Viability, Storage Stabilityand In Vitro Gastrointestinal Tolerance of Lactiplantibacillus plantarum Grown in Model Sugar Systems with Inulin and Fructooligosaccharide Supplementation

: This study aims to investigate the effects of inulin and fructooligosaccharides (FOS) supplementation on the viability, storage stability, and in vitro gastrointestinal tolerance of Lactiplantibacillus plantarum in different sugar systems using 24 h growth and 10 days survival studies at 37 ◦ C, inulin, and FOS (0%, 0.5%, 1%, 2%, 3% and 4%) supplementation in 2%, 3%, and 4% glucose, fructose, lactose, and sucrose systems. Based on the highest percentage increase in growth index, sucrose and lactose were more suitable sugar substrates for inulin and FOS supplementation. In survival studies, based on cell viability, inulin supplementation showed a better protective effect than FOS in 3% and 4% sucrose and lactose systems. Four selected sucrose and lactose systems supplemented with inulin and FOS were used in a 12-week storage stability study at 4 ◦ C. Inulin (3%, 4%) and FOS (2%, 4%) supplementation in sucrose and lactose systems greatly enhanced the refrigerated storage stability of L. plantarum . In the gastrointestinal tolerance study, an increase in the bacterial survival rate (%) showed that the supplementation of FOS in lactose and sucrose systems improved the storage viability of L. plantarum. Both inulin and FOS supplementation in sucrose and lactose systems improved the hydrophobicity, auto-aggregation, co-aggregation ability of L. plantarum with Escherichia coli and Enterococcus faecalis .


Introduction
Inulin and fructooligosaccharides (FOS) are among the most studied and well-established prebiotics. Inulin and FOS consist of a linear chain of fructose, constituted by a monomeric unit of fructose linked by beta glycosidic (2, 1) bonds, with a terminal glucose unit. Inulin has a heterogeneous degree of polymerization ranging (DP) from 6 to 60, while FOS has a DP ranging from 2 to 10 [1][2][3]. The inability of the human digestive system to hydrolyze fructans is due to the lack of effective hydrolytic enzymes that can break β linkages [4]. However, probiotics such as Lactobacillus and Bifidobacterium can degrade these bonds. Furthermore, inulin and FOS are known to modulate intestinal microflora composition and metabolic activity, promoting the growth of bifidogenic bacteria [5]. There are a number of researchers investigating the effect of prebiotics on the viability of probiotics in various food products such as fermented milk [6,7], yogurts [8,9], soy milk [10], fruit juices [11], oat-based products [12], and fermented cream cheese [13]. These studies show inconsistent effects of prebiotics, particularly the effect of oligosaccharides on the growth and viability of probiotics in complex food matrices. Hence, we hypothesized here that the effect of prebiotics on probiotic strains in various food products might be due to the different sugar compositions of these foods.
Lactiplantibacillus plantarum, formerly known as Lactobacillus plantarum [14], is generally regarded as safe (GRAS) and has a long history of safe usage in food products as the average degree of polymerization (DP) for inulin and FOS was reported to be ≥10 and between three to eight, respectively.

Reactivation of Probiotic Culture and Preparation of Inoculum
L. plantarum was activated from glycerol stock according to the method of Nazzaro et al. [37]. Glycerol stock of L. plantarum was streaked on MRS agar and incubated anaerobically in anaerobic jars (Anaerobic Plus System, Oxoid, Hampshire, UK) at 37 • C for 48 h with AnaeroGen sachets (Oxoid, Hampshire, UK). A single colony of L. plantarum from MRS agar was transferred to 10 mL of MRS broth and incubated at 37 • C for 48 h under anaerobic condition followed by centrifugation at 10,000× g for 10 min at 4 • C. The supernatant was removed, and cell pellets obtained were washed thrice with PBS before inoculating into MRS media. The bacterial culture was regularly sub-cultured and maintained on MRS agar plates at 4 • C along with gram staining to detect cross-contamination with other microorganisms.
During inoculum preparation, L. plantarum is grown in MRS broth for 18 at 37 • C, 120 rpm incubator shaker under anaerobic condition followed by centrifugation at 10,000× g for 10 min at 4 • C.

