Quantification of Enterococcus italicus in traditional Italian cheeses by fluorescence whole-cell hybridization

ARTICLE IN PRESS

Systematic and Applied Microbiology 31 (2008) 223–230 www.elsevier.de/syapm

Quantification of Enterococcus italicus in traditional Italian cheeses by fluorescence whole-cell hybridization Maria Emanuela Fornasari, Lia Rossetti, Chiara Remagni, Giorgio Giraffa Consiglio per la Ricerca e la Sperimentazione in Agricoltura, Centro di Ricerca per le Produzioni Foraggere e Lattiero-Casearie, Settore di Ricerca ‘‘Lattiero-Caseario’’, Via Lombardo 11, 26900 Lodi, Italy Received 28 February 2008

Abstract The objective of this work was to investigate the spread of Enterococcus italicus in cheese. For this purpose, a fluorescence whole-cell hybridization protocol (FWCH) with a 16S rRNA probe was optimized to evaluate the presence and abundance of this organism in artisanal Italian cheeses. The FWCH method avoided the quantification problems using classical plate count techniques related to the well-known difficulties to cultivate E. italicus in selective enterococci media. After probe and FWCH optimization, 10 commercially available Italian semi-hard cheeses made with raw ewe or cow milk without starter addition were analyzed. All of them were subjected to FWCH experiments and six of them gave positive results with the probe, i.e. the E. italicus content was 44 log cells/g according to the detection limit of FWCH. Counts showed that E. italicus was present at levels ranging from 5.9170.17 to 7.3470.14 log cells/g; such levels were similar to, or even higher than, the total enterococci counted from the corresponding cheeses using kanamycin aesculin azide agar. The overall reliability of the FWCH method was tested by species-specific PCR. The positive amplification of the expected 323 bp fragment from both a cheese matrix and cell bulks of cheese samples containing high loads of this organism (as determined by FWCH counts) and the successful isolation of E. italicus strains from the above cheeses provided definitive proof of both probe specificity and the presence of this organism in cheeses. Although there is very little available quantitative data on the incidence of E. italicus in cheese, or its role in product quality, this study showed a wide diffusion of this organism in artisanal cheeses, where secondary non-starter lactic acid bacterial microflora, which enterococci belong to, may become dominant during ripening. r 2008 Elsevier GmbH. All rights reserved. Keywords: Enterococcus italicus; (Fluorescence) Whole-cell hybridization; In situ microbial analysis; Species-specific PCR; Traditional cheeses

Introduction The genus Enterococcus (or enterococci) consists of homofermentative and facultatively anaerobic bacteria Corresponding author. Tel.: +39 0371 45011; fax: +39 0371 35579.

E-mail address: [email protected] (G. Giraffa). 0723-2020/$ - see front matter r 2008 Elsevier GmbH. All rights reserved. doi:10.1016/j.syapm.2008.04.002

that are widespread in nature, and present in dairy and other fermented foods. Enterococci are adventitious non-starter lactic acid bacteria (NSLAB) that are able to grow in cheese after manufacturing. They originate from milk and, although pasteurization usually inactivates them, recontamination from the manufacturing environment may easily occur. Therefore, these bacteria are

