Journal of Applied Bacteriology 1992, 73, 294.308

Identification and characterization of helveticin V-I 829, a bacteriocin produced by Lactobacillus helveticus 1829 Elaine E. Vaughan', C. Daly'*= and G.F. Fitzgeraidl 'Department of Food Microbiology and 'National Food Biotechnology Centre, University College, Cork, Ireland 4110/02/92: accepted 28 April 1992 E.E. V A U G H A N . c. D A L Y AND G . F . F I T Z G E R A L D . 1992. Lactobacillus helveticus 1829 produced an antimicrobial agent, designated helveticin V-1829, that demonstrated antagonistic activity against closely-related species. T h e agent was excreted into MRS agar, and was present in the supernatant fluids from both overnight broth and clotted milk cultures. It was heat labile (inactivated by 50°C for 30 min) and was stable over the p H range 2.5 to 6.5. Production of the substance was pH-dependent and maximum yields were obtained in MRS broth cultures maintained at p H 5.5. Helveticin V-1829 was partially purified following growth of the producing strain in a semi-defined MRS medium and precipitating the cell-free filtrate with ammonium sulphate to 30% saturation. The cleared supernatant fluid was then brought to 60% saturation and the resulting precipitate pelleted and dialysed in 0-3 mol/l phosphate buffer. T h e partially purified inhibitor was sensitive to several proteolytic enzymes, and it was bactericidal in its mode of action against indicator cells of Lact. helveticus 1844 and Lact. delbrueckii subsp. bulgaricus 1489, indicating that it was a bacteriocin. A DNA probe specific for the helveticin J structural gene failed to hybridize to total genomic DNA of Lact. helveticus 1829, indicating that helveticin V-1829 is not significantly related to helveticin J.

INTRODUCTION

Lactic acid bacteria are used in the production of a range of fermented food products. Metabolic compounds elaborated by these cultures contribute to the unique and characteristic flavour development of these foods and also play a major role in inhibiting the growth of spoilage bacteria (for a review see Lindgren & Dobrogosz 1990). In addition to organic acids, hydrogen peroxide and diacetyl, many lactic acid bacteria produce bacteriocins which can contribute to the antibiosis. Bacteriocins are defined as antagonistic proteins or protein complexes that show bactericidal activity directed against species closely related to the producer bacterium (Tagg et al. 1976). Lactic acid bacteria of all species used in food fermentations have been demonstrated to produce these compounds (reviewed by Klaenhammer 1988). The lactobacilli, in particular, are notable for the production of antimicrobial compounds, many of which have been characterized as true bacteriocins based on the criteria of Tagg et al. (1976). These generally have a narrow inhibitory host range affecting only related Lactobacilliceae (DeKlerk & Coetzee 1961; DeKlerk 1967; DeKlerk & Smit 1967; Upreti & Hinsdill 1973; Barefoot & Klaenhammer Correspondence to D r G.F. Fitzgerald, Department of Food Microbiology, University College, Cork, Ireland.

1983; Joerger & Klaenhammer 1986). Two bacteriocins have been identified and characterized within the Lact. helveticus species. Lactobacillus helveticus LP27 produces a bacteriocin designated lactocin 27 which exhibits a narrow activity spectrum against Lact. acidophilus and Lact. helveticus spp. (Upreti & Hinsdill 1973). Lactocin 27 was very heat stable but was inactivated by trypsin and pronase, and it had a bacteriostatic effect on the sensitive indicator Lact. helveticus LS18 (Upreti & Hinsdill 1975). The second bacteriocin, helveticin J, inhibited Lact. helveticus, Lact. delbrueckii subsp. bulgaricus and subsp. lactis species and was bactericidal to the indicator Lact. delbrueckii subsp. bulgaricus 1489 (Joerger & Klaenhammer 1986). It was heatsensitive and was destroyed by several proteolytic enzymes demonstrating its proteinaceous nature. After purification of the helveticin J protein (37000 Da), the gene(s) responsible were cloned from their chromosomal location and sequenced (Jorger & Klaenhammer 1990). This study describes the identification and characterization of a bacteriocin, designated helveticin V-1829 which is produced by the homofermentative strain Lact. helveticus 1829. In addition, as helveticin V-1829 shared many of the properties described for helveticin J, hybridization experiments using the hlv gene as a probe were performed to determine if the two helveticins were related.

300 ELAINE E . VAUGHAN ET A L

Table 1 Bacterial strains used in this study and antimicrobial

MATERIALS AND METHODS

spectrum of helveticin V-1829 Source Indicator bacteria

Lactobacillus acidophilus N2*, 216* Lact. delbrueckii subsp. bulgaricus 1489 1 I* Lact. helveticus 1844 223, 303, 32

NCSU

NCDO CNRZ

ucc CNRZ

Non-indicator bacteria

Clostridium tyrobutyricum HKII, HKIII, KI388 KI542, KI51W Enterococcus fuecalis GF590 Lactobacillus acidophilus 88 Lact. brevis 234 1748 Idact. casei 348, 161 383 152 1,act. casei ssp. rhamnosus 202 442 Lact . delbrueckii ssp. bulgaricus 208, 495, 448 Lact. delbrueckii ssp. lactis 252, 239, 326 242, 245, 235, 237 Lact. fermentum 233 Lact. helveticus 1829, 30, CH-1 48 1 384 Lact. viridescens 1615, 1613, 12706 Lactococcus lactis spp. lactis CH001, 497, 496, 495 L . lactis spp. cremoris 495 I,. lactis spp. lactis biovar. diacetylactis 18B, DRCl, 18-16 I'isteria monocytogenes 4b Pediococcus acidilactici 2767 Pseudomonas spp. F120, N2 Streptococcus mutans 65-5

ucc ucc NCSU CNRZ

ucc

ucc NCDO CNRZ

ucc CNRZ

CNRZ CNRZ CNRZ

ucc NCSU NCDO NCDO

ucc ucc ucc DPC NCDO

ucc ucc

All bacteria were examined for inhibition using both cell free filtrates neutralized to pH 6.5, and partially purified helveticin V-1829 (51 200 AU/ml), pH 6.5; controls included semi-defined MRS medium subjected to the partial purification procedure for

