Use of RAPD-PCR and TTGE for the evaluation of biodiversity of whey cultures for Grana Padano cheese

Assessment of Microbial Populations in the Manufacture of Vacuum-Packaged Ready-to-Eat Roast Beef and in a Related Production Plant

Some microbiological criteria were monitored for 6 months in vacuum-packaged roast beef (15 production batches), raw beef (10 batches), and other meat products (12 batches) produced in an Italian small to medium-size enterprise. Fifty-five environmental swab samples also were analyzed. The main bacterial groups were identified by cultural methods according to International Organization for Standardization standards. Listeria monocytogenes was enumerated with the most-probable-number protocol, and species identification was confirmed with a specific PCR assay. Immediately after vacuum packaging, all ready-to-eat (RTE) products had low mean aerobic colony counts (<10² to 2.4 × 10² CFU g^-1), anaerobic colony counts (1.6 to 6.5 × 10¹ CFU g^-1), Enterobacteriaceae counts (1.1 to 1.4 × 10¹ CFU g^-1), and Escherichia coli counts (generally below the detection limit). Nevertheless, the prevalence of L. monocytogenes in these samples was 3.7%. In roast beef samples, the aerobic and anaerobic colony counts reached unacceptable levels (>10⁶ CFU g^-1) after 14 days of refrigerated storage. Because the prevalence of L. monocytogenes increased to 13.3% during storage, a substantial reduction in the shelf life of these products is recommended. Surfaces without direct contact with food (floors and drains) had the highest mean counts for aerobic colonies (8.0 × 10³ to 9.5 × 10⁵ CFU/cm²), anaerobic colonies (2.9 × 10³ to 3.2 × 10⁴ CFU/cm²), Enterobacteriaceae (1.5 × 10¹ to 8.4 × 10¹ CFU/cm²), and E. coli (6.0 to 7.7 CFU/cm²). The levels of L. monocytogenes on direct food contact surfaces were below the detection limit, but more than 25% of floor samples were contaminated. These results reveal the persistence of L. monocytogenes in food processing environments, although at very low levels, posing a high risk of postcooking recontamination for RTE products. To improve hygienic conditions and reduce cross-contamination, an increase in operator awareness and a reassessment of surface sanitization protocols are needed.

New developments in the study of the microbiota of raw-milk, long-ripened cheeses by molecular methods: the case of Grana Padano and Parmigiano Reggiano

Microorganisms are an essential component of cheeses and play important roles during both cheese manufacture and ripening. Both starter and secondary flora modify the physical and chemical properties of cheese, contributing and reacting to changes that occur during the manufacture and ripening of cheese. As the composition of microbial population changes under the influence of continuous shifts in environmental conditions and microorganisms interactions during manufacturing and ripening, the characteristics of a given cheese depend also on microflora dynamics. The microbiota present in cheese is complex and its growth and activity represent the most important, but the least controllable steps. In the past, research in this area was dependent on classical microbiological techniques. However, culture-dependent methods are time-consuming and approaches that include a culturing step can lead to inaccuracies due to species present in low numbers or simply uncultivable. Therefore, they cannot be used as a unique tool to monitor community dynamics. For these reasons approaches to cheese microbiology had to change dramatically. To address this, in recent years the focus on the use of culture-independent methods based on the direct analysis of DNA (or RNA) has rapidly increased. Application of such techniques to the study of cheese microbiology represents a rapid, sound, reliable, and effective way for the detection and identification of the microorganisms present in dairy products, leading to major advances in understanding this complex microbial ecosystem and its impact on cheese ripening and quality. In this article, an overview on the recent advances in the use of molecular methods for thorough analysis of microbial communities in cheeses is given. Furthermore, applications of culture-independent approaches to study the microbiology of two important raw-milk, long-ripened cheeses such as Grana Padano and Parmigiano Reggiano, are presented.

The spatial distribution of bacteria in Grana-cheese during ripening

The microbial composition and its spatial distribution of Grana Trentino, a hard Parmesan-like cheese, was determined, from vat milk to cheese. After cutting along the vertical axis of the cheese wheels, three layers were sampled diagonally across the cheese: under the cheese rind, an intermediate section and the cheese core. After two different ripening periods (9 and 18 months), the cheese samples were analysed using traditional culture dependent and culture independent methods. Milk samples were dominated by mesophilic and psychrophilic bacterial counts. Thermophilic bacteria (Lactobacillus helveticus) were found in high amounts in cooked whey and natural whey starter cultures. After 9 months of ripening, lactic acid bacteria (LAB) counts were higher than those after 18 months. Furthermore, the LAB numbers in the cheese core was lower than those under the rind or in the intermediate section. The main LAB species isolated from milk (Lactococcus lactis, Pediococcus pentosaceus, Streptococcus uberis and Lactococcus garvieae) were not found in the corresponding cheeses. Some differences were observed in the species composition among the three cheese sections. Microbiota under the rind and in the intermediate section was similar and dominated by Lactobacillus paracasei and Lactobacillus rhamnosus. The core, after 18 months of ripening, was characterized by a total absence of LAB. In each sample, all LAB were genotypically grouped and the different biotypes were subjected to several technological tests indicating that some non-starter LAB (NSLAB) displayed technological features that are favorable for the production of Grana Trentino cheese.

