ARTICLE IN PRESS International Dairy Journal 17 (2007) 1115–1122 www.elsevier.com/locate/idairyj Characterization of spontaneous phage-resistant variants of Streptococcus thermophilus by randomly amplified polymorphic DNA analysis and identification of phage-resistance mechanisms A.G. Binettia, V.B. Suáreza, P. Tailliezb,1, J.A. Reinheimera, a Instituto de Lactologı´a Industrial (INLAIN), Facultad de Ingenierı´a Quı´mica (Universidad Nacional del Litoral), Santiago del Estero 2829, 3000 Santa Fe, Argentina b Unité de Recherches Laitières et Génétique Apliquée, URLGA, INRA, Jouy-en-Josas, France Received 26 June 2006; accepted 25 January 2007 Abstract A total of 100 spontaneous phage-resistant mutants isolated from nine commercial Streptococcus thermophilus strains were characterized preliminarily by randomly amplified polymorphic DNA (RAPD) and the nature of their phage-resistance mechanisms was investigated. Only for mutants isolated from one strain, free phages were detected in their culture supernatants when these were titrated on the sensitive strain, suggesting that the mutants could have acquired the resistance phenotype by integrating the phage in their genomes (lysogeny). Adsorption interference was observed in the derivatives isolated from two strains. For mutants isolated from two other strains, restriction–modification (R–M) type systems were detected. In one of these cases, R–M was probably combined with another intracellular anti-phage system. In most cases, the molecular profiles (RAPD fingerprints) obtained with four arbitrary primers showed a high similarity among parent strains and their respective phage-resistant mutants. Some of these mutants were identified as potentially improved strains for industrial use. r 2007 Elsevier Ltd. All rights reserved. Keywords: Streptococcus thermophilus; Phages; RAPD-PCR 1. Introduction Despite the development of a variety of countermeasures (culture rotation, improved sanitation strategies and use of bacteriophage-resistant starter strains) phage infection during product manufacture continues to be the leading cause of failed or retarded dairy fermentations (Brüssow & Desière, 2001; Coffey, Coakley, Mc Garry, Fitzgerald, & Ross, 1998; Forde & Fitzgerald, 1999; Klaenhammer & Fitzgerald, 1994; Neve, 1996; Vadeboncoeur & Moineau, 2004). Several factors, such as lysogenic lactic acid Corresponding author. Tel.: +54 342 4530302; fax: +54 342 4571162. E-mail address: anabinetti@fiqus.unl.edu.ar (A.G. Binetti). Present address: Unité d’Ecologie Microbienne des Insectes et Interactions, Hôte-Pathogène, UMR INRA—Université Montpellier II, France. 1 0958-6946/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.idairyj.2007.01.007 bacteria (LAB) present in raw milk processed daily and the use of non-sterile fermentation media (pasteurized milk) determine the entrance and dynamics of phage populations in dairy plant environments. Thus, the success of commercial lactic starter cultures depends, primarily, on the selection of phage-unrelated strains, which are able to withstand viral infections. The isolation of phage-resistant mutants with satisfactory technological performance from sensitive-strains represents a very interesting approach for obtaining improved strains for industrial purposes (Coffey et al., 1998; Klaenhammer, 1984; Guglielmotti et al., 2006; Quiberoni, Reinheimer, & Tailliez, 1998). Although this technique has the advantage of simplicity and rapidity, it is rarely used for commercial Lactococcus strains since bacteriophage insensitive mutants often exhibit a variety of negative qualities that may exclude them for being used in industrial dairy fermentations (Coffey et al., 1998; Forde ARTICLE IN PRESS 1116 A.G. Binetti et al. / International Dairy Journal 17 (2007) 1115–1122 & Fitzgerald, 1999; Klaenhammer, 1984; Moineau, 1999; Sturino & Klaenhammer, 2004). However, it was successfully employed to isolate phage-resistant variants with good technological abilities from Lactobacillus helveticus, allowing their use in cheese making (Quiberoni, Reinheimer, & Suárez, 1999). Streptococcus thermophilus is one of the most important thermophilic LAB because of its worldwide use in the dairy industry. In Argentina, besides its relevance as a starter in the yoghurt industry, this species is used in the production of other fermented milks and a large variety of hard and semi-hard cheeses (Reinheimer et al., 1997). From this kind of processes many specific phages