Review

Potential of antagonistic

microorganisms and

bacteriocins for the

biological preservation of

foods

U. Schillinger, R. Geisen and W.H. Holzapfel

Interest in novel biological preservation methods has in- creased during recent years, supported by research indicat- ing that antagonistic microorganisms or their antimicrobial metabolites may have some potential as natural preservatives to control the growth of pathogenic bacteria in foods, and also to control mycotoxinogenic fungi. To date, nisin is the only bacteriocin that has found practical application in some industrially processed foods. Many other bacteriocins from lactic acid bacteria have recently been characterized, but their efficacy in foods has not been studied extensively. Bet- ter understanding of the interactions between antimicrobial metabolites and food components is necessary to evaluate their potential as preservatives.

Biological preservation implies a novel scientifically- based approach to improve the microbiological safety of foods. By definition, this concept refers to the use of an- tagonistic microorganisms or their metabolic products to inhibit or destroy undesired microorganisms in foods. Owing to their typical association with food fermen- tation and also their long tradition as food-grade bac- teria, lactic acid bacteria (LAB) are ‘generally recognized as safe’ (GRAS). LAB can exert a ‘biopreservative’ or inhibitory effect against other microorganisms as a re- sult of competition for nutrients and/or of the produc- tion of bacteriocins and other antagonistic compounds

U. Schillinger, R. C&en and W.H. Holzapfel are at the Federal Research Centre for Nutrition, Institute of Hygiene and Toxicology, EngesserstraBe 20, D-761 31 Karlsruhe, Germany (fax: +49-721-6625-l 11; e-mail: ob05@ibm 3090.rz.uni-karlsruhe.de).

58 01996, Elsevier Science Ltd PII: 50924.2244(96)10016-9

such as organic acids and hydrogen peroxide. Bacterio- tins are extracellularly released peptides or protein molecules that are bactericidal (i.e. are destructive) to bacteria that are closely related to the producer micro- organism’. However, some bacteriocins do not kill sus- ceptible bacteria but instead show a bacteriostatic (i.e. inhibitory) mode of action. Bacteriocins produced by LAB may be considered as natural preservatives or biopreservatives.

Biological preservation can also involve the use of antagonistic or biocompetitive microorganisms to inhibit mycotoxinogenic fungi, which often cause problems in foods of plant origin, and to prevent the formation of mycotoxins in foods and agricultural products.

This article will discuss both of these approaches of biological preservation. However, although certain anti- microbial enzymes are also considered as potential com- pounds for use in biological preservation, such enzymes will not be discussed here as this aspect has recently been comprehensively reviewed by Fuglsang et a1.2

Bacteriocin production by LAB The majority of the bacteriocins of LAB studied thus

far have been shown to be small peptides with a mol- ecular weight below 10 000 Da. Their bactericidal activ- ity appears to be limited to other Gram-positive bac- teria; each producer bacterial strain has a mechanism of self-protection against its own bacteriocin3. One group of these low molecular weight bacteriocins contains unusual amino acids, such as lanthionine and B-methyl- lanthionine, which are produced by post-translational formation of thioether linkages between dehydrated amino acids (dehydroalanine and dehydrobutyrine) and cysteine. These bacteriocins are known as ‘lantibiotics’. The most prominent member of this group is nisin, a bacteriocin produced by strains of Lactococcus lactis and the only one to date that has found practical appli- cation in food technology; nisin is permitted as a food additive in approximately 50 countries (for details on applications of nisin, see Barnby-Smith4).

The antibacterial activity of most bacteriocins is di- rected against species that are closely related to the producer and also against different strains of the same species as the producer. However, there are also many bacteriocins that are active against a number of other less closely related Gram-positive bacteria including spoilage bacteria and foodbome pathogens such as Staphylococcus aureus, Listeria monocytogenes and Clostridium botulinum. Those antimicrobials that have a relatively broad inhibitory spectrum are expected to possess the greatest biopreservative potential. Un- fortunately, Gram-negative bacteria are generally not sensitive to bacteriocins from LAB unless the barrier function of their outer membrane is disrupted by treat- ment with chelating agents such as EDTA5. A surpris- ingly high number of bacteriocins from LAB are ac- tive against L. monocytogenes, whereas LAB bacteriocins that show inhibitory activity against other foodbome pathogens such as S. aureus or C. botulinum have been observed less frequently. Some of these antilisterial

Trends in Food Science & Technology May 1996 [Vol. 71

Table 1. Use of bacteriocin-producing lactic acid bacteria in the biopreservation of foods

Product Bacteriocin production in situ by

Meat products

l Italian-type salami

l Turkey summer sausages

l Comminuted cured pork (mettwurst)

l Vacuum-packaged frankfurters

l Vacuum-packaged wieners

Lactobacillus plantarum

Pediococcus acidilactici (three strains)

lactobacillus sake Lb 706

Pediococcus acidilactici JDl -23

Pediococcus acidilactici JBL 1095

Dairy products

l Taleggio cheese

l Cheddar cheese

Vegetable-type foods

l Green-olive fermentation

Fnterococcus faecium 7C5

Lactococcus lactis subsp. cremoris JSlO2

lactobacillus plantarum LPCOl 0

Target organism

listeria monocytogenes

listeria monocytogenes

listeria monocytogenes

listeria monocytogenes

listeria monocytogenes

listeria monocytogenes

Spores of Clostridium sporogenes

Indigenous lactobacilli

Ref.

