Cécile Callon

Diversity of lactic acid bacteria isolated from AOC Salers cheese.

The objective of this work was to describe the diversity of lactic acid bacteria in traditional raw milk Salers cheeses at the species and strain levels. The characterization of 381 strains isolated during ripening and various strain collections was investigated using physiological analysis and molecular techniques: Rep-PCR, species and genus specific amplifications and the sequence analysis of 16S rDNA for strain typing and taxonomic identification. The strains belonged to Lactobacillus plantarum, Lactobacillus paracasei, Lactococcus lactis, Lactococcus garviae, Enterococcus faecalis, Enterococcus faecium, Leuconostoc mesenteroides, Leuconostoc pseudomesenteroides, Streptococcus salivarius, Streptococcus millieri, Streptococcus macedonicus and Pediococcus pentosaceus. A wide phenotypic and genomic heterogeneity was observed within the different species (Lactobacillus plantarum, Lactobacillus paracasei and Leuconostoc mesenteroides) according to the origin and the time of ripening. The natural microflora was different from strain collection and each method must be combined to identify and characterize natural microflora. This study revealed the low selectivity of selective media used for the isolation of different groups of lactic acid bacteria except the Facultatively Heterofermentative lactobacilli medium selecting mesophile lactobacilli and SB medium selective for Enterococcus. The study reveals, for the first time, the microbial lactic acid bacteria community of Salers cheese and its diversity. A better knowledge of microbial flora will be useful to improve understanding of sensory quality of cheeses.

Simplification of a complex microbial antilisterial consortium to evaluate the contribution of its flora in uncooked pressed cheese

A complex microbial consortium derived from raw milk and composed of populations classified in 4 groups (lactic acid bacteria (A), Gram positive catalase positive bacteria (B), Gram negative bacteria (C) and yeasts (D)) can contribute to the inhibition of Listeria monocytogenes in the core of an uncooked pressed cheese. To identify what groups may be involved in the inhibition, the consortium was simplified by successively omitting one group at a time. Pasteurized milk was inoculated with these more or less complex consortia and their effects on L. monocytogenes count, pH, acids and volatile compounds in the core of uncooked pressed cheese were evaluated. The growth of L. monocytogenes was the highest in cheeses prepared with pasteurized milk and only St. thermophilus. Inhibition in other cheeses was expressed by comparison with growth in these ones. All the consortia containing both lactic acid bacteria (group A) and Gram positive catalase positive bacteria (group B)--ABCD, ABD, ABC, AB--were more inhibitory than those containing lactic acid bacteria on its own (A) or associated only with yeasts (AD) or/and Gram negative (ADC). Consortia without lactic acid bacteria were weakly inhibitory or had no effect. Gram positive catalase positive bacteria alone were not inhibitory although most of the species became established in the cheeses. The Lactobacillus population (Lb. casei, Lb. plantarum, Lb. curvatus and Lb. farciminis) was predominant in cheeses (9 log CFU/g) with a higher count than Leuconostoc (7 log CFU/g) and Enterococcus (7 log CFU/g). Lactobacillus counts were negatively correlated with those of L. monocytogenes (r=-0.84 at 18 days) and with the level of D-lactic acid. There was no correlation between L. monocytogenes and Leuconostoc or Enterococcus counts. Complex consortium ABCD and AB not only had a stronger inhibitory power in cheeses than consortium AD, they were also associated with the highest levels of L-lactic and acetic acids. All cheeses inoculated with lactic acid bacteria differed from those without by higher levels of ethyl formiate, pentane and alcohols (2-butanol, 2-pentanol), and lower levels of ketones (2-hexanone, 2,3-butanedione) and aldehydes (2-methyl-butanal). Levels of 2-methyl-butanal, 2-butanol and 2-pentanol were higher in ABCD and AB cheeses than in AD cheeses. Beside their contribution to the inhibition, their effect on cheese flavour must be evaluated.

Microbial biodiversity in cheese consortia and comparative Listeria growth on surfaces of uncooked pressed cheeses.

The study set out to determine how changes in the microbial diversity of a complex antilisterial consortium from the surface of St-Nectaire cheese modify its antilisterial activities. On the basis of the microbial composition of a natural complex consortium named TR15 (Truefood consortium 15), three new consortia of different species and strain compositions were defined: TR15-SC (58 isolates from TR15 collection), TR15-M (pools of isolates from selective counting media) and TR15-BHI (pools of isolates from BHI medium). Their antilisterial activities on the surfaces of uncooked pressed cheese made with pasteurised milk were compared with the activity of complex consortium TR15 and a control cheese inoculated only with starter culture (Streptococcus thermophilus, Lactobacillus delbrueckii). The natural consortium TR15 was the most inhibitory, followed by reconstituted consortium TR15-BHI. The dynamics of the cheese rind microbial flora were monitored by counting on media and by isolate identification using 16S rDNA sequencing and direct 16S rDNA Single Strand Conformation Polymorphism analysis. The combination of these methods showed that rind with natural consortium TR15 had greater microbial diversity and different microbial dynamics than cheese rinds with reconstituted consortia. Cheese rind with the natural consortium showed higher citrate consumption and the highest concentrations of lactic and acetic acids, connected with high levels of lactic acid bacteria such as Carnobacterium maltaromaticum, Vagococcus fluvialis, Enterococcus gilvus, Leuconostoc mesenteroides, Brochothrix thermosphacta and Lactococcus lactis, ripening bacteria such as Arthrobacter nicotianae/arilaitensis, and Gram negative bacteria (Pseudomonas psychrophila and Enterobacter spp.). The highest L. monocytogenes count was on rind with TR15-M and was positively associated with the highest pH value, high succinic and citric acid contents, and the highest levels of Marinilactibacillus psychrotolerans and Gram positive catalase positive bacteria represented by Staphylococcus vitulinus, Brevibacterium linens, Microbacterium gubbeenense and Brachybacterium tyrofermentans. The results show that the species composition of consortium is more important than the number of species. It is likely that inhibition mechanisms differ from one consortium to another; investigating gene expression will be an effective way to elucidate microbial interactions in cheese.