Bacterial cultures and media

The bacterial cultures used in this study are listed in Table 1. All cultures were maintained as frozen stocks at -70°C and before experimental use were propagated twice in broth for 16 h. Lactobacilli and Streptococcus mutans were grown in MRS broth (Difco) at 37"C, Pediococcus acidilactici in MRS at 30"C, lactococcal cultures were grown in M17 broth (Merck; Terzaghi & Sandine 1975) at 30"C, Enterococcus faecalis in M17 at 37"C, and Pseudomonas species and Escherichia coli were grown in Luria-Bertani medium (Maniatis et al. 1982) at 30°C and 37"C, respectively. Clostridia were propagated in Reinforced Clostridial medium (Oxoid). Agar media were prepared by adding 1.5% granulated agar to broth and overlay agar was prepared with 0.7% agar. Chemical reagents were obtained from Sigma. Bacteriocin detection

Lactobacilli were examined for bacteriocin production by both direct (Tagg et al. 1976) and deferred (Kkkessy & Piguet 1970; Barefoot & Klaenhammer 1983) methods with a streak, spot or single colony isolates. Both methods of bacteriocin detection were repeated with the addition of 68 U per ml of filter-sterilized catalase (bovine liver, 40000 U/mg; EC 1.11.1.6) added to all agar media to eliminate antagonism due to hydrogen peroxide. The agar well diffusion method of Tagg & McGiven (1971) was used to test for bacteriocin production in the supernatant fluid from overnight broth and clotted milk cultures. Lactobacillus helveticus 1829 was grown for 16 h in 10% reconstituted skim milk (RSM; Golden Vale, Co. Cork) at 37°C. The clotted culture was centrifuged at 16000 g for 10 min and the supernatant fluid was filtersterilized through a 0.45 pm filter before it was placed in the wells for the assay. Helveticin V-1829 activity was assayed by an adaptation of the critical dilution method used for the assay of bacteriocins (Mayr-Harting et al. 1972; Joerger & Klaenhammer 1986) and Lact. delbrueckii subsp. bulgaricus 1489 was helveticin V-1829. In addition, lactobacilli were examined by direct or deferred (or both) antagonism testing. * Inhibition at 51 200 AU/ml of helveticin V-1829 only. DPC, National Dairy Products Research Centre, Moorepark, Cork; CNRZ, Culture Collection of Station de Recherches Laitieres, INRA, France, courtesy of P.-J. Cluzel; NCDO, National Collection of Dairy Organisms, Reading, England ; NCSU, Department of Food Science Culture Collection, North Carolina State University, Raleigh, NC, courtesy of T. R. Klaenhammer; UCC, Department of Food Microbiology, University College, Cork.

BACTERIOCIN F R O M L A C T . H E L V E T l C U S

routinely used as the indicator culture. T h e bacteriocin titre was defined as the reciprocal of the highest dilution showing complete inhibition of the indicator lawn and was expressed in activity units (AU) per ml. Production studies on helveticin V-1829

Lactobacillus helveticus 1829 cells, washed in MRS broth to remove media constituents, were inoculated (1 %) into 1 1 of the same medium and incubated at 37°C. Samples were aseptically removed over a 24 h period to determine optical density (O.D.) at 580 nm and AU/ml of helveticin V-1829. In order to examine production of helveticin V-1829 at controlled pH, an L.H. fermentation vessel (Model 502D; L.H. Fermentation, Stoke Poges, Bucks) which was connected to an automatic pH controller was used. One 1 each of MRS broth was adjusted to pH 5.0, 5.5, 6.5 and 7.0, and inoculated (3%) with an overnight culture of Lact. helveticus 1829 previously washed in MRS medium. An 8% ammonium hydroxide solution was used to maintain constant pH during fermentation. The temperature was maintained at 37°C and the culture was agitated at 150 rev/min, while being purged with N,. O.D. at 580 nm and AU/ml of helveticin V-1829 were examined over a 24 h period. Partial purification of helveticin V-1829