Characterisation of Lactobacillus helveticus strains producing antihypertensive peptides by RAPD and inverse-PCR of IS elements

Lactobacillus helveticus is used for the manufacture of cheeses and milk-based products. Although it is not considered a probiotic microorganism, some strains demonstrated beneficial effects through the production of antihypertensive peptides from the hydrolysis of casein during milk fermentation. Strain-specificity of bioactive peptide production by L. helveticus makes the availability of reliable typing methods essential for both legal and good manufacturing processes. Accordingly, RAPD and inverse-PCR of five insertion sequence elements were comparatively evaluated for the molecular characterisation of four L. helveticus dairy cultures producing antihypertensive peptides and fourteen reference strains. Calculation of discriminatory indices and cluster analysis of the DNA fingerprints confirmed the suitability of both approaches for acceptable strain differentiation. Although RAPD was more discriminating, for a few test strains a neat discrimination was only achieved through multiplex inverse-PCR, thus suggesting the suitability of a combined analytical approach for a finer strain discrimination.

Polyphasic approach to bacterial dynamics during the ripening of Spanish farmhouse cheese, using culture-dependent and -independent methods

We studied the dynamics of the microbial population during ripening of Cueva de la Magahá cheese using a combination of classical and molecular techniques. Samples taken during ripening of this Spanish goat's milk cheese in which Lactococcus lactis and Streptococcus thermophilus were used as starter cultures were analyzed. All bacterial isolates were clustered by using randomly amplified polymorphic DNA (RAPD) and identified by 16S rRNA gene sequencing, species-specific PCR, and multiplex PCR. Our results indicate that the majority of the 225 strains isolated and enumerated on solid media during the ripening period were nonstarter lactic acid bacteria, and Lactobacillus paracasei was the most abundant species. Other Lactobacillus species, such as Lactobacillus plantarum and Lactobacillus parabuchneri, were also detected at the beginning and end of ripening, respectively. Non-lactic-acid bacteria, mainly Kocuria and Staphylococcus strains, were also detected at the end of the ripening period. Microbial community dynamics determined by temporal temperature gradient gel electrophoresis provided a more precise estimate of the distribution of bacteria and enabled us to detect Lactobacillus curvatus and the starter bacteria S. thermophilus and L. lactis, which were not isolated. Surprisingly, the bacterium most frequently found using culture-dependent analysis, L. paracasei, was scarcely detected by this molecular approach. Finally, we studied the composition of the lactobacilli and their evolution by using length heterogeneity PCR.

Culture-independent methods for identifying microbial communities in cheese

This review focuses on the culture-independent methods available for the description of both bacterial and fungal communities in cheese. Important steps of the culture-independent strategy, which relies on bulk DNA extraction from cheese and polymerase chain reaction (PCR) amplification of selected sequences, are discussed. We critically evaluate the identification techniques already used for monitoring microbial communities in cheese, including PCR-denaturing gradient gel electrophoresis (PCR-DGGE), PCR-temporal temperature gradient gel electrophoresis (PCR-TTGE) or single-strand conformation polymorphism-PCR (SSCP-PCR) as well as some other techniques that remain to be adapted to the study of cheese communities. Further, our analysis draws attention to the lack of data available on suitable DNA sequences for identifying fungal communities in cheese and proposes some potential DNA targets.

Characterization of dominant lactic acid bacteria isolated from São Jorge cheese, using biochemical and ribotyping methods

AIMS

To identify, using phenotypic and genotypic methods, the dominant lactic acid bacteria (LAB) present in São Jorge cheese - one of the 11 Portuguese cheeses currently bearing an Appéllation d'Origine Protegée status.

METHODS AND RESULTS

A total of 225 isolates from milk, curd and cheeses throughout ripening were identified to the genus level, 108 to the species level and ten to the strain level. Phenotypic methods indicated that lactobacilli, followed by enterococci, were the dominant bacteria. The most frequently isolated species were Lactobacillus paracasei, Lactobacillus rhamnosus, Enterococcus faecalis and Enterococcus faecium. Ribotyping differentiated three L. paracasei, two E. faecalis and one Lactobacillus plantarum types. Enterococcus spp. exhibited the highest esterase and beta-galactosidase activities among all isolates.

CONCLUSIONS

The dominant LAB in São Jorge cheese are L. paracasei, L. rhamnosus, E. faecalis and E. faecium. Enterococcus likely plays a leading role upon acidification and aroma development in said cheese.

SIGNIFICANCE AND IMPACT OF THE STUDY

Our results support that a combination of conventional biochemical methods with genotypic methods allows for a thorough characterization and identification of isolates. Despite the limited number of isolates subject to molecular subtyping, a few specific Enterococcus and Lactobacillus strains were found that are promising ones for development of a starter culture. Hence, L. paracasei and E. faecalis are good candidates for a tentative starter culture, designed for manufacturing of São Jorge cheese at large - which takes advantage of actual isolates, in attempts to eventually standardize the quality of said cheese variety.

