Molecular and physiological characterization of ... - Wiley Online Library

Journal of Applied Microbiology 1999, 87, 574–582

Molecular and physiological characterization of dominant bacterial populations in traditional Mozzarella cheese processing M. Morea1, F. Baruzzi1 and P.S. Cocconcelli2 1

Istituto Tossine e Micotossine da Parassiti Vegetali, Bari, 2Istituto di Microbiologia, Universita` Cattolica del Sacro Cuore, Piacenza, Italy 7120/03/99: received 16 March 1999, revised 28 May 1999 and accepted 8 June 1999

The development of the dominant bacterial populations during traditional Mozzarella cheese production was investigated using physiological analyses and molecular techniques for strain typing and taxonomic identification. Analysis of RAPD fingerprints revealed that the dominant bacterial community was composed of 25 different biotypes, and the sequence analysis of 16S rDNA demonstrated that the isolated strains belonged to Leuconostoc mesenteroides subsp. mesenteroides, Leuc. lactis, Streptococcus thermophilus, Strep. bovis, Strep. uberis, Lactococcus lactis subsp. lactis, L. garviae, Carnobacterium divergens, C. piscicola, Aerococcus viridans, Staphylococcus carnosus, Staph. epidermidis, Enterococcus faecalis, Ent. sulphureus and Enterococcus spp. The bacterial populations were characterized for their physiological properties. Two strains, belonging to Strep. thermophilus and L. lactis subsp. lactis, were the most acidifying; the L. lactis subsp. lactis strain was also proteolytic and eight strains were positive to citrate fermentation. Moreover, the molecular techniques allowed the identification of potential pathogens in a non-ripened cheese produced from raw milk. M . M OR E A, F. B AR UZ Z I A ND P .S . C O CC ON C EL LI . 1999.

INTRODUCTION

Large-scale industrial processes that relied on the use of selected starter cultures led to a low variability in the dairy microflora (Salama et al. 1993). However, some traditional dairy products are still fermented using unselected starters, thus obtaining a wide range of products with different flavours, consistencies and microbiological quality (Cogan et al. 1997). These products have been proposed as a source of new and interesting strains for use in food fermentation. The sanitation processes, such as milk pasteurization, which have a fundamental role in the control of pathogenic bacteria, also result in a significant reduction of the natural bacterial populations involved in the production of naturally fermented cheeses. Moreover, the importance of raw milk as a source of strains harbouring genetic diversity has recently been outlined in the traditional cheese produced without pasteurization (Corroler et al. 1998). Correspondence to: Dr Maria Morea, Istituto Tossine e Micotossine da Parassiti Vegetali, Viale L. Einaudi, 51, 70125 Bari, Italy (e-mail: [email protected]).

The need for new strains for the dairy industry, and for a deeper knowledge of the natural microflora present in typical dairy products, led to this study of the microflora involved in traditional Mozzarella cheese fermentation. The natural association of micro-organisms in Mozzarella cheese from water buffalo milk was investigated by Coppola et al. (1988), but little information was available on the bacterial community of traditional Mozzarella cheese made from raw bovine milk. The Lactobacillus community, already proved to be important in the production of Mozzarella cheese (Coppola et al. 1988, 1990; Mukherjee and Hutkins 1994; Yun et al. 1995), has recently been investigated in traditional Mozzarella cheese from raw whole bovine milk (Morea et al. 1998). Although 13 interesting strains of rod-shaped bacteria were found, they were not the dominant microflora. Therefore, to provide a basis for understanding the natural microbial associations, the natural community of dominant strains was studied. The physiological characterization of lactic acid bacteria from dairy products, such as acid production, proteolysis and citrate fermentation, have been studied in detail (Exterkate © 1999 The Society for Applied Microbiology

