Acidification is not involved in the early inhibition of Staphylococcus aureus growth by Lactococcus lactis in milk

ARTICLE IN PRESS

International Dairy Journal 18 (2008) 197–203 www.elsevier.com/locate/idairyj

Acidification is not involved in the early inhibition of Staphylococcus aureus growth by Lactococcus lactis in milk Cathy Charlier, Sergine Even, Michel Gautier, Yves Le Loir Laboratoire de microbiologie, UMR1253 STLO INRA Agrocampus Rennes, 65 rue de Saint Brieuc, CS 84215, 35042 Rennes cedex, France Received 4 February 2007; accepted 7 March 2007

Abstract Seventy-five Lactococcus lactis strains were screened for their inhibitory effect on Staphylococcus aureus growth in milk. Most lactococcal strains had a strong antagonistic effect. Characterization of this effect showed that acidification was not involved in the inhibition observed within the first 24 h of mixed culture. Alternate effects such as bacteriocin- or hydrogen peroxide-production were eliminated. These results question some generally accepted ideas and show that even low acidifying L. lactis strains, widely used in raw milk soft cheeses, can efficiently inhibit S. aureus growth even with initial contamination levels as high as 103 cfu mL1. r 2007 Elsevier Ltd. All rights reserved. Keywords: Staphylococcus aureus; Lactococcus lactis; Mixed cultures; Growth inhibition; Milk

1. Introduction Starter lactic acid bacteria (LAB) diversity is used in combination with different process technologies to offer a wide variety of fermented dairy products. Growth and activity of LAB also have an inhibitory effect on spoiling and pathogenic bacteria (Nes & Johnsborg, 2004; Rossland, Langsrud, Granum, & Sorhaug, 2005) including enterotoxin-producing strains of Staphylococcus aureus, a Gram-positive pathogen frequently involved in food poisoning outbreaks (Le Loir, Baron, & Gautier, 2003). In many countries, some cheeses are still made with raw milk. These traditional cheese-making processes are important to maintain agricultural activity in areas unfavorable to intensive agriculture, frequently associated with the development of organic agriculture and considered as a real gastronomic patrimony. However, they have to face the social demand on food safety leading to severe sanitary criteria and tougher rules. S. aureus is a frequent causative agent of mastitis. It is thus a frequent raw milk contaminant and may grow during the cheese-making process (De Buyser, Dufour, Maire, & Lafarge, 2001; Corresponding author. Tel.: +33 2 23 48 59 04; fax: +33 2 23 48 59 02.

E-mail address: [email protected] (Y. Le Loir). 0958-6946/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.idairyj.2007.03.015

Delbes, Alomar, Chougui, Martin, & Montel, 2006; Jorgensen, Mork, & Rorvik, 2005). Enterotoxin production in foodstuffs occurs (or is detectable) when enterotoxigenic S. aureus population reaches 106 cfu mL1 (Le Loir et al., 2003). Thus, it is quite interesting to select starter LAB able to efficiently inhibit S. aureus growth. LAB–S. aureus interactions have been explored for years (Ammor, Tauveron, Dufour, & Chevallier, 2006; Haines & Harmon, 1973b; Kao & Frazier, 1966; Otero & NaderMacias, 2005; Schellenberg, Smoragiewicz, & KarskaWysocki, 2006). Several parameters were proposed as involved in S. aureus inhibition by LAB, including bacteriocin- (Ammor et al., 2006) and hydrogen peroxide-production (Otero & Nader-Macias, 2005), competition for nutrients (Haines & Harmon, 1973a) and, of course, acidification (Barber & Deibel, 1972; Delbes et al., 2006; Lindqvist, Sylven, & Vagsholm, 2002; Notermans & Heuvelman, 1983). Concerning acidification, several studies have established a direct relationship between pH and level of growth by and survival of S. aureus in model media (Iandolo, Ordal, & Witter, 1964; Minor & Marth, 1970; Notermans & Heuvelman, 1983) or in various fermented products (Delbes et al., 2006; Metaxopoulos, Genigeorgis, Fanelli, Franti, & Cosma, 1981; Olarte, Sanz, GonzalezFandos, & Torre, 2000). A great variability was found in

