Response of Salmonella choleraesuis LT2 Spheroplasts and Permeabilized Cells to the Bacteriocin AS-48

Abstract

The bacteriocin AS-48 was not active against intact cells of Salmonella choleraesuis LT2 at neutral pH, but it was very effective on spheroplasts, suggesting that the outer membrane (OM) acts as a protective barrier. Cells sublethally injured by heat or treated with OM-permeabilizing agents (i.e., EDTA and Tris) became sensitive to AS-48. The combination of two or more treatments decreased the amount of AS-48 required for cell killing. The activity of AS-48 against heat-injured cells did not change significantly in the pH range of 4.0 to 8.0. AS-48 showed bactericidal activity against intact cells of Salmonella at pH 4.0. The potency of AS-48 increased greatly when the bacteriocin was dissolved at pH 9.0.

The antimicrobial efficiency of different methods of food preservation can be improved through application of the hurdle concept: two or more antimicrobial agents acting synergistically at suboptimal levels are more effective than each of them alone at the optimal level (13). This approach is particularly useful not only because it improves the stability of foods that need to be stored under adverse conditions but also because the acceptability of foods is enhanced (13). Microbial cells sublethally injured by different stressing conditions may become sensitive to physical and chemical agents to which healthy cells are resistant (16). As an example, sublethally injured gram-positive and gram-negative bacteria are reported to become sensitive to bacteriocins of lactic acid bacteria (11, 17–19). Nisin, used in combination with a chelating agent (EDTA), exhibited a bactericidal effect on both gram-positive and gram-negative bacteria (1–3, 21). Accordingly, bacteriocins (either alone or in combination with other antimicrobial barriers) may be useful tools for the implementation of methods intended to substantially reduce the load of food borne pathogens and food spoilage bacteria.

The bacteriocin AS-48 produced by Enterococcus faecalis subsp. liquefaciens is a basic peptide (7.14 kDa; pI, 10.5) that exhibits a broad spectrum of activity against gram-positive and gram-negative bacteria, such as Escherichia coli, Myxococcus, Agrobacterium, Rhizobium, etc. (4, 5). However, sensitive gram-negative bacteria (and especially members of the family Enterobacteriaceae) require at least 10-times-higher concentrations of AS-48 for inhibition (7). Since the primary target of AS-48 is the cytoplasmic membrane (6), it is suspected that the outer membranes (OM) of gram-negative bacteria are responsible for their resistance. In this work we used several physicochemical treatments to increase the OM permeability as a way to enhance the activity of AS-48 against Salmonella choleraesuis LT2. The influence of pH on AS-48 activity against this bacterium was also studied.

S. choleraesuis LT2 (CECT 722, from the Spanish Type Culture Collection) was propagated in Luria broth (Gibco, Life Technologies, Paisley, Scotland) or Trypticase soy agar (TSA; Menarini, Barcelona, Spain) at 37°C. Purified samples of bacteriocin AS-48 were obtained from E. faecalis A-48-32 cultures according to the method of Gálvez et al. (4) and stored at −20°C.

Spheroplasts were prepared according to the method described by Rassoulzadegan et al. (15), using a final concentration of lysozyme (Boehringer Mannheim, Barcelona, Spain) of 2 mg/ml. The spheroplasting efficiency was determined to be above 90% by plating on TSA without osmotic stabilizers. Spheroplast suspensions (optical density at 620 nm, ca. 0.345) were treated with AS-48, and lysis was monitored turbidimetrically at 620 nm.