A NA LY S IS OF M OZ ZA R EL LA G RA M- P OS IT I VE BA C TE RI A 575

et al. 1993; Hugenholtz 1993; Klijn et al. 1995). In addition, rapid and reliable protocols based on the polymerase chain reaction (PCR) have been developed to obtain random amplified polymorphic DNA (RAPD) from lactic acid bacteria and enterococci (Cocconcelli et al. 1995; Tailliez et al. 1996; Erlandson and Batt 1997). Moreover, rDNA sequence analysis has been used for the taxonomic identification of lactic acid bacteria from dairy products (Cocconcelli et al. 1997). In order to study the ecology of natural bacterial communities involved in cheese fermentation, and to isolate new strains to be used as starter cultures, a pilot-scale production of Mozzarella was employed using raw milk and natural whey cultures. The dominant microflora involved in the traditional process was examined using physiological tests, RAPD analysis to type the dominant strains, and the sequence of 16S rDNA for taxonomic identification. This molecular approach was also used to provide further knowledge on the presence of potentially pathogenic bacteria in a non-ripened cheese produced without a pasteurization process. MATERIALS AND METHODS Sampling

For the purpose of this study, the Mozzarella cheese was produced using raw whole cow’s milk inoculated with 10% natural whey culture. These natural starter cultures were obtained by spontaneous fermentation, at 18 °C, of whey derived from the cheese production of the previous day. After rennet addition, the curd was maintained at 37 °C for about 1 h, and then stretched in hot water at 80 °C. During the stretching, the temperature of the curd increased to 60 °C. After shaping, the Mozzarella cheese was placed in cold water (10 °C) and stored at 4 °C for 24 h. Samples of acid whey, fermented curd before the stretching process, Mozzarella cheese after shaping (Mozzarella T-0) and Mozzarella cheese after 24 h of storage at 4 °C (Mozzarella T-24), were collected. Immediately after collection, the samples were frozen in liquid nitrogen and, once transferred to the laboratory, stored at −80 °C. Counts, isolation and maintenance of bacterial strains

The frozen samples were thawed, serially-diluted in sterile saline solution and plated on M17 agar (Difco) containing 0·5% lactose (LM17) in order to isolate presumptive Grampositive cocci. Before dilution, samples of Mozzarella T-0 and Mozzarella T-24 were ground in a food processor with a 2% sodium citrate solution (1 : 9 w/w). The plates were incubated at 30 °C and 42 °C under anaerobic conditions (Anaerocult A, Merck, Darmstadt, Germany) for 48 h.

For each collection step, 40 randomly selected colonies from M17 agar plates were isolated in a liquid medium, incubated overnight at appropriate temperatures, frozen at −80 °C and used for further characterization. DNA preparation

Genomic DNA for all PCR reactions was extracted from a single colony, resuspending the cells by vortexing for 30 s in 15 ml of a synthetic resin (Gene Releaser, Bioventures, TN, USA) specifically developed to purify DNA. The cell suspension, placed in a PCR tube, was overlaid with 50 ml mineral oil. Lysis was performed directly in the amplification tube by a microwave oven treatment. Strain typing

Strain typing was performed using RAPD as previously described (Cocconcelli et al. 1995). PCR amplification was carried out in a Thermal Cycler 480 (Perkin Elmer, Alameda, CA, USA). Taq polymerase, deoxynucleoside triphosphates and DNA molecular weight markers were purchased from Boehringer Mannheim, and the gel filtration purified oligonucleotides were obtained from Gibco BRL Life Technologies. Taxonomic identification

rDNA amplification and purification have already been described (Cocconcelli et al. 1997). The reaction sequences were obtained using an ABI PRISMTM dye terminator cycle sequencing kit (Perkin Elmer) and the primer P1 (S) [5?GCGGCGTGCCTAATACATGC-3?] position 41–60 (Klijn et al. 1995). The reaction products were analysed using an Applied Biosystem 310 automated DNA sequencer (Perkin Elmer). The software package of the University of Wisconsin Genetic Group was used for analysis and comparison of DNA sequences. Phylogenetic analyses and calculation of S—ab were performed with the RDP programme, as described by Maidak et al. (1997). In order to identify some Enterococcus strains not well determined by rDNA sequencing, acid production from carbohydrates was investigated as previously reported (Devriese et al. 1993) to ascertain the Enterococcus belonging group. All carbohydrates were obtained from Difco Laboratories. Physiological characterization of isolated strains

