L.M.T. Dicks, Department of Microbiology, University of Stellenbosch, Stellenbosch, South Africa. E-mail: firstname.lastname@example.org
Aims: Screening of five bile salt-resistant and low pH-tolerant lactic acid bacteria for inhibitory activity against lactic acid bacteria and bacterial strains isolated from the faeces of children with HIV/AIDS. Determining the effect of prebiotics and soy milk-base on cell viability and adhesion of cells to intestinal mucus.
Methods and Results: Lactobacillus plantarum 423, Lactobacillus casei LHS, Lactobacillus salivarius 241, Lactobacillus curvatus DF 38 and Pediococcus pentosaceus 34 produced the highest level of antimicrobial activity (12 800 AU ml−1) when grown in MRS broth supplemented with 2% (m/v) dextrose. Growth in the presence of Raftilose®Synergy1, Raftilose®L95 and Raftiline®GR did not lead to increased levels of antimicrobial activity. Cells grown in the presence of Raftilose®Synergy1 took longer to adhere to intestinal mucus, whilst cells grown in the absence of prebiotics showed a linear rate of binding.
Conclusions: A broad range of gram-positive and gram-negative bacteria were inhibited. Dextrose stimulated the production of antimicrobial compounds. Adhesion to intestinal mucus did not increase with the addition of prebiotics.
Significance and Impact of the Study: The strains may be incorporated in food supplements for HIV/AIDS patients suffering from gastro-intestinal disorders.
Lactic acid bacteria play a key role in maintaining the balance of normal gastro-intestinal microflora (Fuller 1989). However, factors such as diet, stress, microbial infection and other diseases disturb the balance, which often leads to a decrease in the number of viable lactobacilli and bifidobacteria (Fuller and Gibson 1997). The subsequent uncontrolled proliferation of pathogenic bacteria may lead to diarrhoea and other clinical disorders such as cancer, inflammatory disease and ulcerative colitis (Fooks et al. 1999).
One of the key properties of probiotic lactic acid bacteria is the adhesion of cells to epithelial cells or intestinal mucus. This requires strong interaction between receptor molecules on epithelial cells and bacterial surfaces (Salminen et al. 1996a). Adhesion of probiotic cells prevent the adhesion of pathogens (Salminen et al. 1996b) and stimulate the immune system (Isolauri et al. 1991,1995; Rolfe 2000). The latter is achieved by interacting with mucosal membranes, which in turn sensitises the lymphoids (Salminen et al. 1996a). Another important characteristic of probiotics is survival at low pH and high bile salts (Bezkorovainy 2001).
The role of bacteriocins (antimicrobial peptides), and their significance in controlling the proliferation of pathogenic bacteria in the intestinal tract, is questionable (Fuller, personal communication). However, from recent reports on bacteriocins active against gram-negative bacteria (Ivanova et al. 1998; Messi et al. 2001; Caridi 2002; Todorov and Dicks 2004), we may experience renewed interest in these peptides and their interaction with intestinal pathogens. Many of the later bacteriocins are produced by lactic acid bacteria normally present in the intestinal tract, viz. Lactobacillus plantarum, Lactobacillus pentosus, Lactobacillus paracasei subsp. paracasei and Enterococcus faecalis (Abrionel et al. 2001).
Many lactic acid bacteria with probiotic properties have been reported and many methods for cell propagation and probiotic preparations have been described (Fuller 1989). In most cases, the cells are freeze-dried or spray-dried (Saarela et al. 2000; Mattila-Sandholm et al. 2002). Many strains do not survive the drying process, have a long lag phase in recovering from the process, and do not remain viable during extended storage at room temperature (Myllärinen et al. 1998). Several methods have been described to stabilise probiotic cells (Kamaly 1997; Mattila-Sandholm et al. 2002). Fooks et al. (1999) suggested the use of prebiotics, defined as ‘‘non-digestible food ingredients that beneficially affects the host by selectively stimulating the growth and/or the activity of one or a limited number of bacteria in the colon, especially Lactobacillus and Bifidobacterium spp.’’ (Gibson and Roberfroid 1995), as stabilising agents.