Growth Curve Study
Each medium with OD 600 (7 log CFU/mL) of inoculum was individually dispensed into a 96-well plate in an anaerobic chamber. The plate was then incubated and measured simultaneously inside a TECAN Spark ® 10M microplate reader (TECAN, Grödig, Austria) at 37 • C for 24 h (Figure 1). Microbial growth was monitored by measuring the absorbance at the OD 600 at 60 min intervals using the TECAN automated microplate reader, with 15 s auto-shaking at 1440 rpm before each measurement. The change obtained at OD 600 was then plotted against time, and the growth index was calculated using the equation below from the sum of all single OD readings and compared with the OD values obtained with the cultures grown in MRS-broth according to modification by Bevilacqua et al. [38]: where OD MAX was the maximum absorbance attained, OD NC was the absorbance of negative control (MRS broth without any sugar and FOS), and OD PC was the absorbance of positive control (MRS broth with 4% glucose).
The OD values of samples with inulin or FOS supplementation were corrected with the OD measurement of MRS broth with inulin or FOS without any bacterial culture. The OD values of samples without inulin or FOS were corrected with the OD measurement of MRS broth without any bacterial culture.
Percentage of increase in growth index = (Growth index of a sugar with inulin/FOS supplementation-Growth index of the sugar without supplementation)/Growth index of the sugar without supplementation ×100.

Survival Study
The survival assay was performed according to Bevilacqua et al. [39]; modified by Parhi et al. [35]. MRS broth (20 mL) was dispensed in Schott bottles and inoculated at 5% w/v with L. plantarum ( Figure 1). The number of viable cells in culture per mL was determined by spread plating 0.1 mL of serially diluted cultures on MRS agar media and incubated at 37 • C for 48 h under anaerobic conditions. Viable cell count was expressed as log CFU/mL. The cell viability (%) and pH (pH-meter F-71, LAQUA, Irvine, CA, USA) at 25 • C were measured at 2-days intervals for 10 days.
Cell viability (%) = CFUmL −1 Day−T CFUmL −1 where CFUmL −1 Day−T was the viable cell count at the day of analysis and CFUmL −1 was the initial viable cell count. Sugar solutions were filter-sterilized using 0.22 µm polyethersulfone membrane syringe filters and added aseptically to Schott bottles with sterilized MRS broth. After inoculating L. plantarum at a concentration of 5% w/v and 24 h fermentation at 37 • C under anaerobic conditions, cultures were stored for 12 weeks at 4 • C. Storage viability (log CFU/mL), pH, auto-aggregation, co-aggregation, hydrophobicity, sugar analysis, organic acid analysis, ethanol analysis, and gastrointestinal assay were analyzed every alternate week. The pH of a medium was measured using a pH-meter (F-71, LAQUA, Irvine, CA, USA) at 25 • C. Storage viability was expressed as log CFU/mL, and colonies were counted after allowing them to grow at 37 • C for 48 h under anaerobic conditions.

Sugar and Organic Acid Analysis
Sugars such as glucose, fructose, sucrose, lactose, and organic acids such as lactic acid and acetic acid were quantified using high-performance liquid chromatography (HPLC) (Agilent, Santa Clara, CA, USA). Samples were diluted with deionized water by a factor of two and filtered through 0.22 µm dual syringe filters (Thermo-line, Sydney, Australia) before they were injected. Ten µL samples were injected into a Hi-Plex Ca column (Agilent, Santa Clara, CA, USA; 300 × 7.7 mm) using Milli-Q water as mobile phase at a flow rate of 0.6 mL/min at 80 • C for sugar determination, and Hi-Plex H column (Agilent, Santa Clara, CA, USA; 300 × 7.7 mm) using 0.01 M H 2 SO 4 as mobile phase at a flow rate of 0.6 mL/min at 75 • C for organic acid. Sugars and organic acids were detected using a refractive index detector (RID). The concentration of each sugar in the medium was determined from their respective calibration curves based on their peak areas obtained from their standard solutions.

Acid Tolerance Assay
The isolates were incubated overnight in MRS broth at 37 • C. Actively grown cells were harvested by centrifugation (10,000× g, 4 • C, 10 min). The pH of MRS broth was adjusted at pH 1.0, 1.5, and 2.0 with 1N HCl [10]. MRS broth adjusted to pH 6.5 was used as a control. Harvested cells were resuspended in MRS broth with acidic pH and incubated at 37 • C. After a time interval of 0, 1, 2, and 3 h, samples were withdrawn and serially diluted in phosphate buffer saline (PBS). Samples were plated on MRS agar plates and incubated at 37 • C for 48 h. Cell viability was assessed by the spot plate count method, and the results were expressed as log CFU/mL.