ARTICLE IN PRESS 224

M.E. Fornasari et al. / Systematic and Applied Microbiology 31 (2008) 223–230

expected to be present at high levels in cheeses derived from both raw and pasteurized milk [19]. A relatively limited number of species are able to survive adverse cheese process conditions, such as extreme pH, temperature, and salinity. Since enterococci form a significant portion of the NSLAB of most cheese varieties during ripening, it is believed that they can contribute positively to flavor development during such ripening. This assumption is corroborated by several reports describing certain functional properties associated to enterococci, such as acidifying, proteolytic, and esterase activities, citrate utilization, and production of aromatic volatile compounds. Moreover, enterococci are capable of producing bacteriocins with activity against pathogenic bacteria such as Listeria monocytogenes and Staphylococcus aureus [16,17,19,20]. Enterococci have also been recognized as emerging nosocomial pathogens, since several virulence factors causing infections such as bacteremia and endocarditis have been described [18,24]. Although a balanced view on both the beneficial and negative traits of enterococci has been presented in several reviews [16,17,20], the genus Enterococcus remains the most controversial group of lactic acid bacteria (LAB) and the acceptability of its presence in fermented food is still a matter of debate. This task is further complicated by both taxonomic and technical problems. At the moment, 31 species are validly published within the genus Enterococcus. However, not all the recently described species meet the physiological and biochemical traits of the typical enterococci. Consequently, a plethora of techniques and modifications of selective media have been reported to solve frequently encountered isolation and quantification problems for enterococci [9]. Jackson et al. [22] showed that media type, incubation temperature, and enrichment period influenced recovery, selection, and antibiotic susceptibility patterns of enterococcal species. A number of ‘in situ’ methods have been introduced to study microbial communities [2]. The common trait of these methods is that morphologically intact cells (both cultivable and uncultivable) can be directly identified and counted. It is generally accepted that ‘in situ hybridization’ is restricted to hybridization in which viable cells are detected within their natural microhabitat. When organisms have been taken from a habitat or grown in laboratory media, the term ‘whole cell’ rather than ‘in situ’ hybridization is preferred [2,32]. Fluorescence in situ hybridization (FISH) or whole-cell hybridization (WCH) with rRNA probes are widely applied in microbial ecology, thus providing microbial identification, physical detection of uncultivable microorganisms, and distribution of microbial populations in several environments, including food products. FISH has been used to evaluate bacterial community structure and location in Stilton cheese [11,12], to detect brevibacteria on the surface of Gruye`re cheese [25], to

detect Lactobacillus plantarum on natural fermented olives [13], and to quantify Leuconostoc populations in mixed dairy starter cultures [29]. In a previous study, the biosafety of Enterococcus italicus, a recently emerged dairy-associated enterococcal species, was investigated [26]. Although no virulence determinants were detected, the finding of tetracyclineresistant strains prompted us to consider the dissemination and quantification of E. italicus in cheeses. Consequently, in this study, a fluorescence WCH (FWCH) protocol was optimized to evaluate the presence and abundance of E. italicus in artisanal Italian cheeses.

Materials and methods Bacterial strains and growth conditions E. italicus LMG 22039T and ILC 1189, Enterococcus saccharolyticus LMG 11427T, Enterococcus sulfureus LMG 13084T, Enterococcus faecium ATCC 19434T, Enterococcus faecalis ATCC 19433T, Enterococcus durans ATCC 19432T, and Enterococcus casseliflavus LMG 10745T were used as reference strains to optimize the FWCH protocol. Reference strains and the other cheese isolates (see later) were maintained as frozen stocks at 80 1C in the presence of 150 ml/l glycerol as a cryoprotective agent. M17 broth (Merck, Darmstadt, Germany) was used to reactivate bacteria after overnight growth at 37 1C.

Cheese sampling, microbial counts, and cell harvesting for bulk preparation Ten samples of commercially available Italian semihard cheeses made with raw ewe or cow milk without starter addition were analyzed (Table 1). Cheese samples (10 g) were homogenized with 90 ml of a sterile 2% sodium citrate solution (pH 7.5) by means of a Stomacher 400 Circulator (Seward Laboratory, London, England). Decimal dilutions were prepared in Ringer solution and pour-plated in duplicate for microbial counts with the following media: M17 agar (Merck) at 37 1C for 48 h for total streptococci; MRS agar (Scharlau Chemie, Barcelona, Spain) at 37 1C for 48 h under anaerobic conditions for total lactobacilli; kanamycin aesculin azide agar (KAA; Merck) at 37 1C for 24 h for enterococci. Bulk formation was performed according to Ercolini et al. [10] from all the duplicated M17 agar plates from the serial dilution 4 to the last.

DNA extraction from isolates, bulk cells and cheese Cheese suspensions in sodium citrate buffer were centrifuged at 8000g for 15 min at 4 1C and, after

ARTICLE IN PRESS M.E. Fornasari et al. / Systematic and Applied Microbiology 31 (2008) 223–230

Table 1.

225

Microbial analysis of the 10 samples of commercially available Italian semi-hard cheeses

Cheese sample (commercial denomination)

Total lactobacilli (MRS agar, 37 1C) (log cfu/g)

Total streptococci (M17 agar, 37 1C) (log cfu/g)

Total enterococci (KAA agar, 37 1C) (log cfu/g)

E. italicus (microscopy counts)a (log cells/g)

V1 (Vastedda cheese) V2 (Vastedda cheese) V3 (Vastedda cheese) P1 (Pecorino cheese) C2 (Cofanaro cheese) E1 (Ericino cheese) E2 (Ericino cheese) E4 (Ericino cheese) P3 (Pecorino cheese) T3 (Tuma Pantesca cheese)