have been isolated over the past years (Suárez, Quiberoni, Binetti, & Reinheimer, 2002). Low frequencies at which S. thermophilus spontaneous phage-resistant mutants occur were previously reported (Moineau, 1999; Viscardi, Capparelli, & Iannelli, 2003; Viscardi, Capparelli, Di Mateo et al., 2003). Notwithstanding, we have recently isolated (Binetti, Bailo, & Reinheimer, 2007) such variants from sensitive commercial strains used in Argentinean dairy plants by the secondary culture method with a frequency that was strain-dependent. Since some of these mutants showed excellent levels of phage resistance and stability, as well as acidifying and proteolytic activities, they could be used as improved strains for industrial purposes. However, the mechanisms involved in their resistance have not been elucidated yet. The aim of this work was to investigate, by means of a primary identification, the resistance mechanisms present in phage-resistant mutants of S. thermophilus and characterize these derivatives based on randomly amplified polymorphic DNA (RAPD) fingerprints. 2. Materials and methods 2.1. Bacterial strains, bacteriophages and culture conditions Spontaneous phage-resistant variants were obtained from nine S. thermophilus strains (identified as 4-C, 5-C, YDS10-C, Jo1-C, M1-C, M8-C, M11-C, MiC1 and MiC7), isolated from commercial starters used in Argentinean dairy industries (INLAIN Culture Collection), and three Italian S. thermophilus strains (identified as I49, I53 and I54) belonging to the Istituto Sperimentale Lattiero Caseario (ISLC, Lodi, Italy) Culture Collection. Bacteriophages used were nine autochthonal S. thermophilus phages (f021-4, f031, fCYM, fCYS1, fQP2, fQPL2, fQPL10 , fMi2 and fMi1) isolated from Argentinean dairy plants (INLAIN Phage Collection) and three Italian S. thermophilus phages (f49, f53 and f54) (ISLC Phage Collection) (Table 1). To isolate spontaneous phage-resistant mutants, the secondary culture method was used (Carminati, Zennaro, Neviani, & Giraffa, 1993). Strains and phageresistant derivatives were grown in M17 broth or M17 agar (Biokar, Beaubois, France) at 42 1C and stored ( 80 1C) in M17 broth supplemented with 15% (v/v) glycerol and in Table 1 Phage-resistance mechanisms present in phage-resistant mutants isolated from commercial S. thermophilus strains Sensitive strain Phage nRa 4-C 5-C YSD10 M1-C M8-C M11-C Jo1-C MiC1 MiC7 I49 I53 I54 f021-4 f031 fCYM fQP2 fQLP2 fQLP10 fCYS1 fMi2 fMi1 f49 f53 f54 3 3 2 6 10 1 1 4 8 26 17 9 Lysogenyb + Adsorption ratec 85.5 93.3 89.9 82.1718.1 80.673.4 99.9 95.7 96.173.6 48.573.2 96.276.6 83.177.1 49.279.0 a nR: Number of phage-resistant mutants isolated; : absence; +: presence. b Spontaneous induction of free phages, detected in the supernatants of mutants cultures. c % (mean value of each group) of adsorbed phages in M17-Ca broth after 30 min at 45 1C. Standard deviation was calculated when nR43. non-fat dry skim milk (Merck, Darmstadt, Germany). Phage enumerations were carried out by double-layer plaque titration method from IDF Standards (1991) using M17 soft agar on M17 agar supplemented with 10 mM CaCl2 (M17-Ca) and 100 mM glycine (Lillehaug, 1997). 2.2. Characterization of strains For S. thermophilus strains and their phage-resistant variants, cell (phase contrast, 1000  , Microscope Jenamed 2 Carl Zeiss, Jena, Germany) and colony (on M17 agar) morphologies were observed. To evaluate sugar fermentation patterns, API 50 CHS (Bio Merieux, Marcy l’Etoile, France) galleries were used, according to the manufacturer’s instructions. 2.3. Phage-resistance mechanisms Lysogeny and adsorption rates were determined for all phage-resistant mutants as previously described (Quiberoni et al., 1998). All assays were performed in triplicate. In the case of mutants that exhibited a relatively high adsorption rate of phage particles and a late lysis in broth, the presence of restriction–modification (R–M) type resistance mechanisms was investigated according to de los Reyes-Gavilán, Limsowtin, Tailliez, Séchaud, and Accolas (1992) modified as follows: a phage suspension was titrated on the sensitive strain and on the phageresistant mutant, and Efficiency of Plaquing (EOP) (first value) was calculated. One or two lysis plaques obtained from the titration on the phage-resistant variant were picked up and suspended in 5 mL of M17-Ca broth. Phage suspensions were kept 24 h at 4 1C and then inoculated with ARTICLE IN PRESS A.G. Binetti et al. / International