Lactobacillus helveticus 1829 was inoculated (3%) into an L.H. fermentation vessel containing 1 1 of a semi-defined MRS medium. The basal semi-defined medium contained (g/l): casamino acids (Difco), 10; yeast extract, 0.5; glucose, 10; sodium-acetate, 5; K,HP04, 2; MgSO, : 7H,O, 0.08; FeSO, : 7H20, 0.004; MnCI, : 4H,O, 0.0012; and Tween 80, 0.1 %; this was sterilized at 121°C for 15 min. The other ingredients were prepared as stock solutions, filter-sterilized and added aseptically to the basal medium in the following concentrations (mg/l) : trytophan, 100; adenine-sulphate, 5 ; guaninehydrochloride, 5; xanthine, 5; uracil, 5; biotin, 0.01 ; calcium-pathothenate, 1 ; nicotinic acid, 1 ; pyridoxalhydrochloride, 0.2; pyridoxine-hydrochloride, 1 ; and riboflavin, 0.1. T h e p H of the medium was maintained at 5.5, otherwise conditions of fermentation were as previously described. After incubation for 18 h, the cells were removed by centrifugation at 16 000 g for 15 min at 4°C. Ammonium sulphate was added slowly to the supernatant fluid to 30% saturation at 4°C. After gentle stirring, the mixture was allowed to rest for 1 h, followed by centrifugation at 16000 g for 1 h at 4°C to remove the protein precipitate. The supernatant fluid was brought to 60% saturation under the same conditions. The precipitate was collected after centrifugation, dissolved in 0.1 mol/l sodium acetate buffer, p H 5.3 and dialysed against the same buffer

301

in Visking dialysis tubing (pore exclusion limit-I0 000 Da; Medicell, London) to remove the ammonium sulphate. Effect of heat treatment and pH on bacteriocin activity

Cell-free supernatant fluids containing helveticin V-1829 and helveticin J were dialysed in 50 mmol/l phosphate buffer (pH 6.5) and adjusted to give bacteriocin samples of equal titre. One ml volumes of each bacteriocin were placed in a boiling water bath. Samples were removed at 0, 1, 5, 15, 30 and 45 min and titred for loss of activity on Lact. delbrueckii subsp. bulgaricus 1489. In addition, samples of helveticin V-1829 (200 AU/ml) were heated at 45, 50 and 60°C. Samples were removed at 0, 5, 15, 30, 45, 60, 90 and 120 min, and titres were determined. T h e p H of cell-free supernatant fluids was adjusted from approximately 4.0 to 1.0, 2.5, 5.0, 6.0, 7.0, 7.5, 8.0 and 9.0 with HCI or NaOH. In addition, after incubation at room temperature for 5, 15, 30 and 45 min, portions of samples at p H 7.5, 8.0 and 9.0 were readjusted to p H 4.0. T h e titres of all samples were determined. Sensitivity of helveticin V-1829 to dissociating agents and proteoiytic enzymes

Crude helveticin V-1829 was treated with the following reagents at a final concentration of 1 % : Tween 20, Tween 80, Triton X-100, N-lauryl sarcosine and sodium dodecyl sulphate (SDS). The effect of 0.1% SDS, 8 mol/l urea, 1 mmol/l dithiothreitol (DTT), 0.2% /?-mercaptoethanol and combinations of these reagents were also examined. Controls consisted of helveticin V-1829 or detergent, in 50 mmol/l sodium phosphate buffer, p H 6.2. All samples and controls were incubated at 37°C for 6 h and titred for helveticin V-1829 activity. T h e following enzymes ( 1 mg/ml) were dissolved in the appropriate buffers as follows; proteinase K (type XI; Sigma), 0.05 mol/l sodium-acetate, p H 6.2; trypsin (EC 3.4.21.4), 0.04 mol/l Tris-hydrochloride, 0.01 mol/l CaCI, , pH 6.2; papain (EC 3.4.22.2), 0.05 mol/l sodium-acetate, at both p H 4.5 and 5 . 5 ; ficin (EC 3.4.22.3), 0.02 mol/l cysteine-hydrochloride, 0.01 mol/l disodium EDTA, 0.15 mol/l NaCI, pH 6.2; pronase (Boehringer, Mannheim), 0.1 mol/l sodium borate, 5 mmol/l CaCl,, 1 mmol/l CoCl,, p H 6-2; lysozyme (Grade 1 ; Sigma), and phospholipase C (EC 3.1.4.3), 50 mmol/l potassium phosphate, pH 6-3. Samples containing partially purified helveticin V-1829 (800-1600 AU/ml) were dialysed in the appropriate buffers before being treated with the enzyme. Controls included buffer ; buffer and enzymes ; buffer and heat-inactivated enzymes; and buffer plus helveticin V- 1829. Helveticin V1 8 2 h n z y m e reaction mixtures and all controls were

302 ELAINE E. VAUGHAN ET A L .

incubated at 37°C for 2 h, after which titres of helveticin V-1829 activity were determined. Bactericidal action of heivetlcin V-1829

Cell-free supernatant fluid containing helveticin V-1829 was dialysed in 0.1 mol/l sodium acetate, p H 5.3 for 18 h at 4°C and diluted to obtain a titre of 200 AU/ml. Cells from a log-phase culture of Lact. helveticus 1844 ( 5 h, 37°C) were washed with 0.1 mol/l sodium acetate buffer, pH 5-3 and added to 10 ml of the helveticin V-1829 preparation at an initial population of 1.35 x lo6 cfu/ml. After 0, 3 and 5 h at 37"C, O.D. at 580 nm and cfu/ml were determined. I n a separate experiment, cell-free supernatant fluid was dialysed in 50 mmol/l potassium phosphate buffer, pH 6.2 and diluted to obtain 5, 25 and 200 AU/ml of helveticin V-1829. Log-phase Lact. delbrueckii subsp. bulgaricus 1489 cells were washed in the same buffer and added to the preparation to yield 2.4 x 10' cfu/ml. T h e effect of helveticin V-1829 on stationary phase cells was also investigated. Stationary phase cells of Lact. delbrueckii subsp. bulgaricus 1489 were washed twice in 50 mmol/l potassium phosphate buffer and diluted to obtain 1 x lo8 cfu/ml prior to treatment with 25 AU/ml of bacteriocin. All assays were incubated at 37°C. O.D. at 580 nm and cfu/ml were determined at 0, 2 and 4 h. Viable counts were performed on M R S agar. Controls included incubating the indicator cells with the helveticin V-1829 preparation which had been boiled to destroy activity.

