Effect of gastric and small intestinal digestion on lactic acid bacteria activity in a GIS1 simulator.

The selection of probiotic strains resistant to gastrointestinal transit is an important stage when developing supplements that contain viable biomass. A total of six strains belonging to different genotypes were tested and compared with both a positive and negative control (Lactobacillus plantarum 5s). Significant differences were found between strains as a result of gastrointestinal transit using the in vitro GIS1 static simulator. The Lactobacillus rhamnosus 428ST strain showed maximum viability as a result of in vitro transit, featuring a survival capacity value, Cs, of over 50 ± 0.01%. The remaining genotypes that were tested showed significant reductions in the enzymes and bile salts at the time of action. The value of the survivability capacity was directly correlated with the synthesis of exopolysaccharides and lactic acid. The test results of the GIS1 system have been compared with those of other studies on gastrointestinal transit resistance that used dynamic models.

Introduction

The digestion process is essential for maintaining one’s health status, as components of ingested food become assimilated. This process begins in the mouth, continues in the stomach and the small intestine, and is finalized in the colon, both by the absorbed substances and by the action of the microbiota that are developed at this level (Guerra et al., 2012). Thus, according to the degree of resistance to this transit process, and to the effects of the interaction between the microbiota that already exist in the colon of each person, positive health-related effects can be achieved. These effects are manifested by a high tolerance to lactose, the correction of episodes of diarrhoea, a reduction in the frequency of bloating, or an inhibition of pathogenic bacterial strains (Hertzler and Clancy, 2003).

The use of functional supplements that are based on the biomass of some probiotic strains is widespread today, even though the in vivo effect of these agents is poorly known. Their administration is usually performed according to the results of an in vitro test, which is based on each agent’s degree of resistance in various conditions given that it partly stimulates the gastrointestinal environment (either by affecting the pH level or bile salt concentration) (Pitino et al., 2010). Currently, one particular method that is employed to test these strains is the use of in vitro simulators; this accurately simulates the physiological conditions associated with gastrointestinal transit. These models, dynamic or static, precisely reproduce the time of retention, pH level, bile salt concentration, and quantity of the digestive enzymes. The systems can be controlled in real time and show the exact changes that appear following the specific action of each tested strain (Hur et al., 2011, Vamanu et al., 2012).

This study determined the viability of six strains belonging to types Weissella and Lactobacillus, which were used to formulate a probiotic product; we subsequently compared these strains to positive and negative controls. The two controls served to accurately analyse the data. Freeze-dried biomass was used because it is regarded as the best preservation method for probiotic microbial strains (Rathnayaka, 2013). In fact, the final form of the tested probiotic strains will be freeze-dried.

Prior research has used a static system to achieve gastrointestinal simulation in vivo, otherwise known as GIS1. Survival capacity (Cs) was determined to highlight the influence of gastrointestinal transit in vitro on the viability of these probiotic strains, as well as to assess its influence on secondary parameters – namely, viscosity, lactic acid synthesis, and exopolysaccharides.

Materials and methods

Samples

The Weissella paramesenteroides FT1a, Lactobacillus sp. 34.1, Lactobacillus rhamnosus E 4.2, Lactobacillus sp. 18.1, Lactobacillus fermentum 428ST, and Lactobacillus plantarum FCA3 were obtained from the collection of the Faculty of Biotechnology, Bucharest, Romania. The L. plantarum 5s strain was used as a positive control (Vamanu et al., 2006), while the Escherichia coli CBAB 2 strain was used as a negative control; these were obtained from the collection of the Faculty of Biotechnology, Bucharest, Romania (Vamanu et al., 2006, Vamanu, 2014). The bacterial strains were maintained in MRS (Oxoid, Basingstoke, Hampshire, UK) and LB broth (Sigma–Aldrich Co., St. Louis, MO, USA), and they were freeze-dried in a Christ-Alpha 1–2 LD freeze dryer (Vamanu et al., 2012).

In vitro simulation system

All in vitro experiments were performed using the GIS1 system (www.gissystems.ro). The in vitro system consisted of a 1000 mL Duran bottle with a removable screw cap; the bottle was made of borosilicate glass. The Duran bottle screw cap had four entry points: the first entry point was used for the sample and gastric juice (pump 1) and HCl (Fluka Biochemica, Switzerland) 1 N solutions (pump 2); a second entry was used for the NaHCO3 (Fluka Biochemica, Switzerland) 1 N solutions (pump 3), as well as for the bile salts and pancreatin solution (pump 4); a third entry point was used for the pH electrode, 3 mm in diameter, from Hanna Benchtop pH Meters HI 2211; and the fourth entry point was used for the thermometer, which maintained the contents at a temperature of 37 °C. To enter the system, four laboratory peristaltic pumps (Behrotest, Type PLP 33; flow rate: 0.4–2.0 L h−1) were used. The Duran vessel was placed on the magnetic stirrer with a ceramic heating plate (IKA C-MAG HS 7) to constantly shake the contents at 50 rpm (Vamanu et al., 2012).