Cell Auto-Aggregation and Co-Aggregation
Cell auto-aggregation and co-aggregation were performed according to Kos et al. [40]. Harvested cells were washed, resuspended in PBS, and adjusted to an absorbance of 0.5 at 600 nm every alternate week. The suspension was incubated at 37 • C for 2 h. One mL of the upper phase was removed carefully to measure the absorbance at 600 nm. Cell auto-aggregation was measured by a decrease in absorbance and measured by using the following equation: where A% represents the percentage of auto-aggregation, A 0 represents the initial value (0 h), and A t represents the final value (2 h).
Equal volumes (2 mL) of L. plantarum and pathogen (E. coli and E. faecalis) suspensions for the co-aggregation assay were divided into glass test tubes and mixed by vortexing. Control tubes containing 2 mL of suspension of each bacterial species. Absorbance was measured after 5 h. The percentage of co-aggregation was determined according to Kos et al. [40]: where A represents absorbance, x and y represent each of the two strains in the control tubes, and (x + y) represents their mixture.
2.6.6. Hydrophobicity of Bacteria Hydrophobicity was determined following the method of Kimoto-Nira et al. [41] with some modifications. First, five mL aliquots of cultures were collected every alternate week in storage assay, centrifuged at 10,000× g for 10 min at 4 • C, and suspended in PBS to obtain an OD 620 of 1.0. Next, one mL of xylene was added to 1.0 mL of cell suspensions ( Figure 1). The solution was incubated at 30 • C for 10 min, mixed for 60 s, and then left to stand for 15 min. The aqueous phase was removed, and OD 620 was determined. The percentage of hydrophobicity was calculated using the following equation: where H% represents the percentage of hydrophobicity, H o represents the initial value (0 min), and H t represents the final value (15 min).

Gastrointestinal Tolerance Assay
The tolerance of probiotics during storage of in vitro digestion was determined by a modified method of Valero-Cases, Frutos [42]. The ringer solution (1000 mL) containing NaCl (6.2 g), KCl (2.2 g), CaCl 2 (0.22 g), and NaHCO 3 (1.2 g) was used as an electrolyte solution to prepare simulated saliva, gastric juice, and intestinal fluid. One mL of cultures were collected over two weeks intervals and then centrifuged 10,000× g for 10 min, at 4 • C. The cells were washed thrice using PBS buffer. The cell pellet was resuspended in 1 mL of PBS pH 7.0 ± 0.2.
Step 1: 1 mL of simulated saliva (100 mg/mL lysozyme in ringer solution) with pH 6.5 ± 0.2 was added to resuspended cells and incubated for 2 h at 37 • C in a water bath.
Step 2: 100 µL aliquots were removed from samples for serial dilution and spot plated on MRS agar. CFU was calculated after incubation at 37 • C for more than 48 h. Step 3: 1 mL of simulated gastric juice (3 mg/mL pepsin in 0.85% NaCl) with pH 2.5 ± 0.2 was added to the same mixture and incubated for 4 h at 37 • C in a water bath.
Step 4: 100 µL aliquots were removed from samples for serial dilution and spot plated on MRS agar. The viable count was calculated after incubation at 37 • C for more than 48 h. Step 5: 1 mL of simulated intestinal fluid (0.3% bile salt + 1 mg/mL pancreatin in 0.85% NaCl solution) with pH 8.0 ± 0.2 was added to same mixture and incubated for 6 h at 37 • C in water bath.
Step 4: 100 µL aliquots were removed from samples for serial dilution and spot plated on MRS agar. The viable count was calculated after incubation at 37 • C for more than 48 h ( Figure 1). The percentage of survival was calculated using the following formula: Bacterial Survival Rate (%) = Log 10 CFU after gastrointestinal assay Log 10 CFU before gastrointestinal assay × 100

Statistical Analysis
Experiments were performed in three independent replicates, and results were expressed as means and standard deviations. These results were statistically analyzed using t-test (week 0 and 12), one-way analysis of variance (ANOVA), and Tukey's test for post-hoc analysis. Statistical significance was determined at p < 0.05 using SPSS (Statistical Package for the Social Sciences) version 23 from IBM Corporation (New York, NY, USA)

