An In Vitro Approach to Study Effects of Prebiotics and Probiotics on the Faecal Microbiota and Selected Immune Parameters Relevant to the Elderly

The aging process leads to alterations of gut microbiota and modifications to the immune response, such changes may be associated with increased disease risk. Prebiotics and probiotics can modulate microbiome changes induced by aging; however, their effects have not been directly compared. The aim of this study was to use anaerobic batch culture fermenters to assess the impact of various fermentable carbohydrates and microorganisms on the gut microbiota and selected immune markers. Elderly volunteers were used as donors for these experiments to enable relevance to an aging population. The impact of fermentation supernatants on immune markers relevant to the elderly were assessed in vitro. Levels of IL-1β, IL-6, IL-8, IL-10 and TNF-α in peripheral blood mononuclear cell culture supernatants were measured using flow cytometry. Trans-galactooligosaccharides (B-GOS) and inulin both stimulated bifidobacteria compared to other treatments (p<0.05). Fermentation supernatants taken from faecal batch cultures supplemented with B-GOS, inulin, B. bifidum, L. acidophilus and Ba. coagulans inhibited LPS induced TNF-α (p<0.05). IL-10 production, induced by LPS, was enhanced by fermentation supernatants from faecal batch cultures supplemented with B-GOS, inulin, B. bifidum, L. acidophilus, Ba. coagulans and Bac. thetaiotaomicron (p<0.05). To conclude, prebiotics and probiotics could lead to potentially beneficial effects to host health by targeting specific bacterial groups, increasing saccharolytic fermentation and decreasing inflammation associated with aging. Compared to probiotics, prebiotics led to greater microbiota modulation at the genus level within the fermenters.


Introduction
Currently, there is an increase in life expectancy, thus a rapidly aging population. According to WHO, the population of adults aged 60 and over has doubled since 1980, and by 2050 this Compared to inulin, short chain FOS has also been shown to improve immune function in older persons [34,40,41]. Two synbiotics containing mixtures of Bifidobacterium bifidum BB-02, Bifidobacterium lactis BL-01 and inulin, and a mixture of Lactobacillus acidophilus and lactitol, were shown to exert positive effects on microbiota composition in healthy elderly persons [16,42]. Bacillus coagulans GBI-30, 6086 (GanedenBC 30 (BC30)) has the potential to suppress the growth of pathogens [43]. In an in vitro study, both the cell wall and the metabolite fractions of BC30 were shown to possess immune modulation properties, anti-inflammatory effects and direct induction of IL-10 [36].
Few studies have directly compared the efficacy of both probiotics and prebiotics in modulation of gut microbiota composition and immune function within the same setting. By targeting a population aged 60-75 it may be possible to target the microbiota and the immune changes in their early stages. Therefore, the aim of this study was to use an in vitro approaches with samples from donors aged 60-75 years to compare the impact of prebiotics and probiotics on the gut microbiota and selected immune markers relevant to the elderly. Common commercial prebiotic and probiotic products were used. Inulin and B-GOS were used as prebiotics; Bifidobacterium bifidum, Lactobacillus acidophilus, and Bacillus coagulans were used as probiotics. Placebos were microcrystalline cellulose and maltodextrin. Bac. thetaiotaomicron and S. typhimurium were also used to investigate the influences of commensal bacteria and a pathogen respectively on the test parameters.

Materials and Methods Chemicals
Bacteriological growth medium supplements were obtained from Oxoid Ltd. (Basingstoke, Hants, U.K.). Inulin was obtained from BENEO GmbH (Mannheim, Germany) and B-GOS from Clasado Ltd (Milton Keynes, UK). All nucleotide probes used for fluorescent in situ hybridisation (FISH) were commercially synthesised and labelled with the fluorescent dye Cy3 at the 5 0 end (Sigma-Aldrich Co. Ltd., Spain). Sterilisation of media and instruments was carried out by autoclaving at 121°C for 15 min.

