The following bacterial strains were used in this study: Bifidobacterium breve, Bifidobacterium infantis, Bifidobacterium longum, Lactobacillus acidophilus, Lactobacillus casei, Lactobacillus delbrueckii subsp. bulgaricus, Lactobacillus plantarum and Streptococcus thermophilus.

Bifidobacterium and Lactobacillus strains were grown anaerobically (Anaerobic System, Model 2028, Forma Scientific, Marietta, OH, USA) in MRS medium (Difco, Detroit, MI, USA) supplemented with 0.05% L-cysteine hydrochloride monohydrate (Merck, Darmstadt, Germany), at 37°C. S. thermophilus was cultured anaerobically in M17 medium (Difco) at 37°C. Salmonella aboni NCTC 6017, Escherichia coli ATCC 11105 and Bacillus licheniformis BGSC 5A24 were used as controls and grown aerobically in LB medium (Difco) at 37°C.

Ten milliliter of the bacterial mid log cultures, counted by plating technique on the previously mentioned media, was collected by centrifugation (5000 × g at 4°C for 5 min), washed, and resuspended in 1 mL of De Jalon buffer (155 mmol/L NaCl, 5.6 mmol/L KCl, 0.5 mmol/L CaCl₂, 6.0 mmol/L NaHCO₃, 2.8 mmol/L glucose)[25], obtaining a final bacterial concentration of 1 × 10⁹-1 × 10¹⁰ colony forming units (CFUs/mL). Subsequently the bacterial suspensions were sonicated (Bronson Sonifier W-250, Heineman, Schwäbisch, Germany) at power level 2 at 20% duty for 8 min and centrifuged at 8000 × g for 30 min to separate cell debris from crude extract. Cell debris was resuspended in 1 mL of De Jalon buffer. The crude extract was ultracentrifuged at 250 000 × g at 4°C for 2 h: the supernatant represents the cytoplasmatic fraction while the pellet, resuspended in 1 mL of 50 mmol/L TRIS.HCl pH 7.6, represents the fraction enriched in membrane proteins.

In order to obtain the bacterial cell wall fraction, bacterial cultures were centrifuged at 5000 × g at 4°C for 5 min, washed in 50 mmol/L Tris HCl pH 7.6, resuspended in 1 mL of protoplast buffer [50 mmol/L Tris HCl pH 7.6, 1 mol/L sucrose, 50 mL of complete^TM protease inhibitor (Roche, Mannheim, Germany), 15 mg/mL lysozyme] and incubated at 37°C for 20 min. The supernatant was recovered by centrifuging at 5000 × g at 4°C for 5 min.

Proteinaceous components were precipitated by addition of 9 volumes of acetone:HCl (10:0.1) to one volume of the cytoplasmatic fraction and collected by centrifuging at 12 000 × g at 4°C for 5 min. Supernatant (non-proteinaceous fraction) and protein pellet were dried on ice under vacuum for removing acetone and resuspended in the initial volume of 50 mmol/L TRIS.HCl pH 7.6. The very low value of pH (about 1) of the non-proteinaceous fraction was adjusted to a value of 7.4 to overcome the unspecific colonic contractile response due to the acid pH. Similarly, the cytoplasmatic fraction was incubated with proteinase K (500 mg/mL) at 50°C for 1 h for the enzymatic digestion of the proteins. All the bacterial fractions of cell debris (i.e., cell wall fraction, membrane proteins, cytoplasmatic fractions, proteinaceous and non-proteinaceous cytoplasmatic components) were aliquoted and stored at -80°C before to be used in the in vitro stimulation of ileum and proximal colon.

Ten μL of B. infantis cytoplasmatic fraction, non-proteinaceous fraction and cytoplasmatic fraction deprived of proteins by proteinase K digestion was analyzed by SDS-PAGE, as described by Laemmli[26], using a 12% polyacrylamide running gel. The gels were stained with silver nitrate.