7.9970.04 8.2570.03 7.5470.04 o6.00 7.6170.09 7.8370.07 7.5370.19 7.6970.01 8.0870.09 8.3470.06

8.4470.08 8.4270.14 7.5970.09 8.9170.05 7.2170.05 7.2470.08 7.7470.01 7.6870.01 8.1570.10 8.6470.06

6.9770.06 5.7770.15 5.8670.14 7.0570.08 6.1170.21 6.0170.03 7.5270.01 7.2870.05 7.7670.03 6.2070.12

5.9970.28 6.3270.24 6.7670.16 6.6670.06 o5.00b o5.00b 5.9170.17 o5.00b o5.00b 7.3470.14

Mean values (7 standard deviation) are indicated. a By epifluorescence microscopy after fluorescence whole-cell hybridization (FWCH) using ESA452 [5] as the DNA probe. b 4104 cells/g was the lowest detection limit of the FWCH method (see results).

repeated washings (3–5 times) with the same buffer to eliminate residual casein particles and milk fat, pellets were dissolved in 10 ml of sterile distilled water. Total DNA was extracted from 100 ml resuspended pellets, according to the protocol described by the Wizard DNA purification kit (Promega Italia, Milano, Italy). Total DNA from bulk suspensions (100 ml) and colonies picked from M17 agar plates used to enumerate total streptococci was extracted by a chelex-based method according to the procedure for Gram-positive (and acidfast) bacteria described in the MicroSeqTM protocol (Applera Italia, Monza, Italy).

PCR identification Total DNAs extracted from cheese, bulk suspensions, and streptococcal isolates were used to amplify an intragenic fragment of 323 bp from the 16S rRNA gene of E. italicus using species-specific PCR primers with the sequences: 50 TACCGCATAATACTTTTTCTCT 30 and 50 GTCAAGGGATGAACATTCTCT 30 [15]. Amplifications were performed in 25 ml volumes with 2 mM of each primer (Biotez, Berlin, Germany), 0.5 units of AmpliTaq polymerase (Applera Italia), 2 mM MgCl2, dNTP mixture (200 mM each), and 1 ml of total DNA. Reactions were carried out in a Perkin Elmer thermal cycler (model 9700; Applied Biosystems). Amplification conditions were as follows: 94 1C for 2 min, and 25 cycles of 94 1C for 30 s, 55 1C for 30 s, and 72 1C for 40 s. The final extension was at 72 1C for 10 min. PCR products were analyzed by electrophoresis through 1% (w/v) agarose gels at 10 V/cm for 1 h in Tris-acetate EDTA (TAE) buffer (TAE: 40 mM Tris acetate, 1 mM EDTA, pH 8.0). E. italicus LMG 22039T was used as a positive control.

Enumeration of E. italicus in cheese by FWCH Cell fixation and permeabilization Cheese pellets treated as described above to extract total DNA were dissolved in 10 ml of phosphate buffered saline (PBS) (130 mM NaCl, 10 mM sodium phosphate buffer, pH 7.4). Cells were fixed according to Amann et al. [1]. Briefly, 1 volume of cell suspension was mixed with 3 volumes of freshly prepared cold paraformaldehyde solution (4% in PBS). The mixture was homogenized by vortexing and then incubated at 4 1C for 16 h. After this step, cells were centrifuged (12,000g, 2 min, 4 1C), washed with PBS and resuspended in an equal volume of the same buffer. Fixed cells were mixed with 1 volume of cold absolute ethanol and stored at 20 1C until use. Cells were permeabilized according to Blasco et al. [7] but with minor modifications. Fixed cells were spotted (125 ml) on a 125 ml Geneframe (AB-0578; Abgene, Epsom, United Kingdom), immobilized by successive transfers (3 min each) in 50%, 80%, and 100% cold ethanol, and left to dry in air. Slides were then covered with 300 ml lysis solution (Sigma Aldrich, Milano, Italy) and incubated at room temperature for 10 min. The enzymatic treatment was stopped by rinsing with 100 ml TE buffer and the slides were left to dry in the air.

Probe design Fluorescence whole-cell hybridization (FWCH) experiments were carried out by using the 16S rRNA probe ESA452, previously designed by Behr et al. [5]. The sequence of the probe is 50 CATTCTCTTCTCATCCTT 30 and corresponds to positions 453–470 of the Escherichia coli 16S rRNA gene. The probe was originally designed to be specific for E. saccharolyticus.