The strains were examined for acid production, and for proteolytic activity and citrate fermentation. The acidifying capacity was determined by pH measure-

© 1999 The Society for Applied Microbiology, Journal of Applied Microbiology 87, 574–582

576 M . M OR E A E T A L .

ments after 6 and 24 h of growth in skimmed milk at 30 and 42 °C, for mesophilic and thermophilic bacteria, respectively. The prtP gene, coding for a cell envelope proteinase (Vos et al. 1989; Exterkate et al. 1993), was investigated using specific primers in PCR reactions, as described by Klijn et al. (1995). Positive strains were tested for proteolytic activity in skimmed milk following the o-phthaldialdehyde method of Church et al. (1983). The reported values were calculated as the average of 10 readings at A340. Citrate utilization by isolated strains was detected using the medium of Kempler and McKay (1980). Citrate-positive colonies were blue after 48 h of incubation at appropriate temperatures under anaerobic conditions. In each test, L. lactis subsp. cremoris CNBL 1162 and L. lactis subsp. lactis CNBL 1163 were used as controls. RESULTS

1·16 × 106 cfu g−1. During stretching, when the water temperature was 80 °C, the bacterial community decreased as a result of the high temperature treatment. In Mozzarella T-0, mesophilic populations were 7·60 × 104 cfu g−1 and at 42 °C, 3·54 × 104 cfu g−1. After 24 h of storage at 4 °C, the mesophilic microflora decreased slightly to 3·50 × 104 cfu g−1 while the thermophilic bacteria showed a moderate growth (4·77 × 104 cfu g−1). RAPD analysis and study of population dynamics

The application of RAPD methodology enabled the growth kinetics of 25 different strains (Fig. 2) deriving from 320 colonies to be followed. In Fig. 3, the strain composition of the samples is shown. When the same strain was found in both mesophilic and thermophilic populations, the highest value was taken. In the acid whey after 24 h of fermentation, five different biotypes, A, 1St, 3Ct, 39v and E, were identified

Bacterial dynamics in the Mozzarella cheese

In Fig. 1, the growth dynamics of presumptive Gram-positive cocci in a traditional Mozzarella cheese-making process is shown. In the acid whey (pH 4·00) after 24 h of fermentation at room temperature, a strong prevalence of mesophiles (1·60 × 107 cfu ml−1) was detected compared with the thermophilic microflora (1·47 × 105 cfu ml−1). After milk heating, inoculum of whey, fermentation at 37 °C for 60 min and milk curdling, the mesophilic microflora remained steady (1·07 × 107 cfu g−1), while thermophiles increased to

8

log cfu g–1

7

6

5

4 Whey

Curd

T-0

T-24

Samples

Fig. 1 Presumptive Gram-positive cocci growth dynamics from Mozzarella sample cheese. Bacterial counts on M17 agar at 30 °C ( ) and 42 °C ( ). T-0, cheese after shaping; T-24, cheese after 24 h at 4 °C

Fig. 2 RAPD patterns of different biotypes of lactic acid bacteria. A: Lanes 1, 7 and 13: DNA molecular weight standards (pBR328 DNA Bgl I-Hinf I digested); 2, G; 3, 33c; 4, F; 5, 34c; 6, 39v; 8, 37c; 9, A; 10, 39z; 11, 51v; 12, 1Zt; 14, 1St; 15, 2Zt; 16, M; 17, 6Ct. B: Lanes 3 and 11: DNA molecular weight standards (pBR328 DNA Bgl I-Hinf I digested); 1, 2Ct; 2, H; 4, 3Ct; 5, 4Ct; 6, 5Ct; 7, 1Ct; 8, 22z; 9, 2Vt; 10, 1Vt; 12, E; 13, 24z



(a) 7

log cfu g

–1

6

5

4

3

2 1St

A

3Ct

39v

E

1Ct

H

(b) 7

log cfu g

–1

6

5

4

3

2 33c

34c

37c

F

G

2Ct

4Ct

5Ct

6Ct

(c) 7 T-24 T-0 curd whey

log cfu g

–1

6

5

4

3

2 39z

1Zt

22z

24z

1Vt

2Vt

51v

(Fig. 3a); they were present in similar amounts. In the microbial association of the curd, 14 different patterns were recognized (Fig. 3a, b); biotype G and biotype F constituted about 60% of the mesophilic microflora (Fig. 3b). Among the bio-

