Previous Article | Next Article ![]()
Applied and Environmental Microbiology, February 2007, p. 1248-1255, Vol. 73, No. 4
0099-2240/07/$08.00+0 doi:10.1128/AEM.01667-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.
Department of Dairy Science and Technology, Faculty of Agriculture, Alexandria University, Alexandria, Egypt,1 Unité Bactéries Lactiques et Pathogènes Opportunistes, INRA, Jouy-en-Josas, France2
Received 18 July 2006/ Accepted 14 December 2006
|
|
|---|
|
|
|---|
During the last decades, several investigators have isolated and identified different lactic acid bacterial species from Domiati cheese, such as Lactococcus lactis subsp. lactis, Lactobacillus delbrueckii subsp. bulgaricus, L. casei (21), L. farciminis, L. alimentarius, Enterococcus faecalis, E. faecium (16), Lactobacillus plantarum, and L. paracasei (15). Other bacterial species were also isolated, including coliforms (2), Micrococcus spp. (20), Arthrobacter spp. (14), Propionibacterium jensenii, Microbacterium lacticum, Brevibacterium linens (16), Staphylococcus aureus (9), and Aeromonas spp. (8). In all the above-mentioned studies, results were obtained using culture methods.
Recently, molecular methods such as temporal temperature gel electrophoresis (TTGE) and/or denaturing gradient gel electrophoresis (DGGE) (5, 17, 38) were successfully used to identify the bacterial biodiversity of different types of cheese such as artisanal Sicilian (43), Stilton (18), mozzarella (6, 19, 34), Beaufort, Saint Nectaire, Morbier, Epoisse (41), Mish (10), Karish (11), and hard Ras cheeses (13). Large-scale analyses of dairy samples in INRA, France, have led to the establishment of a reference database, allowing comparative identification of some 170 bacterial species, including some food pathogens (41). The aim of the present work was to use these molecular methods to characterize the bacterial biodiversity of the popular Egyptian Domiati cheese.
|
|
|---|
Genomic DNA extraction.
Total genomic DNA was extracted from each Domiati cheese sample (7 g) as previously described (41). After undergoing a protein digestion step by the pronase (Boerhinger, Mannheim, Germany), bacterial cells were mechanically lysed. DNA purification was performed as previously described (7). Pellets of DNA were then dissolved in 100 µl Tris-EDTA buffer plus RNase (Sigma, Saint Quentin Fallavier, France) and then analyzed by 0.8% agarose gel electrophoresis.
PCR amplification.
Amplicons for TTGE and DGGE analyses were prepared by performing two successive PCRs using a Gene Amp system model 2400 (PerkinElmer, France) and appropriate primers. First, a 700-bp fragment of the 16S rRNA gene including the V3 region was amplified with primers W01 and W012. Second, the V3 region of 200 bp was amplified using primers HDA1-GC and HDA2. The PCR mixtures and amplification programs were as previously described (40). Sizes and quantities of PCR products were determined by 2% agarose gel electrophoresis (Seakem CTG agarose; TEBU, France) against a standard containing DNA fragments of defined lengths (Smart Ladder, France).
TTGE and DGGE analyses.
TTGE and DGGE analyses of V3 amplicons were applied on 16-cm by 16-cm by 1-mm gels (Bio-Rad DCode universal mutation detection system; Marnes La Coquette, France) and performed as previously described (41). After runs, gels were stained for 15 min with an ethidium bromide solution (0.5 µg/ml of 1x Tris-acetate-EDTA [TAE] buffer), rinsed for 20 min in 1x TAE buffer, and photographed on a UV transillumination table.
Gel analysis and band identification using species database.
TTGE and DGGE gels were analyzed by GelCompar software (Applied-Maths, Belgium) as previously described (41). The software standardizes TTGE and DGGE profiles to minimize migration differences between gels by alignment of the identification ladder with a standard gel (40). Band identifications are performed by comparison to a species database which includes TTGE and DGGE fingerprints of about 170 bacterial species isolated from dairy ecosystems (41). In some cases, specific PCR tests and/or cloning and sequencing were undertaken to confirm species assignments or to distinguish between comigrating species.