Dairy Journal 17 (2007) 1115–1122 0.2 mL of phage-resistant mutant overnight culture and incubated at 42 1C until total lysis, to allow the propagation of potentially modified phages. The lysate obtained was filtered and titrated on both strains (sensitive and resistant) for determining the second EOP value (in the presence of a R–M system, this value must be approximately equal to the unit). From the titre plate of the sensitive strain, one or two lysis plaques were picked up and suspended in 5 mL of M17-Ca broth and, after 24 h at 4 1C, inoculated with 0.2 mL of sensitive strain culture and incubated at 42 1C until total lysis. The resulting lysate was filtered and titrated on both strains (sensitive and resistant). The respective EOP-value was also determined for this stage (if it is similar to that obtained in the first stage, it indicates the presence of an active R–M type phage-resistance mechanism in the mutant studied). All assays were performed in triplicate. 2.4. RAPD analysis Total DNA of strains was obtained by phenol-chloroform extraction as was described previously (de los ReyesGavilán et al., 1992) and quantified by electrophoresis on 0.8% (w/v) agarose gels (Seakem, Tebu, France). Optimized polymerase chain reaction (PCR) amplification reactions were performed in a total volume of 100 mL (10 mmol L 1 Tris-HCl buffer pH 9.0, containing 1.5 mmol L 1 MgCl2) with 1 mL of template DNA (20– 100 mg), 0.5 mmol L 1 primer (Bioprobe, Montreuil-sousBois, France), 2.5 U Taq Polymerase (Qbiogène, Illkirch Cedex, France) and 200 mmol L 1 of each dNTP (Boehringer Mannheim, Mannheim, Germany). Four single arbitrary primers, P1 (50 TGCTCTGCCC 30 ), P2 (50 GGTGACGCAG 30 ), P3 (50 GTCCACACGG 30 ) and P4 (50 CTGCTGGGAC 30 ) were used in separate PCR reactions. A Perkin-Elmer (Courtaboeuf, France) thermo cycler (model 9600) was used to submit DNA samples to 30 cycles of amplification (94 1C for 1 min, 36 1C for 2 min and 72 1C for 2 min). Amplification products were analysed by electrophoresis in 1% (w/v) GTG agarose gels (Seakem, Tebu, Le Perray-en-Yvelines, France) containing 200 mg L 1 of ethidium bromide (Sigma, Saint Quentin Fallavier, France) and viewed by ultraviolet (UV) transillumination at 254 nm. A DNA molecular weight marker, 123 bp DNA Ladder (Gibco BRL, Cergy Pontoise, France) was used as a standard. Photographs of gels under UV light were taken using Polaroid film type 665 and negative pictures were digitalized using a Hewlett Packard (Issy les Moulineaux, France) SCanJet IIcx/T. Digitalized pictures were analysed using the Gel Compare software (AppliedMaths, Sint-Martens-Latem, Belgium). Band profiles obtained with the four primers were normalized and subsequently combined. Densitometric traces were grouped in clusters using the Unweighted Pair Group Method with Arithmetic Average (UPGMA; Romersburg, 1984). 1117 3. Results 3.1. Characterization of strains All phage-resistant variants were identical to their parent strains in cell- and colony-morphologies and sugar fermentation patterns. 3.2. Phage-resistance mechanisms 3.2.1. Lysogeny Only for phage-resistant mutants isolated from the sensitive strain S. thermophilus I49, free phages able to infect the parent strain, but not themselves, were detected in their broth culture supernatants. This fact suggested that the resistance phenotype could be linked to integration of f49 in the streptococcal genomes, leading to a possible phage-resistance mechanism associated to lysogeny (by lysogenic immunity; Table 1). 3.2.2. Adsorption rates In general, a significant inhibition of phage adsorption was not detected for resistant variants since their mean adsorption rates were higher than 82%, except for those isolated from S. thermophilus MiC7 and I54 that were partially unable to bind phage particles (mean adsorption rates of 48.5% and 49.2%, respectively, Table 1). 