride and those with 4% (wt/vol) NaCl added were performed in triplicate to show the reproducibility of the experiments. Standard deviations were calculated both on the experimental values and on the biokinetic parameters derived from the primary model. To determine whether the effect of the added NaCl was due solely to a reduction in aw, an additional fermentation was carried out in the presence of 9.9% (wt/vol) glycerol (sterilized separately) in the absence of added salt. Based on extrapolation of data from previous studies (12, 46), addition of 9.9% (wt/vol) glycerol to the basal growth medium should result in the same decrease in aw as addition of approximately 4% (wt/vol) NaCl (42).
Fermentations were carried out in a 15-liter laboratory fermentor (BiostatC; B. Braun Biotech International, Melsungen, Germany) with a working volume of 10 liters of mMRS as previously described (41). Briefly, temperature (25°C) and pH (held constant at 5.5) control was performed online (Micro-MFCS for Windows NT; B. Braun Biotech International). For preparation of the inoculum, 10 ml of MRS medium was inoculated with 0.5 ml of a freshly prepared L. curvatus LTH 1174 culture and then incubated at 30°C for 12 h. Five milliliters of this preculture was added to 100 ml of MRS medium. After 13 h of growth at 30°C, this culture was used to inoculate the fermentor.
To examine the effect of added induction factor (IF), two additional fermentations were performed using mMRS containing 6% (wt/vol) NaCl at a controlled temperature of 25°C and a constant pH of 5.5. To obtain a source of IF, a fermentation was performed in MRS at a controlled temperature of 22°C and a constant pH of 5.4, a set of conditions optimal for the production of curvacin A (47). After 24 h of growth, cells were removed by centrifugation (at 20,500 x g and 4°C for 30 min), and 50 ml of the cell-free culture supernatant (CFS; curvacin A activity, 1,600 arbitrary units [AU] ml1) was used as a source of IF for the first fermentation containing 10 liters of mMRS. This fermentation using the CFS was also performed in triplicate, and standard deviations were calculated. As an alternative source of the IF, a precipitate of curvacin A and possibly its IF was prepared as follows. The pH of the CFS was adjusted to 6.5, 300 g of ammonium sulfate per liter (48% saturation) was added, and the suspension was stirred at 4°C for 12 h. After centrifugation (at 20,500 x g and 4°C for 30 min), the pellicle and pellet were harvested and dissolved in 50 mM sodium phosphate buffer at pH 6.5 (5 ml per liter of CFS). The second fermentation was carried out by adding 1.5 ml of this protein solution (ammonium sulfate precipitate; curvacin A activity, 153,600 AU ml1) to the fermentor (10 liters of mMRS).
Assays.
At regular time intervals, samples were withdrawn aseptically from the fermentor to determine cell counts (CFU), biomass (cell dry mass [CDM]), levels of soluble bacteriocin activity in CFS, lactic acid concentrations, and residual glucose concentrations. Briefly, CDM was determined gravimetrically after membrane filtration (41); the amount of lactic acid produced and the residual glucose concentration were determined by high-performance liquid chromatography (18); and the level of bacteriocin activity was determined by a modified critical dilution method using L. innocua LMG 13568 as the indicator strain (18). The standard deviations for the CDM, glucose, and lactic acid measurements were 0.11, 0.04, and 0.02 g liter1, respectively.
Primary and secondary modeling.
Primary modeling of cell growth, glucose consumption, lactic acid production, and bacteriocin production and inactivation was performed both to fit the data and to estimate the values of the biokinetic parameters representative of growth and curvacin A production. Abbreviations for biokinetic parameters are explained, and the equations used are given, in Table 1. These equations are the same as those reported by Messens et al. (47), except that bacteriocin production was made dependent on XB, defined as the minimum biomass concentration required for the onset of bacteriocin production due to induction (19, 44). For fermentations performed in triplicate, standard deviations were calculated on all biokinetic parameters estimated via primary modeling.