Species-specific PCR tests.
Specific PCR tests were carried out using different species-specific primers (Table 1) with DNA obtained from the cheese samples. Primers (MWG Biotech AG, Ebersberg, Germany) were prepared at a final concentration of 60 µM in deionized, autoclaved water. PCR was performed in a GenAmp system model 2400 (PerkinElmer, France), and all reactions were carried out following conditions previously provided by the authors (Table 1). Sizes of PCR products were determined using 1.5% agarose gel electrophoresis (Seakem CTG agarose; TEBU, France).
|
View this table: [in a new window] |
TABLE 1. Primers used in this study for the species-specific PCR assays
|
Enumeration of different bacterial groups.
Domiati cheese samples (11 g) were emulsified in 99 ml of sterile 2% (wt/vol) tri-sodium citrate solution (Merck) and homogenized using an Ultra-Turrax mechanical blender (19,000 rpm for 45 s). Serial dilutions were prepared in sterile 1% (wt/vol) peptone water, plated on selective agar medium using a spiral platter (Spiral system; Cincinnati, OH), and incubated at the appropriate temperatures (Table 2). Bacterial enumerations were done on M17 (Difco, Elancourt, France) for lactococci and streptococci, on MRS (Difco) (pH adjusted to 5.2) for mesophilic and thermophilic lactobacilli, on MSE (35) for Leuconostoc, on BEA (Difco) for enterococci, on MSA (Difco) for staphylococci, on BHI (Difco) supplemented with 5% (wt/vol) NaCl for salt-tolerant flora, on YEL (yeast extract-sodium lactate medium) for Propionibacterium spp., on VRBA (Difco) for total coliform, and on cetrimide, fucidin, and cephalosporin (CFC) agar (Oxoid) for Pseudomonas spp.
|
View this table: [in a new window] |
TABLE 2. Range of viable counts of different bacterial groups in Domiati cheese samples
|
|
|
|---|
![]() View larger version (118K): [in a new window] |
FIG. 1. TTGE electrophoresis of V3 16S rDNA fragments from different samples of Domiati cheese. Samples 1 to 11 correspond to V3 16S rDNA regions PCR amplified from genomic DNA extracted from cheese samples (see Materials and Methods). After standardization of the gel by GelCompar software, bands are identified by comparison with the species database. Band a, PCR artifact; b, unidentified band; c, Lactococcus garvieae; d, Aerococcus viridans; e, unidentified band; f, Lactobacillus johnsonii, Lactobacillus gasseri; g, Lactobacillus plantarum, Lactobacillus pentosus; h, Acinetobacter lwoffii; i, Leuconostoc citreum; j, Enterococcus casseliflavus, Staphylococcus lentus; k, Leuconostoc mesenteroides; l, Staphylococcus epidermidis, Staphylococcus aureus, Staphylococcus simulans; m, Lactococcus raffinolactis, Staphylococcus equorum, Staphylococcus capitis; n, unidentified band; o, Enterococcus faecium group, Pseudomonas fluorescens, Leuconostoc pseudomesenteroides; p, Leuconostoc lactis, Staphylococcus xylosus, Lactobacillus brevis; q, Pseudomonas putida; r, Lactobacillus acidophilus group; s, Staphylococcus saprophyticus, Pseudomonas aeruginosa; t, Pediococcus pentosaceus, Macrococcus caseolyticus; u, Streptococcus uberis, Moraxella bovis; v, Enterococcus faecalis, Staphylococcus warneri; w, unidentified band; x, unidentified band; y, Staphylococcus chromogenes, Hafnia alvei, Pseudomonas alcaligenes; z, unidentified band; a', Acinetobacter spp.; b', Lactococcus lactis; c', Streptococcus thermophilus. Markers: 1, Lactococcus garvieae CNRZ1323; 2, Lactococcus raffinolactis CNRZ1214; 3, Enterococcus faecalis CE17; 4, Lactococcus lactis subsp. lactis bv. diacetilactis CNRZ260. Lane M, TTGE standardization ladder; lane 1 to lane 11, cheese samples from 1 to 11.
