Prebiotics are non-digestible food ingredients that are selectively fermented by beneficial bacteria and allow spe-cific changes, both in the composition and/or activity of the intestinal microbiota, conferring benefits to the host4. Because of their chemical structure these compounds are neither absorbed in the upper part of the gastrointestinal (GI) tract nor hydrolyzed by digestive enzymes. Among prebiotic polysaccharides, inulin-type fructans are well cha-racterized for their physiological and biochemical features, maintaining healthy GI functions10. For the last 10 years, Jerusalem artichoke (Helianthus tuberosus L.) has been considered one of the most important candidates to be used as raw material for the industrial production of biological inulin due to its agronomic and industrial potential becoming an interesting crop at the regional level.

Furthermore, the combination of probiotics and prebi-otics is known as symbiotic, which has additive or synergistic benefits with enhanced survival and implantation of live microbial cells in the host’s gut3. Another strategy to enhance probiotic survival is microencapsulation which has been proposed as a useful technology to increase the survival of probiotic cells up to their ingestion, storage and through the GI tract6. Different types of polymers are proposed as coating materials, among them, pectin and its derivatives have proved to be potential prebiotics with improved properties causing positive effects in the distal part of the colon⁵.

The main objectives of the present work were (1) to screen the effect of inulinrich carbohydrates (IRCs) extracted from native Jerusalem artichoke on Lactobacillus paracasei subsp. toleraos F2 growth and (2) to evaluate the combined effect of symbiotic and pectin microbeads to improve bacterial survival over GI conditions and refrigera-tion storage.

F2 isolated from the digestive tract of Ramnogaster arcu-ata and selected for its probiotic properties in Oncorhynchus mykiss7 was the chosen strain. The isolation methodology and molecular identification are described in Sica et al.12.

Food grade commercial inulin (CI) [(C6H10O6)n, carbohy-drate content >90g/100g] was kindly donated by Granotec

S.A. (Chile) and used as reference standard for compar-ison. Jerusalem artichoke inulin (JAI) was obtained from H. tuberosus L. tubers grown and harvested in Córdoba province (Argentina). The handling of Jerusalem artichoke tubers and inulin carbohydrate extractions were done as described in Rubel et al.11.

All experiments were carried out in triplicate, and the mean values are presented as means ± standard deviations (SD). Data were analyzed using analysis of variance (ANOVA) followed by the Tukey’s multiple comparison tests when nec-essary. Statistical significance was established at p < 0.05 for all the comparisons.

The prebiotic effect of inulin upon growth of F2 was analyzed in Man, Rogosa and Sharpe (MRS) modified broth supplemented with 1% (w/v) of CI (MRS-CI) and JAI (MRS-JAI) as principal carbon sources to supplement its growth. MRS supplemented with glucose (Merck, Germany) at 1% (w/v) (MRS-G) was employed as positive growth control, while MRS carbohydrate-free basal medium (MRS-B) was employed as negative control without any of the mentioned carbon sources. Inulin solutions were sterilized through 0.2 ^m fil-ters (Gamafil S.A., Argentina) before being added to the corresponding growth media.

Assays in each modified MRS medium (MRS-CI, MRS-JAI, MRS-B and MRS-G) were performed with an initial cell den-sity adjusted by optical density (OD) at 600 nm of 0.250 (~108 cells/ml) using a spectrophotometer (Thermo Spec-tronic Genesys 20, Thermo Electron Corporation, USA). To screen the effect of inulin upon F2 growth, samples were withdrawn at a 2 h interval along 24 h for total viable counts in MRS agar. In all cases, the plates were incubated at 25 ±1 °C for 48 h and colony forming units (CFU)/ml were determined. The final pH of all cultures was measured. The incubation temperature never exceeded 25 °C due to the sample origin and its potential use in aquaculture7,12.

F2 had the ability to grow and ferment inulin with a significant pH decrease of 6.2 ±0.2 to 4.0 ±0.5 after 24h incubation. The fastest growth rate was achieved for MRS-G (1.22 h-1) followed by MRS-JAI (1.33 h-1), MRS-CI (5.71 h-1) and MRS-B (7.43h-1). No significant differences (p>0.05) were evidenced when analyzing F2 viable counts after 24 h incubation in the presence of glucose and JAI. As shown inFigure 1 log CFU of F2 obtained after 24 h did not exhibit significant differences between MRS-G and MRS-JAI, being 12.94±0.37 and 12.50±0.21 logCFU/ml, respectively, followed by MRS-CI with 10.89 ±0.49 log CFU/ml. As expected MRS-B showed the lowest counts (8.77±0.03log CFU/ml, p < 0.01) as it lacked carbohydrates to support the growth.