9

10

7

11

12

17

16

22

1

l Ready-to-use mixed salads lactobacillus casei (two strains), Pediococcus (two strains)

Spoilage-associated microflora 19 ~

bacteriocins are reported to show very specific bacterici- dal activity against strains of Listeria spp., and not to have any adverse effects on LAB and other Gram- positive bacteria. For example, mesentericin Y 105 from Leuconostoc mesenteroides is reported to be active ex- clusively against strains of Listeria spp. and not against bacteria that are closely related to Leuconostoc6. Such antilisterial bacteriocins might be advantageous in foods and food fermentations in which starter cultures and other desirable bacteria involved in fermentation should not be inhibited and the multiplication of only L. mono- cytogenes is to be prevented.

tinction should be made between this generally accepted practice and the deliberate addition of a bacteriocin as a ‘processing aid’, which would need to be regulated on a legal basis as in the case of nisin.

Biological approaches using bacteriocins to inhibit bacterial growth in foods

Several possible strategies for the application of bacteriocins in the preservation of foods may be con- sidered:

Extrapolation of results obtained from experiments in vitro with laboratory media to food products is not straightforward as foods are very complex, multicom- ponent systems consisting of different interconnecting microenvironments. Various interactions between the bacteriocin molecules and food components may result in a decrease in the efficacy of the biopreservative.

l inoculation of the food with LAB (starter or protective cultures) that produce the bacteriocin in the product (production in situ);

l addition of the purified or semipurified bacteriocin as a food preservative;

l use of a product previously fermented with a bac- teriocin-producing strain as an ingredient in food processing.

During recent years, the production in situ of some bacteriocins by LAB has been studied in several foods and model food systems. The majority of these inocu- lation experiments were performed with meats and meat products (including, vacuum-packaged raw meat or processed meat products as well as fermented sausages), and with the intention of demonstrating the effects of the bacteriocins against L. monocytogenes (Table 1).

Meat and fish products

One hurdle that needs to be overcome before the com- mercial use of a new bacteriocin as a biopreservative can begin is its legal acceptance as a food additive. To date, only nisin has been accepted as a food additive by the Food and Agriculture Organization/World Health Organization Joint Expert Commitee on Food Additives. Bacteriocin-producing cultures that are used as starter cultures in fermented foods can be regarded as process- ing aids or ingredients rather than additives. This is exemplified by the general application of nisin-producing strains of L. lactis in cheesemaking. However, a clear dis-

L. monocytogenes may be a problem in fermented meat products that are subjected to a very short drying process resulting in products that have a relatively high water activity and produce insufficient acid to in- hibit its growth. Cornminuted cured pork (German-type fresh mettwurst) exemplifies such a quickly ripened product in which the multiplication of L. monocytogenes may occur under certain conditions of elevated tempera- tures and slow acid production. Inoculation experiments with a strain of Lactobacillus sake that produces an anti- listerial bacteriocin demonstrated a reduction in the levels of L. monocytogenes by about 1 log cycle in comminuted

Trends in Food Science & Technology May 1996 [Vol. 71 159

Table 2. Use of bacteriocins as biopreservatives in foods

Product

Vacuum-packaged beef

Brined shrinips

Mozzarella cheese

Bacteriocin

Pediocin AcH, nisin

Carnocin Ul49, nisin Z, bavaricin A

Lactococcus lactis bacteriocin

Target organism Refs

Leuconostoc mesenteroides 13

Spoilage bacteria 14

Listeria monocytogenes 18

Kimchi fermentation Sakacin A, pediocin M I

cured pork of pH 5.7 (Ref. 7). A bacteriocin-negative mutant of L. sake did not affect the number of Listeria inoculated into this product.

When other types of fermented sausages typically characterized by a rapid decrease in pH during fermen- tation were inoculated with strains of L. monocytogenes and bacteriocin-producing LAB, a general decrease in the viable numbers of Listeria was observed in the presence of a bacteriocin-producing strain8-lo. However, in most cases the numbers of L. monocytogenes also decreased when non-bacteriocin-producing strains were used, indicating that inhibition was mainly due to the reduction of the pH. For instance, L. monocytogenes levels were decreased to a similar extent when either bacteriocin-producing or bacteriocin-negative strains of Lactobacillus plantarum were added to Italian-type salami9. On the other hand, clear differences in the efficacy of a bacteriocin-producing strain and a non- bacteriocin-producing strain of Pediococcus acidilactici in reducing L. monocytogenes counts were observed during the fermentation of turkey summer sausageslO. In the presence of the pediocin-producer, Listeria counts were reduced by 3.4 log cycles, but by only 0.9 log cycle with the pediocin-negative starter culture.