In this paper, we report on lactic acid bacteria with a broad spectrum of antibacterial activity, including gram-negative bacteria, growth in the presence of prebiotics, and the effect of soy milk-base on freeze-dried cells. Survival at low pH, growth in the presence of bile, and adhesion of the cells to intestinal mucus have been studied.
Material and methods
Bacterial strains and growth conditions
Lactobacillus plantarum 423, Lactobacillus salivarius 241, Lactobacillus curvatus DF38, Lactobacillus casei LHS and Pediococcus pentosaceus 34 were screened for antibacterial properties against gram-positive and gram-negative bacteria, including strains isolated from patients diagnosed with HIV/AIDS (Table 1). Lactic acid bacteria were cultured in MRS medium (Biolab, Biolab Diagnostics, Midrand, SA) and all other strains, except Propionibacterium acidipropionici, in brain heart infusion (BHI) medium (Biolab). Propionibacterium acidipropionici was cultured in GYP medium (5 g l−1 glucose, 3 g l−1 yeast extract, 10 g l−1 peptone, 5 g l−1 NaCl, pH 7.0). Pure cultures were stored at −80°C in growth medium, supplemented with glycerol (15%, v/v, final concentration).
Table 1. Spectrum of antimicrobial activity of lactic acid bacteria
Lactic acid bacteria
Lact. plantarum 423
Lact. curvatus DF38
Ped. pentosaceus 34
Lact. salivarius 241
Lact. casei LHS
LMG: Laboratorium voor Microbiologie, University of Ghent, Belgium.
†Isolated from faeces of patients diagnosed with HIV/AIDS.
Antibacterial activity in the absence and presence of prebiotics
Lactobacillus plantarum 423, Lact. salivarius 241, Lact. curvatus DF38, Lact. casei LHS and Ped. pentosaceus 34 were grown in MRS broth (Biolab) for 18 h at 37°C. Ten μl of the culture was spotted onto MRS agar (Biolab), incubated for 24 h at 37°C and then overlaid with active growing cells of the target strains (Table 1), imbedded in BHI or GYP medium (0·8% agar, w/v). The plates were incubated at 37°C for 24 h and the colonies examined for formation of inhibition zones.
In a separate experiment the lactic acid bacteria were inoculated into 10 ml MRS broth (Biolab), MRS broth (De Man et al. 1960) without dextrose (MRS-D), and MRS-D supplemented with 2·0% (v/v) of 5·0% (w/v) filter-sterilised Raftiline®GR (92% inulin and 8% glucose/fructose/sucrose), Raftilose®L95 (95% oligofructose and 5% glucose, fructose, sucrose) and Raftilose®Synergy1 (92% inulin and oligofructose, and 8% glucose, fructose, sucrose), respectively. All prebiotics were obtained from SAVANNAH Fine Chemicals (ORAFTI, Tienen, Belgium). After 18 h of growth at 37°C, the cells were harvested (8000 g, 10 min, 4°C), the pH of the cell-free supernatant adjusted to 6·0 with 1 mol−1 NaOH and then sterilized by passing through a 0·22 μm pore-size nitrocellulose filter. The cell-free supernatant was tested for antimicrobial activity against the target strains listed in Table 1. The spot-on-lawn method, described by Van Reenen et al. (1998), was used and the antimicrobial activity expressed as arbitrary units (AU) per ml. One AU is defined as the reciprocal of the highest dilution showing a clear zone of growth inhibition (Van Reenen et al. 1998). Changes in the optical density of the culture was measured at 600 nm and changes in pH recorded at specific time intervals throughout the experiment. The experiment was repeated with increasing concentrations (0·5–3·0%, w/v, with 0·5% intervals) of the prebiotic that yielded the highest level of antimicrobial activity. All experiments were conducted in triplicate.
To determine if the antimicrobial activity recorded was bacteriostatic or bactericidal, the lactic acid bacteria were grown in MRS broth (Biolab) at 37°C to OD 600 nm = 1·4, the cells harvested, the cell-free supernatant adjusted to pH 6.0 with 1 mol l−1 NaOH and then filter-sterilized, as described elsewhere. Twenty ml of the filtrate was added to a 100 ml-culture of Clostridium tyrobutyricum LMG13571 in early exponential growth (OD 600 nm = 0·12). Changes in optical density of the cells were determined every hour for 9 h. The experiment was repeated in triplicate.