mixture was overlayed with mineral oil and subjected to 30 cycles of amplification. T h e cycle profile included : denaturation for 2 min at 94"C, 2 min at 45°C for annealing, and 4 min at 68°C for extension of the primers. Where required, the amplified DNA was labelled with digoxigenindUTP by adding 4 p1 of d N T P labelling mixture from the nonradioactive D N A labelling and detection kit (Boehringer, Mannheim) to the PCR reaction. For purification purposes, 20 pl of the amplified labelled DNA was run on a 0.7% agarose gel and was recovered using the Geneclean I1 Kit (BIO 101, La Jolla, CA). The procedures for hybridization and detection of homologous sequences were as described in the instruction manual for the nonradioactive DNA labelling and detection kit. RESULTS

Detection of helveticin V-1829

During a routine screening of Lactobacillus strains prior to use in conjugation experiments, Lact. helveticus 1829 and

DNA Isolation and manipulations

Total genomic DNA was extracted from Lactobacillus strains by the procedure described by Shimizu-Kadota et al. (1983) with the following modifications : the cuItures were grown in 500 ml of MRS broth containing 40 mmol/l DL-threonine to an O.D. at 580 nm of 0.7; and 80 pg/ml of mutanolysin was used instead of J l-induced endolysin. Restriction enzymes were obtained from Boehringer (Dublin) and were used according to the manufacturer's instructions. DNA was transferred from agarose gels to nitrocellulose filters by the method of Southern (1975) as modified by Wahl et al. (1979). Primers 18 nucleotides long were synthesized with a Beckman (San Ramon, CA) System 200A DNA synthesizer. A DNA thermal cycler (Perkin Elmer Cetus, Norwalk, C T ) was used for PCRmediated DNA amplification. The polymerase chain reaction contained: genomic template DNA, 0.1 pg; 4 pl of a d N T P stock (containing dATP, dCTP, d G T P and d T T P at a concentration of 2.5 pmol/l each); 300 ng each of sense and antisense primers; 5 pl of 10 x PCR buffer, and 0.5 pl (5 U/pl) of Tay polymerase (Perkin Elmer Cetus) made up to a final volume of 50 pl with sterile distilled water. T h e

Flg. 1 Antagonistic activity of Lactobacillus helveticus 1829 against Lactobacillus helveticus 1844 demonstrated by (a) the deferred antagonism procedure and (b) the agar well diffusion method as

described in Materials and Methods

BACTERIOCIN FROM L A C T . HELVETICUS

303

Lact. helveticus 1844 were examined for the production of bacteriocins. When the deferred procedure of antagonism was used, Lact. helveticus 1829 showed antimicrobial activity against strain 1844 (Fig. 1). As there is no direct contact between the producer and indicator in this method, the inhibition was not due to bacteriophage. Addition of catalase to anaerobically incubated MRS plates did not affect bacteriocin activity indicating that the antagonism was not caused by hydrogen peroxide. The inhibitory agent was also secreted into supernatant fluids of cultures grown in either MRS broth (neutralized to pH 6.0) or milk as demonstrated by the well diffusion assay (Fig. 1). Furthermore, lactic acid was ruled out as the inhibitory factor as MRS broth brought to pH 3.8 with lactic or acetic acid had no effect on strain 1844, and the compound was also nondialysable in dialysis tubing with a molecular exclusion limit of 10 000 Da.

culture when tested on the sensitive indicator strain 1844. Production of helveticin V-I829 during anaerobic growth under constant p H conditions was examined in an effort to increase this yield (Fig. 2). At two of the pH values tested (pH 5.0, Fig. 2a, and 6.5, Fig. 2c), helveticin V-1829 production was detectable from the middle to late log phase, but halted once the cultures were in the stationary phase of growth. Production at p H 5.0 gave a twofold greater titre than under uncontrolled p H conditions. At p H 6.5, production was low and the bacteriocin activity was unstable. At p H 7.0 (Fig. 2d), after 5 h growth, the producing strain lysed and a minimal amount of helveticin V-1829 was produced. Maximum yields were obtained at pH 5-5, where bacteriocin production was observed from the middle log phase into the stationary phase of growth and a titre of 800 AU/ml was detected after incubation of the culture for 12 h (Fig. 2b).

Optimization of production of helveticin V-1829

Effect of dissociating agents on helveticin V-1829 activity

Helveticin V-1829 was secreted into MRS broth during the logarithmic phase of growth and 200 AU/ml was the highest titre obtained in the supernatant fluid of a 16 h

-

0

T o determine if helveticin V-1829 was present in an aggregated form, crude bacteriocin preparations were treated

800

-

- 600 -

+

d

- 400 >aE -

- 200 I

I

I

0

Time I h )

Time I h 1

800

I

600

600

E 0 n 0

d

< 2

%

o

400

400

1

0

E ; d

m

A 0

200

200

0

0

-I

10

20 Tlmc ( h 1

30

20

I0

30

Time I h I

Fig. 2 Growth of Lactobacillus helveticus 1829 ( 0 )and production of helveticin V-1829 (+) in MRS maintained at a constant pH of 5.0 (a), 5.5 (b), 6.5 (c) and 7.0 (d)

304

E L A I N E E. V A U G H A N E T A L .