Variability of bacterial biofilms of the "tina" wood vats used in the ragusano cheese-making process

Ragusano cheese is a "protected denomination of origin" cheese made in the Hyblean region of Sicily from raw milk using traditional wooden tools, without starter. To explore the Ragusano bacterial ecosystem, molecular fingerprinting was conducted at different times during the ripening and biofilms from the wooden vats called "tinas" were investigated. Raw milks collected at two farm sites, one on the mountain and one at sea level, were processed to produce Ragusano cheese. Raw milk, curd before and after cooking, curd at stretching time (cheese 0 time), and cheese samples (4 and 7 months) were analyzed by PCR-temporal temperature gel electrophoresis (PCR-TTGE) and by classical enumeration microbiology. With the use of universal primers, PCR-TTGE revealed many differences between the raw milk profiles, but also notable common bands identified as Streptococcus thermophilus, Lactobacillus lactis, Lactobacillus delbrueckii, and Enterococcus faecium. After the stretching, TTGE profiles revealed three to five dominant species only through the entire process of ripening. In the biofilms of the two tinas used, one to five species were detected, S. thermophilus being predominant in both. Biofilms from five other tinas were also analyzed by PCR-TTGE, PCR-denaturating gradient gel electrophoresis, specific PCR tests, and sequencing, confirming the predominance of lactic acid bacteria (S. thermophilus, L. lactis, and L. delbrueckii subsp. lactis) and the presence of a few high-GC-content species, like coryneform bacteria. The spontaneous acidification of raw milks before and after contact with the five tinas was followed in two independent experiments. The lag period before acidification can be up to 5 h, depending on the raw milk and the specific tina, highlighting the complexity of this natural inoculation system.

Isolation and molecular characterization of antibiotic-resistant lactic acid bacteria from poultry and swine meat products

The transfer via the food chain from animals to humans of microbes that are resistant to antimicrobial agents is of increasing concern. To determine the contributions of nonpathogenic microflora to the occurrence and spread of antibiotic resistance (AR) genes in the food chain, 123 lactic acid bacteria were isolated from 29 samples of raw and processed pork and chicken meat products that had previously tested positive for one or more AR genes that encode clinically relevant ARs: tet(M), tet(O), tet(K), erm(A), erm(B), erm(C), aac (6')-Ie aph (2")-Ia, mecA, and blaZ. All of the isolates were initially tested for their AR gene profiles by PCR. The 59 isolates carrying a tet, erm, or blaZ gene were taken through molecular identification, analyzed by determination of the MIC, and subjected to genetic fingerprinting. Lactococcus garvieae was the predominant species (28 isolates), followed by Lactobacillus plantarum (11 isolates) and L. salivarius (6 isolates), whereas Lactococcus lactis subsp. lactis, Lactobacillus johnsonii, L. reuteri, L. crispatus, and L. brevis were identified at lower frequencies. The tet(M) and erm(B) genes were the most frequently detected. Assessment of multiple resistances in 18 tet positive (tet+) isolates revealed that tet(M) plus erm(B) and tet(K) plus erm(B) were the most frequent AR gene patterns. Partial sequencing of the tet(M) open reading frame of three selected strains showed high sequence similarities (> 99%) with tet(M) genes previously found in human pathogens (Listeria monocytogenes and Neisseria meningitidis). Southern hybridization with plasmid profiles revealed these strains contained tet(M)-carrying plasmids.

Intraspecific diversity of Lactobacillus curvatus, Lactobacillus plantarum, Lactobacillus sakei, and Leuconostoc mesenteroides associated with vacuum-packed meat product spoilage analyzed by randomly amplified polymorphic DNA PCR

The intraspecific diversity of Leuconostoc mesenteroides, Lactobacillus curvatus, Lactobacillus sakei, and Lactobacillus plantarum was analyzed by randomly amplified polymorphic DNA (RAPD) PCR with universal primers M13 and T3. The study included 100 reference strains and 210 isolates recovered from two vacuum-packed Spanish meat products, fiambre de magro adobado and morcilla, previously identified by rDNA-restriction fragment length polymorphism profiles. The RAPD-M13 profiles identified isolates at species level in L. plantarum and L. mesenteroides, while RAPD-T3 provided profiles in L. sakei. The combination of RAPD-M13 and RAPD-T3 fingerprints revealed a total of 17 profiles in L. mesenteroides, 6 in L. sakei, 12 in L. plantarum, and 6 in L. curvatus. Of these, six profiles corresponding to L. mesenteroides and one corresponding to L. sakei were found in both products. The Shannon-Weaver diversity index (H'), calculated according to RAPD-M13 and RAPD-T3 profiles during storage, revealed that most profiles appeared only in single samplings in both products, indicating a high strain substitution rate during chilled storage of vacuum-packed meat products. When bloating appeared, only one profile corresponding to L. mesenteroides subsp. dextranicum was present throughout the storage period.