Isolation of genomic DNA from pure cultures of the probiotic bacteria was performed as previously described[27]. In order to obtain the complete cell disruption, the method was slightly modified by prolonging the enzymatic lysis for 1 to 3 h and grinding with glass beads (150-212 μm, Sigma, St. Louis, MO, USA). Concentration and purity of all DNA preparations were determined by measuring OD₂₆₀ absorbance and OD_260/280 ratio, respectively. Only DNAs with an OD_260/280 ratio > 1.8 were used.

Guinea-pigs of either sex (200-400 g) obtained from Charles River (Calco, Como, Italy) were used. The animals were housed according to the ECC Council Directive regarding the protection of animals used for experimental and other scientific purposes. All procedures followed the guidelines of The Animal Care and Use Committee of The University of Bologna (Bologna, Italy). The animals were sacrificed by cervical dislocation, and the organ (ileum or proximal colon) required was set up rapidly under a suitable resting tension in 15-mL organ baths containing appropriate physiological salt solution (PSS) consistently warmed (see below) and buffered to pH 7.4 by saturation with 950 mL/L O₂ and 50 mL/L CO₂ gas.

The terminal portion of ileum (3-4 cm near the ileo-caecal junction) was cleaned with Tyrode solution of the following composition: 118 mmol/L NaCl, 4.75 mmol/L KCl, 2.54 mmol/L CaCl₂, 1.20 mmol/L MgSO₄, 1.19 mmol/L KH₂PO₄·2H₂O, 25 mmol/L NaHCO₃, and 11 mmol/L glucose. The mesenteric tissue was removed. The ileum tissue was cut in two segments of 2-3 cm taken in the longitudinal direction along the intestinal wall. The segments were set up upright under 1-g tension in a jacketed tissue bath (15 mL, 37°C) containing Tyrode solution buffered to pH 7.4 by saturation with 950 mL/L O₂ and 50 mL/L CO₂ gas. Tissues were allowed to equilibrate for at least 30 min, during which time the bathing solution was changed every 10 min. After stabilization, the strips were challenged with 1 μmol/L carbachol (Sigma, Italy) to assess the responsive capacity of the tissue[28].

Starting approximately 1 cm distal from the caecocolonic junction, a segment of about 3 cm was excised, cleansed by rinsing it with De Jalon solution[25], and the mesenteric tissue was removed. The proximal colon segment was cut in two segments of about 1 cm each taken in the longitudinal direction along the intestinal wall. The segments were set up upright under 1-g tension at 37°C in a jacketed tissue bath (15 mL, 37°C) containing De Jalon solution buffered to pH 7.4 by saturation with 950 mL/L O₂ and 50 mL/L CO₂ gas. Tissues were allowed to equilibrate for at least 30 min during which time the bathing solution was changed every 10 min. After stabilization, the strips were challenged with 5 μmol/L 5-hydroxytryptamine (5-HT) (Sigma, Italy) in the presence of 1 μmol/L atropine (Sigma, Italy) to assess the responsive capacity of the tissue.

After stabilization and assessment of the responsive capacity of the tissue in the organ bath, concentration-response curves were constructed by cumulative addition of the above described bacterial preparations (1 to 1500 μL). Each successive addition of bacterial preparations was performed after the response to the previous addition reached its maximum level and remained steady. Longitudinal muscle contractions or relaxations were recorded isometrically by securing one end of the tissue segments to a tissue holder and the other end to a force displacement transducer (FT. 03, Grass Instruments, Quincy, MA) using Power Lab software (ADInstruments Pty Ltd, Castle Hill, Australia). Each tracing was obtained by using separate intestinal strips.

Experiments were performed in duplicate with tissue from the same animal, and mean values were recorded. All data are presented as mean ± SE (n = 3-5 for ileum and n = 5-7 for proximal colon). Differences between means were calculated with Student t-test and P-values < 0.05 were considered statistically significant[29-31].

The motility response of the guinea-pig ileum segment was investigated with increasing concentrations of the following components of the probiotic VSL#3 mixture: live bacteria, bacterial cell debris, and crude extracts. VSL#3 live bacteria and cell debris did not modify the guinea-pig ileum motility (data not shown), whereas crude extracts dose-dependently increased both spontaneous phasic and tonic contractions of the ileum (Figure 1A). The effects on motility induced by crude extracts persisted even after adjusting the pH of the solution from around 5.0, as found in the crude extract, to pH 7.4 and were reversible as the basal tone returned to initial values after several washing steps (data not shown).