Viable counting

Cell viability was determined by serial dilution in physiological saline solution. The highest dilutions were then plated on MRS/LB agar and evaluated by an automated colony counter (ColonyQuant; Schuett-biotec GmbH, Göttingen, Germany) using the corresponding software (Vamanu et al., 2012, Vamanu, 2014). To compare bacterial survival at the gastrointestinal level, the Cs was calculated (Vamanu et al., 2012, Vamanu et al., 2013).

Determination of lactate concentrations

Lactate concentrations were measured in vitro, during gastrointestinal digestion, using the Lactate Assay Kit (Sigma–Aldrich Co., St. Louis, MO, USA) (Curtis et al., 2000). The reaction mix consisted of 1000 μL of phosphate buffer (pH 7.5), 100 μL of lactate dehydrogenase, and 100 μL of the sample. The mix was kept in the dark at the room temperature for 30 min. The absorbance was read at 570 nm with a Helios λ spectrophotometer. The amount of lactate from the samples was determined from the standard curve.

Determination of exopolysaccharides

Exopolysaccharide synthesis was determined by precipitation (24 h, at 4–6 °C) with cold absolute ethanol after the bacterial cells were removed (3500g, for 30 min and at 9 °C) by centrifugation at 3000g for 15 min (Salazar et al., 2008, Polak-Berecka et al., 2015, Wasko et al., 2013).

Determination of viscosity

Viscosity (cP) was measured on an Ubbelohde viscometer. The analysed sample was introduced via capillary laboratory equipment. The sample was maintained at 20 °C, and the flow time of the liquid sample between the two parts of the equipment was determined. Viscosity was calculated using the following formula:where η = dynamic viscosity; k = viscometer constant; t = flow time (seconds); and ρ = density at 20 °C (Frengova et al., 2002, Polak-Berecka et al., 2015).

Statistical analysis

All of the parameters were assessed in triplicate, and the results were expressed as the mean ± standard deviation (SD) of the three observations (P < 0.05). Mean values and standard deviations were calculated using the Excel program from the Microsoft Office 2010 software package.

Results and discussion

The viability of the strains used during gastric and intestinal transit is shown in Fig. 1, in parallel with that of the two controls, L. plantarum 5s and E. coli CBAB 2. After 1 h in the presence of an average pH of 2–3, a significant decrease in cell viability was observed; this was mainly found for the two controls, although it was also noted for the L. plantarum FCA3 and W. paramesenteroides FT1a strains. No significant change was observed in the Cs of the L. rhamnosus E 4.2 and Lactobacillus sp. 18.1 strains as a result of the gastric transit. The most significant decrease in viability following digestion at the gastric level was noted for the W. paramesenteroides FT1a and L. plantarum 5s strains, which lost approximately 50 ± 0.01% of the number of viable cells, particularly during the first hour of action in the presence of pepsin and hydrochloric acid. The negative control was stronger by approximately 10 ± 0.1%.

Figure 1

Survival capacity of lactic acid bacterial strains during simulated gastrointestinal digestion.

The W. paramesenteroides FT1a, L. plantarum 5s, and L. rhamnosus E 4.2 strains showed the most significant decrease in Cs by approximately 12 ± 0.5% during the last hour of digestion at the gastric level (P < 0.1). Strains of the Lactobacillus sp. 18.1 and Lactobacillus sp. 34.1 showed constant Cs values. The remaining strains did not exhibit variations in viability greater than 5 ± 0.01% during the final stage of gastric digestion; thus, the general loss in cell viability was barely 3 ± 0.001% for the Lactobacillus sp. 18.1 between the initial and final moments in the gastric digestion process.

A significant difference was found for most strains tested between the start of digestion and the end of the in vitro simulation. The differences in viability were generally over 50 ± 0.01%, meaning that there was a significant decrease in the number of viable cells. The L. fermentum 428ST strain and the negative control (E. coli CBAB 2) were exceptions to this finding; they showed an average loss of approximately 40 ± 0.01% (Fig. 1).

In general, the various strains showed the most significant decrease in viability after 2–3 h of digestion at the duodenal level. The greatest loss was found for Lactobacillus sp. 18.1, which had lost 50 ± 0.01% of its viable cells during the last hour of digestion, resulting in a final Cs value of 0.34 ± 0.02. In contrast, L. fermentum 428ST had lost only 6 ± 0.001% of its viable cells during the same period. The remaining strains showed losses in viability during the last 2 h of development at the duodenal level between 25% and 46 ± 0.01%. Lower losses were observed for W. paramesenteroides FT1a, which exhibited a reduction of approximately 12%, while losses of 13 ± 0.03% were found for the negative control (E. coli CBAB 2) (Vamanu, 2014).