Table 2 Effect of dissociating agents on helveticin V-1829 activity

Activity (AU/ml)

Helveticin V-1829 Dissociating agent

+ treatment

None

1600

Dissociating agent control* -

Nonionic detergents

Tween 20 Tween 80

Triton X-100

1600 1600 1600

0 0 0

800 400 800

200

1600 1600 1600

0 0 0 0

400

0

Anionic detergents

1% SDS 0.1'X SDS

N-lauryl sarcosine 8 mol/l Urea 1 mol/l DTT 0.1O h SDS + 1 mmol/l DTT 0.2% fl-mercaptoethanol 0.1% SDS + 0.2% P-mercaptoethanol

* When

400

0

100

inhibition was observed, it was measured as equivalent

AU/ml. with a variety of dissociating agents, including nonionic and anionic detergents. None of the reagents resulted in an increase in bacteriocin activity (Table 2). In contrast, the anionic agents SDS and N-lauryl sarcosine significantly reduced helveticin V-1829 activity (Table 2). T h e fact that the titres remained the same or decreased indicates that these agents are not capable of dissociating any aggregates of helveticin V-1829 if they are present. In control experiments, treatment with 1% SDS alone had an inhibitory effect equivalent to 200 AU/ml on the indicator (0.1% SDS had no effect) which explains the higher titres obtained after treatment with 1% SDS compared with 0.1 Yo. Susceptibility of partially purlfied helveticin V-1829 to proteoiytic enzymes

Initial attempts at digesting helveticin V-1829 in M R S supernatant fluids with proteases were inconclusive, possibly because of interference due to the high level of proteins in the MRS medium. An alternative growth medium was sought for the producer strain which would contain lower levels of proteins and peptides but would allow good growth of the producer to result in reasonable yields of helveticin V-1829. The chemically defined medium of Thomas et al. (1979), modified for lactobacilli by the addition of Tween 80 (0.1%) and sodium acetate (0.5?40), did not support enough growth of the producer strain for sufficient

yields of helveticin V-1829. However, the addition of casamino acids, tryptophan and yeast extract (0.05%) allowed growth of Lact. helveticus 1829 and production of helveticin V-1829 in the semi-defined medium, at constant p H 5.5, comparable with that obtained in MRS broth. Helveticin V-1829 precipitated between 30 and 60% ammonium sulphate saturation with a recovery of 90 to 100% and this procedure was used to make concentrated preparations of the bacteriocin. 800-1600 AU/ml of partially purified helveticin V-1829 was totally inactivated by treatment with proteinase K, ficin, trypsin and pronase but papain had no effect on activity. In addition, phospholipase C and lysozyme did not inactivate the bacteriocin. None of the controls inhibited the indicator. T h e sensitivity of helveticin V-1829 to the proteolytic enzymes indicate that it is proteinaceous in nature. pH and heat stability

T h e p H of culture supernatants of Lact. helveticus 1829 were adjusted to a range of values from p H 1.0 to 9.0. After incubation at room temperature for 45 min, full activity was retained in samples at p H 2-5 to 6.5; at p H 7.0, activity was significantly reduced, and at p H 1.0 and above p H 7.0, no activity was observed. Readjustment of samples at pH 7.5, 8.0 and 9.0 back to p H 4-0 failed to restore activity. T h e inhibitory activity in cell-free supernatants (200 AU/ml) was totally destroyed by incubation at 60°C for 15 min but full activity was present after incubation at 45°C for 120 min. Incubation at 50°C for 15 rnin reduced the activity by 50% and after 30 rnin, the inhibitor was totally inactivated, indicating its heat-sensitive nature. Since helveticin J has also been reported to be heat-labile (Joerger & Klaenhammer 1986), both bacteriocins were examined for heat stability under identical conditions. I n three independent trials, helveticin V-1829 was found to be more heatsensitive than helveticin J. Regardless of the initial titre, helveticin V-1829 (200-800 AU/ml), was totally inactivated within 1 rnin at 100°C. However, the helveticin J sample (800 AU/ml) still retained 12.5% (100 AU/ml) of activity after incubation at 100°C for 30 min. Helveticin V-1829 retained its activity for several months when stored at 4°C or -20°C. Bactericidal actlvlty of helveticin V-1829

T h e effect of helveticin V-1829 on sensitive indicators was examined to establish if it demonstrated a bactericidal or a bacteriostatic mode of action. T h e addition of 25 and 200 AU/ml of helveticin V-1829 to Lact. delbrueckii subsp. bulgaricus 1489 cells resulted in a similar reduction in the cell population by 96 and 95%, respectively, while the lower concentration of 5 AU/ml killed 77% of the population in

B A C T E R I O C I N F R O M L A C T . HELVETICUS

4 h (Fig. 3a). Twenty-five AU/ml of bacteriocin killed 72% of a population of stationary phase Lact. delbrueckii subsp. bulgaricus 1489 cells (Fig. 3b) indicating that stationary phase cells of this culture were more resistant to the effect of helveticin V-1829 than log-phase cells. In another experiment, 200 AU/ml of the bacteriocin reduced a population of lo6 cfu/ml of Lact. helveticus 1844 by 98% in 5 h

305

(Fig. 3c). T h e O.D. remained constant throughout these experiments. Thus, helveticin V-1829 exhibited a bactericidal mode of action against both Lact. helveticus 1844 and Lact. delbrueckii subsp. bulgaricus 1489 indicator cells but cell lysis was not detected.