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Recorder tracing of the cumulative dose of VSL#3 (A), Bifidobacterium (B), Lactobacillus (C), Streptococcus thermophilus (D), Salmonella aboni (E), Escherichia coli (F), and Bacillus licheniformis (G) crude extracts on the contractility of Guinea-pig isolated ileum. Arrows indicate the addition of 50, 100, 150, 200, 250, 300 and 350 μL of each crude extract tested.

To identify the microbial genera within the VSL#3 cocktail responsible for the effects on motility in the guinea-pig ileum, crude extract mixtures of Bifidobacterium (B. longum, B. infantis, B. breve), Lactobacillus (L. acidophilus, L. casei, L. plantarum, L. bulgaricus), and Streptococcus thermophilus were individually tested. The following differences in the motility response by the crude extracts of the three probiotic genera were observed: (1) Lactobacillus strains induced a concentration-dependent contraction of the ileum, which was rapid in onset; (2) Bifidobacterium promoted a weak and transient stimulation of the ileum in a non concentration-dependent manner; (3) Streptococcus did not promote any significant response (Figures 1B-D).

The intestinal bacteria Salmonella aboni and Escherichia coli and the non-intestinal bacterium Bacillus licheniformis were used as controls to verify the specificity of the probiotic effects on gut motility (Figures 1E-G). Ileum motility was not affected by the exposure to E. coli and B. licheniformis crude extracts (Figures 1F, G), whereas S. aboni crude extracts were able to trigger the ileum contraction in a dose-dependent manner (Figure 1E). Furthermore, the amplitude of contraction induced by S. aboni was significantly higher than that promoted by the VSL#3 mixture and Lactobacillus crude extracts.

In order to verify the involvement of muscarinic receptors in the observed contraction response, the muscarinic antagonist atropine (1 μmol/L) was added to the organ bath. Atropine did not modify the motility response of ileal tissue exposed to all the bacterial crude extracts tested (data not shown).

The motility response of guinea-pig proximal colon segment was investigated using a similar protocol as with the guinea-pig ileum by adding identical concentrations of the bacterial cell preparations tested in the ileum stimulation study. Similar to the guinea-pig ileum, only the crude extracts of the probiotic VSL#3 mixture showed a significant effect on colon motility. However, the effect on the guinea-pig proximal colon was opposite to that observed in ileum (Figure 2A). In particular, the crude extract of the VSL#3 mixture elicited a rapid and sustained relaxation of the colon tissue, characterized by peaks in response to each addition, together with a progressive lowering of the basal tone. As previously carried out with the ileum, crude extract preparations at pH 5 and 7.4 were tested obtaining the same colon relaxation response (data not shown).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Recorder tracing of the cumulative dose of VSL#3 (A), Bifidobacterium (B), Lactobacillus (C), Streptococcus thermophilus (D), Salmonella aboni (E), Escherichia coli (F), and Bacillus licheniformis (G) crude extracts on the contractility of Guinea-pig isolated proximal colon. Arrows indicate the addition of 50, 100, 150, 200, 250, 300 and 350 μL of each crude extract tested.

The VSL#3 extract mixtures of the genera Bifidoba-cterium, Lactobacillus, and S. thermophilus were assessed (Figures 2B-D). All the bacterial groups provoked a dose-dependent relaxation of the colon tissue, but no significant motility effect was observed by exposure to the control bacteria (Figures 2E-G). Clear colon relaxation effects were promoted by the crude extracts of Bifidobacterium strains and S. thermophilus and were quite similar to those induced by the crude extracts of the VSL#3 mixture (Figure 3). Lactobacillus strain’s crude extract showed a lower relaxation response. Washings of the colon segment exposed to successive additions of all crude extracts tested restored the basal tone of the tissue (data not shown). A further investigation was carried out with the crude extracts of VSL#3 B. infantis and L. casei strains, representative of Bifidobacterium and Lactobacillus genera. Colon relaxation patterns induced by these samples were identical to those obtained with the crude extracts of Bifidobacterium and Lactobacillus mixtures (data not shown).