Growth at different pH values and bile concentrations
Lactobacillus plantarum 423, Lact. salivarius 241, Lact. curvatus DF38, Lact. casei LHS and Ped. pentosaceus 34 were grown in MRS broth (Biolab) adjusted to pH 3·0, 4·0, 5·0, 7·0, 9·0, 11·0 and 13·0, respectively. The pH was adjusted with 1 mol l−1 HCl or 1 mol l−1 NaOH before autoclaving and re-adjusted after autoclaving if the pH changed by more than 0·2 units. Resistance to bile was tested by growing the cells in MRS broth (Biolab) adjusted to 0·3, 0·6, 0·8, 1·0, 2·0 and 5·0% (m/v) oxbile (Oxoid, Basingstoke, England). The pH was adjusted to 6·4 with sterile 1 mol l−1 NaOH after autoclaving. The experiment was repeated with the same concentrations (v/v) freshly isolated porcine bile. All tests were conducted in STERELINTM micro titer plates. Each well was filled with 180 μl of the bile-containing medium and inoculated with 20 μl culture (OD 600 nm = 0·3). Optical density readings (at 600 nm) were recorded every hour for up to 11 h. The experiment was repeated with the same strains grown in MRS broth (Biolab) supplemented with Raftilose®Synergy1 (1% w/v). Cultures grown in MRS broth (Biolab) without bile served as control.
Adhesion to intestinal mucus
Porcine ileum, collected from animals immediately after slaughtering, were aseptically dissected into 3 cm-long sections and kept on ice for a maximum of 9 h. Lactobacillus plantarum 423, Lact. salivarius 241, Lact. curvatus DF38, Lact. casei LHS and Ped. pentosaceus 34 were inoculated (2%, v/v) into 250 ml MRS broth (Biolab) and MRS broth (Biolab) supplemented with 1% (w/v) Raftilose®Synergy1, respectively, and incubated at 37°C to OD 600 nm = 1·2. A section of ileum was added to each of the latter cultures and incubated for 6 h at 8°C on a rotary shaker (30 g). Samples of the culture were withdrawn every 2 h, serially diluted and plated onto MRS agar (Biolab). Colonies were counted after 14 h of incubation at 37°C.
The ileum sections were aceptically removed from the flasks and the mucus layer carefully scraped off with a sterile glass slide. Preparations of the mucus samples on microscopic slides were treated with the BacLightTM viability probe (Molecular Probes Inc., Eugene, Oregon, USA) and left for 10 min in the dark at room temperature. Images of adhering bacterial cells were captured using a high-performance CCD camera (Cohu Inc., San Diego, CA, USA) mounted on a Nikon Eclipse E400 epi-fluorescence microscope, equipped with a ×60/1·4 Dic H oil objective and filters. Sections of ileum suspended in MRS broth and not inoculated served as control. Twenty optical fields were selected at random per ileum and the percentage viable and nonviable cells per optical field calculated. Corrections were made for spectral overlap and background fluorescence and the images analyzed with Scion Image software (U.S. National Institutes of Health; http://rsb.info.nih.gov/nih-image). The number of cells that adhered to the mucus, after 6 h of exposure at 8°C, was determined by suspending 1 g of the mucus scrapings into sterile saline (0·75%, w/v, NaCl). A dilution series was made in sterile saline and plated out onto MRS agar (Biolab). Colonies were counted after 24 h of incubation at 37°C.
Survival after freeze-drying in soy milk-base
Commercially available soy milk-base powder was suspended in distilled water (10%, w/v), autoclaved and inoculated with Lact. plantarum 423, Lact. salivarius 241, Lact. curvatus DF38, Lact. casei LHS and Ped. pentosaceus 34 (1·0%, v/v). Viable cell numbers were recorded on MRS agar (Biolab) as described elsewhere. After 24 h of growth at 37°C, the cells were harvested (9000 g; 30 min, 4°C), the pellet resuspended in soy milk-base powder (10%, w/v) and freeze-dried. The number of cells surviving the freeze-drying process was determined by plating onto MRS agar (Biolab), as before, and the percentage surviving cells determined.