|
View this table: [in a new window] |
TABLE 1. Equations used for primary model development
|
By use of secondary modeling, all biokinetic parameters (µmax, Xmax, n, YX/S, ms, YL/S, kB, and kinact) derived from the primary model were expressed as functions of the salt concentration. For this purpose, empirical models were used. Standard deviations were calculated where appropriate.
|
|
|---|
![]() View larger version (13K): [in a new window] |
FIG. 1. Modeling of biokinetic parameters for L. curvatus LTH 1174 in mMRS at a controlled temperature of 25°C and a constant pH of 5.5. (a) Biomass (in grams [CDM] per liter) ( ) and bacteriocin production (in MAU per liter) ( ). (b) Residual glucose concentration ( ) and lactic acid formation () (both in grams per liter). Symbols represent experimental values; lines are drawn according to the model. Experimental values are averages from three fermentations. Error bars, standard deviations.
|
|
View this table: [in a new window] |
TABLE 2. Values of biokinetic parameters and empirical-model equations showing the relationship between biokinetic parameters and sodium chloride concentrationsa
|
![]() View larger version (15K): [in a new window] |
FIG. 2. Secondary modeling of the biokinetic parameters derived from the primary model. Shown are the effects of different NaCl concentrations (expressed in percentages [wt/vol]) added to mMRS on the different parameters for L. curvatus LTH 1174 growing at a controlled temperature of 25°C and a constant pH of 5.5. (a) Bacterial growth (µmax) ( ) per hour and inhibition exponent (n) ( ). (b) Sugar metabolism (YX/S) ( ), expressed in grams of CDM per gram of glucose, and mS ( ), expressed in grams of glucose per gram of CDM per hour. (c) Maximum biomass production (Xmax) ( ), expressed in grams of CDM per liter, and maximum bacteriocin activity (Bmax) ( ), expressed in MAU per liter. (d) Bacteriocin production (kB) (), expressed in MAU per gram of CDM, and apparent rate of bacteriocin inactivation (kinact) ( ), expressed in liters per gram of CDM per hour. For 0 and 4% added NaCl, biokinetic parameters are means from three fermentations; error bars, standard deviations. Lines are drawn according to the model.
|
![]() View larger version (13K): [in a new window] |
FIG. 3. Effects of sodium chloride (concentrations expressed in percentages [wt/vol]) added to mMRS on the production of biomass (expressed in grams [CDM] per liter) (a) and bacteriocin production (expressed in MAU per liter) (b) as a function of time. Symbols: , 0% NaCl; , 4% NaCl; , 6% NaCl. Symbols represent experimental values; lines are drawn according to the model. For 0 and 4% added NaCl, the experimental values and the model derived are means from three fermentations; error bars, standard deviations.
|
Glucose was consumed more slowly at higher concentrations of NaCl, resulting in retarded production of lactic acid (Fig. 4).
![]() View larger version (13K): [in a new window] |
FIG. 4. Effects of different concentrations of sodium chloride (expressed in percentages [wt/vol]) added to mMRS on lactic acid production. Symbols: , 0% NaCl; , 4% NaCl; , 6% NaCl. Symbols represent experimental values; lines are drawn according to the model. For 0 and 4% added NaCl, the experimental values and the model derived are means from three fermentations; error bars (shown only for 4% added NaCl), standard deviations.
|
When the different biokinetic parameters used for the primary model were compared, the values obtained for fermentation in the presence of glycerol (9.9%, wt/vol) were comparable to those obtained for fermentation with 4% (wt/vol) added NaCl. Only the value of Xmax (2.06 versus 2.65 ± 0.18 g [CDM] liter1) was somewhat lower. The difference in Xmax may be ascribed to the formation of cell aggregates, resulting in a nonhomogenous fermentation liquor.
Effect of added IF.
