Microbial solutions to microbial problems; lactococcal bacteriocins for the control of undesirable biota in food


Prof. Colin Hill, Microbiology Department, University College Cork, Cork, Ireland (e-mail: c.hill@ucc.ie).

1. Summary

This review will discuss the potential for certain lactococcal bacteriocins to be used as tools for influencing food safety and quality. The focus will be on the potential use of a broad-spectrum bacteriocin, lacticin 3147, as a tool to control nonstarter lactic acid bacteria (NSLAB) in cheese, to inhibit pathogens in fermented and nonfermented foods and to extend the shelf life of certain products. Different strategies used to incorporate this bacteriocin into foods will also be described, as will possible problems that may arise when using a bacteriocin in foods. The narrow spectrum bacteriocins, lactococcin ABM (which actually corresponds to three bacteriocins, lactococcins A, B and M produced in a single strain) and the medium-spectrum bacteriocin lacticin 481, also have certain potential applications (albeit limited) in the food industry. In particular, these bacteriocins exhibit a bacteriolytic effect on target cells and so can be used for inducing lysis on sensitive cells during cheese manufacture which concomitantly can lead to accelerated flavour development.

2. Introduction

Bacteriocins are antibacterial peptides produced by a wide range of bacteria. The majority of Gram-positive bacteriocins identified to date are produced by lactic acid bacteria (LAB), and a number of these are produced by members of the genus Lactococcus. Because lactococci and other LAB are generally regarded as safe (GRAS) organisms, their bacteriocins have attracted particular attention in recent years in an attempt to develop their potential commercial applications.

Although the bacteriocins identified to date are diverse, both structurally and genetically, they are usually assigned to one of three broad subgroups (Klaenhammer 1993). All lactococcal bacteriocins characterized thus far belong to either group I or group II. Group I bacteriocins are termed lantibiotics. These are ribosomally synthesized peptides characterized by post-translational modification of primary residues into modified amino acids such as lanthionine (Ala-S-Ala) and β-methylanthionine (Abu-S-Ala). Members of this group include three lactococcal bacteriocins; nisin, lacticin 481 and lacticin 3147, in addition to those produced by other Gram-positive genera, which include subtilin, epidermin and mersacidin. Nisin, first discovered in the 1920s (Rogers 1928), is the prototype lantibiotic. It is a 34 amino acid peptide produced by some strains of Lactococcus lactis with a wide spectrum of inhibition against Gram-positive micro-organisms in favourable conditions. Nisin has a wide range of applications because of its broad bactericidal mode of action and because it can be easily broken down by digestive proteases so as to not disturb gut biota. It is the only lantibiotic to date that has been approved for commercial use. It was added to the positive list of EU food additives in 1983 as additive number E234 and was approved by the Food and Drug Administration (FDA) in 1988. Nisin is now approved for use as a food preservative in approximately 50 countries (Delves-Broughton 1990).

Although nisin is the only commercially exploited lantibiotic to date there has been an extensive effort devoted to developing applications for other lantibiotics. Lacticin 3147 is also a broad-spectrum lantibiotic with potential uses in the food industry and in medicine. It is a two-component lantibiotic (two separate peptides act in concert for full activity) produced by L. lactis subsp. lactis DPC3147, which was first isolated from an Irish kefir grain (Ryan et al. 1996). Lactococcus lactis IFPL105 is another lacticin 3147 producer and was isolated from goats milk in Spain (Casla et al. 1996; Martínez-Cuesta et al. 2000a). The genes required for DPC3147 to produce lacticin 3147 are carried on a 60·2-kb conjugative plasmid, pMRC01 whereas IFPL105 relies on a 46-kb nonconjugative plasmid, pBAC105. Lacticin 3147 has a broad spectrum of inhibition exhibiting a bactericidal mode of action against all Gram-positive bacteria tested to date; including food spoilage bacteria such as Clostridium sp., food pathogenic organisms such as Listeria monocytogenes, the mastitis-causing pathogens Staphylococcus aureus and Streptococcus dysgalactiae (Ryan et al. 1996; Casla et al. 1996) and human pathogens such as methicillin-resistant Staph. aureus, vancomycin-resistant Enterococcus faecalis, penicillin-resistant Pneumococcus, Propionibacterium acne and Streptococcus mutans (Galvin et al. 1999). It is also extremely heat-stable, active over a broad pH range and, as the genetic determinants are encoded on a self-transmissible plasmid, pMRC01, it can be conjugally transferred to different strains. These features make lacticin 3147 attractive for developing potential applications in the food industry, veterinary medicine and possibly in the treatment of human diseases.