All data represent an average of three repeats. The values recorded in each experiment did not vary by more than 5%. Single data points are, therefore, presented in the figures without standard deviation bars.
Viable cells (colonies) of Lact. plantarum 423, Lact. salivarius 241, Lact. curvatus DF38, Lact. casei LHS and Ped. pentosaceus 34 inhibited a broad range of gram-positive and gram-negative bacteria, including strains isolated from patients diagnosed with HIV/AIDS (Table 1). Cell-free supernatants of the latter strains, adjusted to pH 6.0, revealed the same spectrum of antibacterial activity (data not shown).
The antimicrobial activities detected in the cell-free supernatants of strains cultured in MRS broth (Biolab) were in general higher than recorded for strains cultured in MRS broth without dextrose (MRS-D). Addition of prebiotics to MRS-D resulted in increased antimicrobial activity, but differed from strain to strain. Lactobacillus casei LHS yielded an antimicrobial activity of 12 800 AU ml−1 when cultured in MRS broth (Biolab). A much lower level of activity (1600 AU ml−1) was recorded when the cells were grown in the absence of dextrose, i.e. in MRS-D broth. However, when the dextrose in MRS broth was substituted with Raftiline®GR or Raftilose®Synergy1, 6400 AU ml−1 was recorded. Growth in the presence of Raftilose®L95 yielded only 3200 AU ml−1. Similar results were recorded for Lact. plantarum 423 and Lact. salivarius 241. Both strains produced 12 800 AU ml−1 when cultured in MRS broth (Biolab). Lactobacillus plantarum 423 cultured in MRS-D and MRS-D, supplemented with Raftilose®L95 or Raftilose®Synergy1, yielded much lower activity levels (800 AU ml−1). Growth in the presence of Raftiline®GR yielded only 400 AU ml−1. Lactobacillus salivarius 241 cultured in MRS-D yielded only 400 AU ml−1. However, when MRS-D was supplemented with Raftiline® GR or Raftilose®Synergy1, the activity increased to 1600 AU ml−1. A level of 3200 AU ml−1 was recorded in MRS-D supplemented with Raftilose®L95. Highest antimicrobial activity (12 800 AU ml−1) was recorded when strain 241 was grown in MRS broth (Biolab). No antimicrobial activity was recorded in the cell-free supernatants of Lact. curvatus DF 38 and Ped. pentosaceus 34, irrespective of the growth medium.
Cell-free supernatants obtained from Lact. plantarum 423, Lact. casei LHS and Lact. salivarius 241 cultured in MRS broth (Biolab) repressed the growth of Cl. tyrobutyricum (Fig. 1). However, no inhibition of Cl. tyrobutyricum was recorded when the experiment was repeated with cell-free supernatants obtained from strains grown in MRS-D broth or MRS-D broth supplemented with Raftiline®GR, Raftilose®L95 or Raftilose®Synergy1 (Fig. 1).
Growth at different pH values and bile concentrations
At pH 3.0 growth of all five strains was suppressed for the first 10 h of incubation. Growth was much more vigorous between pH 5.0 and 6.5 (results not shown). Growth at pH 3.0 and in the presence of ox bile, or porcine bile, was slower compared to growth in the absence of bile. Similar results were recorded with ox bile and porcine bile. Porcine bile at 5·0% (v/v) resulted in growth inhibition. Supplementing the medium with Raftilose®Synergy1 yielded slightly better growth (Fig. 2). In the case of Lact. curvatus DF 38 and Ped. pentosaceus, the OD-values increased from 0·3 to 1·6 after 10 h of incubation. Of all strains, Lact. curvatus DF38 was the most resistant to 5·0% (v/v) porcine bile, as revealed by the high-cell density (OD 600 nm = 2·0) recorded after 6 h of growth in the presence of Raftilose®Synergy 1.
