Genotypic intraspecies heterogeneity of Enterococcus italicus: data from dairy environments

Sample material and bacteriological analysis

Fifty six samples of commercially available cheeses from different region of Italy, four from Lithuania and other two respectively from Perù and Mexico, made with ewe, goat or cow milk were analyzed. At the same time ten samples from cow, pork, poultry meat, two samples from human and cat excrements respectively, and one from olives and their brine were analyzed. The presence of E. italicus was verified in cheese samples directly by total DNA extraction. For the other matrices a conventional enrichment step in M17 broth (Difco Laboratoires, Detroit, USA) supplemented with 1% (w/v) glucose (M17g) was firstly carried out (incubation at 37 °C). Samples (10 g) tested positive were homogenized with 10 volumes of a sterile saline (0.9% NaCl, w/v) solution in a laboratory paddle blender (Stomacher 400, Seward, London, UK). Decimal dilutions were prepared in saline solution and plated in duplicates for viable counts on M17g agar. After 24 h of incubation at 37 °C colony numbers was calculated, characteristic colonies were randomly selected and subcultivated (purified) on M17g plates.

Bacterial strains and growth conditions

Enterococcus italicus strains included in this study and reference strains used for comparison are listed in Table 1 All strains were maintained as frozen stocks at −80 °C using glycerol as cryoprotective agent; reactivation of strains was done in M17g at 37 °C overnight.

DNA extraction

Total DNA of each strain was extracted as described previously 28, plasmidic DNA was isolated by the alkaline extraction procedure as described by Anderson and McKay 29. Further purification of DNA was performed according to published procedures 30. Genomic DNA for all PCR reactions was extracted from a 100 μl aliquot of an overnight culture diluted with 300 μl TE 1 × buffer (10 mM Tris-HCl, 1 mM Na2EDTA, pH 8.0) as described by Mora et al. 31.

Ribotyping

The genomic DNA of Enterococcus italicus isolates (ca. 10 μg) was digested by incubation with 30 U of HpaII endonuclease according to manufacturer's instructions (MBI-Fermentas, Vilnius, Lithuania). A 20 μl aliquot of the digestion mixture was combined with 5 μl of loading buffer and was electrophoresed on a 1.2% (w/v) agarose gel at 100 V for 2 h. Subsequently, DNA fragments were transferred to a nylon membrane (Roche Diagnostics GmbH, Mannheim, Germany) by Southern Blot, according to the manufacturer's instructions 30. Hybridization was performed at 60 °C using E. italicus DSM 15952^T16S rDNA. The probe was produced with primers and conditions previously reported 32. For digoxigenin labeling of the 1500 bp fragment the DIG DNA Labeling and Detection kit (Roche) was used. Pre-hybridization and hybridization were performed in 50% (w/v) formamide at 42 °C overnight and stringency washes in 0.1 × SSC buffer at 65 °C (10 SSC buffer consisted of 1.5 M NaCl and 150 mM sodium citrate). Digoxigenin-labelled probes that hybridized to a target sequence were detected with a digoxigenin luminescence detection kit (Roche) using CSPD {3-/4-methoxyspiro[1,2-dioxetane-3,2′-(5′-chloro)tricyclo[3.3.1.1(3,7)decan]-4-yl) phenyl phosphate; signals were recorded on X-ray film for 2 h.

Genotypic analysis

Isolates were grouped by combined analysis of repetitive element (REP) typing using primer BOXA1R (5′- CTACGGCAAGGCGACGCTGACG-3′) and (GTG)₅ (5′-GTGGTGGTGGTGGTG-3′) (PRIMM, Milan, Italy) as reported by Colombo et al. 33 and Versalovic et al. 34 respectively, and random amplification of polymorphic DNA-PCR (RAPD) typing using primers M13 (5′-GAGGGTGGCGGTTCT-′3) as described by Rossetti and Giraffa 35.

Data analysis

The experiments were repeated a minimum of three times to confirm the band patterns and only consistent and reproducible band patterns were used for data analysis. The gels were marked for the presence (1) or absence (0) of the corresponding band in the different isolates. The binary data generated was analyzed for genetic similarity using unweighted pair group arithmetic mean (UPGMA) program of NTSYSpc software, version 2.11 36.

Primers

The identification of potentially new E. italicus strains was conducted with species-specific primers targeted to 16S rRNA gene 37. Primers used for the detection of specific genes coding for ISEf1 (IS905-like transposase), IS256, traK (conjugal transfer protein, NZ_AEPV01000090), traB (pheromone shutdown protein, NZ_AEPV01000079), repE (replication plasmid protein, NZ_AEPV01000111), int (phage integrase/recombinase, NZ_AEPV01000033), sur (cell wall surface protein, NZ_AEPV01000004), hem (hemolysin, NZ_AEPV01000066), mrcB (penicillin-binding protein 1B, NZ_AEPV01000037), pbpB (penicillin-binding protein 2B, NZ_AEPV01000041), pbp1A (penicillin-binding protein 1A, NZ_AEPV01000044), pbp2A (peni-cillin-binding protein 2A, NZ_AEPV01000087) and pbpC (penicillin-binding protein C, NZ_AEPV01000003) were designed on the basis of the E. italicus DSM15952^T whole genome shotgun sequence (Accession: PRJNA61487, ID: 61487). The sequences of primers are specified in Table 2.