M

2Zt

Fig. 3 Community development of strain populations during Mozzarella cheese fermentation. The 25 different RAPD biotypes present in whey, curd, Mozzarella T-0 and T-24 are shown in (a), (b) and (c), respectively.

types isolated in the curd, A, 3Ct and 1St derived from whey. After thermal treatment, the number of different profiles decreased to 10 RAPD patterns; three (1St, H and 1Ct) were also present in the curd (Fig. 3a). After keeping the


578 M . M OR E A E T A L .

Mozzarella cheese at 4 °C for 24 h, nine biotypes were isolated; two of them, 51v and 1Vt, were new strains (Fig. 3c), 22z, 2Vt, M and 2Zt were also present in the T-0 sample (Fig. 3c), and biotypes 1St, H and 1Ct were found again. Taxonomic characterization of isolated strains

The taxonomic position of all 25 biotypes was determined by sequence analysis of at least 400 bp of the 5? region of the 16S rDNA gene. The S—ab values and current classification of strains from the samples of whey, curd and cheese, collected during the traditional Mozzarella cheese-making process, are given in Table 1. In whey, the strains were identified as Aerococcus viridans, Leuconostoc mesenteroides subsp. mesenteroides, Streptococcus thermophilus, Enterococcus faecalis and Staphylococcus epidermidis. The community in curd was composed of Carnobacterium divergens and C. piscicola, Leuc. mesenteroides subsp. mesenteroides, Lactococcus lactis subsp.

lactis, L. garviae, Strep. thermophilus, Strep. uberis, Ent. faecalis, Ent. sulphureus and Enterococcus spp. After stretching, the community was composed of Leuc. lactis, Strep. thermophilus, Strep. bovis, L. lactis subsp. lactis, Enterococcus spp. and Staphylococcus carnosus. The strains detected after 24 h at 4 °C were identified as Leuc. lactis, Strep. thermophilus, Strep. bovis, L. lactis subsp. lactis and Enterococcus spp. Strains 1Ct, 22z, 1Vt and 2Vt, not completely identified by 16S rDNA sequencing, were partially classified as reported in Materials and Methods. This analysis indicated that strains 1Ct, 1Vt and 22z belonged to the Ent. faecium species group, whereas 2Vt belonged to the Ent. gallinarum species group (Table 2). Characterization of isolated strains

Physiological features of the strains of interest for application in dairy processing, such as acid producing ability, proteolytic

Table 1 Taxonomic identification of strains isolated during Mozzarella cheese processing by means of partial 16S rDNA sequencing. The S—ab values were determined as described in Section 2 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– — ITEM* Isolated in no. Strain S—ab value† Reference in Ribosomal Database Project –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– — 500 G 0·833 Carnobacterium divergens Curd 501 33c 0·811 Carnobacterium divergens Curd 502 F 0·891 Carnobacterium piscicola Curd 503 34c 0·886 Carnobacterium piscicola Curd 504 39v 0·770 Aerococcus viridans Whey 505 A 0·941 Leuconostoc mesenteroides subsp. mesenteroides Whey, Curd 506 37c 0·842 Leuconostoc mesenteroides subsp. mesenteroides Curd 507 39z 0·795 Leuconostoc lactis T-0 508 51v 0·831 Leuconostoc lactis T-24 509 1Zt 0·829 Leuconostoc lactis T-0 510 1St 0·760 Streptococcus thermophilus Whey, Curd, T-0, T-24 511 2Zt 0·843 Streptococcus thermophilus T-0, T-24 512 M 0·818 Streptococcus bovis T-0, T-24 513 6Ct 0·987 Streptococcus uberis Curd 515 2Ct 0·720 Lactococcus lactis subsp. lactis Curd 514 H 0·923 Lactococcus lactis subsp. lactis Curd, T-0, T-24 516 4Ct 0·806 Lactococcus garviae Curd 517 3Ct 0·964 Enterococcus faecalis Whey, Curd 518 5Ct 0·780 Enterococcus sulphureus Curd 519 1Ct 0·801 Enterococcus sp. Curd, T-0, T-24 520 22z 0·802 Enterococcus sp. T-0, T-24 521 1Vt 0·804 Enterococcus sp. T-24 522 2Vt 0·805 Enterococcus sp. T-0, T-24 523 E 0·866 Staphylococcus epidermidis Whey 524 24z 0·886 Staphylococcus carnosus T-0 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– — * Istituto Tossine e Micotossine Bacterial Collection accession number. † The S—ab values were determined as described in Materials and Methods. © 1999 The Society for Applied Microbiology, Journal of Applied Microbiology 87, 574–582