Growth Study
The growth index described by Bevilacqua et al. [38] was used where growth index > 75% stands for growth kinetics similar to those under optimal conditions; growth index in the range of 25-75% underlines a partial inhibition; growth index < 25% stands for potent inhibition of the microorganism [38]. The significant increase in viable count (log CFU/mL) of L. plantarum (Supplementary Tables S1 and S2) correlates with the growth index of L. plantarum (Tables 1 and 2). The growth index of L. plantarum increased with increasing inulin and FOS concentration from 0.5% to 3% as a sole carbon source in MRS broth, suggesting a dose-dependent effect of inulin and FOS on the growth, but a further increase in concentration (4%) showed partial inhibition (Tables 1 and 2). A positive growth effect of 1% inulin of L. plantarum ST16 Pa has been reported by da Silva Sabo et al. [43]. A similar dose-dependent effect of inulin and FOS was reported by Parhi et al. [35] on the growth of L. casei. Moreover, Munoz et al. [44] reported that CFU increased with increasing FOS (1%, 2%) concentration for L. casei LE8, but CFU decreased with increasing FOS (5%) concentration for L. plantarum LE27. The chain lengths of inulin and FOS affect the fermentability of L. plantarum. The short chains of FOS (DP < 10) were rapidly fermented, and long chains of inulin (DP > 20) were steadily fermented [1], resulting in a positive effect on the growth of L. plantarum. However, the growth index of L. plantarum was observed to be below 75% in MRS broth, with 0.5%, 1%, and 4% inulin as the sole carbon source (Table 1). This might be due to the accumulation of partially hydrolyzed inulin and fructose moieties on the outer cell wall, limiting mass transfer in the system. Table 1. Growth index (%) of L. plantarum grown in MRS broth containing 2%, 3%, and 4% of glucose, fructose, sucrose, and lactose supplemented with 0%, 0.5%, 1%, 2%, 3%, and 4% of inulin during 24 h growth at 37 • C.  In comparison to a positive control (growth index = 100%), inulin as the sole carbon source could not support the growth of L. plantarum well. The percentage increase in growth index was determined by comparing the growth index of the inulin or FOS supplemented sugar system with the growth index of the non-supplemented respective sugar system. The growth media, MRS broth containing 2%, 3%, 4% glucose, and fructose, showed growth index >75% while 2%, 3%, and 4% sucrose, and lactose showed growth index < 75%, this suggests that L. plantarum prefers glucose and fructose than sucrose and lactose (Tables 1 and 2). The percentage increase in growth index, which ranged from 5.0-65.4% and 5.5-73.7% with inulin and FOS supplementation, respectively, in all sugar systems, shows the positive effect of inulin and FOS supplementation on L. plantarum. The increase in growth index is most likely due to the release of fructose as a result of partial hydrolysis of inulin and FOS by an extracellular enzyme β-fructofuranosidase produced by L. plantarum [30]. Partial hydrolysis of inulin releases shorter sucrose and fructose, which were subsequently metabolized as an additional carbon and energy source [45]. Perrin et al. [46] reported that the standard for prebiotic action is that probiotics possess cell-associated glycosidases that hydrolyze prebiotics such as oligosaccharides to form monomers of fructose. The highest percentage increase in growth index for inulin supplementation was 65.4% in 2% lactose with 3% inulin, followed by 49.3% in 3% sucrose with 4% inulin. As for FOS, the highest percentage increase in growth index was 73.7% in 3% lactose with 4% FOS followed by 46.5% in 3% sucrose with 2% FOS (Tables 1 and 2). This indicates that for positive effect in growth index of L. plantarum, sucrose and lactose were more suitable sugar substrates for inulin and FOS supplementation. Since FOS showed the highest percentage increase in growth index, it was the better prebiotic supplementation in sucrose and lactose systems. Growth-promoting effects of inulin and FOS on L. plantarum have been reported in the presence of lactose-rich food matrices such as fermented milk [7], yogurts [8], whey [47], and cheese [48]. The positive effect was also exhibited in glucose and fructose systems but at a lower percentage of increase. In the glucose system, the highest percentage growth index increase of 8.3% and 8.5% was observed in 2% glucose with 2% inulin and 2% FOS supplementation, respectively (Tables 1 and 2). This indicates that the positive effect of inulin was similar to FOS in the glucose system. The highest percentage growth index increases 18.0% and 11.4% in the 2% and 3% fructose system was with 2% inulin and 0.5% FOS supplementation, respectively (Tables 1 and 2). This indicates that inulin was a better prebiotic than FOS in the fructose system. L. plantarum preferred growth on glucose compared to fructose, sucrose, and lactose without any inulin and FOS supplementation, but with inulin and FOS supplementation, a significant number of positive effects on growth index were observed in fructose and lactose systems. The results on the growth index of L. plantarum using fructose as the carbon source is consistent with the studies of Corcoran et al. [49], Hedberg et al. [50], and Kneifel [51], in which Lactobacillus spp. utilized fructose efficiently but less than glucose. Several researchers reported that glucose is transported by the phosphotransferase system (PTS), which regulates uptake and metabolism of other carbon sources [52][53][54]. In contrast, fructose metabolism requires the induction of specific enzymes before sugars enter the Embden-Meyerhof pathway (EMP); hence, a slower fructose utilization than glucose was exhibited in the substrate uptake without affecting inulin and FOS metabolism. The results in Tables 1 and 2 also suggest that to achieve the positive effect of inulin or FOS supplementation in the growth index of L. plantarum, a higher concentration of sugar (>4%) is not desirable. Several in vitro studies have reported the positive effect of inulin and FOS on Lactobacillus and Bifidobacterium strains [6,55,56]. This study demonstrates that the concentration and type of sugar and prebiotic are essential factors for stimulating or suppressing the growth of L. plantarum. The cell survivability of L. plantarum increased until day two of incubation and then decreased exponentially, indicating that L. plantarum was experiencing a death phase. Figures 1-3 show the comparison of the survival of L. plantarum in different model sugar systems with inulin and FOS supplementation on day eight and day ten. On day eight and day ten, L. plantarum did not survive in non-supplemented 2% sugar systems media (Figure 2), suggesting a protective effect of inulin and FOS supplementation. On day eight, with inulin supplementation, 40-50% cell viability was observed with 2% sugar systems, with few exceptions (Figure 2A). On the other hand, 35-40% cell viability was observed with 2% fructose with 1-4% FOS supplementation, 2% sucrose, and 2% lactose with 0.5-4% FOS supplementation on day eight ( Figure 2B). Thus, inulin supplementation was better than FOS on day eight in the 2% sugar system based on cell viability results. Interestingly, the cell viability of L. plantarum was 11.1% and 22.1% in 2% fructose with 1-4% inulin supplementation ( Figure 2A) and 2-4% FOS supplementation ( Figure 2B), suggesting FOS was better in sustaining cell viability than inulin in 2% fructose system on day ten. The accumulation of organic acids and other secondary metabolites and the depletion of carbohydrates in the media may increase stress, thereby hindering the survival of L. plantarum. The >20% cell viability of L. plantarum on day 10 in inulin and FOS supplemented media suggest a protective effect. Livingston, Henson [57] earlier reported a similar protective effect of inulin and FOS against stresses, which may be due to plausible interaction of inulin or FOS and phospholipids of membrane resulting in higher membrane stability [58]. For higher concentrations of sugars at 3% and 4%, L. plantarum showed 35-40% cell viability on day eight with inulin supplementation (Figures 3A and 4A). As for FOS supplementation, 25-40% cell viability was observed on day eight (Figures 3B and 4B). Supplementation of inulin resulted in higher cell viability of L. plantarum than FOS supplementation on day eight in 3% and 4% sugar systems. On day 10, cell viability was significantly higher (23-28%) with 3%, 4% sucrose, and lactose supplemented with 0.5%, 1%, 2%, 4% inulin compared to other systems ( Figures 3A and 4A). As for FOS supplementation, 16-22% cell viability of L. plantarum was observed with 3% and 4% fructose, sucrose, and lactose ( Figures 3B and 4B), with a few exceptions. Overall, inulin supplementation showed a 4-5% higher percentage increase in cell viability of L. plantarum than FOS supplementation in 3%, 4% sucrose, and lactose systems. In fermented foods, L. plantarum is often used as a starter culture since it utilizes lactose with high conversion rates and other nutrients such as protein present in whey [47]. This study shows a similar result whereby 25% cell viability of L. plantarum was observed in higher concentrations of 4% lactose (Figure 4). Slow transport of lactose [59] and slower utilization of lactose [60] in Lactobacillus might be advantageous, whereby lactose was available for a more extended period in comparison to other sugars resulting in viable cells until day ten. Sucrose is a disaccharide made up of fructose and glucose, previously reported as unable to be utilized efficiently or even metabolized by L. rhamnosus [49,61,62]. But here in L. plantarum, slower uptake and slower assimilation may have increased carbon source availability in the media similar to lactose and improved cell viability in inulin and FOS supplemented media. As sucrose was metabolized into glucose, a primary carbon source, and fructose, an inducer for β-fructofuranosidase, may have contributed to partial utilization of inulin and FOS, but this requires further investigation.