3.2.3. R–M mechanisms The variants isolated from strains S. thermophilus 5-C and I53 with phages f031 and f53, respectively, showed high adsorption rates (93.3% and 83.1%, respectively, as mean value, Table 1) and a late lysis in broth. Additionally, they were the only group of phage-resistant mutants which exhibited the ability to form visible lysis plaques when they were infected with phages (EOP mean values of 4.6  10 7 and 2.2  10 6, respectively, Binetti et al., 2007). Therefore, they were investigated for the presence of R–M systems. Based on the results obtained (Table 2), it was possible to detect the presence of active R–M type mechanisms in mutant R5-C031-10 (derived from S. thermophilus 5-C). At first, it exhibited a high resistance level against f031 (EOP value: 5.0  10 7). After titrating the potentially modified phage (f031m) on strains S. thermophilus 5-C and R5C031-10, the resulting EOP was near the unit. Finally, when f031m was replicated on the parent strain and then titrated again on both strains, the EOP decreased to a value (3.5  10 6) near to that obtained at a first stage. Also, four mutants isolated from S. thermophilus I53 showed high resistance levels against f53 (EOP values from 1.6  10 7 to 1.2  10 6, Table 2). At the second stage, when the potentially modified phages were propagated on the resistant variants and then titrated on both strain types, EOP values slightly lower than 1 (from 2.2  10 1 to 8.6  10 1) were obtained. Finally, at the third stage of the assay, the EOPs appeared not near the ARTICLE IN PRESS 1118 A.G. Binetti et al. / International Dairy Journal 17 (2007) 1115–1122 Table 2 Evidence of restriction–modification mechanisms in phage-resistant mutants isolated from commercial S. thermophilus strains 1a EOPb Sensitive strain/phageresistant mutant Phage Assay stage PFU mL 5-C R5031-10 f031 1st 1.2  10871.1  108 6.0  10174.0  101 5.0  10 5-C R5031-10 f031c 5.1  10771.3  107 5.6  10773.5  107 1.1 5-C R5031-10 f031 I53 RI53-9 RI53-10 RI53-12 RI53-21 fI53 I53 RI53-9 I53 RI53-10 I53 RI53-12 I53 RI53-21 fI53(RI53-9)c I53 RI53-9 I53 RI53-10 I53 RI53-12 I53 RI53-21 fI53(RI53-9) 2nd 7 3rd 1st 6 1.0  10 78.7  10 3.5  10172.2  101 3.5  10 6 2.7  10871.4  108 9.0  10176.7  101 3.3  10271.4  101 4.3  10172.1  101 1.5  10271.9  102 3.3  10 1.2  10 1.6  10 5.6  10 7 2.2  10 1 8.6  10 1 2.3  10 1 6.6  10 1 5.5  10 2 1.3  10 1 6.5  10 2 6.5  10 2 9.0  10875.4  107 2.0  10873.0  108 2.2  10974.1  108 1.9  10971.1  109 1.9  10971.7  109 4.5  10873.9  109 8.7  10772.9  107 5.7  10772.8  107 2nd c fI53(RI53-10) c fI53(RI53-12) c fI53 (RI53-21) 3.3  10873.5  108 1.8  10771.0  107 1.3  10979.1  108 1.7  10871.1  108 1.1  10979.9  108 7.3  10777.2  107 1.5  10971.2  109 1.0  10871.3  108 3rd fI53(RI53-10) fI53(RI53-12) fI53(RI53-21) 7 6 7 7 a Mean value of three determinations. Efficiency of plaquing. c Modified phage, obtained by replication of the lytic phage on the resistant variant. b unit but higher (approximately 5 log orders) than those obtained at the first stage (from 5.5  10 2 to 1.3  10 1). 3.3. RAPD analysis Individual dendrograms of RAPD profiles were obtained for S. thermophilus phage-sensitive strains and their respective phage-resistant variants (Fig. 1 shows five examples). For variants isolated from S. thermophilus YDS10, one of them (RYDS10-3) appeared almost identical (93% similarity) to the parent strain, while the other one (RYDS10-17) showed higher molecular diversity (83% similarity), mainly in the genome region amplified with primer P2 (Fig. 1A). For S. thermophilus M8-C and its derivatives, two clusters (82% similarity) were obtained from their RAPD profiles (Fig. 1B), principally based on the differences detected with primer P4. Within both groups, a high (490%) similarity coefficient was observed among the respective strains. For S. thermophilus Jo1-C and its single mutant, the similarity coefficient was 84% and the molecular diversity was also detected by primer P4 (Fig. 1C). In the case of the S. thermophilus I53-variants group, the similarity with the parent strain was 84%; the molecular diversity was most evident in the RAPD profiles obtained with the P2 primer (Fig. 1D). Finally, the dendrogram of the S. thermophilus I54-variants group enabled to distinguish two clusters, linked by a similarity coefficient of 79%. The strains belonging to each cluster showed a molecular diversity lower than 15%, which was most significant in the RAPD profiles obtained with primer P2 (Fig. 1E). For the other strains (S. thermophilus 4-C, 5-C, M1-C, M11-C, MiC1, MiC7 and I49) and their respective phage-resistant mutants, the similarity coefficients were also higher than 70% (data not shown). 