ARTICLE IN PRESS 226

M.E. Fornasari et al. / Systematic and Applied Microbiology 31 (2008) 223–230

The specificity of the probe was subsequently checked in the NCBI database using the BlastN algorithm (http:// www.ncbi.nlm.nih.gov/blast/Blast.cgi). Duplex FWCH reactions with ESA452 and the universal Eub338 probe 50 GCTGCCTCCCGTAGGAGT 30 targeting Eubacteria [1] were carried out. The ESA452 and Eub338 probes were labelled in the 50 position with Cy3 and fluorescein isothiocyanate (FITC), respectively. Both probes were oligosynthesized and labelled by Biotez. Hybridization conditions Hybridization experiments were carried out according to the method of Ercolini et al. [11] but with minor modifications. Geneframe slides were transferred to a dark humid room and each well was covered with 300 ml preheated hybridization buffer (0.9 M NaCl, 0.01% SDS, 20 mM Tris–HCl pH 8, 10% formamide, 0.5 pmol ml1 of each labelled probe) and incubated for 5 h at 42 1C. After hybridization, unbound probes were removed from the slide by washing twice at 42 1C for 15 min with pre-warmed washing solution (40 mM NaCl, 0.01% SDS, 5 mM EDTA, 20 mM Tris–HCl at pH 8). Slides were then rinsed with distilled water, air dried, and microscopically observed. Microscopy Bacterial cells were examined under an Axioskop 40 FL microscope (Carl Zeiss S.p.A., Milano, Italy) equipped with a HBO 50 W mercury lamp. Carl Zeiss filter set 20 was used for Cy3 (excitation wavelength, 546/ 12 nm; emission wavelength, 575–640 nm). Carl Zeiss filter set 10 was used for FITC (excitation wavelength, 450–490 nm; emission wavelength, 515–565 nm). At least 10 fields with approximately 50 cells in each were counted for each cheese at a magnification of  100. For each field, the probe-conferred fluorescent cells were counted (Cy3 and FITC) and the resulting values were averaged. To express counts as cells/g, the average values were multiplied by 6.3892  105  a dilution factor to account for the area of the Geneframe (476 mm2), the area of each field (0.005,963 mm2), the volume of the sample deposited onto the Geneframe (125 ml), and the sample dilution. Axiovision software (version 4.1, Carl Zeiss) was used for image analysis and cell counts. The software allowed separated images of Cy3- and FITC-bound cells to be superimposed.

Results and discussion Probe specificity BlastN sequence alignment of the ESA452 probe targeting the 16S rRNA gene showed 100% identity with E. saccharolyticus NCDO 2594T, E. sulfureus A36202, E. italicus TP1.5T( ¼ LMG 22039T), and

E. saccharominimus LMG 21727T. This was not surprising because E. saccharominimus, which is a later synonym of E. italicus [28], E. sulfureus, and E. saccharolyticus constitute a separate phylogenetic species group on the basis of the 16S rRNA gene sequence [27]. No other known bacterial species matched with the ESA452 sequence after BlastN analysis (data not shown). Probe specificity was also tested by separate FWCH experiments from overnight M17 cultures of E. italicus LMG 22039T, E. saccharolyticus LMG 11427T, E. sulfureus LMG 13084T, E. faecium ATCC 19434T, E. faecalis ATCC 19433T, E. durans ATCC 19432T, and E. casseliflavus LMG 10745T using Eub338 and ESA452 as probes. All strains gave positive hybridization with the universal Eub338, whereas only E. italicus LMG 22039T (shown as an example in Fig. 1, panel A) and E. saccharolyticus LMG 11427T reacted positively with ESA452. This result was not surprising either because the ESA452 probe was shown to be effective for the specific detection of E. saccharolyticus [5]. Importantly, no cross-hybridization took place with E. faecium, E. faecalis, and E. durans, which are the most commonly found enterococci in cheese [16,17,19,23,31].

Cell target accessibility To evaluate possible interferences of the cheese matrix, a M17 broth culture of E. italicus LMG 22039T was concentrated by centrifugation and, after 1 resuspension in a 100 volume of Ringer solution, cells were added to cheese sample P3 to a level (approximately 9 log cfu/g of cheese) designed to overrate total endogenous streptococci (8.1570.10 log cfu/g; Table 1). After cell separation and fixation, coccus-shaped cells were enumerated by Eub338- and ESA452-labelled probes in separate hybridization experiments and resulting counts were compared to M17 plate enumeration. Eub338 and ESA452 counts were not significantly different (9.2170.08 and 9.0170.15 log cells/g, respectively) and were comparable to the total streptococci counted in M17 (9.1470.13 log cfu/g). When serial dilutions of the concentrated cell suspension of LMG 22039T were added to the cheese sample and cells were counted by the ESA452 probe, the minimum detectable potential load of E. italicus was approximately 105 cells/ g, i.e. the detection limit of the method was 44 log cells/ g of cheese (data not shown). On the basis of these results, the ESA452 probe can be considered valid for detecting the species E. italicus by FWCH when present in cheese at levels 44 log cells/g, without significant interference on target cell accessibility and count efficiency possibly caused by the cheese matrix.