For the combined treatments with AS-48 and physicochemical agents, cells from 2 ml of an overnight culture of S. choleraesuis LT2 in Luria broth were harvested by centrifugation, washed twice in the same volume of sterile 10 mM sodium phosphate buffer (PB; Probus, Barcelona, Spain), pH 7.2, and finally resuspended in 10 ml of PB (optical density at 620 nm, 0.155; ca. 10⁸ CFU/ml), vortexed, and dispensed in aliquots (1 ml). AS-48 was added at different concentrations to cell suspensions immediately before treatment with the different agents. Cell viability was determined immediately after treatment (time zero [T₀]) and after 6 h of incubation at 37°C (T₆). Samples were serially diluted in sterile saline solution and plated on TSA plates in triplicate. The average number of CFU was determined after 24 h of incubation at 37°C. The logarithmic reduction factor (LRF) was calculated by the following formula: LRF = log CFU per milliliter in controls at T₀ minus log CFU per milliliter in treated samples at T₆ (16).

For heat treatment, cell suspensions were mixed with different bacteriocin concentrations (ranging from 10 to 200 μg/ml) and immediately heated at 60°C for 3 min, after which samples were cooled on ice.

To test the effect of pH on sensitivity to AS-48, intact cell suspensions in different buffers (i.e., citric acid-sodium citrate [pH 4.0], sodium phosphate [pH 7.0], and glycine-NaOH [pH 9.0]; each buffer was at a concentration of 100 mM) (citric acid and sodium citrate were from Sigma Chemical Co., St. Louis, Mo.; glycine and NaOH were from Panreac, Barcelona, Spain) were treated with AS-48 (50 μg/ml, final concentration) dissolved in deionized water or in the above-mentioned buffers. When the combined effect of heat and pH was tested, intact cells (2 × 10⁸ CFU/ml) in 10 mM PB, pH 7.2, were treated with AS-48 dissolved in water or in citrate buffer (pHs 3.0 and 4.0), sodium acetate buffer (pH 5.0; Probus), PB (pHs 6.0, 7.0, and 8.0), or glycine-NaOH (pH 9.0) and then heated at 60°C for 3 min.

EDTA (Boehringer Mannheim) was used as a chelating agent (at a final concentration of 50 mM in 10 mM PB, pH 7.2, or 20 mM in 50 mM Tris-HCl, pH 7.2, plus 0.1 M NaCl). As the OM-permeabilizing cation we used 100 mM Tris(hydroxymethyl)aminomethane, adjusted to pH 8.0 with 6 N HCl. Tris was purchased from Boehringer Mannheim. The other chemicals (HCl, magnesium sulfate, NaCl, and sucrose) were from Panreac.

Our initial results indicated that the bacteriocin AS-48 had no appreciable effect on the viability of S. choleraesuis LT2 when tested on intact cell suspensions at neutral pH even at high concentrations (200 μg/ml with 2 h of incubation at 37°C; data not shown). However, AS-48 induced a rapid lysis of spheroplasts from this bacterium (Fig. 1), even at the lowest concentration tested (10 μg/ml). These data are a clear indication of the protective role of the cell OM.

FIG. 1.

Effect of AS-48 (10-μg/ml final concentration) on spheroplasts prepared from S. choleraesuis LT2. Lysis was monitored at 620 nm. Symbols: ∗, control spheroplasts; ■, spheroplasts plus deionized water; ▴, spheroplasts plus 10 μg of AS-48 in deionized water.

We tested alternative procedures of spheroplast formation to increase the sensitivity of Salmonella to AS-48, such as heat treatment or incubation with permeabilizing agents, as well as the effect of pH on the antimicrobial activity of AS-48. First, we determined the time and temperature required to cause sublethal cell injury: heating at 60°C for 3 min reduced the viable population to 2% of the initial cell numbers (which decreased from 10⁸ to 2 × 10⁶ CFU/ml). Similar heat treatments are applied to some foods, such as liquid egg products. Nevertheless, we noted that a large fraction of heat-injured cells were able to repair the heat damage during the 6-h incubation period at 37°C in PB, and the viable counts increased to ca. 3 × 10⁷ CFU/ml. When heat treatment was carried out in the presence of increasing concentrations of AS-48, the number of viable cells was reduced in proportion to both the amount of bacteriocin added and the time of incubation, especially for concentrations of 25 μg/ml and above (Fig. 2 and Table 1).

FIG. 2.