Effect of Inulin and FOS Supplementation on Storage Viability (log CFU/mL) and pH during 12 Weeks of Storage at 4 • C
Most commercial probiotic food products are often sold under refrigeration for limited shelf life. Therefore, it is essential to understand the effect of inulin or FOS supplementation on the viability of L. plantarum stored under refrigeration. Four selected combinations with more than 40% growth index increase were investigated in a 12-week storage study at 4 • C. All analyses in week 0 were made after 24 h growth of L. plantarum in MRS broth. L. plantarum grew well in MRS broth at week 0, but storage viability (log CFU/mL) and pH depended on the concentration of sugar and prebiotics (Table 3). In week 0, the storage viability of L. plantarum was significantly higher in MRS broth with inulin and FOS supplementation (10.21 ± 0.02-10.40 ± 0.12 log CFU/mL) than those of non-supplemented media (9.07 ± 0.01-9.29 ± 0.06 log CFU/mL) at 37 • C (Table 3). However, the storage viability of L. plantarum declined week until week four in all combinations of media. Due to the low initial pH values, MRS broth was subjected to mild lactic acidification. Exhaustion of essential growth factors or their limited availability and acidification of this complex broth are postulated as possible reasons for the decline of storage viability at week two and week four. These seem similar to the decrease in viable cell count of L. plantarum stored for week four at 4 • C in pineapple, tomato, carrot, and cherry juices [63].   Prebiotic supplementation improved storage viability and significantly preserved L. plantarum by almost three log cycles compared to non-supplemented media (Table 3). In general, the concentration of probiotics in supplemented media should be above the lowest recommended therapeutic level of 6 log CFU/mL [64] and within the minimum recommended daily dose of 10 8 to 10 9 cells [65]. In this study, the storage viability of L. plantarum at the end of 12-week storage at 4 • C was in the range of 7.29 ± 0.01 log CFU/mL to 8.40 ± 0.04 log CFU/mL for 3S+4I, 2L+3I, and 3S+2FOS, 3L+4FOS (Table 3). We speculated that inulin and FOS might form a gel matrix around the probiotic cells [66] and function as a thickener by contributing to a higher total solid content and protecting probiotics from injury during storage.
The ability of L. plantarum to tolerate pH 3.4-8.8 and temperature 12-40 • C makes it a significant commercial strain. However, the production, distribution, and storage of probiotics creates a harsh environment resulting in high mortality and low efficacy of microorganisms. The survival assay and refrigerated storage assay mimic the processing and storage environment of L. planatrum, respectively. The decrease in the cell viability of L. plantarum in the survival assay (Figures 2-4) seems to be more drastic than those in the refrigerated storage assay ( Table 3). The survival assay was performed for ten days at 37 • C with continuous shaking at 120 rpm. This leads to faster consumption of sugar and prebiotic, resulting in the accumulation of organic acids and other secondary metabolites and the depletion of carbohydrates in the media that may increase stress, thereby hindering the survival of L. plantarum. In the refrigerated storage assay at 4 • C, metabolism is slowed down, thus preventing the higher accumulation of organic acids and complete exhaustion of carbohydrates in the media, albeit for a duration of 12 weeks.