lnhlbltory spectrum of helvetlcln V-1829

0

0

0

2

I

Control 200AU/ml 25AU/ml 5AU/ml

d 5

3

0 Control 25AU/ml

Helveticin V-1829 exhibited antagonistic activity against closely-related species particularly other Lact. helveticus strains (Table 1). Partially purified helveticin V-1829 inhibited two Lact. acidophilus strains and Lact. delbrueckii subsp. bulgaricus 11 but only when concentrated to 51 200 AU/ml. T h e producer strain Lact. helveticus 1829 also lost its immunity to helveticin V-1829 at this higher concentration. All of the other bacteria tested were insensitive to the bacteriocin regardless of the assay procedure. Three different previously-characterized bacteriocins, namely lactacin B (Barefoot & Klaenhammer 1983, 1984) and lactacin F (Muriana & Klaenhammer 1987, 1991), produced by Lact. acidophilus N2 and 11 088, respectively, and helveticin J, were examined for their effect on helveticin V-1829-sensitive indicator strains. Single colony isolates of all the bacteriocin producers inhibited the indicators of helveticin V- 1829.

DNA homology between the helveticin J gene and genomlc DNA of the helveticin V-1829 producer

0

0

I

2

3 Time ( h

4

5

Control EOOAU/ml

6

1

Fig. 3 Effect of varying concentrations of helveticin V-1829 and boiled helveticin V-1829 (controls) on indicators (a) Lactobacillus delbrueckii subsp. bulgaricus 1489, (b) stationary phase cells of Lactobacillus delbrueckii subsp. bulgaricus 1489, and (c) Lactobacillus helveticus 1844

The helveticin J structural gene (hlv) was used as a probe against total genomic DNA from Lact. helveticus 1829 to determine conclusively if Lact. helveticus 48 1 and 1829 produce the same or related bacteriocins. It is noteworthy that no plasmid DNA has been detected in Lact. helveticus 1829 and attempts to obtain non-producing variants with a variety of plasmid-curing agents was unsuccessful (results not shown). Probe DNA encoding hlv was obtained by employing the polymerase chain reaction on total DNA isolated from Lact. helveticus 481 with specific primers which were synthesized based on the published nucleotide sequences of hlv (Joerger & Klaenhammer 1990). Using pH1, 5’ATT CCA AGA TAA C T T ATA3’ and pH2, 5’AAG CAT T T A AAT GAA ACA3’ for amplification of the sense and antisense strands of hlv, the polymerase chain reaction resulted in a 1 kb product. No DNA was detected when Lact. helvericus 1829 total DNA was used as the template in an identical reaction. T h e 1 kb fragment, labelled with digoxigenin-dUTP during the amplification procedure, was examined for hybridization to total DNA isolated from Lacr. helveticus 1829, digested separately with restriction endonucleases EcoRI and Hind11 (Fig. 4). Lac-

306

ELAINE E. VAUGHAN E r A L .

(b)

( 0 )

1

2

3

4

5

6

7

I

2

3

4

5

6

7

Fig. 4 (a) Agarose gel electrophoresis of genomic DNA from Lactobacillus helveticus strains. Lanes 1, 2 and 3, EcoRI-digested total genomic DNA from Lact. helveticus 481, 1829 and 384, respectively; lane 4, EcoRIIHindIIIdigested bacteriophage 1 DNA; lanes 5 , 6 and 7, HindII-cleaved genomic DNA from Lact. helveticus 481, 1829 and 384, respectively. (b) Hybridization of the labelled 1 kb PCR product to the DNA in panel a

tobacillus helveticus 481 and 384 (non-bacteriocin producer) total DNAs were included as positive and negative controls, respectively. There are no HindII sites in the 1 kb region but since there is a single EcoRI site, two fragments of an EcoRI-digested chromosome should be resolvable after hybridization with the probe. The 1 kb hlv probe hybridized strongly to two bands in EcoRI-digested DNA from 481 (Fig. 4) in agreement with the restriction map of hlv. Although the probe does not possess a HindII site, a second hybridizing band in Hind 11-digested 481 DNA was observed. However, hybridization to Lact. helveticus strains 1829 and 384 was extremely faint (not detectable on photograph), indicating lack of significant homology between hlv and total DNA from these strains.

DISCUSSION

Lactobacillus helveticus 1829 demonstrated antagonistic activity against Lact. helveticus 1844 under conditions that eliminated antibiosis due to acids, pH, hydrogen peroxide and bacteriophage, and therefore the inhibition was most