ARTICLE IN PRESS M.E. Fornasari et al. / Systematic and Applied Microbiology 31 (2008) 223–230

227

imposition, the Cy3-bound, pink-coloured E. italicus cells in sample V3 were visible against the background on FITC-bound, green-coloured coccus- and rodshaped cells (Fig. 1, panel B). This was not surprising since both total streptococci and lactobacilli, counted on M17 and MRS agar, respectively, were 1–2 logs higher than E. italicus in all cheeses (Table 1). Counts showed that E. italicus was present at levels ranging from 5.9170.17 to 7.3470.14 log cells/g, and such levels were similar to, or even higher than (e.g. in cheese samples V2, V3, and T3), the total enterococci counted from the corresponding cheeses on KAA (Table 1). It is worth noting that FWCH and plate counts are not fully comparable since the latter can only detect the cultivable cell fraction. However, given the difficulty to cultivate and enumerate E. italicus on KAA, or similar selective media for enterococci [14,33], the comparison between FWCH with the ESA452 probe and KAA counts gave a helpful indication of the relatively high proportion of this organism within the cultivable enterococcal population. The comparison between E. italicus, enumerated by FWCH, and total streptococci and lactobacilli, enumerated at 37 1C in M17 and MRS, respectively, was also highly informative. Since counts of E. italicus (when present) accounted for 1–10% of both streptococci and lactobacilli populations (Table 1), it may be concluded that this organism often belongs to the subdominant bacterial species within the LAB community.

Validation of Enterococcus italicus enumeration Fig. 1. Enumeration of Enterococcus italicus from an overnight M17 culture of strain LMG 22039T (panel A) and in cheese sample V3 (panel B) detected by fluorescence whole-cell hybridization with the ESA452 probe labelled with Cy3 (pinkcolored cells). Total eubacteria in the cheese sample were detected by the EU338 probe labelled with FITC (greencolored cells). Bars in each panel ¼ 10 mm.

Enumeration of Enterococcus italicus in cheese by FWCH The validated probe was applied by FWCH to search for the presence of E. italicus in cheese. Ten commercially available Italian semi-hard cheeses made with raw ewe or cow milk without starter addition were analyzed. All the cheeses were subjected to FWCH experiments and six of them gave positive results with probe ESA452, i.e. the E. italicus content was 44 log cells/g, according to the detection limit of FWCH. Fig. 1 shows a duplex FWCH experiment of the V3 cheese sample, where E. italicus was present at considerable levels (6.7670.16 log cells/g; Table 1). After image super-

The overall reliability of the FWCH method for E. italicus detection and quantification was tested. As a first approach, total DNA extracted from the six E. italicus positive cheeses was amplified by speciesspecific PCR, according to Fortina et al. [15]. The expected amplicon of 323 bp, specific for E. italicus, was obtained from all tested cheeses (Fig. 2; cheeses V1, V2, and V3 are shown as examples). This prompted the bulk collection of the microbial colonies from serial dilutions 5 and 6 (5, 6 and 7 for cheese sample T3) of the M17 agar plates of the six E. italicus positive cheeses (Table 1). The DNA was then extracted from the collected colonies, as described in the Materials and methods, and tested again by PCR for the presence of E. italicus. Generally, the expected PCR product was observed from all six cheeses at all dilutions (Fig. 2; cheeses V2, V3, T3, and P1 are shown). This experiment confirmed the overall loads of E. italicus as detected by WFCH. With the aim of verifying the specificity of the ESA452 probe in practical conditions, all the colonies from the last two dilutions of the M17 plate counts for P1 and T3 cheeses were isolated. After strain