Lacticin 481 is another lantibiotic produced by certain L. lactis strains and exhibits a medium spectrum of activity, inhibiting a broad range of other LAB and Clostridium tyrobutyricum. This peptide can have a bacteriolytic effect on sensitive organisms (O'Sullivan et al. 2002). It has been studied for its potential use in cheese ripening to induce lysis of starter strains and therefore deliver lactococcal enzymes during cheese manufacture to improve both flavour and quality.

The Group II bacteriocins are small, unmodified, cationic, hydrophobic peptides. Their inhibitory spectrum is usually narrow, limited to those with a low G + C content such as Listeria sp., Clostridium sp. or other LAB. In some cases, a single strain can produce multiple bacteriocins. For example, some L. lactis strains can produce lactococcins A, B and M, three class II bacteriocins (Morgan et al. 1995). Although these have a narrow spectrum of activity, inhibiting only other lactococci, they have potential applications in the dairy industry as, like lacticin 481, they have a bacteriolytic mode of action which can be exploited for the accelerated lysis of starter cultures during cheese manufacture and ripening.

This review will discuss the potential applications for certain lactococcal bacteriocins. Primarily, the focus will be on the possible use of lacticin 3147 to control NSLAB in cheese manufacture, as well as its potential to inhibit pathogens in foods and to extend the shelf life of certain products. The potential applications of lacticin 481 and lactococcin ABM to improve the overall flavour and quality during cheese ripening will also be reviewed.

3. Applications of lactococcal bacteriocins in the food industry

3.1 Use of lacticin 3147 to control nonstarter lactic acid bacteria in cheese

Lacticin 3147 has been investigated for use in the dairy industry to control NSLAB during the fermentation of dairy products (Ryan et al. 1996; Fenelon et al. 1999). NSLAB are primarily lactobacilli which, although not deliberately added to the fermentation process, can enter through cheese vats and become the dominant flora in a six month ripened cheese. The role of NSLAB in the development of cheese flavour is still not fully clear. Although they can contribute positively to the overall quality of the cheese, they are also associated with off-flavours and with the formation of calcium-lactate crystals and slit-defects often found in cheese. The control of NSLAB in cheese ripening would allow for a more predictable end product.

The most efficient method of introducing lacticin 3147 to a fermented dairy product is to use a lacticin-producing culture as the starter or as a starter adjunct in the fermentation process. DPC3147, the natural lacticin 3147 producer, is unsuitable as a starter culture as the strain is associated with an off-flavour. However, this problem can be eliminated by taking advantage of the conjugative nature of pMRC01 (Ryan et al. 1996; Coakley et al. 1997; O'Sullivan et al. 1998). This approach, which is a frequently used method for genetic improvement (Gasson and Fitzgerald 1994), allows the directed transfer of the plasmid responsible for producing the bacteriocin, pMRC01, to a commercially used lactococcal starter (the newly producing strain is termed a transconjugant). Over 30 lacticin 3147 transconjugants have been created to date (Ryan et al. 1996; Coakley et al. 1997; O'Sullivan et al. 1998). An important consideration when creating and using transconjugants as starters is to ensure that their important industrial traits have been retained and that they remain suitable for cheese manufacture. Nisin-producing starters, for example, have often been associated with slow acid production, reduced proteolysis and poor heat resistance when compared with a commercial starter (Lipinska 1973, 1977). Ryan et al. (1996) showed that a lacticin 3147-producing transconjugant starter (DPC4275) produced acid at comparable rates with the parental commercial starter. It was also shown that the level of bacteriocin produced was constant throughout the ripening process and at the end of cheese production the NSLAB population had been reduced by at least 100-fold (Ryan et al. 1996). A reduction in the NSLAB population was also seen when DPC4275 was used to make low fat cheese at increased ripening temperatures (Fenelon et al. 1999). This offers the cheese-maker far greater control over the developing adventitious flora without the need to resort to expensive alternatives such as cold-ripening. However, as mentioned earlier NSLAB may contribute positively to the flavour and quality of the cheese and therefore it may be desirable that certain NSLAB develop to high numbers during the ripening process. This objective was achieved when a lacticin 3147-resistant variant of Lactobacillus paracasei subsp. paracasei DPC5336 was isolated after repeated exposure to low levels lacticin 3147 (Ryan et al. 2001). This strain, which still remains sensitive to high levels of lacticin, was then used in conjunction with a lacticin 3147-producing starter in Cheddar cheese manufacture. While other NSLAB were inhibited during ripening, the resistant mutant could tolerate the levels of lacticin 3147 and became the dominant microflora in the cheese (Ryan et al. 2001). Developments like this give the cheese-maker greater control of the microbial biota.