Bacteriocin-producing cultures were also used to in- hibit the multiplication of L. monocytogenes in vacuum- packaged wiener or frankfurter sausages stored at 4”C, both of which are model systems for products in which post-processing contamination may occur between peel- ing or slicing and packaging of the product. Addition of P. acidilactici JDl-23 at an inoculation level of 107cfu (colony-forming units) per gram to frankfurter sausages inhibited L. monocytogenes for more than 60days, whereas pathogen counts increased from 104cfu/g to 106cfu/g in control sausages without added pedio- cocci”. A bacteriocin-negative variant of P. acidilactici was also able to suppress the growth of L. monocyto- genes. When the pediococci were added at a lower level (103-lo4 cfu/g), multiplication of Listeria was delayed but not completely inhibited by the bacteriocin producer”. Degnan et al. l2 observed a clear antilisterial effect of another bacteriocin-producing strain of P. acidilactici in temperature-abused vacuum-packaged wieners. The addition of the protective culture resulted in a reduction of L. monocytogenes counts by 2.7 log cycles within 8 days at 25°C. In contrast, Listeria counts increased by 3.2 log cycles in sausages without added pediococci.

Listeria monocytogenes 23

The biopreservative effects of two different bacterio- tins added directly to the product have been studied in vacuum-packaged beef (Table 2). The shelf life of re- frigerated vacuum-packaged meat is dependent on the growth of psychrotrophic bacteria including lactobacilli, leuconostocs, carnobacteria and Brochothrix thermo- sphacta. Rozbeh et al. reported an immediate bac- tericidal effect of nisin and also of pediocin AcH, a bacteriocin produced by P. acidilactici AcH, on L. mesenteroides that had been inoculated into vacuum- packaged beef? a reduction of the bacterial population by 0.6-2.0 log cycles was observed after only 1 day. During storage at 3°C for 8 weeks, LAB counts in- creased in all samples, but remained about 2 log cycles lower in those samples treated with the bacteriocins than in the controls.

Another interesting application of bacteriocins m ight be for the biopreservation of brined shrimps. Brined shrimps are usually produced with the addition of sorbic or benzoic acid to extend their shelf life. Einarsson and Lauzon14 added three different bacteriocins (nisin Z, camocin U149 from Carnobacterium piscicola U149 and bavaricin A from Lactobacillus bavaricus M I 401) to brined shrimps; a delay of bacterial growth was ob- served in the presence of nisin Z. The shelf life of this product was extended by the biopreservative from 10days (control without preservatives) to 31 days. The other two bacteriocins were less effective. However, the strongest antimicrobial effect was exerted by sodium benzoate and potassium sorbate. Addition of these chemi- cal preservatives (at a concentration of 0.05-o. 1%, w/w) to brined shrimps inhibited m icrobial growth completely for 59 days.

Dairy products In cheese production, nitrate is frequently used to in-

hibit the growth of Clostridium tyrobutyricum and other clostridia that cause spoilage by the production of gas and butyric acid (‘late blowing’). Nisin, which is able to prevent the outgrowth of clostridia spores, can be used as an alternative to nitrate, and it is commonly added to pasteurized processed cheese spreadP. Zottola et a1.16 proposed a novel approach for the application of nisin in processed cheese. Instead of adding purified nisin directly, they utilized Cheddar cheese that had been manufactured with n&in-producing lactococci as an in- gredient in the production of processed cheese spreads, which resulted in the extension of the shelf life of these

60 Trends in Food Science & Technology May 1996 [Vol. 71

Factors ~~~p~~ LAB

l Inadequate environment (e.g. pH, temperature, nutrients) for bacteriocin production

l Spontaneous loss of bacteriocin-producing ability

l Phage infection

l Antagonism by other microorganisms present in foods

Factors affecting the efficacy of bacteriocins

l Emergence of bacteriocin-resistant pathogens or spoilage bacteria

l Conditions that destabilize the biological activity of proteins such as proteases or oxidation processes

. Binding to food components such as fat particles or protein surfaces

l Inactivation by other additives

l Poor solubility and uneven distribution in the food matrix

l pH effects on bacteriocin stability and activity

“After DaescheP

products. Spoilage by Clostridium sporogenes was re- duced significantly in the nisin-containing cheese spreads.

Contamination by L. monocytogenes can also cause problems in the production of cheese, especially in those products in which the pH increases during ripen- ing, such as the Italian cheeses Taleggio, Gorgonzola and Mozzarella. Giraffa et al.” demonstrated the pro- duction and stability of an antilisterial bacteriocin from a strain of Enterococcus faecium during the manufac- ture of Taleggio cheese. When this strain was inoculated as an additional starter culture, its bacteriocin did not af- fect the activity of the fhermophilic starter bacteria used in the cheesemaking process. Post-process contami- nation of Mozzarella cheeses can be controlled by bacteriocin-producing strains of L. lactis18. The addition of heat-treated cultures of such strains to Mozzarella cheese inoculated with L. monocytogenes and packaged in small bags resulted in a decrease in the initial counts of Listeria. During a storage period of 2-3 weeks at 5”C, Listeriu counts remained significantly below those of the samples prepared without the addition of bio- preservatives.