|
![]() View larger version (115K): [in a new window] |
FIG. 2. DGGE electrophoresis of V3 16S rRNA gene fragments from different samples of Domiati cheese. Samples 1 to 11 correspond to V3 16S rRNA gene regions PCR-amplified from genomic DNA extracted from cheese samples (see Materials and Methods). After standardization of the gel by GelCompar software, bands are identified by comparison with the species database. Band a, Klebsiella pneumoniae; b, Enterobacter amnigenus, E. coli; c, Citrobacter freundii; d, unidentified band; e, unidentified band; f, Serratia marcescens, Raoultella planticola; g, Corynebacterium variabile; h, Enterobacter cloacae, Klebsiella oxytoca; i, Serratia liquefaciens; j, unidentified band; k, Corynebacterium flavescens; l, Microbacterium gubbeenense; m, Lactobacillus casei; n, Corynebacterium vitaeruminis, Kluyvera ascorbata; o, Clostridium sporogenes; p, Micrococcus spp., Microbacterium spp.; q, unidentified band; r, Corynebacterium ammoniagenes; s, Arthrobacter spp., Brachybacterium tyrofermentans; t, unidentified band; u, Brevibacterium spp.; v, unidentified band; w, Brevibacterium linens (50%); x, unidentified band; y, Brevibacterium casei, Corynebacterium bovis; z, Propionibacterium freudenreichii; a', Kocuria kristinae, Brevibacterium linens (50%); b', Propionibacterium acidipropionici, Kocuria spp.; c', unidentified band; d', unidentified band, e', unidentified band. Markers, 1, Bacillus pumilus ATCC 7725; 2, Klebsiella oxytoca ATCC 103434T; 3, Kytococcus sedentarius CNRZ880; 4, Arthrobacter citreus CNRZ928T; 5, Kocuria kristinae CNRZ872; 6, Propionibacterium jensenii Z87. Lane M, DGGE standardization ladder; lane 1 to lane 11, cheese samples from 1 to 11. TTGE*, bands separated by TTGE method.
|
|
View this table: [in a new window] |
TABLE 3. Presence of bacterial species in individual Domiati cheese samples using specific primer tests
|
|
View this table: [in a new window] |
TABLE 4. Bacterial species identified in Domiati cheese samples using cloning and sequencing strategy
|
Many other bacterial species, belonging to different genera (e.g., Lactococcus, Lactobacillus, Leuconostoc, Enterococcus, Streptococcus, Pseudomonas, and Acinetobacter), were identified as subdominant or minor species in Domiati cheese (Fig. 1). Most of the band assignments using the species database were confirmed by specific PCR tests and/or by cloning and sequencing strategies (Tables 3 and 4). These complementary tests were very useful for identifying precisely the bacteria at the species level. For example, band g was identified as Lactobacillus plantarum and L. pentosus by the reference database and also by band sequencing (Fig. 1 and Table 4), whereas, using the specific PCR tests, L. plantarum was absent (Table 3). Consequently, it supposed that this band might correspond to Lactobacillus pentosus.
Biodiversity among high-G+C-percentage species present in Domiati cheese.
The DGGE profile for Domiati cheese samples (Fig. 2) included 31 different bands. The species database allowed the identification of only 21 bands, whereas the other 10 (Fig. 2, bands d, e, j, q, t, v, x, c', d', and e') could not be identified. The major bands (Fig. 2) putatively identified were those of Kocuria kristinae or Brevibacterium linens (band a') and Propionibacterium acidipropionici or Kocuria spp. (band b'). A species-specific PCR assay failed to identify B. linens (Table 3), thus making it likely that band a' was amplified from K. kristinae. Sequencing of the V3 region fragment of band b' identified K. rhizophila as the species (Table 4). Some other frequent bands of moderate intensity identified Klebsiella pneumoniae (band a), Arthrobacter spp./Brachybacterium tyrofermentans (band s), Corynebacterium variabile (band g), and Propionibacterium freudenreichii (band z) as the corresponding species. Other bacterial species were identified by the DDGE approach (Fig. 2) and confirmed, in some cases, Escherichia coli (band b), Citrobacter freundii (band c), L. casei (band m), and B. casei (band y) (Table 3 and 4).
