Lactobacillus acidophilus CRL 1014 improved “gut health” in the SHIME^®reactor

At weekly intervals, a pure culture of L. acidophilus CRL 1014 (CERELA, San Miguel de Tucumán, Argentina) was inoculated into De Man, Rogosa and Sharpe (MRS) broth (Acumedia, Baltimore, USA). Cultures were harvested during the exponential growth, after, they were centrifuged (4000 × g, 10 min, 4°C) and washed with sterile peptone water. The L. acidophilus CRL1014 cells were kept at the concentration of 10⁸ CFU/mL in sterile peptone water until use [17].

The SHIME^® is a simulator of the human intestinal microbial ecosystem [18, 19] in which environmental conditions (pH, residence time, inoculum, and temperature) are controlled to resemble those found in vivo. A SHIME^® system consists of five double-jacketed vessels, simulating the stomach, the small intestine, and the ascending, transverse and descending colon, with a total retention time of 72 h (Additional file 1: Figure S1). The reactor setup and the composition of the liquid feed (Table 1), which entered the system three times per day, were previously described by Possemiers et al. [14].

The three colon vessels of the SHIME^® reactor were inoculated with bacteria from a fecal sample of a healthy 22-year-old adult female with no history of antibiotic treatment 6 months prior to the study. Aliquots (10 g) of fresh fecal samples were diluted and homogenized with 100 mL of sterilized phosphate buffer (0.1 mol/L, pH 7), containing 1 g/L sodium thioglycolate as the reducing agent.

The microbial inoculum was stabilized over a period of 2-wk on a carbohydrate-based medium and allowed to adapt to the specific environmental conditions of the ascending, transverse and descending colon, in terms of pH range, retention time and available carbon sources [14, 15]. Upon stabilization, the SHIME^® run included 2- wk of basal period (to quantify all steady-state bacterial parameters which were used as starting point to evaluate the effect of a specific treatment), and a 4-wk of treatment period, in which 5 mL of 10⁸ CFU/mL of L. acidophilus CRL 1014 were added once per day to the stomach compartment. Finally, a 2-week washout period without the addition of L. acidophilus CRL 1014 was observed.

At weekly intervals, throughout the entire experimental period, (basal, treatment and washout), 5 mL samples were collected from the reactors for microbiological examinations. The analysis of the intestinal microbiota composition was based on the enumeration of total aerobic and anaerobic bacteria, Enterococcus spp., Lactobacillus spp., Bifidobacterium spp., enterobacteria, and Clostridium spp. One mL of a sample taken from each reactor was suspended in 99 mL of peptone water. Serial dilutions were prepared and inoculated into selective culture media, as follows: total aerobic and anaerobic counts: Standard Methods agar (Acumedia, Baltimore, USA; 37°C/48 h); Enterococcus spp.: KF Streptococcus agar (Acumedia, Baltimore, USA; 37°C/48 h) [16]; Lactobacillus spp.: MRS agar (Merck, Germany; 37°C/48 h, under anaerobiosis). For Bifidobacterium spp. counts was used the Bifidobacterium formulated medium BIM-25 (Difco, France; 37°C/72 h, Anaerobic System, Probac, Brazil) according Munoa & Pares [20], Enterobacteria: MacConkey agar (Acumedia, Baltimore, USA; 37°C/48 h) and Clostridium spp.: RCA Agar (Difco, France; 37°C/48 h, Anaerobic System, Probac, Brazil) [21].

Once a week, throughout the entire experimental period (basal, treatment and washout), samples were collected from the reactors for analysis SCFA and ammonium. The analysis was carried out in triplicates.

Every week, the levels of short-chain fatty acids (SCFA) were determined from samples collected from the reactors and frozen to -20°C. The SCFA were extracted with diethyl ether and determined using a gas chromatograph equipped with a flame-ionization gas detector, a capillary split/splitless injector and an HP-INNOWAX column with a 30 m × 0.25 mm × 0.25 μm inlet (Shimadzu GC2010), using hydrogen as the carrier gas at a flow rate of 1.56 mL/min. The temperatures of the column, injector and detector were 170, 250 and 280°C, respectively [22].