Sugar and Organic Acids
As anticipated, the concentration of sucrose significantly (p < 0.05) decreased during storage ( Table 4). The sucrose concentration in 3S+4I was 1.4% (Table 4); this was significantly higher than 3S, which was 1% at week 0. Similarly, at week 0, lactose concentration was significantly higher in 2L+3I (1.8%) and 3L+4FOS (2.8%) than 2L and 3L ( Table 4). The higher concentration of sucrose and lactose with inulin and FOS supplementation suggests that L. plantarum utilized inulin and FOS in the presence of sucrose and lactose at week 0. On week 12, sucrose concentration dropped below 0.1% with both inulin and FOS supplementation (Table 4), while storage viability was 8.1 ± 0.00 log CFU/mL and 8.2 ± 0.0 log CFU/mL (Table 3), respectively. However, on week 12, lactose concentration was significantly higher (p < 0.05) in 3L+4FOS (0.7%) than 3L (0.2%), while no significant difference was observed between 2L and 2L+3I (Table 4). This could be due to a difference in initial lactose concentration and availability of short-chain FOS in (3L+4FOS) supplemented media. The inefficient lactose utilization by some lactic acid bacteria has not been fully investigated; however, L. rhamnosus GG had been described by its slow utilization of lactose compared to glucose in several in vitro studies [49,50] and even its inability to ferment the lactose [61]. Honda et al. [59] described slow lactose transport in L. brevis KB290 as a reason for inefficient utilization of lactose. Furthermore, β-galactosidase, an inducible enzyme, was required to hydrolysis lactose into glucose and galactose [66]. In addition, complexities involved in galactose metabolism were also responsible for the slower utilization of lactose by L. acidophilus [67]. Studies of other Lactobacillus species have identified various genetic systems responsible for utilizing carbohydrates of varying complexity. For example, the simultaneous utilization of sucrose and FOS might be due to the sucrose phosphoenolpyruvate (PEP)dependent phosphotransferase system (PTS), which transports short-chain FOS into the cytosol, and it was further digested by intracellular β-fructofuranosidase in L. plantarum WCSF1 [34]. Also, the L. acidophilus NCFM genome was coded for an ABC transport system and a putative intracellular β-fructosidase and found to hydrolyze sucrose, inulin-type fructans, or inulin [32]. Similar systems may have been induced during fermentation and storage; the simultaneous utilization of sucrose and FOS or inulin may have led to sucrose concentrations dropping to 0.05-0.1% with and without inulin and FOS supplementation while maintaining 8.13-8.21 log CFU/mL at week 12 (Table 3).
Lactic acid was the primary fermentation end product. The highest concentration of 12.52% was found in 3S+2FOS, followed by 3S+4I and 3S at week 12 (Table 4). At week 0, lactic acid concentration after 24 h fermentation was 6.56% in 3S+2FOS, followed by 4.89% in 3S+4I and 4.32% in 3S ( Table 4). The lactic acid concentration was significantly higher in 3S+2FOS (12.5 ± 0.0%) compared to 3S+4I (10.8 ± 0.0%) and 3S (10.8 ± 0.0%) at week 12, suggesting that sucrose supplemented with FOS increased metabolism, and thus resulted in accumulation of lactic acid in the media (Table 4). Sucrose supplemented with inulin (3S+4I) and FOS (3S+2FOS) created a better growth medium that stimulated L. plantarum than non-supplemented media and resulted in higher (p < 0.05) production of lactic acids. This was consistent with the findings of Desai et al. [68], who found the improved metabolic activity of several species of Lactobacillus in the presence of selected prebiotics. Several authors reported that the utilization of prebiotics and the levels of primary metabolites varied depending on the strain [68,69]. The lactic acid concentration depended on the type of sugar, with sucrose giving a higher (p < 0.05) concentration than lactose. Although lactic acid production is desirable in fermented dairy foods, such a high concentration of organic acids showed no detrimental effect to 3S+4I and 3S+2FOS since the cells maintained relatively constant viability throughout storage. Furthermore, inulin and FOS sustained the metabolic activity of the culture during cold storage, maintaining higher cell viability and resulting in an increase in the production of primary metabolites such as lactic acid (Table 4).
In ethanol analysis, no peaks were observed as L. plantarum are facultative heterofermentative. L. plantarum uses glucose through the EMP to produce lactic acid, while they may also possess an inducible phosphoketolase pathway (PK) with pentose acting as inducers [70].