Table 2 Acid production test for ascertaining the Enterococcus species group to which the strain belongs

— –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Strain ADO GLY MAN MLZ D-RAF SOR Species group — –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 1Ct – – – – – – Ent. faecium 1Vt – – – – – – Ent. faecium 22z – – ¦ – – – Ent. faecium 2Vt – ¦ ¦ ¦ – – Ent. gallinarum — –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Ado: adonitol; Gly: glycerol; Man: mannitol; Mlz: meleizitose; D-Raf: D-Raffinose; Sor: Sorbitol.

Table 3 Phenotypic and genotypic properties of strains isolated during Mozzarella cheese processing

— –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Strain pH (6 h) pH (24 h) prtP* Prt† Cit‡ — –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Carn. divergens G 6·50 6·47 – ND – Carn. divergens 33c 6·52 6·17 – ND ¦ Carn. piscicola F 6·46 6·04 – ND ¦ Carn. piscicola 34c 6·47 6·01 – ND – A. viridans 39v 6·59 6·47 – ND – Leuc. mes. subsp. mesenteroides A 6·44 5·71 – ND – Leuc. mes. subsp. mesenteroides 37c 6·44 5·84 ¦ 0·243 ¦ Leuc. lactis 39z 6·46 6·03 – ND ¦ Leuc. lactis 51v 6·46 5·96 ¦ 0·201 – Leuc. lactis 1Zt 6·63 6·44 ¦ 0·222 – Strep. thermophilus 1St 5·53 4·47 – ND – Strep. thermophilus 2Zt 5·46 4·25 – ND – Strep. bovis M 6·09 5·18 – ND – Strep. uberis 6Ct 6·04 5·50 – ND – L. lactis subsp. lactis 2Ct 5·90 4·94 – ND ¦ L. lactis subsp. lactis H 5·23 4·11 ¦ 0·558 – L. garviae 4Ct 6·24 5·62 – ND ¦ Ent. faecalis 3Ct 6·24 5·68 – ND ¦ Ent. sulphureus 5Ct 6·41 5·60 – ND ¦ Enterococcus. sp. 1Ct 6·11 5·02 – ND – Enterococcus sp. 22z 6·25 5·69 – ND – Enterococcus sp. 1Vt 6·19 5·46 – ND – Enterococcus sp. 2Vt 6·30 5·46 – ND – Staph. epidermidis E 6·50 6·23 – ND – Staph. carnosus 24z 6·50 5·94 – ND – L. lactis subsp. cremoris 1162 5·58 4·24 ¦ 0·415 – L. lactis subsp. lactis 1163 6·43 6·23 – 0·242 ¦ Milk Control 6·65 6·60 / 0·220 (30 °C) 0·228 (37 °C) — –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– * Presence of sequences related to prtP gene. † Proteolyic activity according to o-phtaldialdeyde (oPA) absorbance values at 340 nm; ND Not Determined. ‡ Capacity to metabolize citrate.

activity and citrate fermentation ability, were analysed (Table 3). The isolates were also tested for the presence in their genomes of DNA sequences related to the prtP gene coding for cell envelope proteinase (Klijn et al. 1995). Strains