Vegetable products Bacteriocin-producing LAB also show potential for

the biopreservation of foods of plant origin, especially minimally processed foods such as prepackaged mixed salads and fermented vegetables. Vescovo et a1.19 ob- served a reduction of the high initial bacterial loads of ready-to-use mixed salads when bacteriocin-producing LAB were added to the salad mixtures. Furthermore, bacteriocin-producing starter cultures may be useful for the fermentation of sauerkraut20~21 or olives to prevent the growth of spoilage organisms. In the fermentation of Spanish-style green olives, a bacteriocin-producing strain of L. plunturum dominated the indigenous lac- tobacilli without adversely affecting the organoleptic properties of the product z In contrast, a non-bacteriocin- .

producing variant of this strain was outnumbered by the ‘natural’ Lactobacillus population.

The use of biopreservatives that are active against L. monocytogenes has also been proposed for the produc- tion of kimchi, a traditional Korean spiced, fermented cabbage product 23 The two bacteriocins used in the in- . oculation experiments behaved differently. The addition of crude sakacin A from L. sake Lb 706 had no effect on the growth of L. monocytogenes in kimchi fermented at 14”C, whereas the bacteriocin produced by P. acidilac- tici M caused a rapid initial reduction in numbers of Listeria, and was able to control L. monocytogenes throughout the 16-day fermentation.

Factors lim iting the efficacy of bacteriocins in foods Results obtained in inoculation experiments indicate

that many bacteriocins are less effective inhibitors in the food matrix than in vitro. Direct deductions from results obtained in vitro to predict efficacy in practical situ- ations are therefore not justified. Factors that contribute to a reduction of bacteriocin activity in foods are sum- marized in Box 1. Some inactivation effects are the direct consequences of components in foods such as proteolytic enzymes or bacteriocin-binding protein or fat particles. For instance, nisin activity against L. monocytogenes decreases with increasing fat concen- tration24. Nisin probably binds to milk-fat globules, thereby reducing its availability to inhibit Listeria cells. Other limitations of bacteriocin efficacy result from the properties of the target organisms. Not all strains of L. monocytogenes show the same degree of sensitivity to antilisterial bacteriocins, and mutants that are resist- ant to several bacteriocins may occur spontaneously in sensitive populations of Listeria. The instability of the property to produce bacteriocins and the inability to re- lease sufficient quantities of the bacteriocin to suppress undesired bacteria in the food environment may prevent successful application of bacteriocin-producing cultures

Trends in Food Science & Technology May 1996 [Vol. 71 16

in food preservation. In dairy fermentations, problems may arise as a result of phage infections leading to a loss of viability of the bacteriocin producer, in a similar way as is known to occur with cheese starter cultures. Novel approaches are needed to increase the effective- ness of protective cultures and biopreservative agents in foods. Those strategies include the use of recombinant DNA technology for the amplification of bacteriocin genes or their transfer to other strains, such as to starter cultures with desirable properties that are used in food technology. Another approach might be the utilization of bacteriocins in surface-packaging materials, such as sausage casings; or microencapsulation technologies, which may afford protection and more-even distribution of the bacteriocin molecules within foods. For example, it has been shown that the encapsulation of pediocin AcH within liposomes could be used to enhance its activity in meat slurries25.

Future prospects for the use of bacteriocins as biopreservatives

The particular approach of biological preservation that needs to be taken will be determined by the type of product and the intrinsic and extrinsic parameters exist- ing during processing, storage and distribution. In our opinion, bacteriocin production in situ by starter cul- tures has a good chance of finding applications in fer- mented foods. In particular, an additional safety factor seems desirable for mild (low-acid) fermented foods. In non-fermented refrigerated products, such as minimally processed meats or prepacked vegetable salads, only those strains producing sufficient and potent amounts of bacteriocin but no other metabolic compounds at amounts that are detrimental to the sensory quality of the product can be applied. The direct addition of purified bacterio- tins obviously provides a more controllable preservative tool in such products. However, the cost factor at present makes this approach unattractive.

In general, biological preservation approaches seem attractive as a safety parameter in foods with reduced contents of ingredients such as salt, sugar, fat and acid that usually serve as factors that are potentially in- hibitory to microbial growth. It is expected that biologi- cal preservation methods may enjoy better consumer ac- ceptance than preservation methods that use traditional chemical preservatives. Examples are nitrite, which is used for curing meat, and sorbate and benzoate, which are used in many foods including delicatessen salads. The use of bacteriocins that are active against endo- sporeforming bacteria such as C. botulinurn may allow the heat treatment of canned foods to be milder. When nisin was used in low-acid canned vegetable foods, a shorter total heating time was required to kill thermo- philic spores of spoilage organisms such as Bacillus stearothermophilus and Clostridium thermosaccharo- lyticumz6. The reduction of the intensity of the heat processing results in products that have an improved nutritional value, appearance and texture.