To determine the effect of added IF on curvacin A production, fermentations in the presence of 6% (wt/vol) NaCl were performed by adding either CFS or protein solution (ASP) as a source of IF. These fermentations were compared with a fermentation in which no CFS or ASP was added. Bacteriocin activity was determined immediately after inoculation and addition of CFS or ASP. No inhibition of the indicator strain was observed, indicating that the amount of bacteriocin that was added in this way did not contribute to the observed inhibition during these experiments. Growth was comparable in the three cases, although less formation of aggregates was observed during the fermentations with added IF. This resulted in an Xmax of 2.02 ± 0.08 or 2.10 g (CDM) liter1 with added CFS or ASP, respectively.
Modeling of bacteriocin activity for fermentations with and without the addition of IF (CFS or ASP) is represented in Fig. 5. Addition of IF resulted in an increase in Bmax from 0.30 mega arbitrary unit (MAU) liter1 (no IF added) to 0.79 ± 0.10 or 0.52 MAU liter1 when CFS or ASP was added, respectively. These bacteriocin activities were the result of specific bacteriocin production levels of 0.40 MAU (g of CDM)1 (no IF added) and 0.62 ± 0.06 or 0.40 MAU (g of CDM)1 when CFS or ASP was added, respectively, indicating that CFS was a better source of IF than ASP. Possibly the precipitation of the IF was not very successful, or the IF may form inactive complexes in a more isolated form. Another possibility is that other compounds present in CFS stimulate bacteriocin production. The values for kinact were comparable in the three cases. Addition of IF also affected the onset of curvacin A production, as reflected in the XB. When either CFS or ASP was added, bacteriocin production started from the beginning of the fermentation and XB was reduced to only 0.03 ± 0.03 g (CDM) liter1, compared with 0.90 g (CDM) liter1 when no IF was added. Finally, sugar metabolism was not affected by the addition of IF, resulting in comparable values for YX/S and mS.
![]() View larger version (14K): [in a new window] |
FIG. 5. Modeling of bacteriocin production (in MAU per liter) by L. curvatus LTH 1174 either with no IF added ( ) or with the addition of CFS ( ) or ASP ( ) in mMRS with 6% (wt/vol) added sodium chloride at a controlled temperature of 25°C and a constant pH of 5.5. Symbols represent experimental values; lines are drawn according to the model. For the fermentation with CFS added, the experimental values and the model derived are means from three fermentations; error bars, standard deviations.
|
|
|
|---|
Salt affects the growth and bacteriocin production of LAB. In the case of L. curvatus LTH 1174, a low concentration of NaCl (2%, wt/vol) showed no effect on bacterial growth, while an inhibition that increased linearly was evident with higher salt concentrations. The strong negative effect of high salt concentrations on the growth of LAB has been reported previously (18, 54, 64). However, low concentrations of salt (1 to 2%, wt/vol) can sometimes enhance bacterial growth (23, 36, 64, 65).
Although a strong negative effect of salt was observed on the maximum specific growth rate (µmax) of L. curvatus LTH 1174, the maximum attainable biomass concentration (Xmax) did not decrease. This indicates that the aw constraints caused by NaCl addition may not impose large energetic burdens (37). In addition, replacing the sodium chloride with glycerol as an aw-lowering agent resulted in comparable inhibition; hence, the effect exerted by NaCl was probably due to its effect on aw. Although an ionic effect cannot be excluded, such an effect is not likely, because of the natural salt resistance of the LAB present in fermented meat and vegetable products (5, 36). However, Lactobacillus sakei CTC 494 shows a decreasing trend of Xmax with increasing NaCl concentrations (42). This may indicate that adding NaCl to the growth medium interferes with the efficiency of the substrate conversion into biomass for the latter strain. A linear decrease in the growth rate with aw values below the optimum is usually observed (46, 53).