Combined effect of heat treatment and exposure to AS-48 on survival of S. choleraesuis LT2. Cell suspensions in PB, pH 7.2, were heated at 60°C for 3 min without bacteriocin (controls; striped bars) or with increasing AS-48 concentrations. The number of viable cells was determined immediately after heat treatment (T₀; cross-hatched bars) and after 6 h of incubation at 37°C (T₆; solid bars).

TABLE 1.

Effect of AS-48 on viability of S. choleraesuis LT2 under different conditions (heat injury, pH, and permeabilizing agents)

Treatment	LRFa at AS-48 concn (μg/ml) of:
	0	10	25	50	100	200
Heat + AS-48		0.43	3.02	3.98	4.73	4.85
pH 4.0 + AS-48 at pH 4.0	1.10	1.14	1.75	2.36	5.78	6.03
pH 6.0 + AS-48 at pH 6.0	0			0.28
pH 9.0 + AS-48 at pH 9.0	0.03			5.39
pH 4.0 + AS-48 at pH 9.0	1.10	4.16	4.70	5.34	6.06	6.18
pH 7.0 + AS-48 at pH 9.0	0	4.78	5.40	5.42	5.43	6.13
Heat + AS-48 at pH 3.0				1.99
Heat + AS-48 at pH 4.0				3.97
Heat + AS-48 at pH 5.0				4.23
Heat + AS-48 at pH 6.0				4.47
Heat + AS-48 at pH 7.0				4.29
Heat + AS-48 at pH 8.0				2.23
Heat + AS-48 at pH 9.0				6.27b
EDTA (PB) + AS-48	0.22	0.22	1.38	1.93	2.75	5.04
EDTA (Tris) + AS-48	0.58			2.26
Tris + AS-48	0.18	0.95	1.18	1.42	1.55	2.64
Heat + EDTA	2.18
Heat + EDTA + AS-48		5.20b	5.20b

^a Calculated as log CFU per milliliter of controls at T₀ minus log CFU per milliliter of treated samples after 6 h of incubation at 37°C. 

^b No viable cells were detected in 0.3 ml of sample. 

To test the effect of pH on AS-48 activity, exponential-phase cells in buffer at pHs 4.0, 6.0, and 9.0 were treated with AS-48 dissolved in the same buffers. At pH 6.0, the bacteriocin showed no activity against Salmonella. At pH 4.0, the bacteriocin was active against intact cells, especially at concentrations of 50 μg/ml or higher. The initial population was reduced by six log units after 6 h of incubation with 200 μg of AS-48/ml (Table 1). Bacteriocin AS-48 showed the greatest effects on Salmonella at pH 9.0, yielding a 5.39-log unit reduction when tested at 50 μg/ml. Such reduction was attributed to the sole action of AS-48, because control cells remained unaffected by alkaline pH. Very similar results were obtained when we tested a range of concentrations of AS-48 (dissolved at pH 9.0) on cell suspensions in buffer at pH 7.0 or pH 4.0 (Table 1). It is remarkable that alkaline AS-48 solutions had a pronounced bactericidal effect on intact cells of Salmonella at low concentrations (e.g., 10 μg/ml reduced the surviving population by 4.78 log units at pH 7.0) independently of which component (the cell suspension, the bacteriocin solution, or both) was buffered. These data suggest that the bacteriocin is rapidly adsorbed to the cells and also that bacteriocin adsorption may be a key factor enhanced by alkalinity. The probable conformational changes that AS-48 molecules undergo when the pH shifts from 9.0 to 7.0 deserve further study. In this respect, bacteriocin AS-48 is markedly different from nisin, whose potency is attenuated as pH increases (9), probably because its solubility is much lower. By contrast, AS-48 is highly soluble in aqueous buffers over a broad range of pH.