Adhesion to intestinal mucus
The number of free-living cells in the culture decreased over time, irrespective of whether the cells were cultured in MRS or MRS supplemented with Raftilose®Synergy1 (Fig. 3). Staining with the BacLight™ viability probe revealed strong adhesion of the lactic acid bacteria to the ileum mucus, with the best adhesion recorded for Lact. curvatus DF38, Lact. casei LHS and Lact. salivarius 241 (Fig. 4). More than 90% of the cells that adhered to the mucus remained viable (fluorescent green). The number of cells that adhered to the mucus ranged from 150 to 250 per optical field for Lact. curvatus DF38 and Lact. casei LHS, and between 300 and 500 per field for Lact. salivarius 241. Less than 100 cells per optical field were recorded for Lact. plantarum 423 and Ped. pentosaceus 34 (not shown). The number of viable cells that adhered to the mucus of the ileum after 6 h of exposure at 8°C, ranged from 8·0 × 105 to 7·0 × 106 CFU g−1 for Lact. curvatus DF38 and Lact. casei LHS, and from 1·7 × 106 to 1·7 × 107 CFU g−1 for Lact. salivarius 241. Less than 1·0 × 105 CFU g−1 mucus was recorded for Lact. plantarum 423 and Ped. pentosaceus 34.
Survival after freeze-drying
Soy milk-base supported the growth of the lactic acid bacteria, as indicated by an increase in two log-cycles over 24 h (not shown). However, the cell numbers were reduced by two log-cycles when freeze-dried. Cells grown in MRS broth (Biolab) on the other hand, were reduced by only one log cycle after freeze-drying, suggesting that they are more resistant to freeze-drying compared to cells cultured in soy milk-base.
The spectrum of antimicrobial activity recorded for Lact. curvatus DF 38, Ped. pentosaceus 34, Lact. plantarum 423, Lact. casei LHS and Lact. salivarius 241 (Table 1) is broader than recorded for most lactic acid bacteria (De Vuyst and Vandamme 1994). Activity against gram-negative bacteria is unusual and has thus far only been reported for a few lactic acid bacteria. An increasing number of broad-spectrum antimicrobial compounds are being described, most of which are referred to as antibiotic-like (De Vuyst and Vandamme 1994).
The production of short-chain fatty acids such as acetate, propionate and butyrate from the fermentation of inulin and oligofructose has been well documented (Schley and Field 2002). In the large intestine, the increase in short-chain fatty acids and lactic acid leads to a decrease in pH, which provides optimal conditions for the growth of the lactic acid bacteria. The increased number of lactic acid bacteria competes with pathogens for nutrients and receptors on the gut wall (Schley and Field 2002). Increased antimicrobial activity was recorded in MRS-D broth supplemented with 1% Raftilose®Synergy1, which is a combination of inulin and fructo-oligosaccharides. Similar results have been reported by Tzortzis et al. (2004). In the latter study, an increase in antimicrobial activity was recorded when Lactobacillus mucosae, Lactobacillus acidophilus and Lactobacillus reuteri were grown in the presence of malto-oligosaccharides. The authors concluded that organic acids are not always the sole antimicrobial agents. Production of short chain fatty acids from inulin and fructo-oligosaccharides cannot be ruled out. However, in our study, highest antimicrobial activity was recorded when cells were grown in normal MRS broth (Biolab), which suggests that dextrose is a key nutrient in production of the antimicrobial compounds. Glucose is rapidly metabolised in the intestine, which suggests that the diet would have to consist of high concentrations of inulin and fructo-oligosaccharides to present the same results in vivo.
Cell-free supernatants of Lact. plantarum 423, Lact. casei LHS and Lact. salivarius 241, added to log-phase cells of Cl. tyrobutyricum resulted in retarded growth, but only from 4 to 5 h after addition (Fig. 1). The mode of action of these antimicrobial compounds are bacteriostatic, with variable degrees thereof. Cell-free supernatants from cells grown in MRS (Biolab) were the most active, as recorded by the slowest increase in OD-values of Cl. tyrobutyricum after addition of the supernatants (Fig. 1). Supernatants from cells grown in Raftilose®L95 and Raftilose®Synergy1 were the second most effective judged from the changes recorded in OD-values (Fig. 1).