ARTICLE IN PRESS 198

C. Charlier et al. / International Dairy Journal 18 (2008) 197–203

the inhibitory activity among complex starter LAB mixtures and among LAB species (Haines & Harmon, 1973a; Kao & Frazier, 1966) but only two reports estimated the inhibitory efficacy at the intra-species level, one of which included Lactococcus lactis, the model LAB and one of the most-used LAB species in cheese production (Haines & Harmon, 1973b). This study suggested that bacteriocin-independent inhibition is similar among strains within the L. lactis species. Here, a large-scale analysis of L. lactis antagonistic potential against S. aureus clearly questions the homogeneity of the antagonistic potential within L. lactis species as well as the importance of acidification in the early inhibition of S. aureus growth by L. lactis in milk. 2. Material and methods 2.1. Bacterial strains and growth conditions The 75 L. lactis strains screened in this study came from the CNRZ collection of INRA (UMR STLO INRA Agrocampus Rennes, France) and included 36 L. lactis subsp. lactis strains (48%), 19 L. lactis subsp. lactis biovar diacetylactis strains (25%), and 20 L. lactis subsp. cremoris (27%). These L. lactis strains were used in fermented dairy products (in the name of the strains, IL stands for Industrie Laitie`re). Two S. aureus strains were used in this study: S. aureus LM48, a staphylococcal enterotoxin A-producing strain isolated from cow milk, in 2002, in a French dairy farm (laboratory collection, Rennes, France) and S. aureus RN4220, a laboratory strain (Peng, Novick, Kreiswirth, Kornblum, & Schlievert, 1988). S. aureus RN4220 is a wellcharacterized and transformable agr-deficient derivative of an S. aureus strain isolated from a human infection. Mixed cultures with S. aureus and L. lactis strains and pure cultures of S. aureus strains (used as control) were grown in tubes at 30 1C, under static conditions, on M17 broth containing 0.5% lactose (AES, Combourg, France) or low heat skim milk (kindly provided by P. Schuck, UMR STLO INRA Agrocampus Rennes, France) reconstituted in sterile water (10% (w/v); hereafter referred to as LH milk). The same stock of LH milk powder was used for the entire study to avoid any variability in milk composition. LH milk properties are similar to raw milk ones, ensuring mixed culture conditions close to traditional cheese-making process (Garem, Schuck, & Maubois, 2000). Media were inoculated using pre-cultures grown overnight under static conditions on the same medium at 25 1C (L. lactis) or 37 1C (S. aureus). For cultures on LH milk, two successive subcultures in milk were grown to allow adaptation to the medium. Predetermined volumes of each pre-culture were added to mixed and control cultures to obtain initial cell concentrations of 103 and 106 cfu mL1 for S. aureus and L. lactis, respectively. Those concentrations were determined according to field conditions. For L. lactis, a population of 106 cfu mL1 corresponds to the concentration used for starter LAB in cheese-making

process. For S. aureus, a population of 103 cfu g1 of end product corresponds to the level of contamination tolerated in some raw milk cheeses. Cultures in M17 broth or LH milk at constant pH were grown in 2 L fermentor (Setric Ge´nie Industriel, Toulouse, France). Regulation of pH at the initial value (i.e. 6.8 for milk and 6.9 for M17) was achieved by automatic addition of NaOH (5 M). 2.2. Analytical methods Bacterial growth was followed by CFU determination using the micromethod previously described (Baron et al., 2006). S. aureus population was determined on Tryptic Soy Broth (TSA, AES, Combourg, France) agar plates supplemented with 6.5% NaCl and incubated 24 h at 37 1C. Total population was determined on M17 agar plates incubated 24 h at 30 1C. The results reported here are the mean counts of at least two independent experiments. Lactic acid concentration was measured by highpressure liquid chromatography (HPLC). Prior to analysis, proteins were precipitated by adding 0.02 M H2SO4 and one drop of HCl (37%) to 2 mL of milk culture sample. After centrifugation (10000  g, 15 min, 4 1C), the supernatant was filtered through a 0.2 mm Millipore filter. Separation was achieved using an Aminex A.6 column (Biorad, Marnes la Coquette, France) and 0.02 M H2SO4 as eluent at a flow rate of 1 mL min1. The column effluents were monitored using an UV spectrophotometric detector at 210 nm (Michalski et al., 2004). 2.3. Effect of pH/lactic acid on S. aureus growth in LH milk Cultures were grown as described above, except that 0–1% lactic acid was added prior to inoculation, which resulted in milk acidification (pH 6.8–4.1). The effect of lactic acid (0–4%) on S. aureus growth in skim reconstituted milk was also investigated at pH 6.8 after neutralization of pH by addition of NaOH (10 M). 2.4. Detection of bacteriocin-producing strains Bacteriocin production was detected as described previously (Piard, Delorme, Giraffa, Commissaire, & Desmazeaud, 1990). Tested L. lactis strains were grown on M17 medium until late exponential phase. A 10 mL sample of culture supernatant was then laid, after neutralization by NaOH, on a M17 agar plate seeded with L. lactis CNRZ117, a strain known to be sensitive to most bacteriocins. Plates were incubated at 30 1C for 24 h to allow growth of L. lactis CNRZ117. The presence of bacteriocin in the tested supernatant led to an inhibition zone, due to the diffusion of the bacteriocin through the agar. L. lactis CNRZ481, a bacteriocin-producing strain, was used as a positive control. Experiments were repeated twice. Similar experiments were performed using S. aureus RN4220 and LM48 as test strains.