Faecal sample preparation
Faecal samples were collected from three individuals (62-66 years of age). All volunteers were in good health and had not ingested antibiotics for at least 6 months before the study. Samples were collected on site on the day of the experiment and were used immediately. These were diluted 1:10 (w/v) with anaerobic phosphate buffered saline (PBS; 0.1 M; pH 7.4) and homogenised in a stomacher for 2 min (460 paddle beats/min). Resulting faecal slurries from each individual were used to inoculate batch culture vessels.

Sample processing
In preparation for FISH analysis 375 μl batch culture supernatant was taken in duplicate into two tubes of 4°C 1125 μl 4% (w/v) paraformaldehyde solution and fixed at 4°C for 4 hours. After 4 hours, the batch culture supernatant was centrifuged for 5 minutes at 11337 xg (Eppendorf centrifuge minispin, Eppendorf, UK) at room temperature. The supernatant was carefully removed and discarded. The pellet was re-suspended in 1 ml of cold 1×PBS by aspirating carefully using a pipette. Again, the sample was centrifuged for 5 minutes at 11337 xg at room temperature and the supernatant discarded. The sample was washed again in 1 ml cold PBS as above and centrifuged. All supernatant was carefully removed. Finally, the pellet was re-suspended in 150 μl cold 1×PBS and 150 μl ethanol. The sample was mixed by vortexing and then stored at -20°C.
In preparation for SCFA analysis, 1 ml of batch culture supernatant was taken in duplicate and centrifuged for 10 minutes at 11337 xg. The supernatant was stored at -20°C.
For in vitro immunoassays, 1 ml of batch culture supernatant was taken in triplicate, centrifuged for 10 minutes at 11337 xg and filtered through a 0.22 μm filter device (Millipore, Schwalbach, Germany). The cell-free supernatant was finally stored at -20°C.

Organic acid analysis
Organic acid production was determined by GC. Extraction and derivatisation of samples was conducted according to Richardson, Calder [51]. A 5890 SERIES II Gas Chromatograph (Hewlett Packard, UK) using an Rtx-1 10m×0.18mm column with a 0.20μm coating (Crossbond 100% dimethyl polysiloxane; Restek, Buckinghamshire, UK) was used for analysis of SCFA. Temperatures of injector and detector were 275°C, with the column programmed from 63°C for 3 minutes to 190°C at 10°C min -1 and held at 190°C for 3 minutes. Helium was the carrier gas (flow rate 1.2 ml min -1 ; head pressure 90 MPa). A split ratio of 100:1 was used. The SCFA standard was run every 20 samples to update the calibration as necessary. This standard solution contained (mM): sodium formate, 10; acetic acid, 30; propionic acid, 20; isobutyric acid, 5; n-butyric acid, 20; iso-valeric acid, 5; n-valeric acid, 5; sodium lactate, 10; sodium succinate, 20. Peak areas of the standard solution, to which internal standard was added, were used to calculate response factors for each organic acid with respect to the internal standard. Response factor and peak areas within samples were calibrated and calculated using Chemstation B.03.01 (Agilent Technologies, Cheshire, UK). The response factors were calculated using Eq 1. Amount of organic acids in the samples was calculated using Eq 2.

Preparation of peripheral blood mononuclear cells
Fasted blood samples were taken from six healthy volunteers aged 60-73 years, in sodium heparin vacutainer tubes (Greiner Bio-One Limited, Gloucestershire, United Kingdom). The study was conducted according to guidelines laid down in the Declaration of Helsinki, and all procedures involving human subjects were approved by the Ethics Committee of the University of Reading. The ethics approval number was UREC 14/05. Written informed consent forms were obtained from all subjects. Blood was layered over an equal volume of lympholyte (Cedarlane Laboratories Limited, Burlington, Ontario, Canada) and centrifuged at 930 xg for 15 min at room temperature. Peripheral blood mononuclear cells (PBMCs) were harvested from the interface, washed once with PBS, and then resuspended in Roswell Park Memorial Institute (RPMI) 1640 medium (containing glutamine, Autogen Bioclear Ltd., Wiltshire, UK). These steps were repeated to achieve low contamination of erythrocyte. The pellet was finally resuspended in RPMI 1640 medium and cell numbers counted using trypan blue and a cell counter (Coulter, Fullerton, CA, USA). Cells were adjusted to the required concentration.