Even at a concentration of 2% (wt/vol) NaCl, which did not affect the bacterial growth of L. curvatus LTH 1174, the maximum attainable bacteriocin activity (Bmax) decreased by almost 40% from that in the fermentation without added salt. This can be explained by the decrease in specific bacteriocin production (kB), since Xmax values were comparable. At higher NaCl concentrations, kB decreased further. In contrast, Uguen et al. (64) reported an increased lacticin 481 production when the osmolarity of the growth medium increased due to added NaCl. Also, for plantaricin S, the highest production is observed at a sodium chloride concentration of 2.5% (wt/vol) (39). On the other hand, the production of sakacin K by L. sakei CTC 494 is negatively affected by added NaCl (42), as is the case for the antilisterial carnobacteriocin B2 produced by Carnobacterium piscicola A9b (30). Although enterococci are more salt resistant, production of the enterocins A and B by Enterococcus faecium CTC 492 is also inhibited in the presence of NaCl (3). However, enterocin production can be increased in the meat matrix by the addition of a small amount of the bacteriocin, which acts as an inducer (4).
A possible explanation for the inhibition of bacteriocin production by NaCl is interference with the binding of the IF to its receptor (50). The IF is excreted by the producing strain, and when it reaches a critical concentration, the binding to its receptor initiates bacteriocin production. For C. piscicola A9b, the induction capacities of both acetate and carnobacteriocin B2 itself are negatively affected by the addition of NaCl (51). Moreover, for E. faecium CTC 492, addition of IF can overcome the inhibition of enterocin production (3, 50). Even at NaCl concentrations that do not affect growth, the induction of bacteriocin production decreases, indicating that higher concentrations of the inducer are necessary to sustain bacteriocin production (50). Likewise, curvacin A production by L. curvatus LTH 1174 is abolished at elevated temperatures (35°C) due to reduced synthesis of the cationic pheromone peptide Sap-Ph (19, 47). Possibly, addition of NaCl to the growth medium interferes with pheromone stability or binding to its receptor and/or reduces pheromone synthesis. Indeed, addition of CFS or ASP as a source of IF increased Bmax. In contrast, addition of IF could not reverse the negative effect of NaCl on sakacin P production (49), which was explained by the fact that sakacin P production is an "all or nothing process" (20).
Alternatively, in the actual sausage environment, the lower specific bacteriocin production may very well be counterbalanced by an increased relative bacteriocin activity, due to a higher susceptibility of pathogens in the presence of certain sausage ingredients such as salt. For instance, the addition of NaCl acts synergistically on the activity of curvacin A against Escherichia coli O157:H7 (24). Other strains of E. coli tested are also sensitive to nisin and curvacin A at a pH of <5.5 and in the presence of >3% (wt/vol) NaCl (25). For the enterotoxin-producing organism Bacillus cereus, a similar higher susceptibility to bacteriocin AS-48 is observed at an NaCl concentration of 5% (wt/vol) (1). For L. monocytogenes, a higher susceptibility to curvaticin 13 or nisin is also observed at higher NaCl concentrations (8, 62). Fewer bacteriocin molecules seem to be necessary to exert the same lethal effect. Moreover, during fermentation, a pH drop occurs, which will render certain species more susceptible to the bacteriocin that is produced in situ.
L. sakei and L. curvatus are the dominant LAB in European sausage starters (28). However, their adaptive responses to the meat environment differ in that L. curvatus is slightly more sensitive to salt than L. sakei (36), as is also shown in papers on L. curvatus LTH 1174 and L. sakei CTC 494 (42; this study). This difference may at least partially explain the dominance of L. sakei over L. curvatus in fermented sausages, as is often observed upon isolation of these strains from such products (2, 32, 52, 57). The use of bacteriocin-producing meat starter cultures, such as L. curvatus and L. sakei, for sausage fermentation enhances their competitiveness and contributes to a safer and more uniform end product. It is essential to understand the effects of different sausage-related environmental factors on the induction of curvacin A production in order to better understand the potential for application of this strain in the production of fermented sausages. The potential of this strain can be considerably broadened by the addition of CFS containing IF to achieve a higher level of bacteriocin production under unfavorable conditions. The addition of fermentation liquor or lyophilized whole-culture medium is an avenue to be investigated.
Vincent Schrijvers and Tom De Winter are acknowledged for technical assistance. L. curvatus LTH 1174 was kindly provided by W. P. Hammes (Institut für Lebensmitteltechnologie, Universität Hohenheim, Stuttgart, Germany).
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»