ARTICLE IN PRESS 228

M.E. Fornasari et al. / Systematic and Applied Microbiology 31 (2008) 223–230

Fig. 2. Species-specific PCR for Enterococcus italicus. The expected amplicon of 323 bp is shown after amplification of total DNA extracted from: cheeses V1, V2, and V3 (lanes 4, 5, and 6); bulk-collected microbial colonies from M17 agar plates from serial dilutions 5 and 6 (5, 6 and 7 for cheese sample T3) of the E. italicus positive cheeses V2 (lanes 9 and 10), V3 (lanes 11 and 12), T3 (lanes 13, 14, and 15), and P1 (lanes 16 and 17); the strain E. italicus 38, isolated from cheese P1 (lane 18). Lanes 2 and 3: PCR amplification from E. italicus LMG 22039T and ILC 1189 used as PCR positive controls; lanes 7 and 8, cheeses V1 and V2, respectively, added with LMG 22039T and used as controls of positive amplification from the cheese matrix; lanes 19 and 20, negative controls (no DNA added); lanes 1 and 21, molecular size DNA marker (sizing range: 100–12,000 bp; 1 kb Plus DNA ladder; Invitrogen Italia, Milan, Italy).

purification and total DNA extraction, some isolates were shown to belong to E. italicus after species-specific PCR (Fig. 2; lane 18 is shown as an example). E. italicus is difficult to isolate and enumerate using selective media for enterococci because it does not tolerate 6.5% NaCl and 0.4% sodium azide [14,33]. This made it necessary to optimize a FWCH protocol to identify and quantify this organism rapidly in cheese and, to this end, a probe (ESA452) originally designed to detect E. saccharolyticus [5] was selected. It is important to point out that E. saccharolyticus has been isolated from maize- [3,6] and cassava-based fermented foods [4], but it has never been recovered from dairy products [8]. This prompted us to apply ESA452 to detect E. italicus in cheeses. The ESA452 probe specificity for E. italicus was tested in cheese and verified by species-specific PCR of bulk-collected colonies from M17 agar plates used to enumerate total streptococci. The amplification of the expected fragment of 323 bp from both the cheese matrix and cell bulks of cheese samples containing high loads of this organism (as determined by FWCH counts) and the successful isolation of E. italicus strains from the above cheeses

provided definitive proof for both probe specificity and the presence of this organism in cheeses. This study confirmed once more that enterococci form part of the LAB and are of great importance in dairy products. In the present study, their loads in cheeses agreed with the levels reported in the literature, which range from 105 to 107 cfu/g. Cheeses may contain a number of different species of enterococci, but E. faecium and E. faecalis are the species most commonly isolated [16,17,19,23,31]. However, the introduction of highly performing culture-independent techniques is helping to increase our knowledge about enterococcal diversity in food. In particular, in situ methods such as FISH or FWCH, are helpful for quantifying non-cultivable and/or non-dominant species in dairy ecosystems [21]. E. italicus/E. saccharominimus was originally isolated from two artisanal Italian cheeses, Toma piemontese and Robiola piemontese [14], as well as from Belgian, Moroccan, and Romanian dairy products [33] and, more recently, from Italian Pannerone and Slovak Bryndza cheeses [23,26]. Although there is little quantitative data on the incidence of E. italicus in cheese or its role in product quality, the present investigation highlighted a wide and quantitatively high diffusion of this organism in artisanal cheeses, where secondary NSLAB microflora, which enterococci belong to, may become dominant during ripening. It is possible that, until now, the presence of E. italicus in cheeses had been underestimated because of the aforementioned difficulties to cultivate this bacterium on classical enterococcal media. In a previous study, the safety of 30 E. italicus strains isolated from dairy products (especially cheeses) was investigated [26]. None of the strains showed virulence determinants by PCR and were generally susceptible to antibiotics. However, the finding of tetracyclineresistant strains in that study suggests that safety concerns may arise from the frequent presence of this bacterium within the (sub)dominant LAB community of many cheeses, as shown by the present investigation. We provided a culture-independent method based on WCH to evaluate the diffusion of E. italicus in cheeses, and this could be the basis on which to develop similar procedures for applying to other fermented foods. In this regard, more sensitive FISH-based protocols, such as catalyzed reporter deposition (CARD)-FISH, are presently available that yield significantly higher signal intensities than FISH with fluorescently monolabelled oligonucleotide probes [30]. Although we are aware of the low sensitivity in cheese (44 log cells/g) of the FWCH protocol provided here, the method may, however, be useful to detect E. italicus rapidly when present at (sub)dominant levels within the NSLAB microflora of cheeses, thus contributing to a better understanding of the ecology and role of this organism in these products.