Viability assays
To determine the appropriate supernatant concentration, PBMC viability, at different supernatant concentrations was determined using the trypan blue test. PBMCs, adjusted to 2×10 6 cells/ml, were incubated in twenty-four-well plates in the presence of RPMI 1640 medium, pure batch culture medium supernatant, 0h and 24h supernatant from B. bifidum treated and S. typhimurium treated vessels separately for 24 h at 37°C in an air-CO 2 (19:1) atmosphere. The tested supernatant amounts of each treatment were 1%, 1.5%, 2%, 3%, 4%, 5% and 10% (v/v) of 2ml (final working volume). At the end of the incubation, cell numbers were counted using trypan blue test. According to the results, 1% (v/v) was appropriate to use for different treatment supernatants.

Cytokine stimulation and detection
PBMCs, adjusted to 2×10 6 cells/ml, were incubated in twenty-four-well plates in the presence of 1 mg/ml lipopolysaccharide (LPS; L4516, Sigma-Aldrich Co. Ltd. UK), 1% (v/v) pure batch culture medium, 1 mg/ml LPS with 1% (v/v) pure batch culture medium or 1 mg/ml LPS with 0h, 5h and 24h 1% (v/v) supernatants from ten vessels for 24 h at 37°C in an air-CO 2 (19:1) atmosphere. At the end of the incubation, cell culture supernatants were collected and stored at -20°C for later analysis of cytokine production. Non-stimulated cultures were used as blank controls.

Statistical analysis
All statistical tests were performed with the use of SPSS version 18 (SPSS Inc, Chicago, IL). Results are presented as means (n = 3) ± SD.
For bacterial populations and SCFA concentrations, within the same treatment, differences from 0-h value were tested using paired Student's t test. At the same time point, differences among treatments were analysed by one-way ANOVA. For cytokine production, differences from LPS value were tested using an independent t test. Within the same fermentation treatments, variations from 0-h values were tested using paired Student's t test. At the same time point, differences among treatments in cytokine production were analysed by one-way ANOVA. Significant differences were determined by post hoc Tukey HSD (Honestly Significant Difference) test. A value of P <0.05 indicates a significant difference.

Enumeration of bacterial populations by FISH
Bacterial populations are shown in Fig 1 and S1 Fig. In the control vessel, growth of Atopobium group (p<0.05, paired Student's t test) and total bacteria (p<0.05, paired Student's t test) were stimulated compared to 0h. Growth of bifidobacteria was significantly stimulated by B-GOS, inulin and maltodextrin during fermentations compared to control (p<0.05, ANOVA), with higher levels following B-GOS fermentation. B. bifidum, L. acidophilus and Ba. coagulans were also shown to significantly stimulate bifidobacterial numbers compared to time 0h (p<0.05, paired Student's t test). Numbers of lactobacilli/enterococci were significantly increased following B-GOS, inulin, L. acidophilus and Ba. coagulans at 30h and 48h compared to other treatments (p<0.05, ANOVA). Numbers of Eubacterium rectale-Clostridium coccoides were increased following B-GOS fermentations at 48h compared to others (p<0.05, ANOVA). In addition, the Clostridium histolyticum group was reduced following B-GOS fermentation at 30h compared to other treatments (p<0.05, ANOVA). Following maltodextrin fermentation, levels of Bacteroides-Prevotella spp. and Clostridium histolyticum group were significantly stimulated compared to other treatments (p<0.05, ANOVA). Following the different treatments, there was no significant change in total bacterial numbers, indicating that overall bacterial numbers remained constant following prebiotic (B-GOS and inulin) and probiotic (B. bifidum, L. acidophilus and Ba. coagulans) use. In the control vessel, as a carbon source, potato starch stimulated the production of all SCFAs compared to 0h (p<0.05, paired Student's t test). Acetate production was significantly stimulated following B-GOS and maltodextrin fermentation compared to other treatments (p<0.05, ANOVA). Propionate production was significantly stimulated following maltodextrin fermentation compared to other treatments (p<0.05, ANOVA). Levels of butyrate were significantly higher in vessels with B-GOS (p<0.05, ANOVA) and inulin (p<0.05, ANOVA) compared to others. Production of branched chain fatty acids, iso-butyrate and iso-valerate, were repressed by prebiotics (B-GOS and inulin) and probiotics (B. bifidum, L. acidophilus and Ba. coagulans) (p<0.05, ANOVA). However, they were significantly higher in vessels with maltodextrin (p<0.05).