likely to be caused by a bacteriocin-like substance. The inhibitory agent, helveticin V-1829, designated in accordance with the recommendations for the naming of antibiotics (Hobby 1964), was shown to be a true bacteriocin according to the criteria outlined by Tagg et al. (1976). Helveticin V-1829 exhibited a narrow spectrum of activity against related species including Lact. helveticus and Lact. delbruecki; subsp. bulgaricus strains. It was bactericidal in its mode of activity against the sensitive indicators Lact . helveticus 1844 and Lact. delbrueckii subsp. bulgaricus 1489. Both 25 and 200 AU/ml reduced the viability of the Lact. delbrueckii cell population by a similar value and at similar rates. A concentration of approximately 25 AU/ml may be the saturating level of helveticin V-1829 which can act on a population of lo* cfu of Lact. delbrueckii subsp. bulgaricus 1489 rendering higher bacteriocin concentrations redundant. Unlike many other bacteriocins, its activity was weak and slow, taking hours rather than minutes to exert its lethal action. Caseicin 80 is similar to helveticin V-1829 in this respect requiring days to demonstrate its activity on the slow-growing indicator Lact. casei B109 (Rammelsberg et al. 1990). The physiological state of the indicator culture has been shown to have a strong influence on susceptibility to the lethal action of a bacteriocin, with actively multiplying cells being the most sensitive (Tagg et al. 1976). The bacteriocin has a more rapid bactericidal effect on Lact. delbrueckii subsp. bulgaricus 1489 than on Lact. helveticus 1844 which has a relatively slow doubling time of 168 min (results not shown). Although helveticin V-1829 could also kill cells in the stationary phase of growth, its bactericidal effect was significantly reduced. A number of studies have shown that control of the medium p H is a critical factor in bacteriocin production (Goebel et al. 1955; Joerger ?i Klaenhammer 1986). Optimal helveticin V-1829 yields were obtained at pH 5.5 in the semidefined MRS medium designed for Lact. helveticus 1829. Despite several attempts, the producer could not be grown successfully to stationary phase at p H 7.0 as lysis occurred after one log cycle of growth. This phenomenon is most likely due to the acidophilic nature of the microorganism, as lactobacilli tend to be inhibited by alkali or neutral medium (Kandler i? Weiss 1986). There is also a possibility that these growth conditions may induce a lysogenic phage, although a similar lytic effect could not be induced using mitornycin C (results not shown). A low level of bacteriocin was released during lysis and may represent the initial synthesis of helveticin V-1829 within the producer cell. T h e higher titre of helveticin V-1829 at pH 5.5 may be partly attributed to its stability at lower pH values but other factors may be involved such as decreased enzymatic digestion either in the cells or medium at this specific pH. It was interesting to speculate that helveticin V-1829

BACTERIOCIN FROM L A C T . HELVETlCUS

may be related to helveticin J (Joerger & Klaenhammer 1986) even though both bacteriocins are produced by different Lact. helveticus strains. They are both bacteriocins with a narrow activity spectrum and both are heat-sensitive in comparison to the heat-stable lactocin LP27 produced by Lact. helveticus LP27 (Upreti & Hinsdill 1973). In addition, the two bacteriocins have been inactivated by treatment with pronase, trypsin, ficin and proteinase K . An attempt to differentiate between the two bacteriocins with the host spectrum as a criterion was not successful. I n fact, all the bacteriocin producers examined in this study were capable of inhibiting the helveticin V-1829 indicators suggesting that certain cultures are more sensitive to bacteriocins in general. A closer examination of some of the physical properties of helveticin J and helveticin V- 1829 revealed that there were some differences between them. Although both bacteriocins have been reported as heat-sensitive, it was observed that helveticin V-1829 was significantly less stable at high temperatures than helveticin J when tested under identical conditions. Many bacteriocins are initially isolated as large macromolecular complexes that can be dissociated to release more active units. Lactocin 27, helveticin J and lactacin F were dissociated in the presence of SDS (Upreti & Hinsdill 1973; Joerger & Klaenhammer 1986; Muriana & Klaenhammer 1991) while treatment with 8 mol/l urea resulted in a 200-fold increase in activity for lactacin B (Barefoot & Klaenhammer 1984). An increase in activity was not detected when helveticin V-1829 was treated with a variety of dissociating agents. In addition, the detergent SDS had an adverse and therefore opposite effect on helveticin V-1829 compared with helveticin J. Anionic detergents often unfold proteins by complexing to the interior ‘hydrophobic core’ of their native structure which may affect their three-dimensional conformation. The reduction in helveticin V-1829 activity following treatment with SDS and Nlauryl sarcosine may be due either to partial denaturation of helveticin V-1829 or to disruption of its association with other molecules which have a stabilizing effect on its activity. DNA-DNA homology studies were used as being a rigorous method of comparing helveticin V-1829 with helveticin J. Since the gene for helveticin J has been cloned and sequenced (Joerger & Klaenhammer 1990), it was possible to amplify specifically hlv from genomic DNA of Lact. helveticus 481 by employing suitable primers. The PCR product generated was the correct size as determined from the hlv nucleotide sequence. An amplified product was not obtained in a similar reaction with Lact. helveticus 1829 genomic DNA as the template providing preliminary evidence that this strain does not have hlv or a related gene. The 1 kb probe hybridized to EcoRI-cleaved 481 giving the

307

pattern anticipated from the hlv restriction map. Very weak hybridization was observed to EcoRI and HindII-digested 1829 and 384 DNAs but these bands were a different size to the strong bands obtained for 481 DNA. Some weak hybridizing bands were also present in the 481 DNA and instead of one strong band lighting up in the Hind11 digest, two strong bands were detected. These extra bands are most likely due to repeated sequences in the 481 genome. Thus, the results of both a biochemical and genetic approach indicate that helveticin V-1829 is a novel bacteriocin and is not significantly related to helveticin J. ACKNOWLEDGEMENTS

We thank T.R. Klaenhammer and P.-J. Cluzel for providing strains. This work was supported by the European Community Biotechnology Action Programme (Contract no. BAP-0008-IRL). REFERENCES

BAREFOOT, S . F . & K L A E N H A M MTE.RR, . (1983) Detection and activity of lactacin B, a bacteriocin produced by Lactobacillus acidophilus. Applied and Environmental Microbiology 45, 180% 1815. BAREFOOT, S.F. & K L A E N H A M M ETR . R, . (1984) Purification and characterization of the Lactobacillus acidophilus bacteriocin lactacin B. Antimicrobial Agents and Chemotherapy 26,