Another advantage of using lacticin 3147 transconjugants as dairy starters is that the plasmid pMRC01 encodes bacteriophage resistance, specifically an abortive infection mechanism (Coakley et al. 1997). The susceptibility of lactococcal starter cultures to phage attack is a serious problem in the dairy industry and can result in production and economic losses. Thus, the introduction of the lacticin-3147 genetic determinants to a starter culture has the added benefit of conferring protection from lactococcal phage attack as a result of the linked abortive infection mechanism. (Coakley et al. 1997; O'Sullivan et al. 1998). This is in contrast to nisin-producing starters which are associated with phage susceptibility (Lipinska 1977).

3.2 Use of bacteriocins in the acceleration of cheese ripening

Bacteriocins such as lacticin 3147 and nisin, which inhibit a broad range of food spoilage and food pathogenic bacteria, have obvious applications in the food industry. Class II bacteriocins, such as lactococcin ABM, only inhibit other lactococci and therefore their associated applications are limited. However, it has been shown that lactococcin ABM can have both a bactericidal and bacteriolytic mode of action on target cells (Morgan et al. 1995). In fact, sensitive strains undergo complete lysis following exposure to an ABM producer, resulting in the release of intracellular enzymes. The lytic abilities of these bacteriocins may have an application in the acceleration of cheese ripening as explained below.

Cheddar cheese usually has a maturation time of at least six months, during which gradual autolysis of the starter cultures occurs. Lysis results in the release of intracellular enzymes such as lactate dehydrogenase (LDH) and postproline dipeptidyl aminopeptidase (PepX) which break down the casein in the cheese to small peptides and amino acids. The amino acids released are the precursor compounds responsible for flavour development in cheese (Fox and Wallace 1997). As cell lysis is a slow process and often a limiting step in cheese maturation, controlled early lysis is known to be advantageous for improved flavour development.

The potential of using lactococcin ABM to induce starter cell lysis has been investigated (Morgan et al. 1997, 2002). Addition of these bacteriocins to growing sensitive cells is associated with a gradual decrease in optical density (O.D.) in laboratory media and a concomitant increase in the release of intracellular enzymes (Morgan et al. 1995). Cheese making trials using L. lactis subsp. lactis DPC3286 (an overproducer of lactococcin ABM) as a starter adjunct with the cheese-making strain L. lactis subsp. cremoris HP demonstrated increased starter cell lysis, elevated enzyme release and an overall reduction in bitterness when compared with the control cheeses (Morgan et al. 1997). However, a problem that may be encountered when using bacteriocins to lyse starters is that the rate of acidification may be compromised if the target lactococcal strain is also the primary acidifier. To overcome this problem with lactococcin ABM, a three-strain system was developed (Morgan et al. 2002). This included a lactococcin producing adjunct (DPC3286), a bacteriocin-sensitive starter HP as a target and Streptococcus thermophilus, a bacteriocin-resistant species. Streptococcus thermophilus is included as it is not inhibited by lactococcin ABM, is insensitive to lactococcal phage and is an efficient acid producer. As this strain continued to grow throughout ripening overall acidification was not affected and elevated enzyme release was achieved by lysis of the starter (Fig. 1) (Morgan et al. 2002). This indicates a potential use for lactococcin ABM as a tool in cheese manufacture to increase enzyme release, thereby accelerating the rate of ripening while not negatively affecting the quality of the cheese.

Figure 1.