ARTICLE IN PRESS C. Charlier et al. / International Dairy Journal 18 (2008) 197–203

2.5. Detection of proteolytic activity, total DNA extraction and PCR experiments for the detection of prtP gene in L. lactis Lactococcus lactis strains were tested for proteinase phenotype by a plate assay as previously described (Huggins & Sandine, 1984). The presence of prtP gene in the strains was checked by a PCR test using primers designed in conserved regions of prtP sequence as determined by the alignment of several publicly available sequences (GenBank accession numbers: AF247159, AY542690, AY542691 and J04962). The primers used were PRT1 50 -ACAGTTGGCGGCTAAAGGTA-30 for forward primer and PRT2 50 -ACCGAGACAACTGTGCCTTC-30 for reverse primer. PCR were run on total DNA extracted from L. lactis strain. Briefly, samples of 50 mL of the L. lactis overnight cultures were collected and spun down in PCR tubes. Culture supernatants were discarded and cell pellets were resuspended in 50 mL of distilled water. Samples were submitted to heating (10 min at 95 1C). After heating, 10 mL of each supernatant containing total DNA extract was used for PCR experiments with the primer pair specific for prtP. Primers specific for ldh (L. lactis gene encoding the lactate dehydrogenase; LDH1, 50 -ATGGTGTCGCTGTAGCTCTTG-30 forward and LDH2, 50 -GCAGTCGCTTACGCCATATTG-30 reverse) and L. lactis 16S rRNA (16S1 50 -ACTGGATGAGCAGCGAACGG-30 forward and 16S2, 50 -ACAACGCGGGATCATCTTTGA-30 reverse) were used as positive controls. Taq Polymerase was used as recommended by the supplier (New England Biolabs, Hitchin, Hertfordshire, UK) and PCR were run on a BioRad (Marnes la Coquette, France) thermocycler (95 1C during 10 min/25 PCR cycles as follow: 95 1C during 30s, 55 1C during 30s, and 72 1C during 1 min 30/72 1C during 10 min). 3. Results and discussion 3.1. Variability in antagonistic potential of L. lactis strains against S. aureus Seventy-five L. lactis strains were screened for their ability to inhibit S. aureus growth in LH milk. Overnight cultures were used to inoculate the mixed cultures at 106 and 103 cfu mL1 for L. lactis and S. aureus, respectively. In parallel, pure cultures of S. aureus strains were grown in LH milk and used as controls. After 24 h of growth, pH was measured and S. aureus and total populations were estimated by enumeration as described above. As the results for LM48 and RN4220 in mixed cultures with L. lactis strains were similar, only the LM48 data are discussed hereafter. Inhibition levels were variable among L. lactis strains and ranged from bacteriostatic to no inhibitory effect on S. aureus growth. However, a great majority of the lactococcal strains tested (93%) had a strong inhibitory effect. L. lactis strains could be classified into three distinct groups, taking account of both S. aureus