Future approaches should consider the application of bacteriocins in combination with treatments that

enhance their effectiveness in foods. Examples of the synergistic effects that can be obtained using mild traditional preservation techniques in conjunction with novel food processing technologies are known, but require further investigation. Ultrahigh hydrostatic pressure (UHP) and pulsed electric field (PEF) treat- ment are examples of new non-thermal methods of food preservation that cause loss of viability and sub-lethal injury to bacterial cells. Stressed and sub-lethally in- jured bacteria are more susceptible to the bactericidal action of bacteriocins because of their impaired cell- wall barriers. Even Gram-negative bacteria that are usu- ally insensitive to bacteriocins from LAB may become susceptible following treatments that result in sub-lethal injury*‘. Kalchayanand et a1.28 observed an increase in the cell death of L. monocytogenes, Escherichia coli 0157:H7 and Salmonella typhimurium when either UHP or PEF treatment was used in combination with either nisin or pediocin AcH.

The antimicrobial efficiency of a bacteriocin may also be enhanced or broadened by using it in combination with other bacteriocins or other compounds including surfactants, chelating agents or other metal-complexing compounds such as siderophores. Even in a sensitive bacterial population some cells may be insensitive to one bacteriocin but may be inhibited by other bacterio- tins. Thus, cells escaping the bactericidal action of the first bacteriocin may be killed by a second one. Indeed, a combination of sakacin A and nisin inhibited the growth of L. monocytogenes much more strongly than either of the bacteriocins used singly (U. Schillinger, unpublished). Similarly, Hanlin et a1.29 observed the increased antibacterial activity of a combination of nisin and pediocin AcH against several Gram-positive bacteria.

Crossing the ‘Gram barrier’ will probably present the greatest challenge with respect to bacteriocins from LAB. On account of their cell-wall structure, Gram- negative bacteria such as Salmonella and Shigella spp. are resistant to bacteriocins from LAB. Future ap- proaches may be directed towards an extension of the mode of action (e.g. by the modification of bacteriocin structure or the sensitization of the Gram%egative cell wall using components of food-grade additives). It may be possible to design bacteriocin molecules with im- proved stability and solubility, an enhanced host range and higher biological activity by using existing mol- ecular biological methods (recombinant DNA technol- ogy). Furthermore, site-directed mutagenesis of bacteriocin structural genes would allow the construction of various mutants with altered functional properties. Protein engineering techniques have already been applied in the case of nisin30. To date, the use of nisin is limited to acidic foods because of the low solubility and stability of the nisin molecule at neutral and high pH values. The introduction of lysine residues considerably enhanced the solubility of nisin in the neutral pH range30. Moreover, the construction of engineered bacteriocin molecules will contribute to an understanding of the structure-function relationships essential for estimating

62 Trends in Food Science & Technology May 1996 [Vol. 71

the potential of such compounds for the preservation of foods.

Biological approaches for the inhibition of mycotoxin formation by fungi

Mycotoxinogenic fungi can be identified in almost all food commodities, particularly in foods of plant origin such as cereals, maize, nuts, soft fruit and vegetables as well as in foods processed from them. Mycotoxins are a heterogeneous class of secondary metabolites that are toxic both to animals and humans. More than 300 differ- ent mycotoxins are known, of which some important ex- amples include: the aflatoxins produced by Aspergillus flaws, Aspergillus parasiticus and Aspergillus nomius; the ochratoxins produced mainly by Penicillium verru- cosum and Aspergillus ochraceus; the trichothecenes produced by Fusarium spp.; the fumonisins produced mainly by Fusarium moniliforme and Fusarium prolif- eratum; and patulin produced, from amongst others, by Penicillium spp. and Aspergillus spp. The plant or fruit may be infected either in the field or during storage, depending on the type of plant and on the particular fungal species involved. A primary fungal at- tack can lead to a manifested infection with resulting mycotoxin production; however, this depends on the en- vironmental conditions, which are often uncontrollable. For this reason, measures that control the growth of fungi are advantageous for the production of toxicologi- cally safe foods. Like the new approaches developed for the suppression of pathogenic and spoilage bacteria in foods, biological strategies have been developed to control the growth and mycotoxin production of fungi within or on plants or plant food products. The use of biocompetitive microorganisms is one approach that can be employed to inhibit mycotoxin formation in certain food commodities, and can be achieved in two different ways.

The first type of biocompetitive control involves con- trolling aflatoxin formation by the use of biocompetitive non-aflatoxinogenic strains. Nearly all of the strains of A. parasiticus that have been isolated are aflatoxino- genie and produce the aflatoxins B,, B,, G, and G, (Ref. 31), whereas 56% of the A. fravus strains isolated produce aflatoxin Bz only. These non-aflatoxinogenic strains may have lost some aflatoxin biosynthetic genes3*. Nevertheless, such strains can be found in the same habitat as aflatoxinogenic strains because they have similar growth characteristics. Non-aflatoxino- genie strains with the potential to be highly competitive have been isolated”. These strains have been applied as biocontrol agents for cottonseed”“, peanuts’4 and maizeXs. In the case of maize, the atoxinogenic biocon- trol strain was shown to reduce aflatoxin formation by SO-95% when it was inoculated either at the same time as or before the aflatoxinogenic strain. Aflatoxin reduc- tion using such biocontrol methods was achieved under both pre- and postharvest conditions, and in the case of peanuts, aflatoxin contamination was shown to decrease from 531 ppb to 11 ppb in the first year and from 241 ppb to 40ppb during the next year. Biocompetition

with non-aflatoxinogenic strains does not require exten- sive growth of these strains on the plant; rather, they oc- cupy the same infection sites on the plant, and protect against infection by the aflatoxinogenic strains.