Lactate dehydrogenase (LDH) activity in cheese juice made with Lactococcus lactis HP (open bars), L. lactis HP + L. lactis DPC3286 (Bac+) + Streptococcus thermophilus DPC1842 (black bars) over 180 days (Morgan et al. 2002)

Lacticin 3147, produced by L. lactis IFPL105 can also have a bacteriolytic effect on sensitive cells. Lactococcus lactis IFPL105 and the lacticin 3147-producing transconjugant L. lactis IFPL3593 have been successfully used in the acceleration of cheese ripening (Martínez-Cuesta et al. 1998, 2001). When the lacticin 3147 transconjugant, IFPL3593 was used as a starter culture in cheese manufacture, increased lysis of starter adjuncts allowing for the release of elevated levels of PepX resulted (Martínez-Cuesta et al. 2001). Furthermore, as the starter strain had the lacticin 3147 immunity genes, it was not inhibited throughout ripening and therefore complete acidification was allowed and the resultant cheese was not adversely affected (Martínez-Cuesta et al. 2001). It has been noted, however, that lysis with lacticin 3147, nisin and lactococcin ABM requires the presence of the autolysin gene, acmA, in the sensitive strain (Martínez-Cuesta et al. 2000b).

Lacticin 481, produced by a number of lactococcal strains, also has applications in the ripening of cheese. This is a medium spectrum lantibiotic, inhibiting other LAB and Cl. tyrobutyricum and also has a bacteriolytic mode of action. It was demonstrated that LDH release caused by the addition of lacticin 481 (L. lactis subsp. lactis DPC5552) to target cells was higher than levels found with either lacticin 3147 (DPC3147) or lactococcin ABM (DPC3286) (O'Sullivan et al. 2002). It was also shown, in laboratory-scale cheese trials, that use of a lacticin 481-producing culture as a starter adjunct results in elevated levels of LDH from the starter HP (O'Sullivan et al. 2002). Another interesting and novel phenomenon with lacticin 481 was that the O.D. of exposed cells did not decrease; in fact it increased even while intracellular enzymes were released. This suggests that lacticin 481 could be used to induce early cell lysis but also the target cell would continue to grow leaving its acid-producing abilities unaffected (O'Sullivan et al. 2002). Furthermore, as lacticin 481 is effective against other LAB, it may have an application in inhibiting NSLAB during cheese manufacture. Indeed, it was found that cheese made with HP and a lacticin 481 producer as an adjunct resulted in up to a 5-log reduction in the growth of NSLAB compared with the corresponding control (O'Sullivan et al. 2003a). This indicates a potential dual role for lacticin 481 in cheese manufacture.

To conclude, given that Cheddar cheese ripening can be a long and costly process, it would be economically advantageous to accelerate the process. Results suggest that the use of bacteriocins as tools to induce early cell lysis would appear to be a possible option. Also, as the use of bacteriocin-producing cultures involves the natural alteration of starter cultures, there would not be any additional technology costs or genetic modification events associated with the process.

3.3 Inhibition of food pathogens and extension of shelf life of foods

In recent years there has been an increase in consumer concern about the use of chemical additives to ensure product safety and to extend the shelf life of foods. In response to these concerns efforts have been made to introduce minimal processing technologies and to find alternative food grade preservatives. As lacticin 3147 inhibits a large number of food pathogenic organisms it would appear to be particularly suited to use as a biopreservative in foods. Delivering lacticin 3147 into foods may be accomplished in a number of ways. As discussed, a lacticin 3147-producing strain may be used as a starter or a starter adjunct in fermented foods. This approach has the advantage that lacticin 3147 can be produced by food grade strains and is not considered an additive, and has no cost implications for the manufacturer. Lacticin 3147 may also be added in the form of powdered whey or a milk fermentate (whey or milk is fermented with a lacticin producer and subsequently spray-dried) or as a concentrated lacticin 3147 preparation. Powdered lacticin fermentate would be added as a food ingredient (most probably replacing skim milk or whey powder in an existing formulation). While these fermentates are potentially useful the leglislative position in Europe governing their use is unclear. The use of concentrated or purified lacticin preparations may also be problematic as such preparations would be regarded as food additives and, furthermore, may not be economically feasible.

One of the food pathogens that lacticin 3147 may be useful against is L. monocytogenes. This organism can be a difficult pathogen to control because of its ubiquitous distribution, tolerance to high levels of salt and its ability to grow at a relatively low pH and at refrigeration temperatures. While fermented foods are generally considered to be relatively free of pathogens, Listeria has been found to survive and possibly grow in fermented foods (Berry et al. 1990; Benkerroum et al. 2002) made with raw materials containing the organism. The susceptibilities of a large range of L. monocytogenes strains were tested to varying levels of lacticin 3147. Importantly, while the level of sensitivity varied among different isolates, a completely resistant strain was never encountered (C. Guinane unpublished data). This is contrary to certain class IIa bacteriocins such as enterocin A, mesentericin Y105, divercin V41 and pediocin AcH, where natural resistance among Listeria strains has been shown to occur (Ennahar et al. 2000).