199

population and pH value: group I comprised 36 strains (48%) that exhibited a strong acidification (pH24hp4.5) and a strong inhibition (X2.5 log, i.e. after 24 h, the S. aureus population cultured with group I strains was X2.5 log lower than the S. aureus population in the control pure culture); group II comprised 34 strains (45%) with low acidification (pH24hX5.8) and strong inhibition (X2.5 log); group III comprised five strains (7%) with low acidification (pH24hX5.9) and low inhibition (p0.5 log). Although we screened a large panel, only few strains were found in group III. This might explain why previous studies performed on small numbers of strains concluded that there was only a slight inter-strain variability in terms of inhibitory potential in the L. lactis species (Haines & Harmon, 1973b). We report here for the first time that the early inhibition of S. aureus growth by L. lactis in milk is strain-dependent. This result is consistent with other studies exploring the inhibitory potential of L. lactis strains against Bacillus cereus, another Gram-positive pathogenic bacterium involved in food poisoning outbreaks. The authors found that an intra-species variability exists among L. lactis strains when screened for their inhibitory capacities against B. cereus growth in reconstituted skimmed milk (Rossland, Andersen Borge, Langsrud, & Sorhaug, 2003). The different subspecies of the L. lactis panel were evenly distributed among the three groups. Characteristics of a representative strain of each group are presented in Table 1. Group II and III strains did not grow as high as the group I strains in milk (maximum population, in pure culture, 108 versus 109 cfu mL1). Using a PCR test targeted on the prtP gene encoding the unique lactococcal cell-wall protease required for an optimal growth in milk medium (de Vos, Vos, de Haard, & Boerrigter, 1989), this was related to the lack of PrtP (Table 1). PCR targeted on ldh and L. lactis 16S rRNA were used as positive controls. Similar results were obtained with two other strains in each of the three groups.

Table 1 Characteristics of mixed cultures with Lactococcus lactis strains representative of the three groups

IL13 (group I) IL417 (group II) IL411 (group III) a

Total population (cfu mL1)a

Staphylococcus aureus inhibitiona,b

pH24h

prtPc

1.4  109 8  107 6.8  106

3.3 logs 2.7 logs 0.2 log

4.3 6.0 6.4

+  

Data are mean values of at least two independent experiments. Given as the difference in S. aureus population between pure and mixed cultures. c Presence (+) or absence () of prtP, determined by a PCR test as described in material and methods. b

ARTICLE IN PRESS C. Charlier et al. / International Dairy Journal 18 (2008) 197–203

200

1.0E+10

3.2. S. aureus growth inhibition occurs at the beginning of the mixed culture

Total population (cfu.ml-1)

1.0E+09 1.0E+08 1.0E+07 1.0E+06 1.0E+05 1.0E+04 1.0E+03 1.0E+02 0

5

10

15 Time (h)

20

25

30

0

5

10

15 Time (h)

20

25

30

S. aureus population (cfu.ml-1)

1.0E+07

1.0E+06

1.0E+05

1.0E+04

1.0E+03

1.0E+02

7 6.5

pH

6 5.5

Kinetics of the inhibition were analyzed for a total of three L. lactis strains, one representative of each group (Table 1). Mixed cultures were performed in LH milk as described above. S. aureus counts slightly increased (1 log cfu mL1) during the first 6 h of mixed culture. Afterwards, growth inhibition occurred in the mixed cultures with L. lactis IL13 (group I) and IL417 (group II) and was maintained throughout the experiment (Fig. 1). These results are consistent with other studies where S. aureus was even found to peak (3 log increase) within the first hours of three different semi-hard cheese-making processes (Delbes et al., 2006). When cultivated with L. lactis IL411 (group III), S. aureus growth was only delayed. Until 14 h of mixed culture, L. lactis IL411 was predominant over the S. aureus population (42 log) and total population corresponded to L. lactis population. Interestingly, although IL411 growth was similar to the strains’ growth in groups I and II during the first 6 h, IL411 population dropped after 9 h. Additional experiments showed that when IL411 population reached 1.108 cfu mL1 in LH milk, both in pure and in mixed culture, a great loss of viability was observed. This may result from autolysis of this lactococcal strain. After 24 h of mixed culture, the total population represented in fact the S. aureus population, suggesting that S. aureus benefits from L. lactis IL411 loss of viability or lysis to overcome its growth inhibition or limitation. Such inhibition of S. aureus growth was previously reported in a variety of semi-hard cheeses (Delbes et al., 2006; Lamprell, 2003) and a positive correlation was found between pH at 6 h and the subsequent staphylococcal growth between 6 and 24 h suggesting that pH has a modulating effect on subsequent growth until 24 h (Delbes et al., 2006; Lindqvist et al., 2002). Here, inhibition of S. aureus growth occurred rapidly in mixed culture with L. lactis IL13 and IL417 and persisted regardless their acidification potentials (Fig. 1C). The inhibition may be due either to the acidification for group I and other feature(s) (to be determined) for group II or to feature(s) other than acidification for both groups.