The ammonia content was determined using a selective ion meter (710A model, Orion) coupled to an ammonia selective-ion electrode (Orion 95–12). The apparatus was calibrated using 0.1 M standard ammonium chloride solutions, at the concentrations of 10, 100, and 1000 mg/L of ammonia. To every 25 mL of sample, 0.5 mL of ISA solution (Ionic Strength Adjuster, Orion – a pH-adjusting and ionic force solution) was added. All measurements were carried out at 25°C [23].

DNA was extracted from 2 mL of sample using the QIAamp DNA Stool Mini Kit (Qiagen, Hilden, Germany), according to the manufacturer’s protocol. DNA yield was quantified using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Willmington, USA).

The diversity of the Lactobacillus community in samples taken throughout model operation was assessed by DGGE. To prevent a low amplicon yield a nested PCR approach was used as previously described [24]. This involved a first round of PCR with primers Bact27f (5′-GTTTGATCCTGGCTCAG-3′) [25] and 1492R (5′- CGG CTA CCT TGT TAC GAC-3′) [26], followed by a second PCR with primers Lab159f (5′-GGAAACAGATGCTAATACCG-3′) and Lab677-GCr (5′-GCCCGGGGCGCGCCCCGGGCGGGGCGGGGGCACGGGGGGGCACCGCTACACATGGAG-3′) [24].

PCR was performed using the GoTaq^® Green Master Mix kit (Promega, USA). Samples were amplified in a Veriti^® 96-Well Thermal Cycler (Applied Biosystems, USA) by using the following program: initial denaturation at 94°C for 2 min; 35 cycles of denaturation at 94°C for 30 s, annealing at primer-specific temperature for 40 s, elongation at 72°C for primer-specific time, and extension at 72°C for 5 min, followed by a final cooling to 4°C. The annealing temperature and elongation time was set at 52°C/1.30 min with primers Bact27f and 1492r and 60°C/1 min with primers Lab159f and Lab677-GCr. The PCR products that were used as templates in nested PCR were purified with the QIAquick PCR Purification Kit (Qiagen, USA).

DGGE analysis of PCR amplicons was based on the protocol described by Muyzer et al. [27] by using the DCode System (Bio-Rad Laboratories, Hercμles, CA, USA). Electrophoresis was done as described previously [24] in an 8% polyacrylamide gel with a denaturant gradient of 30–50% (100% was defined as 40% formamide and 7 M urea) for 16 h at 85 V in a 0.5× TAE buffer at a constant temperature of 60°C. Gels were stained with silver nitrate by the method of Sanguinetti et al. [28], scanned at 400 d.p.i., and further analyzed by the BioNumerics 6.0 software (Applied Maths). The distance matrices of each DGGE based on the Pearson correlation similarity coefficient to cluster the samples was analyzed using the BioNumerics software (Applied Maths).

Bands of interest in the lactobacilli community fingerprint were excised from the gel and transferred into 25 μl of TE buffer, and incubated overnight at 37°C to allow diffusion of the DNA. Two microliters of the eluted DNA were used for reamplification with the GC-clamped primer by using the conditions described above and the PCR products generated were checked by DGGE. Only PCR products which yielded a single band and comigrated with the original band were purified by the QIAprep spin miniprep kit (Qiagen, USA) and were subjected to DNA sequence analysis at the genomic facilities of The Human Genome Research Center (HGRC), at the University of São Paulo. BLAST searches were performed to determine the closest known relatives of the partial rRNA gene sequence obtained in GenBank.