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Relaxation induced by crude extract of VSL#3 mixture, Lactobacillus, Streptococcus, Bifidobacterium, Escherichia coli, Salmonella aboni and Bacillus licheniformis preparations on Guinea-pig proximal colon. Relaxation was expressed as change in mg-tension per cm of chart paper in resting tone (A). Relaxation was expressed in percentage, by considering VSL#3 mixture-induced maximal relaxation as 100% (B). The volume (μL) indicates the minimal value exerting maximal relaxation effects. Each crude extract was prepared from a bacterial suspension equivalent to a concentration of 3.5 x 10⁹ CFUs/mL. Each point is the mean ± SE of 5 to 7 observations. Mean ± SE was given (P < 0.05).

To confirm that bacterial cytoplasmatic components are mainly responsible for colon relaxation, a more refined fractioning of the B. infantis cells was performed by separating cell wall fraction, membrane proteins, and cytoplasmatic fraction. A concentration-response curve was constructed for each of these bacterial portions (Figure 4). Indeed, the cytoplasmatic fraction showed a sustained relaxation of the colon segment equal to that demonstrated by the B. infantis crude extract. As expected, the cell wall fraction and the membrane proteins did not produce any significant effect.

Open in New Tab   Full Size Figure   Download Figure

Figure 4 Relaxation induced by B. infantis crude extract, cytoplasmatic fraction, membrane fraction and cell wall fraction on Guinea-pig proximal colon. Relaxation was expressed as change in mg-tension per cm of chart paper in resting tone (A). Relaxation was expressed in percentage, by considering crude extract-induced maximal relaxation as 100% (B). The volume (μL) indicates the minimal value exerting maximal relaxation effects. Each crude extract was prepared from a bacterial suspension equivalent to a concentration of 3.5 x 10⁹ CFUs/mL. Each point is the mean ± SE of 5 to 7 observations. Mean ± SE was given (P < 0.05).

The cytoplasmatic fraction of B. infantis was further refined in proteinaceous and non-proteinaceous components. The absence of protein in the non-proteinaceous fraction was verified by SDS PAGE analysis (Figure 5). As reported in Figure 6, the effect on colon relaxation was induced by the non-proteinaceous components, which showed a motility response similar to that exerted by the cytoplasmatic fraction. The relaxation effect exerted by the non-proteinaceous components was confirmed by colon motility experiments carried out with a B. infantis cytoplasmatic fraction deprived of proteins by proteinase K digestion. This preparation, in which the enzymatic protein degradation was demonstrated by SDS PAGE analysis (Figure 5), evoked a significant colonic relaxation response similar to that obtained with the buffered non-proteinaceous preparation (Figure 6). To further characterize the B. infantis non-proteinaceous cytoplasmatic components, the possible involvement of bacterial genomic DNA fraction was tested. Addition to the organ bath of B. infantis DNA ranging from 0.5 μg to 10 μg did not modify the colon motility (data not shown).

Open in New Tab   Full Size Figure   Download Figure

Figure 5 SDS PAGE elec-trophoresis of B. infantis cytoplasmatic fraction (lane 1), cytoplasmatic non-proteinaceous components (lane 2), cytoplas-matic fraction proteinase K digested (lane 3), molecular weight marker (lane 4).

Open in New Tab   Full Size Figure   Download Figure

Figure 6 Relaxation induced by B. infantis cytoplasmatic fraction, cytoplasmatic proteinaceous components, cytoplasmatic non-proteinaceous components and cytoplasmatic fraction proteinase K digested on guinea-pig proximal colon Relaxation was expressed as change in mg-tension per cm of chart paper in resting tone (A). Relaxation was expressed in percentage, by considering cytoplasmatic fraction-induced maximal relaxation as 100% (B). The volume (μL) indicates the minimal value exerting maximal relaxation effects. Each crude extract was prepared from a bacterial suspension equivalent to a concentration of 3.5 x 10⁹ CFUs/mL. Each point is the mean ± SE of 5 to 7 observations.