5

3.3. Inhibition of S. aureus growth in milk by lactic acid/ low pH

4.5 4 0

5

10

15

20

25

30

Time (h) Fig. 1. (A) Growth kinetics of mixed cultures between Staphylococcus aureus LM48 and Lactococcus lactis strains of the three groups. Total (A) and S. aureus LM48 (B) population counts were determined as described in the text. (C) Acidification curves. Symbols correspond to LM48 in pure culture (~), and in mixed culture with L. lactis IL13 (group I) (m), IL417 (group II) (K), or IL411 (group III) (’). Data are the averages of two independent experiments.

We evaluated the impact of pH alone, on S. aureus LM48 growth in pure culture in pre-acidified LH milk (pH 6.8–4.1) as described above. Pre-acidification was performed by adding 0–1% lactic acid, which is predominantly produced by the tested L. lactis strains as determined by HPLC (Michalski et al., 2004) (data not shown). For a pH range corresponding to group II strains (pHX5.8), the S. aureus growth rate gradually decreased, yet for no more than 30–40% (Fig. 2A). For pH lower than 5, which corresponded to lactate concentrations above

ARTICLE IN PRESS C. Charlier et al. / International Dairy Journal 18 (2008) 197–203

0.5% (found with group I strains), the maximum specific growth rate strongly decreased and no growth occurred at pH 4.4–4.5. Previous works reported similar effects of pH and or lactate concentration on S. aureus growth rate in laboratory media (Haines & Harmon, 1973a; Iandolo et al., 1964; Notermans & Heuvelman, 1983). Acidification may play a long-term role by altering S. aureus survival and preventing subsequent growth. We showed here that the early S. aureus growth inhibition was not due to acidification with group II and III strains as the pH of the mixed cultures was always permissive for S. aureus growth (pHX6). S. aureus population should have reached a level similar to the control, even at reduced growth rate (Fig. 2). II and III

I

1.2

1.2

1 1

0.8

0.8

0.4 0.2

0.6

0 -0.2

Lactate (%)

Growth rate (h-1)

0.6

0.4

-0.4 -0.6

0.2

-0.8 0

-1 4

5

6

7

pH II and III

I

1.E+08

Biomass concentration (cfu.ml-1)

1.E+07 1.E+06 1.E+05 1.E+04 1.E+03 1.E+02

201

S. aureus growth inhibition by L. lactis IL13 (group I) occurred within the 5 first hours, while pH was still around 6.4 reinforcing the idea that acidification did not play a major role in the early inhibition. 3.4. Acidification is not involved in the early L. lactis antagonistic potential against S. aureus in milk To elucidate the role of acidification in the early S. aureus growth inhibition (especially for group I strains), cultures were performed in LH milk at constant pH. S. aureus growth inhibition occurred in the early stage of mixed culture and a 3 log inhibition was observed after 24 h, like under unregulated pH conditions (Fig. 3). This demonstrated that acidification does not play an essential role in S. aureus growth inhibition in milk. However, S. aureus inhibition may be due to lactic acid by itself since it reportedly alters bacterial growth independently of the pH value (Hsiao & Siebert, 1999). A maximum of 3% of lactic acid was produced in mixed cultures with group I strains as determined by HPLC (Michalski et al., 2004). However, we found that neither the growth rate nor the biomass reached after 24 h of culture were affected by lactate concentrations lower than 4% at neutral pH (data not shown). Thus, the inhibition observed in mixed cultures in milk under constant pH conditions was not due to the accumulation of lactic acid. Altogether, these results are interesting as, with group I and II L. lactis strains, S. aureus populations did not reach levels where enterotoxin production is detectable and or activated (Le Loir et al., 2003). Significant inhibitory effects on S. aureus growth were observed in mixed cultures even with low-acidifying L. lactis strains (group II). This result should be pointed out since the current trend in softor semi-soft cheese-making process (notably in stabilized cheeses) is to use L. lactis/Streptococcus thermophilus as mixed starter cultures to maintain pH above 5.2 so that ripening starts faster. We showed here that a relevant selection of starter L. lactis strains (in favor of group II strains) makes it possible to keep a rather high pH in the first steps (until the beginning of ripening) of cheesemaking processes and still prevents from the risk of S. aureus growth and enterotoxin production. 3.5. The L. lactis antagonistic potential observed is independent of bacteriocin- and or oxygen peroxideproduction

1.E+01 1.E+00 4

5

6

7

pH Fig. 2. Effect of pre-acidification on Staphylococcus aureus LM48 growth in LH milk at 30 1C under static conditions: (A) exponential growth rate (~) and initial concentration of lactate (W) and (B) S. aureus LM48 populations determined after 24 h of culture in milk pre-acidified at different pH. Data are the averages of two independent experiments. Range of final pH measured at the end of mixed cultures are indicated for group-I, II and III strains.

