Influence of Osmolarity and the Presence of an Osmoprotectant on Lactococcus lactis Growth and Bacteriocin Production

Effect of increased osmotic strength on growth of L. lactis ADRIA 85LO30.The growth of L. lactisADRIA 85LO30 was assayed on chemically defined medium (CDM) (10) containing increasing concentrations of NaCl. The growth rate and the growth yield decreased with the increasing osmolarity of the medium (Table 1). Growth was abolished with NaCl contents higher than 0.5 M. In the presence of 0.4 M NaCl, both parameters were reduced by 70% of their value in the absence of NaCl. Therefore, this salt condition was used for further studies. Growth inhibition was also observed when other osmotica, such as sucrose or KCl, were used to increase the osmolarity of the medium (data not shown).

To determine the major solutes accumulated under hyperosmotic strength, cells grown on CDM alone or with 0.4 M NaCl were harvested in exponential growth phase and ethanol-soluble fractions were analyzed by natural-abundance ¹³C nuclear magnetic resonance (NMR) (6). As expected, no major intracellular compound was accumulated in unstressed cells (Fig.1A). In the presence of 0.4 M NaCl, the main accumulated solute was lactic acid (Fig. 1B). Lactic acid does not have the characteristics required for an osmoprotectant (2), so its accumulation could not efficiently counterbalance the osmotic pressure and was deleterious for bacterial metabolism. Thus, no accumulation of efficient osmolyte was detected by natural abundance ¹³C-NMR in stressed cells. This could explain the lower osmotolerance of L. lactis ADRIA 85LO30 compared to other strains of lactic acid bacteria (11).

• Open in new tab
• Download powerpoint

Fig. 1.

Natural-abundance ¹³C-NMR spectra of ethanol-soluble fractions of L. lactis ADRIA 85LO30 grown on CDM without NaCl (A), with 0.4 M NaCl (B), and with 0.4 M NaCl added to 1 mM GB (C). The peaks were identified in comparison with reference spectra of commercial compounds: GB (b) and lactic acid (l).

Effect of compatible solutes on growth of L. lactisADRIA 85LO30.When present in salt-containing medium, GB is efficiently accumulated by most of the lactic acid bacteria studied (4, 5, 8, 16). Therefore, GB and other compounds known to act as bacterial osmoprotectants (2) were tested on osmotically stressed cells of L. lactis ADRIA 85LO30 (Table2). All of the methylated onium compounds structurally related to GB (i.e., dimethylsulfonioacetate [DMSA], dimethylsulfoniopropionate [DMSP], l-carnitine, and arsenobetaine) allowed the growth rate and the growth yield to increase by 60%. By contrast, no effect of choline, proline, or pipecolate on growth parameters was observed (Table 2). Many bacteria, includingEscherichia coli, Bacillus subtilis, andStaphylococcus aureus, were able to oxidatively convert choline to GB (1, 7, 13). However, even under aerobic conditions, choline was unable to improve the growth of L. lactis ADRIA 85LO30 (data not shown).

¹³C-NMR experiments were performed on cells grown in 0.4 M NaCl CDM supplemented with 1 mM GB. GB was the main solute observed on the spectrum (Fig. 1C), whereas lactate accumulation was abolished. The maximum accumulation levels of osmoprotectants were measured with and without 0.4 M NaCl. In salt-containing medium, bacteria were able to transport GB, DMSA, DMSP, l-carnitine, choline, proline, and pipecolate. The accumulation levels were correlated with the efficiencies of the osmoprotectants at improving growth (Table 2). Presumably, the structures of the osmoprotectants could play a role in their efficient uptake and accumulation. This hypothesis was previously suggested to account for the efficiency of choline transport byS. aureus (9) and for l-carnitine accumulation by Lactobacillus plantarum (12). Thus, GB and its analogs were accumulated at elevated levels in stressing medium, restablishing a positive turgor pressure, which is one of the effects usually attributed to osmoprotectants (2).

Effect of NaCl and GB on lacticin 481 production.We evaluated bacteriocin production by cells grown on CDM with and without NaCl. Lacticin 481 was obtained as previously described and assayed against the indicator strain L. lactis IL1837 (15, 17). In CDM without NaCl, bacteriocin production remained constant during all stages of growth and did not exceed 5 arbitrary units (AU) ml⁻¹ (Fig. 2). When NaCl was supplied, production increased from the beginning of stationary phase, reaching a maximal value of 20 and 12 AU ml⁻¹ in the presence of 0.2 and 0.4 M NaCl, respectively (Fig. 2). In all conditions, a decrease of bacteriocin production was observed in late stationary phase. Proteolytic activity was assayed (3) but was not detected. This result was in agreement with the fact that no decrease of optical density at 570 nm (OD₅₇₀) was observed in stationary phase (Fig. 2), indicating that no cells lysis occurred. Thus, this decline could not be due to proteolytic enzymes released during cell lysis. The adsorption of bacteriocin to producer cells was also assayed (19) and was detected. This phenomenon would explain the decline of lacticin 481 production, as was described for amylovorin L471 (3).

• Open in new tab
• Download powerpoint

Fig. 2.

Effect of NaCl and GB on lacticin 481 production during the growth of L. lactis ADRIA 85LO30. Cells were cultivated at 30°C without shaking on CDM with increasing NaCl concentrations or with 0.4 M NaCl added to 1 mM GB. Turbidimetry was monitored at 570 nm, and bacteriocin production (expressed in arbitrary units per milliliter) was evaluated. To quantify lacticin 481, twofold dilutions of produced bacteriocin were assayed by the supernatant diffusion method (15). The supernatants were dialyzed against CDM without NaCl and concentrated against solid polyethylene glycol 8000 for 48 h at 4°C in dialysis membranes with a molecular mass cutoff of 500 Da. The values are the means of triplicate experiments; the standard deviation was less than 20%. The solid symbols represent bacterial growth for each condition, and the open symbols represent bacteriocin production. Square, 0 M NaCl; diamond, 0.2 M NaCl; circle, 0.4 M NaCl; triangle, 0.4 M NaCl with 1 mM GB.

When 1 mM GB was added to CDM containing 0.4 M NaCl, lacticin 481 production was close to that obtained on nonstressing CDM (Fig. 2). The decrease could be explained by GB accumulation, which would allow a positive turgor pressure to be reestablished and would protect intracellular macromolecules (2). Thus, the failure of turgor or the alteration of macromolecules could be the inducers of bacteriocin production. This suggestion is supported by the results obtained with amylovorin L471, which demonstrated that unfavorable growth conditions stimulated bacteriocin production (3).

In this study, we demonstrated that GB improves the growth parameters of L. lactis ADRIA 85LO30 under hyperosmotic constraints but provokes the decrease of bacteriocin production. Also, as with nisin Z production by L. lactis I0-1 (14), optimal conditions for growth seem to be quite different from those for lacticin 481 production. Therefore, to improve bacteriocin production, it is important to know the differential effects of environmental factors on L. lactis growth and bacteriocin production.