H and 2Ct of L. lactis subsp. lactis, and strains 1St and 2Zt of Strep. thermophilus, were the most able to decrease pH values below 6·00 after 6 h of fermentation. These strains were also the most acidifying after 24 h when curdled milk


580 M . M OR E A E T A L .

could be observed in the test tubes. A 0·393 kb fragment containing sequences related to the prtP gene was found in four of the 25 strains. The positive strains were: two strains of Leuc. lactis, one strain of L. lactis subsp. lactis and one strain of Leuc. mesenteroides subsp. mesenteroides. The prtP¦ strains were analysed for their ability to degrade milk proteins; only L. lactis subsp. lactis H showed a significant proteolytic activity in milk. As expected, citrate fermenting ability was found in two strains of Lactococcus and Leuconostoc; in addition, two strains of Carnobacterium and Enterococcus were also able to utilize citrate.

DISCUSSION

Application of microbiological, molecular and physiological techniques allowed characterization of bacterial associations involved in the fermentation process of an artisanal Mozzarella cheese produced, using traditional techniques, from raw milk and a natural whey culture. The presumptive coccal microflora was higher than the lactobacilli community (Morea et al. 1998) in each sample from traditional Mozzarella cheese processing. Attention was therefore turned to studies of population dynamics using RAPD analysis in order to follow the dominant biotypes from natural whey cultures to Mozzarella cheese after 24 h at 4 °C; 25 different strains of Grampositive bacteria were found. Although other natural whey cultures for Registered Designation of Origin (RDO) cheeses (Cocconcelli et al. 1997) are prepared under thermophilic conditions, Mozzarella whey fermentation is driven by mesophilic strains, which comprise 99% of the total population, because of the low incubation temperature. A similar situation was observed in the curd, where the mesophilic microflora represented 90% of the total bacterial community. When RAPD was used to identify the different strains present in each sample, five strains were detected in the whey, whereas a higher number of strains, belonging to different species, were found in the curd. In the whey culture obtained from the spontaneous fermentation of whey at room temperature, A. viridans, Staph. epidermidis and Ent. faecalis were found in association with dairy strains Strep. thermophilus and Leuc. mesenteroides subsp. mesenteroides. A more complex bacterial community was observed in the curd, with 11 additional strains being detected. This could be explained by the growth of natural bacterial populations of raw whole milk when heated to 37 °C, confirming the relevance of mesophilic lactic cocci from raw milk for typical cheese production, as already reported (Corroler et al. 1998). The analysis of population dynamics revealed that the most important strain was Strep. thermophilus 1St, which derived from whey and was isolated in all the samples as one of the dominant populations. The role of the raw milk microflora

in cheese fermentation was demonstrated by the presence of L. lactis subsp. lactis H and Enterococcus sp. 1Ct, deriving from milk, in all the samples from curd to Mozzarella T-24. The remaining strains, found only in one or two samples, cannot be considered as important in the processing of this fermented food. All the strains belonging to Strep. thermophilus and L. lactis subsp. lactis showed higher acidifying activity when compared with the results obtained by analysis of other European artisanal dairy products (Cogan et al. 1997). The acidifying strains Strep. thermophilus 1St and L. lactis subsp. lactis H, whose colony counts in curd were high, could play an important role as they help to coagulate the curd. Although four prtP¦ strains were found, only L. lactis subsp. lactis H exhibited good proteolytic activity in milk. This could be explained by comparing the viable counts in milk. The counts on LM17 of strain H and L. lactis subsp. cremoris CNBL 1162 (prtP¦ control strain) were about 10 times higher than those of the other prtP¦ strains (data not shown). The presence of strains such as L. lactis subsp. lactis H confirms the importance of high quality raw milk in traditional cheese processing as a source of technological dairy strains, as previously reported (Corroler et al. 1998; Desmasures et al. 1998). The citrate-fermenting strains were present from whey to Mozzarella T-0; in the curd, about 30% of cells were citratefermenting and among these, two strains of Carnobacterium constituted 56%. The presence of these citrate-fermenting strains suggests that further research is needed in order to determine the influence of these populations on the organoleptic characteristics of Mozzarella cheese. Enterococci endowed with technological properties have been found in many artisanal cheeses (Jensen et al. 1975; Sua´rez et al. 1983; Fontecha et al. 1990; Cogan et al. 1997). In each sample from Mozzarella cheese-processing, Enterococcus strains were found. Two of these strains were citrate-fermenting, suggesting an important role in the microbial ecology of Mozzarella cheese fermentation. It was not possible to identify four of the Enterococcus strains by rDNA sequencing. These isolates are being studied to gain further knowledge of their taxonomic identification, as the results of physiological characterization were not in agreement with the data obtained from sequencing comparison, which confirms the presence of genetic diversity in the Enterococcus genus. During this study, the four Carnobacterium strains were found to constitute 70% of the curd population; two of them were citrate-fermenting. As C. divergens and C. piscicola are widely known as bacteriocin producers (Quadri et al. 1997; Holck et al. 1996), and their presence has been recently reported in dairy products (Coventry et al. 1997; Herbin et al. 1997), it would be interesting to determine their ecological role in the dairy natural ecosystem and in the organoleptic characteristics of traditional Mozzarella cheese.