There have also been attempts to biocontrol afla- toxinogenic strains with certain LAB. However, the reported results are contradictory. Gourama and Bullerman36 recently reviewed the antimycotic and anti- aflatoxinogenic effects of LAB. They described the dif- ferent influences of various LAB on aflatoxin biosynthe- sis, which resulted in either decreased or increased production. Karunaratne et a1.37 demonstrated the sup- pression of aflatoxin production by A. flaws in the presence of Lactobacillus acidophilus, Lactobacillus bulgaricus and L. plantarum. However, Luchese and Harrigan3* reported increased aflatoxin levels in cultures of A. parasiticus grown in the presence of L. lactis, which might be due to changes in pH as a result of the production of lactic acid. According to these authors, aflatoxin production by A. parasiticus is higher at an acidic pH than at a neutral pH. This might be the reason for the observed increase in aflatoxin production in the presence of L. lactis. Gourama and Bullerman36 stated that low pH favours aflatoxin production, whereas fun- gal growth is decreased.

The second type of biocompetitive control involves the protection of fruit against mycotoxin formation, predominantly against patulin formation by Penicillium expansum, and against rotting caused by various fungi. The antagonistic microorganisms used for this pur- pose are mainly yeasts such as Cryptococcus laurentiiX9 and Debaryomyces hansenii40, or bacteria such as Pseudomonas jluorescens . 41 Different mechanisms are argued to be responsible for the antagonistic activity of the various biocontrol microorganisms. Penicillium digitatum, the most common cause of citrus fruit rot, is inhibited only by living cells of D. hansenii; the culture filtrate is inactive. Furthermore, the yeast failed to in- hibit P. digitatum in a rich medium and succeeded only in a minimal medium or on grapefruit tissue. Thus, these results suggest that D. hansenii competes for nutrients with P. digitatum under restrictive condi- tions40. The yeast Pichia guilliermondi, which inhib- its Penicillium roqueforti, a species often identified in mouldy bread, obviously produces antagonistic volatiles42. In the case of Pseudomonas spp., small lipopeptides (known as pseudomucins) with antibiotic activity against fungi have been shown to be produced4’. These peptides contain unusual amino acids such as hydroxyaspartic acid and diaminobutyric acid. Other antagonistic bacteria including Enterobacter agglom- erans produce chitinolytic enzymes that degrade fungal cell walls. Four chitinolytic enzymes detected in a particular strain of E. agglomerans were identified as two N-acetyl-P-D-glucosamidases, an endochitinase and a chitobiosidaseti. Flavobacterium aurantiacum and several other bacteria have been shown to degrade aflatoxins enzymatically45.

Approaches for the biological control of fungi are not as well developed as those for pathogenic bacteria.

Trends in Food Science & Technology May 1996 [Vol. 71 163

Future research should concentrate on the identification of safe (GRAS) antagonistic microorganisms, and should be targeted towards a better understanding of the inhibitory mechanisms.

References 1 Tagg, J.R., Dajani, A.S. and Wannamaker, L.W. (1976) ‘Bacteriocins of

Gram-positive Bacteria’ in Bacterial. Rev. 40, 722-756 2 Fuglsang, C.C., Johansen, C., Christgau, S. and Adler-Nissen, J. (1995)

‘Antimicrobial Enzymes: Applications and Future Potential in the Food Industry’ in Trends Food Sci. Techno/. 6, 390-396

3 Jack, R.W., Tagg, J.R. and Ray, B. (1995) ‘Bacteriocins of Gram-positive Bacteria’ in Microbial. Rev. 59, 171-200

4 Barnby-Smith, F.M. (1992) ‘Bacteriocins: Applications in Food Preservation’ in Trends Food Sci. Technol. 3, 133-l 37

5 Stevens, K.A., Sheldon, B.W., Klapes, N.A. and Klaenhammer, T.R. (1991) ‘Nisin Treatment for Inactivation of Salmonella and Other Gram-negative Bacteria’ in A/$. Environ. Microbial. 57, 3613-3615

6 Hechard, Y., Derijard, B., Letellier, F. and Cenatiempo, Y. (1992) ‘Characterization and Purification of Mesentericin Y105, an Anti-Listeria Bacteriocin from leuconostoc mesenteroides’ in 1. Cen. Microbial. 138, 2725-2731

7 Schillinger, U., Kaya, M. and Liicke, F-K. (1991) ‘Behaviour of Listeria monocytogenes in Meat and Its Control by a Bacteriocin-producing Strain of Lactobacilb sake’ in I. Appl. Bacterial. 70,473-478