The results highlight the fact that the functional products used in this study hold promise in the formulation designed herein, especially with respect to the L. fermentum 428ST, Lactobacillus sp. 34.1, L. rhamnosus E 4.2, Lactobacillus sp. 18.1, and L. plantarum FCA3 strains. Comparative studies carried out with the two controls demonstrated that cell viability is a specific parameter that can be accurately assessed using the in vitro simulator, GIS1. The synthesis of secondary metabolites (lactic acid and/or exopolysaccharides) is a direct indicator of the capacity of strains tested in unfavourable environments at the gastric and small-intestine levels (Cock et al., 2013). These findings were confirmed by previous in vitro studies of Lactobacillus acidophilus johnsonii, Lactobacillus casei immunitas, and Lactobacillus casei shirota (Lo Curto et al., 2011). The exact role of exopolysaccharides on the viability of lactic acid bacteria is not well known. It is accepted that it favours colonization and the capacity to remain at the gastrointestinal level, which is evident based on its ability to maintain viability and by its prominent role in human health (Darilmaz et al., 2011). With respect to L. fermentum 428ST, a significant quantity of carbon dioxide appeared to be released during simulation. Moreover, the synthesis of polysaccharides was reduced during simulation, decreasing throughout the various stages of gastrointestinal digestion. Thus, by the end of the digestion process, the quantity of polysaccharides was 0.53 ± 0.005 μg/mL, and during the last 2 h of digestion in the intestine, the value decreased by approximately 30 ± 0.01%. In contrast, the strains of W. paramesenteroides FT1a had the lowest level of exopolysaccharide synthesis, with a value close to that of the control, L. plantarum 5s; the value of W. paramesenteroides FT1a reached a maximum of 0.2 ± 0.0005 μg/mL. The most significant quantity was found for L. rhamnosus E 4.2, which was an average of four times greater than that for L. fermentum 428ST throughout the entire in vitro simulation process. It was further demonstrated that polysaccharide synthesis was correlated with the lack of carbon dioxide released.

In terms of viscosity, there was an evolution proportional to exopolysaccharide synthesis. A decrease was observed in relation to passage from one compartment in the digestion process to another. The most significant decrease, by approximately 10 ± 0.60%, was calculated for the Lactobacillus sp. 34.1 strain; otherwise, the η value decreased by an average of 5 ± 0.40%. The L. fermentum 428ST strain appeared to be the exception to this finding, as a constant average value was calculated (1.15 ± 0.06 cP) during the entire period of in vitro simulation (Fig. 2).

Figure 2

Dynamic viscosity during simulated gastrointestinal digestion.

The Cs value directly correlated with the synthesis of lactic acid, which was interpreted to be a direct indicator of viable cell metabolism. During the in vitro transit process, the lactic acid level decreased, particularly during the intestinal stage, for strains W. paramesenteroides FT1a, Lactobacillus sp. 34.1, and L. plantarum FCA3. The average value determined at the end of the intestinal stage was of 25 ± 0.20 mM. In contrast, the L. fermentum 428ST strains showed double the quantity of lactic acid at the end of the intestinal transit. The correlation between the Cs value and lactic acid synthesis yielded significant values during gastric transit, 0.7–0.86 ± 0.02; this was lower during transit at the intestinal level (0.34–0.48 ± 0.10; P < 0.1). Furthermore, the correlation between cell viability and the dynamic viscosity values was significant (0.85–0.97 ± 0.07; P < 0.1).

The most resistant strain, L. fermentum 428ST, presented with an average viability of 4.73 ± 0.24 log CFU/mL when compared to controls, whose average viability decreased by approximately 25% following transit. When compared with the other strains that were extremely sensitive to the action of pancreatin and bile salts during the last 2 h of digestion, L. fermentum 428ST maintained a constant decrease, not exceeding an average of 0.5 ± 0.03 log CFU/mL. The remaining strains that were tested even showed losses that reached 1 log CFU/mL (L. plantarum FCA3). In this study, the loss of viability for some probiotic strains reached a maximum of 60 ± 0.19%. The results obtained herein were expressed in terms of the Cs value, which fell within the limits reported in the literature for all strains (P < 0.05); the only exception to this finding was observed for W. paramesenteroides FT1a. Previous studies have shown that viability can be improved by the presence of a prebiotic (Vamanu et al., 2012).

In conclusion, this study showed that the value of cell viability was directly correlated with the synthesis of lactic acid and exopolysaccharides, as well as with the value of viscosity during the in vitro simulation of gastrointestinal transit in the GIS1 system. Further studies are needed to determine the Cs in the colon using, in parallel, a static and a dynamic model for the in vitro simulation of segments of the human colon.