Survival of Escherichia coli O26:H11 exceeds that of Escherichia coli O157:H7 as assessed by simulated human digestion of contaminated raw milk cheeses.

Shiga toxin producing Escherichia coli (STEC) are an important cause of human foodborne outbreaks. The consumption of raw milk dairy products may be an important route of STEC infection. For successful foodborne transmission, STEC strains must survive stress conditions met during gastrointestinal transit in humans. The aim of this study was to evaluate the survival of two STEC strains of serotypes O157:H7 and O26:H11 during simulated human digestion in the TNO gastro-Intestinal tract Model (TIM) of contaminated uncooked pressed cheeses. The survival of cheese microflora during in vitro gastrointestinal transit was also determined for the first time. The level of STEC increased from 2 log₁₀ CFU/ml to 4 log₁₀ CFU/g during the first 24h of cheese making and remained stable at around 4 log₁₀ CFU/g during cheese ripening and conservation. During transit through the artificial stomach and duodenum, levels of STEC decreased: 0.2% of E. coli O157:H7 and 1.8% of E. coli O26:H11 were recovered at 150 min in the gastric compartment, compared with 14.3% for the transit marker. Bacterial resumption was observed in the jejunum and ileum: 35.8% of E. coli O157:H7 and 663.2% of E. coli O26:H11 were recovered at 360 min in the ileal compartment, compared with 12.6% for the transit marker. The fate of STEC was strain-dependent, the survival of E. coli O26:H11 being 13 times greater than that of E. coli O157:H7 at the end of digestion in the cumulative ileal deliveries. These data provide a better understanding of STEC behavior during gastrointestinal transit in humans after ingestion of contaminated cheese.

Contribution of hydrogen peroxide to the inhibition of Staphylococcus aureus by Lactococcus garvieae in interaction with raw milk microbial community.

The response of Staphylococcus aureus growth inhibition by Lactococcus garvieae to catalase and milk lactoperoxidase, and its efficiency in raw milk cheese were evaluated. S. aureus and L. garvieae were co-cultivated in broth buffered at pH 6.8, and in raw, pasteurized and microfiltered milk, in presence and absence of catalase. Although H2O2 production by L. garvieae was detected only in agitated broth, the inhibition of S. aureus by L. garvieae was reduced by catalase both in static and shaking cultures by 2.7 log, pasteurized milk (approximately 0.7 log), microfiltered milk (approximately 0.6 log) and raw milk (approximately 0.2 log). The growth of S. aureus alone in microfiltered milk was delayed compared with that in pasteurized milk and inhibition of S. aureus by L. garvieae was stronger in microfiltered milk. The inhibition of coagulase-positive staphylococci (CPS) by L. garvieae in raw milk cheese was similar to that in raw milk (approximately 0.8 log), but weaker than that in pasteurized and microfiltered milks. L. garvieae also had an early antagonistic effect on the growth of several other microbial groups, which lastingly affected populations levels and balance during cheese ripening.

Control of Shigatoxin-producing Escherichia coli in cheese by dairy bacterial strains

Bio-preservation could be a valuable way to control Shigatoxin-producing Escherichia coli (STEC) in cheese. To this end, 41 strains were screened for their inhibitory potential on model cheese curd and on pasteurized and raw milk uncooked pressed cheeses. Strains of Lactococcus lactis, Lactococcus garvieae, Leuconostoc pseudomesenteroides, Leuconostoc citreum, Lactobacillus sp, Carnobacterium mobile, Enterococcus faecalis, Enterococcus faecium, Macrococcus caseolyticus and Hafnia alvei reduced STEC O26:H11 counts by 1.4-2.5 log cfu g(-1) and to a lesser extent STEC O157:H7 counts in pasteurized milk cheeses. Some strains can act in synergy to inhibit STEC in raw milk uncooked pressed cheeses. Inhibitory associations had no adverse effect on the sensory characteristics of these cheeses. The association of H. alvei, Lactobacillus plantarum and Lc. lactis was the most inhibitory: after inoculation of this consortium into milk, STEC O26:H11 and O157:H7, inoculated at 2 log cfu ml(-1), were reduced by up to 3 log cfu g(-1) in ripened cheese. Inhibition in cheese cannot be predicted from H2O2 production in BHI medium, decreased pH or milk reduction. It is not clear what role the rapid decrease in pH during the first 6 h may play in the inhibition. Further studies will be needed to determine the nature of the inhibition.