Along with acidification, bacteriocin- and or oxygen peroxide-production are often associated with LAB inhibitory potential against S. aureus (Cotter, Hill, & Ross, 2005; Otero & Nader-Macias, 2005). Both features were checked as described previously (Nunez de Kairuz et al., 1988; Piard et al., 1990) for eight L. lactis strains (IL13, IL417, IL411 and five other strains chosen over the three groups). They produced neither bacteriocin nor hydrogen peroxide.

ARTICLE IN PRESS C. Charlier et al. / International Dairy Journal 18 (2008) 197–203

202

features, and probably nutrient-related phenomena (e.g. nutritional competition), significantly contribute to the antagonism of L. lactis strains against S. aureus when cultivated in milk. Interestingly, these nutrient-related phenomena seem to be common to group I and group II, corresponding to strong- and low-acidifying L. lactis strains, respectively.

1.0E+11

Total population (cfu.ml-1)

1.0E+10 1.0E+09 1.0E+08 1.0E+07

4. Conclusions

1.0E+06 1.0E+05 1.0E+04 1.0E+03 1.0E+02 0

5

10

0

5

10

15 Time (h)

20

25

30

20

25

30

S. aureus population (cfu.ml-1)

1.0E+09 1.0E+08 1.0E+07 1.0E+06 1.0E+05 1.0E+04 1.0E+03 1.0E+02 15 Time (h)

Fig. 3. Kinetics of Staphylococcus aureus and total population growth in LH milk at constant pH (6.9) during mixed cultures. Total (A) and S. aureus LM48 (B) population counts were determined as described in the text. Signs correspond to LM48 in pure culture (~), and in mixed culture with L. lactis IL13 (group I) (m), IL417 (group II) (K), or IL411 (group III) (’). Data are the averages of two independent experiments.

3.6. The antagonistic potential of L. lactis is medium dependent Another feature, which has been proposed to account for the inhibitory potential of L. lactis is nutritional competition (Haines & Harmon, 1973a). The involvement of such mechanisms was investigated by performing mixed cultures between S. aureus LM48 and L. lactis IL13, IL417, and IL411 on M17 medium, a rich laboratory medium under regulated pH conditions. Surprisingly, inhibition of S. aureus LM48 growth was completely absent in these conditions (data not shown). In all cases, S. aureus reached a biomass concentration around 108 cfu mL1 after 24 h. These results strongly support that medium-dependent

This study was performed in conditions close to field conditions and cheese-making process (medium, temperature, level of inoculation) and revealed a wide range in L. lactis antagonistic potentials against S. aureus growth in milk. Our results contrast with previous studies, which reported a homogenous antagonistic potential within L. lactis species (Haines & Harmon, 1973a,b) and a prevalence of L. lactis acidification capacities in the antagonistic potential against S. aureus growth (Delbes et al., 2006; Meyrand et al., 1998). Although acidification plays an important role in S. aureus survival, we demonstrated here that acidification is not involved in the early L. lactis antagonistic potential against S. aureus growth in milk and that low-acidifying L. lactis strains efficiently inhibit S. aureus growth even in milk at initial contamination levels as high as 103 cfu mL1. This is particularly relevant since a trend in semi-soft and soft cheese technology is to use low-acidifying LAB starters so that ripening starts faster. Numerous studies explored the effect of bacteriocinproducing L. lactis strains on different spoiling or pathogenic bacteria in milk products (Abee, Krockel, & Hill, 1995). In contrast, bacteriocin-independent inhibition is poorly documented. Apart from bacteriocin production or acidification, the involvement of nutrient-related phenomena in the S. aureus growth inhibition was never clearly demonstrated in milk. Nutrient-related phenomena are often considered as direct nutritional competition or limitation. However, indirect inhibitory effects may also be involved. The availability of nutrients may trigger other mechanisms, leading for instance to the secretion of metabolites, peptides or signaling molecules, which would in turn be responsible for the inhibitory effect. Going further in this kind of study is complex when envisioned through a classical microbiological approach. The development of DNA microarrays makes it now possible to study L. lactis or S. aureus gene expression profiles in response to diverse environmental conditions (Resch, Rosenstein, Nerz, & Gotz, 2005; Weinrick et al., 2004). As bacterial inter-species interactions are multifactorial, these microarray tools will be very useful to unravel the mechanisms involved in the interactions. Acknowledgments Authors are grateful to Noe¨l Deschamps, Vincent Juillard and Jean-Christophe Piard (INRA Jouy en Josas, France) for constructive discussions at the beginning of this