4. Discussion Over the last decade, and due to the evolution of the processes and the extensive use of commercial thermophilic cultures, the dairy industry has faced increasing phage problems with S. thermophilus strains (Brüssow, Bruttin, Desière, Lucchini, & Foley, 1998; Coffey et al., 1998; Forde & Fitzgerald, 1999; Klaenhammer & Fitzgerald, 1994; Moineau, 1999; Neve, 1996; Suárez, Quiberoni, Binetti, & ARTICLE IN PRESS A.G. Binetti et al. / International Dairy Journal 17 (2007) 1115–1122 1119 Fig. 1. Dendrograms obtained by comparison (Program Gel Compar, Applied Maths, Sint-Martens-Latem, Belgium) and clustering (UPGMA method: Unweighted Pair Group Method using Arithmetic Averages) of RAPD profiles of S. thermophilus strains (A: YDS-10, B: M8-C, C: Jo1-C, D: I53 and E: I54) and their respective spontaneous phage-resistant mutants, using the primers P1, P2, P3 and P4. Parent strains are indicated in bold. Reinheimer, 2002). In an effort to find strains with an improved phage resistance, 100 spontaneous phage-resistant mutants were isolated from commercial S. thermophilus strains under selective pressure of specific phages (Binetti et al., 2007). In the present work, it was possible to establish that different mechanisms of phage resistance are present among these mutants. Spontaneous phage induction was detected only for the variants derived from S. thermophilus I49, which was probably due to the integration of the f49 genome in their chromosomes, a phage-resistance mechanism linked to super-infection immunity or lysogeny. This resistance mechanism occurs very frequently in lactococci and lactobacilli, but is almost inexistent in S. thermophilus (Carminati & Giraffa, 1992; Josephsen & Neve, 1998; Mercenier, 1990; Neve, 1996). The failure of the challenging phage to adsorb to the cell, presumably due to mutations in the receptor gene, holds very often for bacteriophage-insensitive mutants of lactococci (Moineau, 1999; Sturino & Klaenhammer, 2004; Vadeboncoeur & Moineau, 2004). In case of streptococcal resistant variants, instead, a normal adsorption of phage particles was observed, except for those derived from S. thermophilus MiC7 and I54 that were only partially unable to adsorb viruses. Recently, Viscardi, Capparelli, and Iannelli (2003) and Viscardi, Capparelli, Di Mateo et al. (2003) developed two methods for selection of phageresistant S. thermophilus strains by means of flow cytometry that allowed the isolation of variants in which phage adsorption had been blocked. Interference with phage adsorption as frequently observed in other LAB, ARTICLE IN PRESS 1120 A.G. Binetti et al. / International Dairy Journal 17 (2007) 1115–1122 such as different species of Lactobacillus (Neviani, Carminati, & Giraffa, 1992; Quiberoni et al., 1998; Reinheimer, Morelli, Callegari, & Bottazzi, 1993; Reinheimer, Quiberoni, Tailliez, Binetti, & Suárez, 1996; Ventura, Callegari, & Morelli, 1999) and Lactococcus (for review, see Forde & Fitzgerald, 1999) is in most cases plasmid-encoded. In general, the adsorption inhibition shown by spontaneous phage-resistant mutants presents some drawbacks since it is highly specific and reversion to phage sensitivity can occur at a high frequency (Forde & Fitzgerald, 1999; Limsowtin & Terzaghi, 1976; Moineau, 1999). Consequently, the inherent instability of this system and the persistence of phage particles in the environment may limit the significance of adsorption inhibition as a potent defence mechanism (Forde & Fitzgerald, 1999). Therefore, phageresistant mutants from S. thermophilus MiC7 and I54 may be of restricted value as industrial strains. In contrast, R–M systems are powerful defence mechanisms since they interrupt the infection process prior to the initiation of phage-directed cell death and they remove phage particles from the environment. The combination of R–M and Abi systems can be even more effective by countering the genetic flexibility of phages (usually evolving to escape the restriction) increasing the degree of insensitivity and expanding the range of phage types against which the host is resistant (Forde & Fitzgerald, 1999; Sturino & Klaenhammer, 2004). In this work, one derivative from S. thermophilus 5-C (resistant to f031) could have acquired the phage-resistance phenotype by means of a tentative R–M type system that could be silent or absent in the parent strain. The R–M phenotype in this case was clearly evidenced by the classical methodology proposed by de los Reyes Gavilán, Limsowtin, Séchaud, Veaux, and Accolas (1990). Furthermore, in some mutants derived from S. thermophilus I53 (resistant to f53) the nature of the phage-resistance mechanism was not clear, but our results suggested the presence of R–M type mechanisms that would be combined with other anti-phage systems. The phenotype characteristic typical of R–M systems was verified but, an increase