The efficacy of a lacticin 3147-producing culture in controlling L. monocytogenes in cottage cheese has also been investigated (McAuliffe et al. 1999). The lacticin-producing transconjugant DPC4275 was used as a starter and a high inoculum of L. monocytogenes Scott A (104 cfu g−1 cheese) was added. Results showed a 99·9% reduction in the numbers of Scott A in the bacteriocin containing cheese at 4°C after 5 days compared with when a nonlacticin 3147 producing starter was used where there was no decrease in numbers (McAuliffe et al. 1999). It was also shown that there was no regrowth of Listeria cells upon enrichment in contrast to that which occurred when the bacteriocin piscilin 126 was used (Wan et al. 1997). This may be due to the fact that spontaneous lacticin 3147-resistant Listeria mutants do not arise at high levels of the bacteriocin (Dodd 1996; Klijn 2001). Lacticin 3147 did not prevent the cottage cheese from spoiling, as this is often caused by Gram-negative bacteria, yeasts and moulds (McAuliffe et al. 1999). However, it may be possible to restrict the growth of these micro-organisms by the use of lacticin 3147 in combination with other antimicrobials effective against these organisms. In addition to the above approach, DPC4275 was used to inhibit L. monocytogenes Scott A on the surface of mould-ripened cheese (Ross et al. 2000). As the surface of this cheese ranges between pH 6·5 and 8, nisin proved to be ineffective in controlling the pathogen as it is optimal at an acidic pH (Ross et al. 2000). Lacticin 3147 is effective over a broad pH range and the use of DPC4275 resulted in a 1000-fold reduction in Listeria numbers (Ross et al. 2000).

Combinations of bacteriocins have previously been used by a number of research groups in attempts to increase activity and to limit resistance development (Hanlin et al. 1993; Bouttefroy and Milliere 2000). To test this possibility we initiated investigations to determine if it was possible to co-produce both lacticin 3147 and lacticin 481 in a single food grade strain (O'Sullivan et al. 2003b). A number of transconjugants were created capable of co-producing the two lantibiotics (lacticin 3147 and lacticin 481; O'Sullivan et al. 2003b). It was demonstrated that both bacteriocins together were more efficient in killing Lactobacillus fermentum and inhibiting L. monocytogenes than either bacteriocin alone (O'Sullivan et al. 2003b). This indicates a potential application for lacticin 481 in food safety in addition to NSLAB control and LAB lysis.

There is also potential for the use of lacticin 3147-producing strains as protective cultures in fermented meats. In the production of salami, DPC4275 was used as a starter and was compared with a conventional starter (Coffey et al. 1998). It was found that the salami produced by the lacticin 3147-producing strain was comparable in terms of pH and aw to the control salami. It was also shown that the bacteriocin was produced throughout manufacture and that overall the characteristics of the salami produced by DPC4275 were acceptable in terms of preservation and food safety (Coffey et al. 1998). It has been reported, however, that sodium chloride and sodium nitrite can inhibit the bacteriocinogenic effect of bacteriocin-producing cultures in a meat system (Leroy and De Vuyst 1999). This problem was overcome with a preinoculation enrichment procedure, whereby DPC4275 was treated with 2·5-ppm manganese (Mn) and 250-ppm magnesium (Mg) before the fermentation process (Scannell et al. 2001). Mn and Mg can promote the metabolic activities of LAB, thereby possibly favouring bacteriocin production. The sausage produced by this method had higher acid and bacteriocin production after 10 days when compared with other treatments (Scannell et al. 2001). The protective ability of lacticin 3147 in the meat was also demonstrated as there was a significant reduction in Listeria innocua and Staph. aureus numbers in spiked sausage samples in the presence of DPC4275 (Scannell et al. 2001).