ARTICLE IN PRESS C. Charlier et al. / International Dairy Journal 18 (2008) 197–203

work, and for providing control strains for bacteriocin detection assay. They also thank Judith L. Kengne´ Moube´ for technical assistance and Marina Cretenet for critical reading of the manuscript. Cathy Charlier is recipient of a Ph.D. fellowship from INRA and Re´gion Bretagne. Part of this work was supported by Agence Nationale de la Recherche (‘‘GenoFerment’’ Project). References Abee, T., Krockel, L., & Hill, C. (1995). Bacteriocins: Modes of action and potentials in food preservation and control of food poisoning. International Journal of Food Microbiology, 28, 169–185. Ammor, S., Tauveron, G., Dufour, E., & Chevallier, I. (2006). Antibacterial activity of lactic acid bacteria against spoilage and pathogenic bacteria isolated from the same meat small-scale facility: 2-Behaviour of pathogenic and spoilage bacteria in dual species biofilms including a bacteriocin-like-producing lactic acid bacteria. Food Control, 17, 462–468. Barber, L. E., & Deibel, R. H. (1972). Effect of pH and oxygen tension on staphylococcal growth and enterotoxin formation in fermented sausage. Applied Microbiology, 24, 891–898. Baron, F., Cochet, M. F., Ablain, W., Grosset, N., Madec, M. N., Gonnet, F., et al. (2006). Rapid and cost-effective method for microorganism enumeration based on miniaturization of the conventional plate-counting technique. Le Lait, 86, 251–257. Cotter, P. D., Hill, C., & Ross, R. P. (2005). Bacteriocins: Developing innate immunity for food. Nature Reviews Microbiology, 3, 777–788. De Buyser, M. L., Dufour, B., Maire, M., & Lafarge, V. (2001). Implication of milk and milk products in food-borne diseases in France and in different industrialised countries. International Journal of Food Microbiology, 67, 1–17. Delbes, C., Alomar, J., Chougui, N., Martin, J. F., & Montel, M. C. (2006). Staphylococcus aureus growth and enterotoxin production during the manufacture of uncooked, semihard cheese from cows’ raw milk. Journal of Food Protection, 69, 2161–2167. de Vos, W. M., Vos, P., de Haard, H., & Boerrigter, I. (1989). Cloning and expression of the Lactococcus lactis subsp cremoris SK11 gene encoding an extracellular serine proteinase. Gene, 85, 169–176. Garem, A., Schuck, P., & Maubois, J. L. (2000). Cheesemaking properties of a new dairy-based powder made by a combination of microfiltration and ultrafiltration. Le Lait, 80, 25–32. Haines, W. C., & Harmon, L. G. (1973a). Effect of selected lactic acid bacteria on growth of Staphylococcus aureus and production of enterotoxin. Applied Microbiology, 25, 436–441. Haines, W. C., & Harmon, L. G. (1973b). Effect of variations in conditions of incubation upon inhibition of Staphylococcus aureus by Pediococcus cerevisiae and Streptococcus lactis. Applied Microbiology, 25, 169–172. Hsiao, C. P., & Siebert, K. J. (1999). Modeling the inhibitory effects of organic acids on bacteria. International Journal of Food Microbiology, 47, 189–201. Huggins, A. R., & Sandine, W. E. (1984). Differentiation of fast and slow acid-producing strains of lactic Streptococci. Journal of Dairy Science, 67, 1674–1679. Iandolo, J. J., Ordal, Z. J., & Witter, L. D. (1964). The effect of incubation temperature and controlled pH on the Growth of Staphylococcus aureus Mf 31 at various concentrations of NaCl. Canadian Journal of Microbiology, 10, 808–811. Jorgensen, H. J., Mork, T., & Rorvik, L. M. (2005). The occurrence of Staphylococcus aureus on a farm with small-scale production of raw milk cheese. Journal of Dairy Science, 88, 3810–3817. Kao, C. T., & Frazier, W. C. (1966). Effect of lactic acid bacteria on growth of Staphylococcus aureus. Applied Microbiology, 14, 251–255. Lamprell, H. (2003). La production des ente´rotoxines dans les fromages en fonction de la diversite´ phe´notypique et ge´ne´tique des souches de Staphylococcus aureus. Poligny, France: University of Bourgogne.