Growth of Lact. curvatus DF 38, Ped. pentosaceus 34, Lact. plantarum 423, Lact. casei LHS and Lact. salivarius 241 in MRS (Biolab) at pH 3.0 with dextrose as the only carbon source, and in the presence of bile, was slow (Fig. 2). The addition of Raftilose®Synergy1 to MRS-D resulted in increased growth, suggesting that inulin and oligofructose are used as carbon sources. The largest increase in cell numbers was recorded for Lact. curvatus DF38 (Fig. 2). Similar findings have been reported by Roberfroid and Delzenne (1998), Van Loo et al. (1999) and Roberfroid (1999). Lactobacillus acidophilus 74-2 increased with one log-cycle if cultured in the presence of 1% (w/v) fructo-oligosaccharide (Gmeiner et al. 2000). The authors have shown a drastic inhibition of Escherichia coli and other enterobacteria and the antibacterial activity was ascribed to a substantial increase in production of short-chain fatty acids, accompanied by a decrease in pH. Resistance to low pH and elevated concentrations of bile salts are important to the growth and survival of bacteria in the intestinal tract (Havenaar et al. 1992). Soy milk is rich in soybean-oligosaccharides, raffinose and stachyose and is an excellent growth medium for lactic acid bacteria (Kamaly 1997; Wang et al. 2002,2003; Beasley et al. 2003). However, the strains included in our study were less resistant to freeze-drying when cultured in soy milk-base.
The adhesion of lactic acid bacteria to epithelial cells or intestinal mucus has been well documented (Salminen et al. 1996a; Felley and Michetti 2003; Rinkinen et al. 2003). The role that prebiotics play in adhesion has also been studied (Topping et al. 2003). Based on the data obtained in our study, most of the bacterial cells adhered to the mucus on the ileum during 6 h at 8°C (Fig. 3). The number of cells that adhered to the mucus could not be calculated accurately due to a high level of background interference (Fig. 4). However, concluded from the number of cells visible, Lact. salivarius 241 adhered the strongest to mucus (Fig. 4c). Furthermore, between 1·7 × 106 and 1·7 × 107 viable cells per gram mucus was recorded for Lact. salivarius 241, which is approximately double of that recorded for Lact. curvatus DF38 and Lact. casei LHS (8·0 × 105 to 7·0 × 106 CFU g−1). The adhesion studies were performed at 8°C to avoid the growth of normal intestinal bacteria and are not a true reflection of what may occur in the gastro-intestinal tract at 37°C and during digestion of food particles. Furthermore, only adhesion to mucus was studied and based on these results, we concluded that Raftilose®Synergy1 decreased the rate at which cells bind to mucus. In general, the lower cell numbers developed in the absence of the prebiotic showed a more linear decrease in cell numbers and binding to the mucus. Staining of the mucus with the BacLight™ viability probe indicated that the majority of the adhered cells (>90%) remained viable (Fig. 4).
Freeze-drying is the most popular method to dry lactic acid bacteria for storage purposes, although more expensive than spray-drying. With both these drying methods, cell damage may occur (Porubcan and Sellars 1979). Freeze drying of lactic acid bacteria grown in soy milk-base resulted in a two log-cycle decrease in cell numbers, which could indicate cell damage. It would however be prudent to include protectants such as lactose or sucrose, monosodium glutamate (MSG) and ascorbate, which are usually added to cultures to prevent cell destruction and prolong storage to soy milk-base before freeze-drying (Champagne et al. 1991; Souzu 1992; Mäyrä-Mäkinen and Bigret 1998).
Although this study was, as most others reported, performed in vitro, it provided us with a better understanding of the effect prebiotics have on the production of antimicrobial compounds, growth of probiotic cells at low pH and in the presence of bile salts, and adhesion of probiotic cells to intestinal mucus. Of specific interest in this study was the inhibition of a number of pathogenic strains isolated from patients diagnosed with HIV/AIDS. Although the pathogens are not uniquely associated with HIV-positive individuals, they may cause acute infections due to an overall low immunity.
The study was funded by the Medical Research Council (MRC) and the National Research Foundation (NRF).