DNA amplification procedure and PCR conditions

Each 25 μl reaction mixture contained 100 ng of bacterial DNA, 2.5 μl of 10 × reaction buffer, 200 μM deoxynucleoside triphosphate, 2.5 mM MgCl₂, 0.5 μM of each primer and 0.5 U of Taq polymerase (MBI-Fermentas). The amplification was performed in a Gene Amp PCR System 2400 (PerkinElmer, Norwalk, CT). PCR conditions: initial denaturation step (2 min 94 °C) was followed by 35 cycles of denaturation at 94 °C for 1 min and annealing at 55 °C for 1 min, with extension at 72 °C for 2 min. The final cycle was followed by an additional 7 min elongation period at 72 °C. PCR products were separated on 1.5% (w/v) agarose gel stained with ethidium bromide in 1 X Tris-acetate-EDTA buffer (40 mM Tris-acetate and 1 mM EDTA, pH 8.0) and photographed under UV light.

Isolation and distribution of E. italicus in sample matrices

We analyzed different food matrices, out of 74 samples, for the presence of E. italicus, with particular regards to cheese (62 samples), meat (10), olives and brine (2) and human (1) and cat excrements (1). After species-specific PCR, the expected amplicon of 323 bp was obtained only from total DNA of dairy samples (10%) suggest-ing that dairy habitat may be one favourite ecosystem of E. italicus. Twenty-five randomly selected colonies grown on M17g agar plates for each PCR positive dairy sample were collected. In total, 150 novel isolates were subjected for identification by molecular analyses using E. italicus species-specific probe. Only from Pannerone and Mexican cheddar cheese E. italicus was isolated (Table 1). From other cheese samples, despite the positive signal in PCR using species-specific probe, E. italicus strains were not obtained by cultivation. The distribution of E. italicus is most likely related to the type of cheeses and/or stage of ripeness. In Pannerone and Toma-like cheeses E.italicus seems to be a typical member of the bacterial population, while in other kind of cheeses this species may be only sporadically present.

Plasmid profiling

In all E. italicus isolates we detected bands indicative for plasmids, only the type strain (DSM 15952^T) tested negative. The isolates contained two to nine plasmids, some with high molecular mass (Fig. 1). Two E. italicus strains (TP1.3 and TP2.3) had the same plasmid profile although, as reported in Fortina et al. 10, they are different isolates.

Molecular typing

Three different primer sets (Fig. 2) were used for the evaluation of E. italicus genetic fingerprinting. Primer (GTG)₅ showed low discriminatory ability (Fig. 2), only five electrophoretic profiles were obtained. However, the presence of a common band of about 1200 bp in all strain tested was interesting. More heterogeneous patterns were obtained with the primers BOXA1R and M13 (Fig. 2). Dendrograms obtained from UPGMA analyses demonstrated that these primers are different in their discriminatory power. BOXA1R primer clustered 16 groups (and/or strains), primer M13 grouped 19 strains in 13 clusters. The isolates obtained from Robiola cheese and isolate TP3.D from Toma piemontese cheese were separated from the other isolates at similarity level ranging from 1 to 30% using M13 primer (Fig. 2). Within the main cluster it was possible to observe different level of correlation among the strains. Some isolates showed a unique band profile, whereas other strains (EIP2, 5, 6, 10 Pannerone cheese) showed a band pattern similarity of 100%. REP-PCR using primer BOXA1R allowed a better discrimination at strain level, 16 of the 19 strains showed a unique fingerprint (Fig. 2). UPGMA dendrogram obtained combining M13 and BOXA1R profiles (Fig. 3) further indicated a high-level of biodiversity, represented by two different strain clusters. Cluster B contained the unique isolate TP3.D and cluster A was split into different sub-clusters. For cluster A no correlation between ecological niche and isolates could be observed.

Ribotyping

Two different clusters of E. italicus strains with seven or eight bands in their riboprints could be identified by PCR-ribotyping (Fig. 4). One cluster was characterized by seven bands, including all strains isolated from Toma piemontese and Robiola cheese as well as isolates from Pannerone cheese (EIP1, EIP3, EIP9 and EIP10), Mexican cheddar (ITAC1) and Belgian cheese (ESA). The ribotype fingerprints of the other isolates were characterized by eight bands and formed separate cluster.

Genome analysis

Virulence factor typing showed that all 13 genes screened are detectable in E. italicus strains (Table 3). Whereas, the reference strains of the other tested species gave variable results. Genes for phage integrase/recombinase (int), hemolysin (hem) and the penicillin binding protein (pbpB) were detected in all enterococci.

Gene for surface cell wall protein (sur) was only detected in strains of E. italicus.