203

Le Loir, Y., Baron, F., & Gautier, M. (2003). Staphylococcus aureus and food poisoning. Genetics Molecular Research, 2, 63–76. Lindqvist, R., Sylven, S., & Vagsholm, I. (2002). Quantitative microbial risk assessment exemplified by Staphylococcus aureus in unripened cheese made from raw milk. International Journal of Food Microbiology, 78, 155–170. Metaxopoulos, J., Genigeorgis, C., Fanelli, M. J., Franti, C., & Cosma, E. (1981). Production of Italian dry salami: Effect of starter culture and chemical acidulation on Staphylococcal growth in salami under commercial manufacturing conditions. Applied and Environmental Microbiology, 42, 863–871. Meyrand, A., Boutrand-Loei, S., Ray-Gueniot, S., Mazuy, C., Gaspard, C. E., Jaubert, G., et al. (1998). Growth and enterotoxin production of Staphylococcus aureus during the manufacture and ripening of Camembert-type cheeses from raw goats’ milk. Journal of Applied Microbiology, 85, 537–544. Michalski, M. C., Camier, B., Briard, V., Leconte, N., Gassi, J. Y., Goude´dranche, H., et al. (2004). The size of native milk fat globules affects physico-chemical and functional properties of Emmental cheese. Le Lait, 84, 343–358. Minor, T. E., & Marth, E. H. (1970). Growth of Staphylococcus aureus in acidified pasteurized milk. Journal of Milk and Food Technology, 33, 516–520. Nes, I. F., & Johnsborg, O. (2004). Exploration of antimicrobial potential in LAB by genomics. Current Opinion in Biotechnology, 15, 100–104. Notermans, S., & Heuvelman, C. J. (1983). Combined effect of water activity, pH and sub-optimal temperature on growth and enterotoxin production of Staphylococcus aureus. Journal of Food Science, 48, 1832–1840. Nunez de Kairuz, M. S., Olazabal, M. E., Oliver, G., Pesce de Ruiz Holgado, A. A., Massa, E., & Farias, R. N. (1988). Fatty acid dependent hydrogen peroxide production in Lactobacillus. Biochemical and Biophysical Research Communications, 152, 113–121. Olarte, C., Sanz, S., Gonzalez-Fandos, E., & Torre, P. (2000). The effect of a commercial starter culture addition on the ripening of an artisanal goat’s cheese (Cameros cheese). Journal of Applied Microbiology, 88, 421–429. Otero, M. C., & Nader-Macias, M. E. (2005). Inhibition of Staphylococcus aureus by H2O2-producing Lactobacillus gasseri isolated from the vaginal tract of cattle. Animal Reproduction Science, 96, 35–46. Peng, H. L., Novick, R. P., Kreiswirth, B., Kornblum, J., & Schlievert, P. (1988). Cloning, characterization, and sequencing of an accessory gene regulator (agr) in Staphylococcus aureus. Journal of Bacteriology, 170, 4365–4372. Piard, J. C., Delorme, C., Giraffa, G., Commissaire, J., & Desmazeaud, M. (1990). Evidence of a bacteriocin produced by Lactococcus lactis CNRZ481. Netherlands Milk Diary Journal, 44, 143–158. Resch, A., Rosenstein, R., Nerz, C., & Gotz, F. (2005). Differential gene expression profiling of Staphylococcus aureus cultivated under biofilm and planktonic conditions. Applied and Environmental Microbiology, 71, 2663–2676. Rossland, E., Andersen Borge, G. I., Langsrud, T., & Sorhaug, T. (2003). Inhibition of Bacillus cereus by strains of Lactobacillus and Lactococcus in milk. International Journal of Food Microbiology, 89, 205–212. Rossland, E., Langsrud, T., Granum, P. E., & Sorhaug, T. (2005). Production of antimicrobial metabolites by strains of Lactobacillus or Lactococcus co-cultured with Bacillus cereus in milk. International Journal of Food Microbiology, 98, 193–200. Schellenberg, J., Smoragiewicz, W., & Karska-Wysocki, B. (2006). A rapid method combining immunofluorescence and flow cytometry for improved understanding of competitive interactions between lactic acid bacteria (LAB) and methicillin-resistant S aureus (MRSA) in mixed culture. Journal of Microbiological Methods, 65, 1–9. Weinrick, B., Dunman, P. M., McAleese, F., Murphy, E., Projan, S. J., Fang, Y., et al. (2004). Effect of mild acid on gene expression in Staphylococcus aureus. Journal of Bacteriology, 186, 8407–8423.