As noted earlier the use of lacticin 3147 in the form of a bioactive powder is also an option for the food industry. Indeed, a formulation called Nisaplin® (Danisco, Copenhagen, Denmark) is used in the preservation of foods. However, this has its limitations because of the poor solubility of nisin above pH 6 and therefore is largely restricted to acidic foods. MicroGARD® (Danisco) is a powder produced from a fermentate of Propionibacterium freundenreichii subsp. shermanii and is commonly used commercially as a biopreservative in cottage cheese. A heat-stable peptide of 700 Da has been reported to be among the inhibitory agents present in MicroGARD®, and the powder is reported to be active against Gram-negative bacteria and some fungi (Al-zoreky et al. 1991) which may be linked to its propionic acid content. A lacticin 3147 powder preparation is generated by fermenting demineralized whey powder with DPC3147 (Morgan et al. 1999). The inhibitory effect of this powder was investigated in broth systems against two common food-borne pathogens, L. monocytogenes Scott A and Staph. aureus 10. A 10% solution of the lacticin 3147 enriched powder proved to be very efficient in reducing L. monocytogenes Scott A levels at pH 5 and 7. Staphylococcus aureus 10, although more resistant to the lacticin powder, was also effectively inhibited by a 15% solution at pH 7 and to a lesser extent at pH 5 (Morgan et al. 1999). The efficiency of this powder in a number of food systems was also demonstrated (Morgan et al. 1999, 2001). In an infant formula, designed with vulnerable consumers in mind, there was over a 99% kill of L. monocytogenes Scott A within 3 h, whereas in the control culture without the lacticin 3147-producing powder, there was a 700% increase in Listeria numbers (Morgan et al. 1999). In natural yoghurt and in cottage cheese, a 10% solution of the reconstituted demineralized whey powder was effective in reducing L. monocytogenes Scott A (Fig. 2) (Morgan et al. 2001). In soup, Bacillus cereus was reduced by 80% within 3 h with a 1% (w/w) powder (Morgan et al. 2001). These results show the potential for the use of a lacticin 3147-based powder to inhibit food pathogens and as it can inhibit at pH 7, it can be used in foods with a neutral pH.

Figure 2.

The survival of Listeria monocytogenes ScottA without lacticin 3147 (open bars) and with lacticin 3147 (black bars) in natural yoghurt (a) and in cottage cheese (b) at 30°C (Morgan et al. 2001)

A number of studies have also been carried out to investigate whether the efficacy of lacticin 3147 could be improved through its use in combination with other antimicrobial treatments (Morgan et al. 2000; Scannell et al. 2000a). One example involves the use of high hydrostatic pressure at 150–600 MPa (Morgan et al. 2000). While high hydrostatic pressure can be a very efficient method for food preservation, however, this technology may be expensive and economic factors have to be taken into consideration. The combined efficiency of high pressure with bacteriocins such as nisin and pediocin has been previously investigated (Hauben et al. 1996; Kalchayanand et al. 1998; Lopez-Pedemonte et al. 2003). Lopez-Pedemonte et al. (2003) reported efficient inactivation of B. cereus spores with the combination of high pressure and nisin. The use of bacteriocins in combination with high pressure may allow for lower pressures to be used to inactivate certain pathogens and therefore reduce costs. It was found that for two organisms tested, L. innocua and Staph. aureus, the co-treatment with high pressure and lacticin 3147 had increased lethality compared with either treatment alone (Morgan et al. 2000). It is likely that high pressure impairs the structural integrity of the target cell membrane and thus increases its susceptibility to the action of the bacteriocin. The combinatory effect resulting from the use of both of these antimicrobials could therefore allow for a more economically feasible process. Similarly, the use of lacticin 3147 or nisin in combination with certain organic acids can enhance overall antimicrobial activity in fresh pork sausage (Scannell et al. 2000a). Sodium citrate (SC) and sodium lactate are common organic acid preservatives used by the food industry. It was demonstrated that the activity of lacticin 3147 or nisin was improved by the addition of organic acids against Clostridium perfringens and L. innocua. Gram-negative bacteria are not susceptible to the action of lacticin 3147, however when it was combined with SC there was a significant reduction of Salmonella Kentucky in a pork sausage system (Scannell et al. 2000a). This may be due to the phenomenon whereby chelators, such as citrate, can weaken the lipopolysaccharide layer of the Gram-negative outer membrane making the cell surface more accessible to bacteriocins (Stevens et al. 1992).