Cytokine production
Supernatants from PBMCs cultured without batch culture supernatant were used as controls (+/-). In the absence of LPS, there was no stimulation of IL-1β, IL-6, IL-8, IL-10 and TNF-α mean ± SD from triplicate samples.*, significant differences from the 0h value within the same treatment, p<0.05. Significant differences (p<0.05) among treatments at the same time point are indicated with different letters from the same colour of column.

Discussion
Prebiotics and probiotics have been shown to modulate the intestinal bacterial composition towards a potentially healthy composition in elderly populations in several studies [2,13,42,43]. The current study directly compared the impact of both prebiotics and probiotics on the gut microbiota of elderly volunteers using an in vitro approach; then using an ex vivo approach monitored the potential impact on selected immune parameters.
In the current study B-GOS led to a positive microbial shift, with the potential for reduced inflammation by stimulating bifidobacteria growth, enhancing IL-10 production and inhibiting TNF-α production. Positive effects of B-GOS on colonic bacterial balance with stimulation of bifidobacteria, concurrent with reduced inflammation following intervention was observed by Vulevic et al., [13]. The reduced inflammatory potential observed in the current study was not as dramatic as that observed in the in vivo study of Vulevic, such differences could be related to the PBMC in vitro approach. The impact of B-GOS on the microbiota has been observed in different clinical settings, such as, overweight adults [52] and Irritable Bowel Syndrome patients [53]. In addition, an in vitro study looking at the porcine microbiota also confirmed the positive effects of B-GOS [54]. In the current study, the positive effect of B-GOS on beneficial bacteria at the expense of pathogenic bacteria showed that B-GOS intervention could lead to a potentially beneficial shift of microbiota composition in elderly persons [13,55]. This is relevant when considering the changes that occur in the microbiota during ageing, this includes lower levels of bifidobacteria and increased inflammation. The results from the current study also showed a positive microbial shift following inulin with stimulation of bifidobacteria and lactobacilli which has also been supported by several in vivo and in vitro studies [2,[56][57][58].
The current study confirmed the bifidogenic effects of B. bifidum used as a probiotic, rather than in a synbiotic combination. In previous studies, synbiotics containing B. bifidum were also shown to induce a significant stimulatory effect of the bifidobacteria genus rather than B. bifidum alone [16,59]. Furthermore, the stimulation of lactobacilli by probiotic L. acidophilus in this study was similar to that of a synbiotic containing L. acidophilus, observed to increase faecal lactobacilli levels in healthy elderly [42,60]. This shows that both of these probiotics possess this functionality in the absence of a prebiotic.
Both prebiotics and probiotics may modulate the microbiota composition by targeting different beneficial bacterial groups. Consequently, gut barrier function may be improved, pathogen infections reduced and disease risk decreased. Prebiotics showed the potential to lead to greater microbiota modulation at the genus level compared to probiotics in the current in vitro study. When comparing B-GOS and inulin, B-GOS showed a greater stimulatory effect on positive bacteria and a greater inhibitory effect on harmful bacteria. This indicates that under the current conditions, B-GOS was a more effective prebiotic candidate in modulating microbiota composition. Values are mean ± SD. PBMC from three volunteers was incubated with batch culture supernatants for 24h. #, significant differences from LPS value p<0.05. *, significant difference from 0-h value within the same fermentation treatments. At the same time point, differences among different treatments in cytokines production were analysed by one-way ANOVA. Significant differences (p<0.05) determined by post hoc Tukey HSD test were not found. In Changes in SCFA production were associated with microbiota influences following treatment. As bifidobacteria and E. rectale-C. coccoides are producers of acetic acid [61] and butyric acid [13,61], respectively, B-GOS and inulin showed stimulatory effects on these two acids. Alterations in SCFA and BCFA production suggest proteolytic fermentation was reduced upon fermentation of B-GOS, inulin, B. bifidum, L. acidophilus and Ba. coagulans. Proteolysis is often associated with dysbiosis and negative fermentation end-products, such as ammonia and nitrosamines [62]. As such the results indicate a potential shift in fermentation to the more beneficial saccharolysis.