Auto-Aggregation, Hydrophobicity, and Co-Aggregation
One of the important characteristics of probiotics is the adhesion of microorganisms to the human intestine, preventing their immediate elimination by peristalsis and providing a competitive advantage in this ecosystem [71]. Aggregation properties are important characteristics of bacterial strains used as probiotics and have been linked to the adhesion of Lactobacillus in a previous study [72]. The auto-aggregation ability of bacteria maintains the bacterial population in the gut [73]. At week 0, auto-aggregation rate significantly increased by 28.1% (3S+4I), 37.3% (2L+3I), 35.6% (3S+2FOS), and 32% (3L+4FOS) in comparison to the non-supplemented media 3S, 2L and 3L ( Table 5). As the storage study proceeds, a decline in auto-aggregation rate can be seen in all media comparative to week 0. At week 12, the auto-aggregation rate of 3S+4I, 2L+3I, 3S+2FOS, and 3L+4FOS was 16-26% higher than non-supplemented media, suggesting inulin and FOS supplementation significantly improved auto-aggregation ability of L. plantarum ( Table 5). The level of adhesion determines bacterial membrane hydrophobicity to hydrocarbons. This study observed varying adhesion to xylene for L. plantarum grown in 3S+4I, 2L+3I, 3S+2FOS, and 3L+4FOS (Table 5). Hydrophobicity for L. plantarum in 3S+2FOS (90%) was highest, followed by 3S+4I (89%), 2L+3I (88%), and 3S+2FOS (86%) at week 0 ( Table 5). As the storage study proceeds, a decrease in hydrophobicity was observed, similar to the auto-aggregation rate. However, a 20-40% increase in hydrophobicity was observed when L. plantarum was stored in supplemented media compared to non-supplemented media at week 12 (Table 5). Some authors have suggested that improved surface properties such as auto-aggregation and hydrophobicity correlate with their adhesive capacity [40,74]. Previously, Ramos et al. [36] reported a 61.9% auto-aggregation rate and no hydrophobicity of L. plantarum SAU96. Moreover, Kotzamanidis et al. [74] reported a 44.3% auto-aggregation rate and 61.3% hydrophobicity of L. plantarum 2035 in MRS broth, suggesting auto-aggregation and hydrophobicity of L. plantarum vary with different strains of Lactobacillus. Similar results were shown by L. paracasei 276 and L. plantarum WSFC-1, where auto-aggregation rate and hydrophobicity improved with supplementation of FOS and inulin [75]. The results in this study are also in accordance with Li et al. [76]. The carbohydrate source (sucrose) in the growth medium affected the surface parameters, henceforth influencing membrane hydrophobicity of L. plantarum. Table 5. Auto-aggregation, hydrophobicity, and co-aggregation of L. plantarum with E. coli and E. faecalis in MRS broth during 12-week storage assay at 4 • C.

Co-Aggregation (%) with E. faecalis
Week 0 Week 12 Week 0 Week 12 Week 0 Week12 Week 0 Week 12 The co-aggregation assay is a reliable method to evaluate the close interaction between Lactobacillus and pathogenic bacteria. Co-aggregation of L. plantarum with E. coli (Table 5) was better than E. faecalis (Table 5) at week 0. L. plantarum showed better coaggregation ability in 3S+4I (70.4%), 2L+3I (63.7%), 3S+2FOS (58.6%) and 3L+4FOS (64.7%) at week 0 as compared to non-supplemented media (17-57%) ( Table 5). Interestingly, co-aggregation ability with both E. coli and E. faecalis increased as the storage week proceeds. At week 12, the co-aggregation ability of L. plantarum with E. coli and E. faecalis in supplemented media was 29-55%, and 8.5-35% higher than non-supplemented media suggesting inulin and FOS supplementation of significantly improved co-aggregation ability (Table 5). In this study, L. plantarum showed a high auto-aggregation percentage and hydrophobicity, which might increase the adhesion to intestinal epithelial cells. Also, L. plantarum showed higher co-aggregation with E. coli than E. faecalis.