A priority when considering the commercial use of a bacteriocin is having the ability to produce it on a large scale. The immobilization of LAB for this purpose has received increasing interest in recent years. Immobilization has the advantages of increased productivity and improved long term stability over free-cell fermentations. DPC3147 was immobilized to calcium alginate and bacteriocin production was compared with the process using free cells (Scannell et al. 2000b). It was shown that continuous fermentation of lacticin 3147 was possible and that the calcium alginate beads were stable for 180 h (Scannell et al. 2000b). This indicates that it may be feasible to produce lacticin 3147 at an industrial level. Immobilization of the bacteriocin itself may also be useful in the creation of bioactive packaging (Scannell et al. 2000c). This essentially involves the incorporation of an antimicrobial agent into a food packaging material. Lacticin 3147 was immobilized to cellulose-based packaging (Scannell et al. 2000c). The activity of the bacteriocin remained stable for a period of three months. This method would allow consistent production and distribution of lacticin 3147 in a food system. It was found, however, that lacticin 3147 bound poorly to plastic in comparison with the binding of a nisin powder (Scannell et al. 2000c).

It is evident that there is a role for bacteriocins such as lacticin 3147 in the biopreservation of foods. As lacticin 3147 would easily be broken down by digestive enzymes it does not pose a threat to humans, making it a safe and economically viable alternative to chemical preservatives presently used in the food industry.

4. Resistance

As discussed, lacticin 3147 has potential applications as a biopreservative in foods. Its widespread use would however be compromised were resistant strains to emerge at a high frequency. Natural resistance to certain bacteriocins can occur and, as mentioned earlier, Listeria strains have been found that are resistant to certain class IIa bacteriocins (Ennahar et al. 2000). Acquired resistance in an otherwise sensitive population can also be a potential problem. Acquired resistance can be spontaneous or induced after increased exposure.

Resistance to nisin and class IIa bacteriocins has been well documented. It has been shown that nisin-resistant variants can be readily isolated following exposure to the lantibiotic. In Gram-positive bacteria the frequency of nisin resistance can vary from 10−2 to 10−8 (Ming and Daeschel 1993; Mazzotta et al. 1997; Gravesen et al. 2002). Nisin resistance has been shown to correlate with changes in the cell wall (Davies et al. 1996) and cell membrane (Verheul et al. 1997). A study on class IIa bacteriocins demonstrated that the frequency of resistance of L. monocytogenes ranges from 10−4 to 10−6 (Dykes and Hastings 1998) and that resistant mutants are also often cross-resistant to other class IIa bacteriocins (Dykes and Hastings 1998).

Studies to date on lacticin 3147 resistance would indicate that spontaneously resistant mutants are not isolated at high levels of the bacteriocin or at high frequencies (Dodd 1996; Coakley et al. 1997; Ryan et al. 1998; Klijn 2001). It was found that for 11 L. lactis strains resistance was negligible, with the exception of L. lactis DPC220 which exhibited a low level of resistance (Coakley et al. 1997). This absence of resistance is so marked that lacticin 3147 can be used as a sole selectable marker for conjugations involving the transfer of pMRC01 into other lactococci (Coakley et al. 1997). The frequency at which certain mastitis-causing pathogens develop resistance was also studied and similar results were found (Ryan et al. 1998). Streptococcus dysgalactiae was found to have a frequency of resistance of <10−6 and increased levels of the bacteriocin were able to control the pathogen. No resistant derivatives of Staph. aureus 10 were identified (Ryan et al. 1998). Interestingly, it was also found that acid adapted L. monocytogenes that were significantly more resistant to nisin than the parent strain were only marginally cross-resistant to lacticin 3147 (van Schaik et al. 1999).

It has been shown, however, that mutants resistant to low levels of lacticin 3147 can be created by repeated exposure to the bacteriocin (Fig. 3) (Dodd 1996; Klijn 2001; Ryan et al. 2001). It was found that stable L. lactis (Klijn 2001) and L. monocytogenes (Dodd 1996; Klijn 2001) resistant mutants could be created by a stepwise increase of lacticin 3147. The physiological implications of lacticin 3147 resistance in L. monocytogenes were reduced growth at low temperatures, increased hydrophobicity compared with the parent strain and a slight increased resistance to other bacteriocins (Dodd 1996). As outlined earlier a Lact. paracasei subsp. paracasei-resistant mutant of DPC5336 was created by repeated increased exposure to lacticin 3147 (Ryan et al. 2001). This was used as an adjunct in cheddar cheese manufacture allowing greater control of the microflora throughout ripening. This resistant mutant adsorbed less bacteriocin and was found not to flocculate as well as the parent strain (Ryan et al. 2001). This indicated cell surface alterations in the resistant mutant.