We also tested the combined effect of sublethal heat and pH by adding 50 μg of AS-48/ml (dissolved in buffers in a pH range from 3.0 to 9.0) to cell suspensions in 10 mM PB, pH 7.2, before heating them. The number of survivors at T₆ in samples treated with bacteriocin in the pH range from 4.0 to 8.0 decreased markedly, and no survivors were obtained in samples with bacteriocin added at pH 9.0 (Table 1). AS-48 showed the lowest antimicrobial activity when added in a solution at pH 3.0.

When AS-48 was used in combination with a chelating agent (EDTA), the number of survivors after 6 h of treatment was reduced proportionally to the bacteriocin concentration added (Fig. 3A and Table 1), indicating that permeabilized cells became sensitive to bacteriocin. The addition of EDTA alone produced a very small reduction in the number of cells, either in PB or in 10 mM Tris-HCl, pH 7.0 (LRF_EDTA = 0.22).

FIG. 3.

Combined effect of permeabilizing agents and AS-48 on S. choleraesuis LT2. (A) Cells resuspended in PB, pH 7.2, were incubated for 6 h with 50 mM EDTA and increasing concentrations of AS-48. (B) Cells in 0.1 M Tris-HCl, pH 8.0, were treated with increasing bacteriocin concentrations.

Cell suspensions were also made in 0.1 M Tris-HCl (pH 8.0), since Tris has been reported to increase the permeability of the OM, releasing proteins, lipopolysaccharide (LPS), and alkaline phosphatase from the cell envelope (10, 24), and to enhance the antimicrobial activity of some bacteriocins (2). The combination of Tris and AS-48 also caused a noticeable reduction in the number of viable cells (Fig. 3B). However, although cells in 0.1 M Tris seemed to be more sensitive to low concentrations of AS-48 (especially to a concentration of 10 μg/ml) than EDTA-treated cells, the number of survivors did not decrease so markedly when the bacteriocin concentration increased.

The combined effect of EDTA, sublethal heat, and AS-48 was studied by adding EDTA (50 mM) and/or AS-48 (10 and 25 μg/ml) to cell suspensions right before heating them (Table 1). The viability of cells heated in the presence of EDTA alone was reduced by 2.18 log units after 6 h of incubation. Remarkably, no viable cells were found with a combination of all three treatments, even at the lowest concentration of AS-48 tested (10 μg/ml).

Cell permeation by heat or EDTA treatment has also been used to enhance the antimicrobial activity of other bacteriocins against gram-negative bacteria (1–3, 11, 21). The molecular basis of the integrity of the OM lies in its LPS—a polyanionic charged structure—whose molecules are linked electrostatically by divalent cations, especially Mg²⁺ or Ca²⁺ (14). Thus, chelating agents such as EDTA also have a strong OM-disorganizing effect. Treatment with EDTA induces a massive and instantaneous loss of LPS (8, 14, 20, 21, 23), exposing the OM phospholipids, which in turn facilitate the entry of hydrophobic compounds into the cell. The synergistic action of EDTA and AS-48 may be more complex, because sequestration of divalent cations by EDTA may also enhance the interaction of AS-48 (which can behave as a polycation) with LPS and hence may enhance OM disorganization. Similar results have been reported for polycationic antibiotics, such as the polymyxin group, which complexes strongly with LPS and disorganizes the OM (22) to reach the final target, the cytoplasmic membrane.

The combination of chelating agents with heat treatment should have a greater permeabilizing effect, since cell wall disruption caused by heat alone can be repaired in the presence of sufficient divalent cations (12). In fact, cells that were heat injured in the presence of EDTA became highly sensitive to much lower concentrations of AS-48.

From this work we can conclude that the antimicrobial activity of AS-48 against S. choleraesuis LT2 can be enhanced by choosing an adequate pH (either under acidic conditions or by using alkaline AS-48 solutions) or by combination with other procedures, such as mild heat treatment or OM-permeabilizing agents like EDTA or Tris. In the food industry mild heat treatments are preferable, as the organoleptic properties of foods are better preserved. The combination of sublethal heat with food grade EDTA and AS-48 could also be of value in controlling the proliferation of Salmonella, and future studies should be carried out in this area.