The prebiotic index (/preb) was calculated as the ratio of probiotic growth in the chosen prebiotic to probiotic growth in a control carbohydrate2.

CFU of probiotics in the prebiotic carbohydrate CFU of probiotic in control carbohydrate

/preb expresses a quantitative value for comparison of dif-ferent prebiotic carbohydrates8. The highest /preb was found for JAI with 0.97 ± 0.01 followed by CI with 0.84 ± 0.06. The /preb registered is almost 1 corresponding to the behaviour evidenced for glucose, being a good index for its replace-ment by an economical and natural resource.

Microcapsules were prepared by external cross-linking of a w/o emulsion1 with some modifications. Briefly, a 2% (w/v) pectin solution was prepared in distilled water at 50 °C under magnetic stirring at 300 rpm for 2 h. The pectin used was low methoxyl Genu Pectin LM104 AS purchased from CPKelco (CPKelco, Limeira, Brazil) with a degree of esterification of 27% and degree of amidation of 20% of the original car-boxyl groups. The solution was left to cool down until 30°C and the probiotic suspension prepared from F2 cultures suspended in sterile phosphate-buffered saline, PBS (0.15mol/l NaCl, 0.05mol/l KH2PO4, 0.05mol/l K2HPO4, pH 7.2) or PBS with 1% CI and JAI was added. The mixture was stirred for another 30min and then a w/o emulsion was prepared by blending 20 ml of the former mixture with 20 ml of commer-cial sunflower oil containing 1% (v/v) Tween 80 (Merck KGaA, Germany). A fine emulsion was generated using a homoge-nizer (Pro 250-Homogeneizer, ProScientific, USA) at a speed of 10 000 rpm for 3 min. Microcapsules were then generated by adding 20g of calcium chloride solution (Anedra S.A., Argentina) (0.5%, w/w) in the form of w/o emulsion similar to that containing pectin and probiotics. This emulsion was added dropwise to the original w/o emulsion (40 ml) under the lowest speed of the homogenizer and left stirring for another 30min.

The resulting microcapsules were recovered by a dou-ble washing-filtering sequence (Whatman® Grade 3 filter paper, provided by Sigma-Aldrich) by diluting 60g of the final emulsion with 250 ml of distilled water containing 0.1% (v/v) Tween 80. Then, they were stored under refrigeration temperature at 4 ±1 °C and protected from light.

Particle size distribution and the area basis mean diame-ter (D32) of the pectin microgels were obtained by analyzing the digital micrographs of the emulsions using the Image J 1.50i software [National Institute of Health (NIH), Bethesda, MD, USA]. The image was converted to grayscale (8-bit) and binarized (converted from grey level to black and white) using a thresholding method to differentiate the pectin microspheres. At least 3 micrographs were analyzed per sample so that more than 1000 particles were used.

Microgel particle sizes ranged roughly between 0.2 and 20 ^m with D32 of 7.48 ±1.30, 7.20 ±0.89 and 6.73 ±1.48 ^m for microcapsules with JAI, CI and without prebiotic, respectively.

Bacterial encapsulation efficiency was determined through the entrapment yield (EY) as stated in Reid et al.9:

Log 10N Log 10N0

where N is the number of viable entrapped cells released from the beads and N0 is the number of free cells added to the polymer mixture immediately before entrapment.

To determine the entrapment of the different suspen-sions of F2, 0.1 g of microcapsules were suspended in 0.1 M sodium citrate buffer pH 6.2 for 20min followed by gen-tle shaking until complete dissolution at room temperature. Serial dilutions were performed and the bacterial contents of the capsules were determined by plating on MRS agar. EY of F2 in pectin microbeads was 92.3% without statistical differences between controls and particles containing IRCs.

To evaluate the protective potential of the prebiotic and microencapsulation over F2 survival through GI passage [sim-ulated gastric (SGJ) and intestinal juice (SIJ)] assays were carried out in a two-step model.