ARTICLE IN PRESS M.E. Fornasari et al. / Systematic and Applied Microbiology 31 (2008) 223–230

Acknowledgments The authors are grateful to Dr. D. Carminati for critical reading of the manuscript. This work was partially supported by the Agriculture Research Service of the Sicily Region (Italy), through the research programme VALFORT.

References [1] R.I. Amann, B.J. Binder, R.J. Olson, S.W. Chisholm, R. Devereux, D.A. Stahl, Combination of 16S rRNAtargeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations, Appl. Environ. Microbiol. 56 (1990) 1919–1925. [2] R.I. Amann, W. Ludwig, K.H. Schleifer, Phylogenetic identification and in situ detection of individual microbial cells without cultivation, Microbiol. Rev. 59 (1995) 143–169. [3] F. Ampe, N. ben Omar, C. Moizan, C. Wacher, J.P. Guyot, Polyphasic study of the spatial distribution of microorganisms in Mexican Pozol a fermented maize dough, demonstrates the need for cultivation-independent methods to investigate traditional fermentations, Appl. Environ. Microbiol. 65 (1999) 5464–5473. [4] F. Ampe, A. Sirvent, N. Zakhia, Dynamics of the microbial community responsible for traditional sour cassava starch fermentation studied by denaturing gel gradient electrophoresis and quantitative rRNA hybridisation, Int. J. Food Microbiol. 65 (2001) 45–54. [5] T. Behr, C. Koob, M. Schedl, A. Mehlen, H. Meier, D. Knopp, E. Frahm, U. Obst, K.H. Schleifer, R. Niessner, W. Ludwig, A nested array of rRNA targeted probes for the detection and identification of enterococci by reverse hybridisation, Syst. Appl. Microbiol. 23 (2000) 563–572. [6] N. ben Omar, F. Ampe, Microbial community dynamics during production of the fermented maize dough pozol, Appl. Environ. Microbiol. 66 (2000) 3664–3673. [7] L. Blasco, S. Ferrer, I. Pardo, Development of specific fluorescent oligonucleotide probes for in situ identification of wine lactic acid bacteria, FEMS Microbiol. Lett. 225 (2003) 115–123. [8] S. Coppola, G. Blaiotta, D. Ercolini, Dairy products, in: L. Cocolin, D. Ercolini (Eds.), Molecular Techniques in the Microbial Ecology of Fermented Foods, Springer Science, New York, 2008, pp. 31–90 (Chapter 2). [9] K.J. Domig, H.K. Mayer, W. Kneifel, Methods used for the isolation, enumeration, characterisation and identification of Enterococcus spp. 2. Pheno- and genotypic criteria, Int. J. Food Microbiol. 88 (2003) 165–188. [10] D. Ercolini, G. Moschetti, G. Blaiotta, S. Coppola, The potential of a polyphasic PCR-DGGE approach in evaluating microbial diversity of natural whey cultures for water-buffalo mozzarella cheese production: bias of culture-dependent and culture independent analyses, Syst. Appl. Microbiol. 24 (2001) 610–617. [11] D. Ercolini, P.J. Hill, C.E.R. Hodd, Bacterial community structure and location in Stilton cheese, Appl. Environ. Microbiol. 69 (2003) 3540–3548.