Although a few studies have shown that prebiotics and probiotics could directly modulate cytokine production of elderly people in vitro [63][64][65][66], the current study is the first to directly compare their effects. In addition, the metabolites of pathogenic and commensal bacteria were considered. Cell-free supernatants contain batch culture medium, faecal water and metabolites of substrates. LPS would invoke an immune response and subsequently stimulate production of immune markers. Cell free fermentation metabolites may subsequently have anti-inflammatory effects by inhibiting production of TNF-α and enhancing production of IL-10. The SCFA production would become stable after 24 hours, therefore batch culture supernatants at 0h, 5h and 24h were collected and incubated with LPS and PBMC.
The down-regulation effects of metabolites from prebiotics and probiotics on TNF-α suggest anti-inflammatory potential. A positive impact may be directly associated with fermentation end products of prebiotics and probiotics. A few studies have shown that TNF-α production induced by stimuli in vitro could be inhibited by SCFA, especially butyrate and acetate [67][68][69][70]. This study showed fermentation supernatants from prebiotics and probiotics contained high levels of acetate and butyrate, with anti-inflammatory potential. Therefore, this study indicated the beneficial effects of prebiotics and probiotics metabolites and their beneficial effects on selected immune markers in elderly.
IL-10 is an important anti-inflammatory cytokine, which may counteract the production of proinflammatory cytokines, such as TNF-α [71,72]. In this study, supernatants from prebiotic and probiotic fermentations enhanced production of IL-10 in vitro. There may be several fermentation metabolites associated with this impact, for example SCFA [67,69,72]. Similarly, enhancement of IL-10 production by Bac. thetaiotaomicron may be also linked to its fermentation end products, although the increase was not as dramatic as that produced by prebiotics and probiotics. In this study S. typhimurium has not been found to change inflammation status, although prebiotics and probiotics led to a more positive inflammatory status.
Prebiotics (B-GOS, and inulin) and probiotics (B. bifidum, L. acidophilus and Ba. coagulans) led to a change in the balance of the microbiota to a potentially positive balance, as seen by an increase in bifidobacteria, a group known to be at reduced levels in older people. Furthermore, supernatants from prebiotic and probiotic fermentations showed an anti-inflammatory effect by inhibiting production of pro-inflammatory cytokines and enhancing production of antiinflammatory cytokines which was possibly related to SCFA concentrations. This research indicates that prebiotics and probiotics have huge potential for modulating the microbiota and inflammation status of elderly people. Furthermore, the prebiotic effect observed was more marked than that of probiotics. Such results are important when evaluating the best treatment to use in targeted interventions. addition, cytokines in non-stimulated PBMC (blank) and in pure batch culture medium-treated PBMC (batch) were also determined. There was no significant difference between them. There was also no significant difference between LPS (LPS-stimulated PBMC) and batch+LPS (PBMC incubated with pure batch culture medium and LPS). Values are mean ± SD. PBMC from three volunteers was incubated with batch culture supernatants for 24h. #, significant differences from LPS value p<0.05. Ã , significant difference from 0-h value within the same fermentation treatments. At the same time point, differences among different treatments in cytokines production were analysed by one-way ANOVA. Significant differences (p<0.05) determined by post hoc Tukey HSD test were not found. In addition, cytokines in non-stimulated PBMC (blank) and in pure batch culture medium-treated PBMC (batch) were also determined. There was no significant difference between them. There was also no significant difference between LPS (LPS-stimulated PBMC) and batch+LPS (PBMC incubated with pure batch culture medium and LPS). (TIF)