Probiotic suspensions (~108 cells/ml) with 1% (w/v) JAI, CI or in complete absence of IRCs (controls) were inoculated in sterile SGJ composed of 125 mM NaCl, 7 mM KCl, 45 mM NaHCO3 and 0.3% (w/v) pepsin (Saporiti S.A., Argentina), pH was adjusted to 2.0 with HCl (0.1 mol/l). After 90min under low agitation, samples were centrifuged at 5000 x g for 10 min and washed twice with sterile PBS before being re-suspended in SIJ. SIJ was done by using sterile PBS containing pancreatin 0.09% (w/v) (Saporiti S.A., Argentina) and bile salt 0.6% (w/v), pH was adjusted to 8.0 using NaOH (0.1 mol/l). After 90min, incubated cells were harvested by centrifugation and 100 ^l were re-suspended in 900 ^l of peptone water and serially diluted before plating in MRS agar for enumeration. In all cases the samples were incubated at 25 ±1 °C for 48 h.

The survival percentage of F2 after each GI step was calculated as:

Survival % - Log

JAI significantly improved F2 survival to simulated diges-tive solutions compared to CI and control (without carbon sources). After 1.5h in SGJ, F2+JAI exhibited 72.39% survival (6.11 ± 0.03 log CFU/ml) whereas F2+CI and F2-control the survival considerably dropped to 44.90 (3.79 ±0.69logCFU/ml) and 39.69% (3.35 ±0.72log CFU/ml), respectively. The same trend was evidenced after exposure to SIJ with a decrease of 0.20 log units for F2+JAI, 1.27 log units for F2-control and 1.56 log units for F2+CI. At the end of the passage through the GI tract, final viable cell counts were 5.91 ±0.26 log CFU/ml for F2+JAI, 2.23 ±0.75log CFU/ml for F2+CI and 2.08 ± 1.10 log CFU/ml for F2-control.

Figure 1: Growth curves of Lactobacillus paracasei subsp. toleraos F2 in MRS supplemented with 1% (w/v) glucose (MRS-G), 1% commercial inulin (MRS-CI), 1% (w/v) Jerusalem artichoke inulin (MRS-JAI) and without carbohydrates as negative control (MRS-B). Growth curves were determined by viable counts expressed as log CFU/ml. Samples were incubated at 25 ±1 °C for 24 h. 

For encapsulated cells, simulated GI challenge was per-formed at the end of the trial. For this purpose, a total of 1 g of microparticles was analyzed following the same pro-cedure as previously described for free probiotic cells. At the end of each step, the encapsulated cells were released from the particles for viable count, as previously described for EY.

The presence of IRCs increased the resistance of F2 to SGJ exposure; viable cell reductions were 16% and 18% in the presence of JAI and CI, respectively, versus a 23% cell drop for F2 without prebiotics (F2-control). Similarly, at the end of the SIJ step, the encapsulated F2 cells without prebiotics showed the lowest survival, 4.66±0.01 log CFU/g, whereas the presence of IRCs significantly increased their resistance with viable counts of 6.76 ±0.33 log CFU/g for F2+JAI followed by F2+CI with 5.74±0.12CFU/g. No signif-icant differences between the IRCs used were observed.

Furthermore, microencapsulation with pectin and JAI as prebiotic is an excellent tool to preserve the concentration and viability of probiotic cells until they reach the intestine.

To evaluate the combined effect of symbiotic and pectin microbeads to improve bacterial survival in refrigeration storage, pectin microparticles were kept at refrigeration temperature (4°C) for an eight-week (~56 days) interval. The number of culturable probiotic cells was determined once a week. Supplementary Figure 1 shows that the survival rate of encapsulated F2 cells in the presence of IRCs over time was significantly higher (p<0.01) than for free probiotic cells. The viability of F2 entrapped in beads con-tinuously decreased from day 0 to day 56: 2.50, 1.85 and 1.61 log units for control, CI and JAI, respectively.

It was observed that microparticles with the addition of JAI as prebiotic, prepared by emulsification/internal gelation using pectin, represent an efficient system for promoting the resistance of F2 during the simulation of pas-sage through the GI tract, also being an efficient means to promote viability of the microorganism at 4 °C for 56 days.

Conflict of interest

The authors have no conflicts of interest to declare.

Acknowledgments

We thank the Agencia Nacional de Promoción de la Investigación, el Desarrollo Tecnológico y la Innovación, Fondo para la Investigación Científica y Tecnológica (FONCYT), Ministerio de Ciencia, Tecnología e Innovación de la Argentina (PICT 2015, No. 0156) for funding this research and Dr. Rubel Irene for kindly providing the inulin used from Jerusalem artichoke tubers.