229

[12] D. Ercolini, P.J. Hill, C.E.R. Dodd, Development of a fluorescence in situ hybridization method for cheese using a 16S rRNA, Probe. J. Microbiol. Methods 52 (2003) 267–271. [13] D. Ercolini, F. Villani, M. Aponte, G. Mauriello, Fluorescence in situ hybridization detection of Lactobacillus plantarum group on olives to be used in natural fermentations, Int. J. Food Microbiol. 112 (2006) 291–296. [14] M.G. Fortina, G. Ricci, D. Mora, P.L. Manachini, Molecular analysis of artisanal cheeses reveals Enterococcus italicus sp. Nov, Int. J. Syst. Evol. Microbiol. 54 (2004) 1717–1721. [15] M.G. Fortina, G. Ricci, F. Borgo, P.L. Manachini, Rapid identification of Enterococcus italicus by PCR with primers targeted to 16S rRNA gene, Lett. Appl. Microbiol. 44 (2007) 443–446. [16] C.M.A.P. Franz, M.E. Stiles, K.H. Schleifer, W. Holzapfel, Enterococci in foods – a conundrum for food safety, Int. J. Food Microbiol. 88 (2003) 105–122. [17] M.R. Foulquie´ Moreno, P. Sarantinopoulos, E. Tsakalidou, L. De Vuyst, The role and application of enterococci in food and health, Int. J. Food Microbiol. 106 (2006) 1–24. [18] G. Giraffa, Enterococci from foods, FEMS Microbiol. Rev. 26 (2002) 163–171. [19] G. Giraffa, Functionality of enterococci in dairy products, Int. J. Food Microbiol. 88 (2003) 215–222. [20] G. Giraffa, Enterococci and dairy products, in: H. Huy (Ed.), Handbook of Food Production Manufacturing, vol. II, Wiley, New York, 2007, pp. 85–97 (Chapter 50). [21] G. Giraffa, D. Carminati, Molecular techniques in food fermentation: principles and applications, in: L. Cocolin, D. Ercolini (Eds.), Molecular techniques in the microbial ecology of fermented foods, Springer Science, New York, 2008, pp. 1–30 (Chapter 1). [22] C.R. Jackson, P.J. Fedorka-Cray, M.C. Jackson-Hall, L.M. Hiott, Effect of media, temperature and culture conditions on the species population and antibiotic resistance of enterococci from broiler chickens, Lett. Appl. Microbiol. 41 (2005) 262–268. [23] D. Jurkovic, L. Krizkova, M. Sojka, A. Belicova, R. Dusinski, J. Krajcovic, C. Snauwaert, S. Naser, P. Vandamme, M. Vancanneyt, Molecular identification and diversity of enterococci isolated from Slovak Bryndza cheese, J. Gen. Microbiol. 52 (2006) 329–337. [24] F.H. Kayser, Safety aspects of enterococci from the medical point of view, Int. J. Food Microbiol. 88 (2003) 255–262. [25] B. Kolloffel, L. Meile, M. Teuber, Analysis of brevibacteria on the surface of Gruye`re cheese detected by in situ hybridization and by colony hybridisation, Lett. Appl. Microbiol. 29 (1999) 317–322. [26] L. Maietti, B. Bonvini, G. Huys, G. Giraffa, Incidence of antibiotic resistance and virulence determinants among Enterococcus italicus isolates from dairy products, Syst. Appl. Microbiol. 30 (2007) 509–517. [27] S.M. Naser, F.L. Thompson, B. Hoste, D. Gevers, P. Dawyndt, M. Vancanneyt, J. Swings, Application of multilocus sequence analysis (MLSA) for rapid identification of Enterococcus species based on rpoA and pheS genes, Microbiology 151 (2005) 2141–2150.

ARTICLE IN PRESS 230

M.E. Fornasari et al. / Systematic and Applied Microbiology 31 (2008) 223–230

[28] S.M. Naser, M. Vancanneyt, B. Hoste, C. Snauwaert, K. Vandemeulebroecke, J. Swings, Reclassification of Enterococcus flavescens Pompei et al. 1992 as a later synonym of Enterococcus casseliflavus (ex Vaughan et al. 1979) Collins et al. 1984 and Enterococcus saccharominimus Vancanneyt et al. 2004 as a later synonym of Enterococcus italicus Fortina et al. 2004, Int. J. Syst. Evol. Microbiol. 56 (2006) 413–416. [29] K.N. Olsen, E. Brockmann, S. Molin, Quantification of Leuconostoc populations in mixed dairy starter cultures using fluorescence in situ hybridisation, J. Appl. Microbiol. 103 (2007) 855–863. [30] A. Pernthaler, J. Pernthaler, R. Amann, Fluorescence in situ hybridization and catalyzed reporter deposition for

the identification of marine bacteria, Appl. Environ. Microbiol. 68 (2002) 3094–3101. [31] A. Serio, A. Paparella, C. Chaves-Lo´pez, A. Corsetti, G. Suzzi, Enterococcus populations in Pecorino Abruzzese cheese: biodiversity and safety aspects, J. Food Prot. 70 (2007) 1561–1568. [32] A. Vaid, A.H. Bishop, Amplification of fluorescently labelled DNA within Gram-positive and acid-fast bacteria, J. Microbiol. Methods 38 (1999) 53–62. [33] M. Vancanneyt, M. Zamfir, L.A. Devriese, K. Lefebvre, K. Engelbeen, K. Vandemeulebroecke, M. Amar, L. De Vuyst, F. Haesebrouck, J. Swings, Enterococcus saccharominimus sp. nov., from dairy products, Int. J. Syst. Evol. Microbiol. 54 (2004) 2175–2179.