8 Berry, E.D., Liewen, M.B., Mandigo, R.W. and Hutkins, R.W. (1990) ‘Inhibition of Wisteria monocytogenes by Bacteriocin-producing Pediococcus During the Manufacture of Fermented Semidry Sausage’ in 1. Food Protect. 53, 194-l 97

9 Campanini, M., Pedrazzoni, I., Barbuti, S. and Baldini, P. (1993) ‘Behaviour of Listeria monocyfogenes During the Maturation of Naturally and Artificially Contaminated Salami: Effect of Lactic-acid Bacteria Starter Cultures’ in Int, 1. Food Microbial. 20, 169-l 75

10 Luchansky, 1.8. et a/. (1992) ‘Genomic Analysis of Pediococcus Starter Cultures Used to Control Listeria monocytogenes in Turkey Summer Sausage’ in Appl. Environ. Microbial. 58, 3053-3059

11 Berry, E.D., Hutkins, R.W. and Mandigo, R.W. (1991) ‘The Use of a Bacteriocin-producing Pediococcus acidilactici to Control Postprocessing Listeria monocytogenes Contamination of Frankfurters’ in 1. FoodProtect. 54, 681-686

12 Degnan, A.]., Yousef, A.E. and Luchansky, J.B. (1992) ‘Use of Pediococcus acidilactici to Control Listeria monocytogenes in Temperature-abused Vacuum-packaged Wieners’ in 1. Fooo’ Protect. 55,9&l 03

13 Rozbeh, M., Kalchayanand, N., Field, R.A., Johnson, M.C. and Ray, 8. (1993) ‘The Influence of Biopreservatives on the Bacterial Level of Refrigerated Vacuum Packaged Beef in 1. Food Saf. 13,99-l 11

14 Einarsson, H. and Lauzon, H.L. (1995) ‘Biopreservation of Brined Shrimp (Panda/us borealis) by Bacteriocins from Lactic Acid Bacteria’ in App/. Environ. Microbial. 61, 669-676

15 Delves-Broughton, J. (1990) ‘Nisin and Its Uses as a Food Preservative’ in Food Technol. 44,100-l 17

16 Zottola, E.A., Yezzi, T.L., Ajao, D.B. and Roberts, R.F. (1994) ‘Utilization of Cheddar Cheese Containing Nisin as an Antimicrobial Agent in Other Foods’ in Int /. FoodMicrobiol. 24, 227-238

17 Giraffa, C., Picchioni, N., Neviani, E. and Carminati, D. (1995) ‘Production and Stability of an Enterococcus faecium Bacteriocin During Taleggio Cheesemaking and Ripening’ in Food Microbial. 12, 301-307

18 Stecchini, M.L., Aquili, V. and Sarais, I. (1995) ‘Behavior of Listeria monocytogenes in Mozzarella Cheese in Presence of Lactococcus /a& in Int. 1. Food Microbial. 25,301-310

19 Vescovo, M., Orsi, C., Scolari, C. and Torriani, S. (1995) ‘Inhibitory Effect of Selected Lactic Acid Bacteria on Microflora Associated with Ready-to-use Vegetables’ in Left. App/. Microbial. 21, 121-125

20 Harris, L.J., Fleming, H.P. and Klaenhammer, T.R. (1992) ‘Novel Paired Starter Culture System for Sauerkraut, Consisting of a Nisin-resistant Leuconostoc mesenteroides Strain and a Nisin-producing Lactococcus lactis Strain’ in App/. Environ. Microbial. 58, 1484-l 489

21 Breidt, F., Crowley, K.A. and Fleming. H.P. (1995) ‘Controlling Cabbage Fermentations with Nisin and Nisin-resistant Leuconostoc mesenteroides’ in Food Microbial. 12, 109-l 16

22 Ruiz-Barba, J.L., Cathcart, D.P., Warner, P.J. and Jimenez-Diaz, R. (1994) ‘Use of Lactobacillus plantarum LPCOl 0, a Bacteriocin Producer, as a Starter Culture in Spanish-style Green Olive Fermentations’ in Appl. Environ. Microbial. 60,2059-2064

23 Choi, S-Y. and Beuchat, L.R. (1994) ‘Growth Inhibition of Listeria monocytogenes by a Bacteriocin of Pediococcus acidilactici M During Fermentation of Kimchi’ in FoodMicrobiol. 11,301-307

24 Jung, D-S., Bodyfelt, F. and Daeschel, M.A. (1992) ‘Influence of Fat and Emulsifiers on the Efficacy of Nisin in Inhibiting Listeria monocytogenes in Fluid Milk’ in 1. Dairy Sci. 75, 387-393

25 Degnan, A.J. and Luchansky, J.B. (1992) ‘Influence of Beef Tallow and Muscle on the Antilisterial Activity of Pediocin AcH and Liposome-encapsulated Pediocin AcH’ in 1. Fooo’ Protect 55, 552-554

26 Vas, K., Kiss, I. and Kiss, N. (1967) ‘Use of Nisin for Shortening the Heat Treatment in the Sterilization of Green Peas’ in Z. Lebensm.-Unters. Forsch. 133,141-144