Figure 3.

Zones of inhibition created by colonies of a lacticin 3147 producer in a lawn of a sensitive indicator (left) and increased resistance of the indicator (right) following repeated exposure to lacticin 3147 (Klijn 2001)

In conclusion, resistant mutants do not arise at high levels of lacticin 3147 and mutants with relatively enhanced resistance can only be created with incremental increases of the bacteriocin. As strains need to be preconditioned to lacticin 3147 for this to occur this problem is unlikely to emerge were lacticin to be commercially applied.

5. Future prospects for lacticin 3147

Although this review has focused on the use of lacticin 3147 in the food industry, it also has a number of potential medical applications. With the growing incidence of antibiotic-resistant mutants, there is a growing demand for alternative antimicrobial treatments. As lacticin 3147 inhibits all Gram-positive mastitic pathogens tested, it has been investigated for its use in the treatment/prevention of mastitis as a substitute to intramammary antibiotics (Ryan et al. 1998,1999; Twomey et al. 2000). It has been shown that lacticin 3147 together with a Teat sealTM formulation can protect animals against Strep. dysgalactiae (Ryan et al. 1999) and Staph. aureus (Twomey et al. 2000). The combination treatment appears to be very efficient probably because of the barrier protection of the seal containing the bacteriocin within the site of infection. Additionally, the lacticin and teat seal combination did not cause irritation in the udder (Ryan et al. 1998,1999).

As mentioned earlier, lacticin 3147 inhibits a wide range of human pathogens. Within the inhibitory spectra of lacticin 3147 are important human pathogens such as methicillin-resistant Staph. aureus, vancomycin-resistant Ent. faecalis, penicillin-resistant Pneumococcus, Strep. mutans and P. acne (Galvin et al. 1999). These pathogens include some problematic antibiotic-resistant strains. In time-kill curves, there was an efficient reduction in cells in all cases within 2 h using a concentrated lacticin 3147 preparation of 20 000 AU ml−1 (Galvin et al. 1999). Propionibacterium acne was found to be the most sensitive of the pathogens tested and methicillin-resistant Staph. aureus strains were the most resistant (Galvin et al. 1999). Propionibacterium acne is one of the responsible agents for skin conditions including acne. Acne can be treated with antibiotics such as erythromycin but there has been increasing antibiotic resistance reported for P. acne. As lacticin 3147 is active at a physiological pH, it may be feasible to develop topical creams containing lacticin 3147 as an alternative therapeutic compound.

Streptococcus mutans is the causative agent of dental caries. Tooth decay is one of the most common human disorders. There may be a role for lacticin 3147 in the prevention of dental caries with the possible development of a lacticin 3147-containing toothpaste or mouthwash.

6. Discussion

This review has outlined the potential applications for lactococcal bacteriocins, primarily focusing on lacticin 3147, but also for certain other lactococcal bacteriocins. Although the bacteriocins lacticin 3147, lacticin 481 and lactococcin ABM, and their associated applications differ considerably, it is evident from this review that bacteriocins can be used as tools to influence the microbial population in a food system. Lacticin 3147 has a number of favourable characteristics. It is heat stable, active over a broad pH range, has a wide spectrum of activity, has a low level of associated resistance and because of the conjugative nature of pMRC01, lacticin 3147 production and immunity can be readily transferred from strain to strain. There are a number of delivery systems available for the use of lacticin 3147 but it would appear that cost effective ways have been developed with the use of bacteriocin-producing cultures as starters, as adjuncts in fermented foods or as protective cultures on the surface of foods.

Although quite diverse, there are a number of potential applications for lactococcal bacteriocins. It has been established that lacticin 3147 has a wide range of possible uses with roles certainly in food quality, food safety and also possibly as an alternative antimicrobial in biomedicine. It is hoped that bacteriocins such as those discussed in this review will be exploited to their maximum potential.

7. Acknowledgements

The authors would like to acknowledge the support of Science Foundation Ireland and from the Irish Government under the National Development Plan 2000–2006.