in phage particles following the third stage of the R–M screen was indicative for an extra mechanism (probably another intracellular system like Abi) being also present, which could increase the phage-resistance power of these mutants. This hypothesis is based on the fact that any other mechanism (lysogeny and adsorption interference) would have been detected by means of our preliminary study. Additional studies are being made to elucidate this hypothesis. Most of known R–M and Abi systems have been identified in lactococci and they have mainly a plasmid location (Forde & Fitzgerald, 1999; Klaenhammer, 1984; Klaenhammer & Fitzgerald, 1994; Moineau, 1999). R–M systems have also been detected in other members of LAB, such as L. helveticus (de los Reyes-Gavilán et al., 1990) (combined with Abi systems), L. delbrueckii subsp. lactis and L. fermentum (Moineau, 1999). In S. thermophilus, for which very few phage-resistance mechanism have been reported, the situation may be different as a consequence of the relative absence of plasmids in this species (Moineau, 1997). The use of a molecular technique such as RAPD fingerprinting enables the identification of many organisms at the species level and the study of the diversity among strains belonging to a particular species. The classical method is based on the use of a single oligonucleotide to generate a fingerprint of PCR products for the direct comparison of the genomes (Welsh & McClelland, 1990; Williams, Kubelik, Livak, Rafaliski, & Tingey, 1990). For LAB, it has allowed the identification and characterization of lactobacilli species (L. acidophilus, L. casei, L. delbrueckii and L. helveticus) from dairy products (Cocconcelli, Parisi, Senini, & Bottazzi, 1997; Drake, Small, Spence, & Swanson, 1996; Giraffa et al., 2004; Guglielmotti et al., 2006; Mora et al., 2002; Moschetti et al., 1998). Several authors (Quiberoni et al., 1998; Tailliez, Quénée, & Chopin, 1996; Tailliez, Tremblay, Ehrlich, & Chopin, 1998) have used this technique with three arbitrary primers to study the molecular diversity among L. plantarum, L. pentosus, L. helveticus and L. lactis strains. In this study, RAPD was utilized to detect molecular diversity and relationship among the phageresistant mutants and their respective parent strains. We modified the conventional method used for S. thermophilus (Giraffa & Rossetti, 2004; Mora et al., 2002; Moschetti et al., 1998) to increase sensitivity and reproducibility by means of four primers according to Tailliez et al. (1998). The comparison of profiles obtained for the different groups of phage-resistant strains clearly showed that all of them had molecular profiles highly similar to that of the corresponding phage-sensitive strain. In most cases, the phage-resistance phenotype could arise from changes in the genome that were not amplified by the primers used in this work. However, the differences detected for some phage-resistant mutants (similarity coefficients o85%) could correspond to genetic diversity related to changes in phage receptors (derivatives from S. thermophilus MiC7 and I54), R–M type mechanisms (derivatives from S. thermophilus 5-C) or another possible genetic modification responsible to phage-resistance phenotype (derivatives from S. thermophilus I53 and other mutants for which the phage-resistance mechanism was not identified). 5. Conclusions Even though additional studies are required to clarify the tentative phage-resistance mechanisms identified in this work, some spontaneous phage-resistant strains would be potentially suitable for the formulation of industrial starter cultures with enhanced features. The most favourable representative would be S. thermophilus RI53-9, derivative from S. thermophilus I53 that, based in our results, exhibited the combination of strong phage-resistance mechanisms. Additionally, from a previous report (Binetti et al., 2007), this strain showed an excellent industrial performance and a high stability of its phage-resistance ARTICLE IN PRESS A.G. Binetti et al. / International Dairy Journal 17 (2007) 1115–1122 phenotype. Also, certain variants of S. thermophilus M1-C and Jo1-C, in spite of the undetermined nature of their phage-resistance mechanisms, exhibited high levels and stability of their phenotypes and good technological characteristics (Binetti et al., 2007). Probably they have acquired these phenotypes by means of a combination of strong intracellular systems like Abi since spontaneous