The sequences of the 16S rDNA fragments were deposited in the GenBank database. The accession numbers of the 4 sequences are as follows (band code in parentheses): KF054352 (lac1), KF054351 (lac2), KF054353 (lac3) and KF054350 (lac4).

The significance of all results was investigated with one-way ANOVA, and individual means were compared using the Tukey’s test (p < 0.01), using the statistical software Sigma Stat 5.0.

Table 2 shows the microbial counts obtained for the flasks that simulated the ascendant, transverse and descendant colon of the SHIME^® reactor. Using traditional selective growth media, the microbiological analyses revealed that the administration of L. acidophilus CRL1014 influenced the composition of the intestinal microbial community.

Plate counts were used to assess the capacity of L. acidophilus CRL1014 to temporarily colonize the colon during a simulated long term administration and to investigate the effects on the composition of the indigenous microbial community in the SHIME^®. As reflected in the plate count data (Table 2), the administration of L. acidophilus to the system induced a significant increase (p < 0.01) in lactobacilli and bifidobacteria counts, with a concentration increase of at least 2 log CFU in all colon compartments. However, in all colon compartments, a high increase in clostridia starting from the second week of the basal period in ascending colon and first week of the treatment in transverse and descending colon was observed (Table 2).

For other microbial groups, such as total aerobes, facultative anaerobes, and enterobacteria, there were no significant changes in the populations of these microorganisms during the treatment period.

DGGE analysis was used to monitor qualitative changes in the composition and structure of the Lactobacillus communities in the three compartments simulating the colon conditions (Figure 1).

Clustering of the specific DGGE fingerprints for lactobacilli (Figure 1) indicated that the treatment had effect on the composition of the Lactobacillus community, after 4 weeks of treatment with L. acidophilus CRL1014, higher similarity values (>90%) were found between reactors 3, 4, and 5 (Figure 1). Finally, after two wk the treatment period was finished and the highest similarity values (99 to 91%) were found in all colon regions between the last week of treatment and the washout period (Figure 1).

Based on the DGGE fingerprint analysis of the colon microbial community, several shifts in bands or changes in band intensity were observed. To identify the bacterial species that were responsible for those changes, DNA fragments from bands of interest were excised from the DGGE gel, isolated, and finally sequenced. We were not able to obtain sequences from all bands but four. The successfully sequenced rDNA fragments were marked as lac 1, lac 2, lac 3, and lac 4. The band marked as lac1 migrates to the same position as the fragment obtained from pure cultures of L. acidophilus CRL1014 and showed a 99% identity with L.acidophilus (JQ031741.1). The fragments lac2 and lac3 demonstrated high identity (>98%) to Lactobacillus casei (JQ412731.1) and Lactobacillus johnsonii (AB186343.1), respectively. Finally, the band marked “lac 4” was dominant during the whole experimental period and had 99% of similarity with Lactobacillus sakei (GI,AB609050.1).

Short chain fatty acids (SCFA) analysis and ammonium production are often used to characterize microbial metabolism.

During the treatment period with L. acidophilus, the values of concentration of ammonium ion decreased significantly (p < 0.01) in all the regions investigated (Figure 2).

Figure 3 depicts the production of acetate, propionic, and butyric acids during the basal, treatment, and washout periods in the SHIME^® vessels. During treatment with L. acidophilus CRL 1014, a significant increase (p < 0.01) occurred in the production of acetate acid in all the reactors analyzed. However, the greatest concentration of these acids occurred in vessel three, which simulates the ascending region of the colon. The propionic acid showed a significant increase (p < 0.01) in reactors 3 and 4 (ascending and transverse colon) and the butyric acid increased significantly (p < 0.01) in reactors 4 and 5 (transverse and descending colon). In the washout period, the levels of SCFA diminished in all vessels. The highest SCFA production consisted of acetic acid in all vessels (Figure 3). However, in reactor 5 the highest increase was in propionic acid. Lactic acid was only observed for the treatment period in all vessels (ascending, transverse and descending).