27 Kalchayanand, N., Hanlin, M.B. and Ray, B. (1992) ‘Sublethal Injury Makes Gram-negative and Resistant Gram-positive Bacteria Sensitive to the Bacteriocins, Pediocin AcH and Nisin’ in Left. Appl. Microbial. 15, 239-243

28 Kalchayanand, N., Sikes, T., Dunne, C.P. and Ray, B. (1994) ‘Hydrostatic Pressure and Electroporation Have Increased Bactericidal Efficiency in Combination with Bacteriocins’ in Appl. Environ. Microbial. 60, 4174-4177

29 Hanlin, M.B., Kalchayanand, N., Ray, P. and Ray, B. (1993) ‘Bacteriocins of Lactic Acid Bacteria in Combination Have Greater Antibacterial Activity’ in 1. Food Protect 56,252-255

30 Rollema, H.S., Kuipers, O.P., Both, P., deVos, W.M. and Siezen, R.J. (1995) ‘Improvement of Solubility and Stability of the Antimicrobial Peptide Nisin by Protein Engineering’ in Appl. Environ. Microbial. 61, 2873-2878

31 Bennett, J.W. and Papa, K.E. (1988) ‘The Aflatoxinogenic Aspergillus spp.’ in Adv. Plant Patho/. 6,263-280

32 Klich, M.A., Yu, J., Chang, P-K., Mullaney, E.J., Bhatnagar, D. and Cleveland, T.E. (1995) ‘Hybridization of Genes Involved in Aflatoxin Biosynthesis to DNA of Aflatoxinogenic and Non-aflatoxinogenic Aspergilli’ in App/. Microbial. Biotechnol. 44,439-443

33 Cotty, P.J. (1990) ‘Effect of Atoxigenic Strains of Aspergillus f/avus on Aflatoxin Contamination of Developing Cottonseed’ in Plant Dis. 74, 273-275

34 Dorner, J.W., Cole, R.J. and Blankenship, P.D. (1992) ‘Use of Biocompetitive Agent in Drought Stressed Peanuts’ in /. Food Protect. 55, 888-892

35 Brown, R.L., Cotty, P-J. and Cleveland, T.E. (1991) ‘Reduction in Aflatoxin Content of Maize by Toxigenic Strains of Aspergi//us flavd in 1. Food Protect. 54,623-626

36 Gourama, H. and Bullerman, L.B. (1995) ‘Antimycotic and Antiaflatoxinogenic Effect of Lactic Acid Bacteria: A Review’ in 1. Food Protect. 57,1275-1280

37 Karunaratne, A., Wezenberg, E. and Bullerman, L.B. (1990) ‘Inhibition of Mold Growth and Aflatoxin Production by Lactobacillus spp.’ in 1. Food Protect. 53,230-236

38 Luchese, R.H. and Harrigan, W.F. (1990) ‘Growth of and Aflatoxin Production by Aspergillus parasiticus When in the Presence of Either Lactococcus /a&is or Lactic Acid and at Different initial pH Values’ in 1. Appl. Bacterial. 69, 512-519

39 Roberts, R.G. (1990) ‘Postharvest Biological Control of Gray Mold of Apple by Crytococcus laurentii’ in Phytopathology 80, 526-530

40 Droby, S., Chalutz, E., Wilson, C.L. and Wisnieswski, M. (1989) ‘Characterization of the Biological Activity of Debarymyces hansenii in the Control of Penicillium digitatum on Grapefruit’ in Can. /. Microbiof. 35, 794-800

41 Leifert, C., Sigee, DC, Stanley, R., Knight, C. and Epton, H.A.S. (1993) ‘Biocontrol of Botryfis cinerea and Alternaria brassicola on Dutch White Cabbage by Bacterial Antagonists at Cold-store Temperatures’ in Plant Pathol. 42,270-279

42 Bjiirnberg, A. and Schniirer, 1. (1993) ‘Inhibition of Growth of Grain-storage Molds in vitro by the Yeast Pichia anomala (Hansen) Kurtzman’ in Can. 1. Microbial. 39, 623-628

43 Potera, C. (1994) ‘From Bacteria: A New Weapon Against Fungal Infection’ in Science 265,605

44 Chernin, L., ismailov, Z., Haran, S. and Chet, I. (1995) ‘Chitinolytic Enterobacter aggfomerans Antagonistic to Fungal Plant Pathogens’ in Appl. Environ. Microbial. 61, 1720-l 726

45 Line, J.E., Brackett, R.E. and Wikinson, R.E. (1994) ‘Evidence for Degradation of Aflatoxin B, by Flavobacterium aurantiacum’ in 1. Food Protect. 57, 78&791

46 Daeschel, M.A. (1993) ‘Applications and Interactions of Bacteriocins from Lactic Acid Bacteria in Food and Beverages’ in Bacteriocins of Lactic Acid Bacteria (Hoover, G. and Steenson, L.R., eds), pp. 63-91, Academic Press

Trends in Food Science & Technology May 1996 [Vol. 71