lysogeny, adsorption interference and R–M systems were not detected. The RAPD technique have been shown to be a useful tool to rapidly confirm the parental relationships between phage-sensitive strains and their phage-resistant derivatives and indicated that contaminations did not occur in the isolation of mutants. Acknowledgements This work was supported by the Consejo Nacional de Investigaciones Cientı́ficas y Técnicas (CONICET) of Argentina (Proyecto PIP 2000 No 02035), the Universidad Nacional del Litoral (Santa Fe, Argentina)—Programación CAI+D 2002 (Proyecto No 155) and the Agencia Nacional de Promoción Cientı́fica y Tecnológica de Argentina (Proyecto PICT 2000, No 09-08200). We would like to thank Dr. Domenico Carminati (ISLC, Lodi, Italy) who provides us with phages f49, f53 and f54 and their S. thermophilus host-strains. References Binetti, A. G., Bailo, N. B., & Reinheimer, J. A. (2007). Spontaneous phage-resistant mutants of Streptococcus thermophilus: Isolation and technological performance. International Dairy Journal, 17(4), 343–349. Brüssow, H., Bruttin, A., Desière, F., Lucchini, S., & Foley, S. (1998). Molecular ecology and evolution of Streptococcus thermophilus bacteriophages—A review. Virus Genes, 16, 95–109. Brüssow, H., & Desière, F. (2001). Comparative phage genomics and the evolution of Siphoviridae: Insights from dairy plants. Molecular Microbiology, 39(2), 213–222. Carminati, D., & Giraffa, G. (1992). Evidence and characterization of temperate bacteriophage in S. salivarius subsp. thermophilus 18. Journal of General Microbiology, 59, 71–79. Carminati, D., Zennaro, R., Neviani, E., & Giraffa, G. (1993). Selezione e caratteristiche di mutanti fago—resistenti de L. helveticus. Scienza e Tecnica Lattiero-Casearia, 44, 33–48. Cocconcelli, P. S., Parisi, M. G., Senini, L., & Bottazzi, V. (1997). Use of RAPD and 16S rDNA sequencing for the study of Lactobacillus population dynamics in natural whey cultures. Letters in Applied Microbiology, 25, 8–12. Coffey, A., Coakley, M., Mc Garry, A., Fitzgerald, G. F., & Ross, R. P. (1998). Increasing phage resistance of cheese starters: A case study using Lactococcus lactis DPC4268. Letters in Applied Microbiology, 26, 51–55. de los Reyes Gavilán, C. G., Limsowtin, G. K. Y., Séchaud, L., Veaux, M., & Accolas, J.-P. (1990). Evidence for a plasmid-linked restriction–modification system in Lactobacillus helveticus. Applied and Environmental Microbiology, 56, 3412–3419. de los Reyes-Gavilán, C. G., Limsowtin, G. K. Y., Tailliez, P., Séchaud, L., & Accolas, J. P. (1992). A Lactobacillus helveticus-specific DNA probe detects restriction fragment length polymorphisms 1121 in this species. Applied and Environmental Microbiology, 58, 3429–3432. Drake, M., Small, C. L., Spence, K. D., & Swanson, B. G. (1996). Rapid detection and identification of Lactobacillus spp. in dairy products by using the polymerase chain reaction. Journal of Food Protection, 59, 1031–1036. Forde, A., & Fitzgerald, G. (1999). Bacteriophage defence systems in lactic acid bacteria. Antonie van Leeuwenhoek, 76, 89–113. Giraffa, G., Andrighetto, C., Antonello, C., Gatti, M., Lazzi, C., Marcazzan, G., et al. (2004). Genotypic and phenotypic diversity of Lactobacillus delbrueckii subsp. lactis strains of dairy origin. International Journal of Food Microbiology, 91, 129–139. Giraffa, G., & Rossetti, L. (2004). Monitoring of the bacterial composition of dairy starter cultures by RAPD-PCR. FEMS Microbiology Letters, 237, 133–138. Guglielmotti, D., Reinheimer, J., Binetti, A., Giraffa, G., Carminati, D., & Quiberoni, A. (2006). Characterization of spontaneous phage-resistant derivatives of Lactobacillus delbrueckii commercial strains. International Journal of Food Microbiology, 111, 126–133. IDF. (1991). Methods for phage monitoring. IDF Standard (No 263). Brussels, Belgium: International Dairy Federation. Josephsen, J., & Neve, H. (1998). Bacteriophages and lactic acid bacteria. Lactic acid bacteria microbiology and functional aspects. New York, USA: Marcel Dekker. Klaenhammer, T. R. (1984). Interactions of bacteriophages with lactic cocci. Advances in Applied Microbiology, 30, 1–29. Klaenhammer, T. R., & Fitzgerald, G. F. (1994). Bacteriophages and bacteriophage resistance. Genetics and biotechnology of lactic acid bacteria. London, UK: Blackie Academic & Professional. Lillehaug, D. (1997). An improved plaque assay for poor plaqueproducing temperate lactococcal bacteriophages. Journal of Applied Microbiology, 