Gastrointestinal Tolerance Assay
Probiotics are currently viewed as resistant to specific conditions occurring in the gastrointestinal tract [77]. Here we explore, the effects of exposure to the simulated gastrointestinal tract on the bacterial survival rate of L. plantarum grown and stored in inulin and FOS supplemented media, as shown in Table 3. In vitro tolerance to gastric conditions varied depending on the bacterial strain, pH value, and techniques used to determine the survival rate of bacterial cells. None of the bacterial strains showed viability or ability to grow after exposure to gastric juices at pH 1.5 (data not shown). At week 0, L. plantarum showed a significantly higher percentage difference in bacterial survival rate (BSR) with 3S+4I (21.55%), 3S+2FOS (23.66%), and 2L+3I (23.29%), 3L+4FOS (20.31%) when compared to non-supplemented media (Table 6). Similarly, Pan et al. [78] also reported that 2% FOS as the sole carbon source enhanced the survival of L. plantarum NI2L+3I02 in simulated gastrointestinal juice. In L. plantarum, BSR (%) decreased significantly (p < 0.05) in all media combinations as the storage proceeds; the metabolic and limited nutrient stress might have affected the gastrointestinal tolerance (Table 6). Higher sensitivity has been reported for L. rhamnosus and other L. plantarum strains [79,80]. Buriti et al. [81] reported similar results, who reported higher cell viability of L. acidophilus La-5 during the gastric phase involving HCl and pepsin when the strain was incorporated in a synbiotic light mousse containing sugar and 2% inulin. However, Gomez-Mascaraque et al. [82] observed a viability loss for L. plantarum CECT using whey protein concentrate (WPC) powder as a coating agent during the gastric phase. Schell, Beermann [83] observed an increased survival of microencapsulated L. reuteri DSM 20,016 (in sweet whey and shellac) during the in-vitro gastrointestinal environment, probably due to cell structure recovery of the injured bacteria to the less stressful conditions of the enteric phase.
At week 12, L. plantarum showed a significantly higher percentage increase in BSR of 25.21% in 3L+4FOS followed by 21.33% in 2L+3I, 20% in 3S+2FOS, and 14.33% in 3S+4I as compared to non-supplemented media ( Table 6). Based on BSR (%), FOS supplementation was better than inulin supplementation in the lactose system. The significant increase in BSR (%) in the in vitro gastrointestinal environment of L. plantarum during storage may be due to the slow degradation of inulin and FOS in acidic conditions, which improved probiotic survival. Gardiner et al. [84] suggested that protective extracellular polysaccharides enhance the survival of the probiotic strains during gastric transit. Therefore, prebiotic ingredients may be an alternative to improve probiotic survival through the gastrointestinal tract [85]. Besides, the survival of probiotic cells may be attributed to inulin and FOS resistance to hydrolysis by the gastrointestinal enzymes and the low solubility of long-chain inulin [86].
Moreover, these results establish that inulin and FOS supplementation significantly increased bacterial cultures' in vitro gastrointestinal associated stress tolerance. Specifically, mild salt stress and lower pH adaptation may elicit adaptive responses that reduce and support such stress tolerance, respectively. However, adhesive assay with intestinal cell lines (HT29 and Caco2) with in vitro gastrointestinal tract assay will further elucidate the effect of inulin and FOS on probiotic survival and adhesive properties.

Conclusions
In growth assay and survival studies, the highest percentage increase in growth index and cell viability showed that sucrose and lactose were more suitable sugar substrates for inulin and FOS supplementation. Furthermore, sucrose and lactose systems with inulin or FOS supplementation supported and improved the stability and probiotic potential of L. plantarum during 12-week refrigerated storage at 4 • C. The higher percentage increase in bacterial survival rate showed that FOS supplementation was better than inulin supplementation in sucrose and lactose systems in the gastrointestinal tolerance study. Furthermore, hydrophobicity, auto-aggregation, co-aggregation ability of L. plantarum with E. coli and E. faecalis were improved in sucrose and lactose systems with FOS and inulin supplementation. In conclusion, using the right type and concentration of carbon source and the right kind and concentration of prebiotic for culturing L. plantarum can improve its cell viability and efficacy, as shown in gastrointestinal robustness. This study provides insight to design fermentation and storage conditions that aim to produce probiotic products with improved viability and gastrointestinal tolerance and have a higher potential to achieve their desired health beneficial effects.