Enumeration of different bacterial groups.
Several selective media were used to enumerate the different bacterial groups present in Domiati cheese samples (Table 2). The highest bacterial counts (ranging between 7.1 and 7.6 log10 CFU/g) were obtained with MSA medium, which is generally selective for the staphylococcal populations. Total coliforms were recorded as the lowest number of cells (2.1 to 3.8 log10 CFU/g). The counts of salt-tolerant flora ranged between 6.4 and 7.1 log10 CFU/g. In general, the counts of coccal lactic acid bacteria (LAB) (Lactococcus, Streptococcus, Leuconostoc, and Enterococcus), as presented in Table 2, were higher than those of rod members of LAB (mesophilic and thermophilic lactobacilli).
|
|
|---|
The combination of molecular tools (i.e., PCR-TTGE, PCR-DGGE, species-specific PCR assays, and cloning/sequencing analyses) allowed us to identify 46 different bacterial species that are present in Domiati cheese. Most of the bacteria were first presumably identified by the assignation of the TTGE/DGGE bands to a complex species database (40, 41). But the limitation of this method concerns the species comigrations. Despite sequence differences, melting temperatures of comigrated V3 fragments are similar, and thus, they migrate at the same position in denaturing gels (37). To confirm species assignments or to distinguish between comigrating species, species-specific PCR tests and/or cloning and sequencing were performed. This strategy proved to be very useful to more precisely identify the bacterial species. According to the frequencies and intensities of the bands on TTGE and DGGE gels (TTGE/DGGE methods are considered semiquantitative techniques [53] and generally, band intensities reflect the relative proportion of each species in the total bacterial population [39]), we could differentiate the bacterial population of Domiati cheeses into three groups. (i) The first is composed of dominant bacterial species (they are almost identified in each sample), e.g., Leuconostoc mesenteroides, Lactococcus garvieae, Aerococcus viridans, Lactobacillus versmoldensis, Pediococcus inopinatus/Macrococcus caseolyticus, and Lactococcus lactis subsp. lactis. The bacteria of this group may play the main role in the fermentation and organoleptic properties of Domiati cheeses because of their common presence in the tested samples. (ii) The second group is composed of frequently encountered bacterial species. These bacteria belonged to both LAB species, e.g., Leuconostoc citreum, Lactobacillus casei, Lactobacillus johnsonii, Staphylococcus thermophilus, E. faecalis, E. faecium/Leuconostoc pseudomesenteroides, and to non-LAB species, e.g., Vibrio spp., K. kristinae, C. variabile, K. rhizophila, Arthrobacter spp./B. tyrofermentans, plus numerous species of coagulase-negative staphylococci. (iii) The third group consists of occasionally encountered bacterial species, e.g., Lactococcus raffinolactis, Acinetobacter lwoffii, Staphylococcus lentus, S. chromogenes, Enterobacter cloacae, and Klebsiella oxytoca. The bacteria of the second and third groups may have a secondary activity in the fermentation process of Domiati cheeses, but the origins and potential roles of these as well as the bacteria of the first group would need further investigation.
Domiati cheese manufacture involves several technological steps, including natural fermentation of milk, salting, renneting, and ripening (pickling) in brine or salted whey solutions. The salt added to cheese milk plays an important role during manufacturing by favoring or inhibiting bacterial growth (the salt content in our cheese samples ranged between 5.46 and 9.50% of wet weight). Many of the identified bacterial species could be recognized as partially or totally salt-tolerant bacteria, e.g., Leuconostoc mesenteroides, Lactobacillus versmoldensis (up to 14% salt tolerance) (28), Lactococcus lactis subsp. lactis, Staphylococcus saprophyticus, K. rhizophila, K. halotolerans, Brachybacterium tyrofermentans (up to 16% salt tolerance) (48), K. kristinae, and Microbacterium gubbeenense. In some cases, the identified species were directly related to a marine environment, e.g., Lactococcus garvieae (which often occurs in dairy environments [3, 12, 23] and is considered one of the major pathogens responsible for fish mortality [44]), Aerococcus viridans, Vibrio spp., and marine sediment bacterium. As previously described (47), our results indicate that salt-tolerant and marine bacteria may play a role in the ripening process of Domiati cheese. Other studies of red-smear soft cheeses have demonstrated that a fraction of cheese flora was composed of microorganisms related to a marine environment (22, 32).