The use of selected starters in large-scale industrial processes is causing a continuous decrease in natural microbial biodiversity. Traditional dairy products from unselected microflora and non-pasteurized milk are still the source of unknown strains that could be used to differentiate dairy products. Moreover, the application of molecular techniques has allowed the selection of interesting strains of lactic acid bacteria, within the complex microbial communities, and of potential pathogenic microbes present at lower levels and eventually associated with raw milk.

ACKNOWLEDGEMENTS

This work was supported by the EC structural funds FESR no. 94.05.09.013. The authors thank G. Stea for his technical assistance.

REFERENCES Church, F.C., Swaisgood, H.E., Porter, D.H. and Catignani, G.L. (1983) Spectrophotometric assay using o-phthaldialdehyde for determination of proteolysis in milk and isolated milk proteins. Journal of Dairy Science 66, 1219–1227. Cocconcelli, P.S., Parisi, M.G., Senini, L. and Bottazzi, V. (1997) Use of RAPD and 16S rDNA sequencing for the study of Lactobacillus population in natural whey culture. Letters in Applied Microbiology 25, 8–12. Cocconcelli, P.S., Porro, D., Galandini, S. and Senini, L. (1995) Development of RAPD protocol for typing of strains of lactic acid bacteria and enterococci. Letters in Applied Microbiology 21, 376–379. Cogan, T.M., Barbosa, M., Beuvier, E. et al. (1997) Characterization of the lactic acid bacteria in artisanal dairy products. Journal of Dairy Research 64, 409–421. Coppola, S., Parente, E., Dumontet, S. and Peccerella, A. (1988) The microflora of natural whey cultures utilized as starters in the manufacture of Mozzarella cheese from water-buffalo milk. Lait 68, 295–310. Coppola, S., Villani, F., Coppola, R. and Parente, E. (1990) Comparison of different starter systems for water-buffalo Mozzarella cheese manufacture. Lait 70, 411–423. Corroler, D., Mangin, I., Desmasures, N. and Gueguen, M. (1998) An ecological study of lactococci isolated from raw milk in the Camembert cheese Registered Designation of Origin area. Applied and Environmental Microbiology 64, 4729–4735. Coventry, M.J., Gordon, J.B., Wilcock, A. et al. (1997) Detection of bacteriocins of lactic acid bacteria isolated from foods and comparison with pediocin and nisin. Journal of Applied Microbiology 83, 248–258. Desmasures, N., Mangin, I., Corroler, D. and Gue´guen, M. (1998) Characterization of lactococci isolated from milk produced in the Camembert region of Normandy. Journal of Applied Microbiology 85, 999–1005.