83, 85–90. Limsowtin, G. K. Y., & Terzaghi, B. E. (1976). Phage resistant mutants: Their selection and use in cheese factories. New Zealand Journal of Dairy Science and Technology, 11, 251–256. Mercenier, A. (1990). Molecular genetics of Streptococcus thermophilus. FEMS Microbiology Reviews, 7(1–2), 61–77. Moineau, S. (1997). Bacteriophages and phage resistance in Streptococcus thermophilus: An update. Marshall Italian & Specialty Cheese Seminars. Moineau, S. (1999). Applications of phage resistance in lactic acid bacteria. Antonie van Leeuwenhoek, 76, 377–382. Mora, D., Fortina, M. G., Parini, C., Ricci, G., Gatti, M., Giraffa, G., et al. (2002). Genetic diversity and technological properties of Streptococcus thermophilus strains isolated from dairy products. Journal of Applied Microbiology, 93, 278–287. Moschetti, G., Blaiotta, G., Apponte, M., Catzaddu, P., Villani, F., Deiana, P., et al. (1998). Random amplified polymorphic DNA and amplified ribosomal DNA spacer polymorphifsm: Powerful methods to differentiate Streptococcus thermophilus strains. Journal of Applied Microbiology, 85, 25–36. Neve, H. (1996). Bacteriophage. Dairy starter cultures. New York, USA: VCH. Neviani, E., Carminati, D., & Giraffa, G. (1992). Selection of some bacteriophages- and lysozyme-resistant variants of Lactobacillus helveticus CNRZ892. Journal of Dairy Science, 75, 905–913. Quiberoni, A., Reinheimer, J. A., & Suárez, V. B. (1999). Performance of Lactobacillus helveticus spontaneous phage-resistant mutants in hard cheese production. International Dairy Journal, 8, 941–949. Quiberoni, A., Reinheimer, J. A., & Tailliez, P. (1998). Characterization of Lactobacillus helveticus spontaneous phage resistant mutants by RAPD-PCR fingerprints and phenotypic parameters. Food Research International, 31, 537–542. Reinheimer, J., Binetti, A., Quiberoni, A., Bailo, N., Rubiolo, A., & Giraffa, G. (1997). Natural milk cultures for Argentinean cheese production. Journal of Food Protection, 60, 59–63. ARTICLE IN PRESS 1122 A.G. Binetti et al. / International Dairy Journal 17 (2007) 1115–1122 Reinheimer, J. A., Morelli, R., Callegari, M. L., & Bottazzi, V. (1993). Phage resistance in Lactobacillus helveticus CNRZ 328. MicrobiologieAliments-Nutrition, 11, 235–240. Reinheimer, J. A., Quiberoni, A., Tailliez, P., Binetti, A. G., & Suárez, V. B. (1996). The lactic acid microflora of natural whey starters used in Argentina for hard-cheese production. International Dairy Journal, 6, 869–879. Romersburg, H. C. (1984). Cluster analysis for research. Belmont, CA, USA: Life-time Learning Publications. Sturino, J., & Klaenhammer, T. R. (2004). Bacteriophage defense systems and strategies for lactic acid bacteria. Advances in Applied Microbiology, 56, 331–378. Suárez, V. B., Quiberoni, A., Binetti, A. G., & Reinheimer, J. A. (2002). Thermophilic lactic acid bacteria phages isolated from Argentinean dairy industries. Journal of Food Protection, 65, 1597–1604. Tailliez, P., Quénée, P., & Chopin, A. (1996). Estimation de la diversité parmi les souches de la collection CNRZ: application de la RAPD à un groupe de lactobacilles. Le Lait, 76, 147–158. Tailliez, P., Tremblay, J., Ehrlich, S. D., & Chopin, A. (1998). Molecular diversity and relationship within Lactococcus lactis, as revealed by ramdomly amplified polymorphic DNA (RAPD). Systematic and Applied Microbiology, 21, 530–538. Vadeboncoeur, C., & Moineau, S. (2004). The relevance of genetic analysis to dairy bacteria: Building upon our heritage. Commentary. Microbial Cell Factories, 3(15). Ventura, M., Callegari, M. L., & Morelli, L. (1999). Surface layer variations affecting phage adsorption on seven Lactobacillus helveticus strains. Annali de Microbiologia ed Enzimologia, 49, 45–53. Viscardi, M., Capparelli, R., Di Matteo, R., Carminati, D., Giraffa, G., & Iannelli, D. (2003). Selection of bacteriophage-resistant mutants of Streptococcus thermophilus. Journal of Microbiological Methods, 55, 109–119. Viscardi, M., Capparelli, R., & Iannelli, D. (2003). Rapid selection of phage-resistant mutants in Streptococcus thermophilus by immunoselection and cell sorting. International Journal of Food Microbiology, 89, 223–231. Welsh, J., & McClelland, M. (1990). Fingerprinting genomes using PCR with arbitrary primers. Nucleic Acids Research, 18, 7213–7218. Williams, J. G., Kubelik, A. R., Livak, K. J., Rafaliski, J. A., & Tingey, S. V. (1990). DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Research, 18, 6531–6535.