This study allows us to clarify reports of the occurrence of staphylococcal species in Domiati cheese, as numerous species have been identified by molecular tools. These results reflected the high bacterial counts using MSA medium (7.1 to 7.6 log10 CFU/g). Previous studies focused on the occurrence of Staphylococcus aureus because of its important role in food poisoning. Our results indicate a relative predominance of S. sciuri and S. saprophyticus (both coagulase-negative staphylococci) in Domiati cheese. Fortunately, S. aureus was detected in only one cheese sample, indicating either that good sanitation procedures were applied during Domiati cheese manufacturing or that S. aureus is not a good competitor with other bacterial species (36). This study also revealed the presence of other coagulase-negative staphylococci: S. simulans and S. chromogenes were identified for the first time in Domiati cheese. These species are common causes of subclinical mastitis (27, 29). However, it is notable that S. simulans produces lysostaphin, a cell wall-degrading enzyme that lyses practically all known staphylococcal species (45), and was recently developed for its bactericidal antistaphylococcal effects (49). Its presence might actually control development of populations of susceptible staphylococci in Domiati cheese. Other bacterial species known to cause mastitis were also identified (e.g., Streptococcus parauberis, Staphylococcus epidermidis, Citrobacter freundii, and Serratia marcescens) but were detected as minor species.
In the present study, several bacterial species (never previously isolated by culturing from Domiati cheese) were detected for the first time in Domiati cheese samples, e.g., LAB (Lactococcus garvieae, A. viridans, Pediococcus inopinatus, Lactobacillus versmoldensis, L. johnsonii, Leuconostoc citreum, and S. parauberis), and non-LAB (Pseudomonas putida, Staphylococcus sciuri, S. chromogenes, S. simulans, Macrococcus caseolyticus, Citrobacter freundii, Corynebacterium variabile, Microbacterium. gubbeenense, Propionibacterium freudenreichii, K. rhizophila, K. halotolerans/Rothia spp., and marine sediment bacterium). These results confirm interest in using the molecular methods for exhaustive and precise identification of complex microbial biodiversity occurring in artisanal cheeses such as Domiati.
The considerable bacterial biodiversity found in Domiati cheeses may explain the heterogeneous production of cheeses available in Egyptian markets. It is expected that each bacterial species present in Domiati cheese may contribute in some way to the ripening process but would vary according to its tolerance toward the probably highly variable cheese production conditions (e.g., salt concentrations, pH, ripening temperatures, and type of pickling solutions [whey or brine]).
Although traditional Domiati cheese is made from raw milk without the addition of any starters, most of the bacteria identified as dominant belonged to LAB. Nevertheless, improvements in Domiati cheese manufacture may be necessary to obtain a safe and homogenous product. This would require the systematic use of clean raw materials, controlled manufacturing steps, and selection of the appropriate LAB for the ripening process. The origin and role of the salt-tolerant and marine bacteria present in Domiati cheese should be further investigated. Finally, the application of culture methods will be valuable for the isolation of "positive" bacteria that characterize Domiati cheese for future use as cheese starter cultures.
We are very grateful to Alexandra Gruss for reading and correcting the English of the manuscript and also to Marina Aigle for very useful technical assistance. G. El-Baradei thanks Pierre Pery for support and encouragement; colleagues at UBLO and VIM Units, INRA, Jouy-en-Josas, France; S. E. El-Rakshy, M. H. Ragab, and the director of the Middle East Bureau (AUF); and the teams in Alexandria, Egypt, and Beirut, Lebanon.
Published ahead of print on 22 December 2006. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»