Devriese, L.A., Pot, B. and Collins, M.D. (1993) Phenotypic identification of the genus Enterococcus and differentiation of phylogenetically distinct enterococcal species and species groups. Journal of Applied Bacteriology 75, 399–408. Erlandson, K. and Batt, C.A. (1997) Strain-specific differentiation of Lactococci in mixed starter culture population using randomly amplified polymorphic DNA derived probes. Applied and Environmental Microbiology 63, 2702–2707. Exterkate, F.A., Alting, A. and Bruinenberg, P.G. (1993) Diversity of cell envelope proteinase specificity among strains of Lactococcus lactis and its relationship to charge characteristics of the substratebinding region. Applied and Environmental Microbiology 59, 3640– 3647. Fontecha, J., Pelaéz, C., Jua´rez, M., Requena, T., Go´mez, C. and Ramos, M. (1990) Biochemical and microbiological characteristics of artisanal hard goat’s cheese. Journal of Dairy Science 73, 1150– 1157. Herbin, S., Mathieu, F., Brule, F., Branlant, C., Lefebvre, G. and Lebrihi, A. (1997) Characteristics and genetic determinants of bacteriocin activities produced by Carnobacterium piscicola CP5 isolated from cheese. Current Microbiology 35, 319–326. Holck, A., Axelsson, L. and Schillinger, U. (1996) Divergicin 750, a novel bacteriocin produced by Carnobacterium divergens 750. FEMS Microbiology Letters 136, 163–168. Hugenholtz, J. (1993) Citrate metabolism in lactic acid bacteria. FEMS Microbiology Reviews 12, 165–178. Jensen, J.P., Reinbold, G.W., Washam, C.J. and Vedamuthu, E.R. (1975) Role of enterococci in Cheddar cheese: organoleptic considerations. Journal of Milk and Food Technology 38, 142– 145. Kempler, G.M. and McKay, L.L. (1980) Improved medium for detection of citrate-fermenting Streptococcus lactis subsp. diacetylactis. Applied and Environmental Microbiology 39, 926– 927. Klijn, N., Weerkamp, A.H. and de Vos, W.M. (1995) Detection and characterization of lactose-utilizing Lactococcus subsp. in natural ecosystems. Applied and Environmental Microbiology 61, 788– 792. Maidak, B.L., Olsen, G.J., Larsen, N., Overbeek, R., McCaughey, M.J. and Woese, C.R. (1997) The RDP (Ribosomal Database Project). Nucleic Acids Research 25, 109–111. Morea, M., Baruzzi, F., Cappa, F. and Cocconcelli, P.S. (1998) Molecular characterization of the Lactobacillus community in traditional processing of Mozzarella cheese. International Journal of Food Microbiology 43, 53–60. Mukherjee, K.K. and Hutkins, R.W. (1994) Isolation of galactosefermenting thermophilic cultures and their use in the manufacture of low browning Mozzarella cheese. Journal of Dairy Science 77, 2839–2849. Quadri, L.E., Kleerebezem, M., Kuipers, O.P. et al. (1997) Characterization of a locus from Carnobacterium piscicola LV17B involved in bacteriocin production and immunity: evidence for global inducer-mediated transcriptional regulation. Journal of Bacteriology 179, 6163–6171. Salama, M.S., Sandine, W.E. and Giovannoni, S.J. (1993) Isolation of Lactococcus lactis subsp. cremoris from nature by colony hybridization with rRNA probes. Applied and Environmental Microbiology 59, 3941–3945.


582 M . M OR E A E T A L .

Sua´rez, J.A., Barneto, R. and In˜igo, B. (1983) Contribution to study of Mahoń cheese. III. Lactic acid bacteria and enterococci. Chemie Mikrobiologie Technologie der Lebensmittel 8, 52–56. Tailliez, P., Queńeé, P. and Chopin, A. (1996) Estimation de la diversite´ parmi les souches de la collection CNRZ: application de la RAPD a` un groupe de lactobacilles. Lait 76, 147–158. Vos, P., Simons, G., Siezen, R.J. and de Vos, W.M. (1989) Primary

structure and organization of the gene for a prokaryotic, cell envelope-located serine proteinase. Journal of Biological Chemistry 246, 13579–13585. Yun, J.J., Barbano, D.M., Kiely, L.J. and Kindstedt, P.S. (1995) Mozzarella cheese: impact of rod: coccus ratio on composition, proteolysis, and functional properties. Journal of Dairy Science 78, 751–760.