32%334. D E K L E R KH.C. , (1967) Bacteriocinogeny in Lactobacillus firmenti. Nature (London) 214, 609. D E K L E R KH.C. , & COETZEE, J . N . (1961) Antibiosis among lactobacilli. Nature (London) 192, 340-341. D E K L E R KH.C. , & S M I T ,J.A. (1967) Properties of a Lactobacillus fermenti bacteriocin. Journal of General Microbiology 48, 309-3 16. GOEBEL,W.F., B A R R Y ,G . T . , J E S A I T I S ,M.A. & M I L L E RE.M. , (1955) Colicin K. Nature (London) 176, 7 W 701. HOBBY,G . L . (1964) Round table on nomenclature of antibiotics. Antimicrobial Agents and Chemotherapy, pp. 1102-1 103. J O E R G E R , M.C. & K L A E N H A M MT E.RR,. (1986) Characterization and purification of helveticin J and evidence for a chromosomally determined bacteriocin produced by Lactobacillus helveticus 481. Journal of Bacteriology 167, 439-146. J O E R G E RM.C. , & KLAENHAMME TR . R, . (1990) Cloning, expression, and nucleotide sequence of the Lactobacillus helveticus 481 gene encoding the bacteriocin helveticin J. Journal of Bacteriology 172, 6339-6347. KANDLER , & WEISS,N . (1986) Regular, nonsporing gram0. positive rods. In Bergey’s Manual of Systematic Bacteriology ed. Sneath, P.H.A., Mair, N.S., Sharpe, M.E. & Holt, J.G. Vol. 2, pp. 120g1418. Baltimore: Williams and Wilkins. KEKESSY,D.A. & P I G U E TJ.D. , (1970) New method for detecting bacteriocin production. Applied Microbiology 14, 6 1 6 622.

308 ELAINE E . VAUGHAN ET A L .

K L A E N H A M M E R , T .R . ( 1988) Bacteriocins of lactic acid bacteria. Biochimie 70, 337-349. L I N D G R E NS,. E . & DOBROGOSZ, W.J. (1990) Antagonistic activities of lactic acid bacteria in food and feed fermentations. F E M S Microbiology Reviews 87, 149-173. M A N I A T I ST, . , F R I T S C H E , . F . & S A M B R O O KJ ,. (1982) Molecular Cloning: a Laboratory Manual. New York: Cold Spring Harbor Laboratory. M A Y R - H A R T I NA., G , HEDGES, A.J. & B E R K L E YR,. C . W . (1972) Methods for studying bacteriocins. In Methods in Microbiology ed. Norris, J.R. & Ribbons, D.W. Vol. 7A, pp. 3 15-422. New York: Academic Press. M U R I A N AP,. M . & K L A E N H A M M ETR. R, . (1987) Conjugal transfer of plasmid+mcoded determinants for bacteriocin production in Lactobacillus acidophilus 88. Applied and Environmental Microbiology 53, 553-560. M U R I A N AP, . M . & K L A E N H A M M ETR. R , . (1991) Purification and partial characterization of lactacin F, a bacteriocin produced by Lactobacillus acidophilus 11 088. Applied and Environmental Microbiology 57, 114-121, R A M M E L S B E RM G ., , M U L L E R ,E. & R A D L E RF. , (1990) Caseicin 80: purification and characterization of a new bacteriocin from Lactobacillus casei. Archives of Microbiology 154, 249-252. S H I M I Z U - K A D O TM A ., , S A K U R A T I ,. & T S U C H I DNA. , (1983) Prophage origin of a virulent phage appearing on fermentations of Lactobacillus casei S-1. Applied and Environmental Microbiology 45, 669-674.

S O U T H E RE N . M, . (1975) Detection of specific sequences among DNA fragments separated by gel electrophoresis. Journal of Molecular Biology 98, 505-517. T A G G J, . R . & M C G I V E NA.R. , (1971) Assay system for bacteriocins. Applied Microbiology 21, 943. .W , . (1976) T A G G J, . R . , D A J A N IA , . S . & W A N N A M A K EL R Bacteriocins of gram-positive bacteria. Bacteriological Reviews 40,722-756. T E R Z A G H IB, . E . & S A N D I N E ,W . E . (1975) Improved medium for lactic streptococci and their bacteriophage. Applied Microbiology 29, 807-8 13. T H O M A ST, . D . , E L L W O O DD, . C . & L O N G Y E A V R ,. M . C . (1979) Change from homo- to heterolactic fermentation by Streptococcus lactis resulting from glucose limitation in anaerobic chemostat cultures. Journal of Bacteriology 138, 10F-117. U P R E T I ,G . C . & H I N S D I L LR,. D . (1973) Isolation and characterization of a bacteriocin from a homofermentative Lactobacillus. Antimicrobial Agents and Chemotherapy 4, 487494. U P R E T I ,G . C . & H I N S D I L L ,R . D . (1975) Production and mode of action of lactocin 27: bacteriocin from a homofermentative Lactobacillus. Antimicrobial Agents and Chemotherapy 7, 139-145. W A H L ,G . M . , S T E R N M , . & S T A R KG , . R . (1979) Eficient transfer of large DNA fragments from agarose gels to diazobenzyloxymethyl paper and rapid hybridization by using dextran sulphate. Proceedings of the National Academy of Science U S A 76,3683-3687.