L. plantarum subsp. plantarum DSM 20174 is the DSMZ reference strain (Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Braunschweig, Germany). Cells were anaerobically grown at 30 °C for 24 h in MRS deMan, Rogosa and Sharpe broth [15] (casein peptone, tryptic digest 10 g/L; meat extract 10 g/L; yeast extract 5 g/L; Tween 80 1 g/L; K₂HPO₄ 2 g/L; Na-acetate 5 g/L; (NH₄)₂ citrate 2 g/L; MgSO₄ × 7H₂O 0.20 g/L; MnSO₄ × H₂O 0.05 g /L), pH 6.2, complemented with 20 g/L of inulin, pectin or glucose, the last used as control. All chemicals were supplied by Sigma (Milano, Italy). All solutions were sterilized and all manipulations were aseptically performed.

The artificial gastric juice contained MRS broth, plus 3 g/L pepsin, with a final pH adjusted to 2.0. The artificial intestinal juice contained MRS broth plus 1 g/L pancreatin and 4.5 g/L bile salts (sodium glycocholate and sodium taurocholate, code number X606C, Oxoid, Basingstoke, UK). Both solutions were filter-sterilized (0.22 μm, Millipore SpA, Milano, Italy). The strain was treated using the protocol of De Giulio et al. [16]. Bacterial suspensions were incubated in artificial gastric juice for 180 min at 30 °C and recovered by centrifugation. After washing with sterile physiological solution (8.5 g/L NaCl), pellet was resuspended in the simulated gastrointestinal juice (4.5 mL/100 μL of cells). Samples were incubated at 30 °C for 60 min. Cell viability (cfu/mL) was evaluated by anaerobic culturing on MRS plates (30 °C, 48 h) before and after incubation in the two simulated gastrointestinal juices. Each separate experiment was performed three times.

Preparation of Heat Killed Cells (HKC) was performed following the method described by Liu et al. [17] with some modifications. Two percent of L. plantarum subsp. plantarum was inoculated into MRS broth (Sigma Aldrich Italia, Milano, Italy) and cultured at 30 °C for 20 hr under anaerobic conditions. Cells were harvested by centrifugation for 5 min at 5,000× g at 4 °C (Biofuge, Beckmann Coulter Italia,Cassina de Pecchi, Italy); the pellet was washed twice, resuspended in phosphate-buffered saline (pH 7.2) buffer, heated at 100 °C for 30 min and dried. Samples were re-suspended again to 5 mg dry cell/mL, autoclaved at 121 °C for 15 min and stored in 4 °C prior to use.

Three replicates were carried out for each analysis. Statistical significance was performed as described in De Giulio et al. [16] and set at P < 0.05.

Several species of the gastrointestinal microbioma have the capability to ferment oligo- and polysaccharides [23]. Such capacity is common within the genus Bifidobacterium, but it is strain-dependent among the genus Lactobacillus [24].

In our experiments, L. plantarum subsp. plantarum was capable of growing on pectin and inulin, reaching values of 10.07_log10 cfu/mL (±0.03) and 10.28_log10 cfu/mL (±0.06), respectively (Figure 1). These values were not consistently different with respect to that obtained in glucose (used as control), 10.42_log10 cfu/mL (±0.05), although lower with respect to those observed when L. acidophilus was treated undeer the same growth conditions [25]. In contrast, the tolerance to the simulated gastric and pancreatic juices was deeply dependent on the initial growth conditions, which caused a more effective viability when the strain was pre-treated in the presence of pectin or inulin (Δ log₁₀ 1.5 and 2.4 cfu/mL respectively at P < 0.05) as compared to the growth in presence of glucose (Δlog₁₀ 4.0 cfu/mL at P < 0.01) (Figure 1). The evident loss of viability confirms once more that glucose, different from inulin and pectin, would be not capable to exercise a significant defensive effect, mainly after the double passage through the simulated gastric and pancreatic juices.

The diverse resistance in the simulated gastro-intestinal juices could be also linked to the difference on the protein profiles, whch were analyzed by the lab-on-a-chip approach (Figure 2).

Growth in the presence of pectin (Figure 2, lanes 4-6) or inulin (Figure 2, lanes 7-9) beneficially influenced the preservation of protein expression patterns, mainly during the subsequent passages of the strain through the simulates of gastric (AGJ) and pancreatic (AGPJ) juices, if compared to the growth in presence of glucose used as control (Figure 2, lanes 1-3): in this last case, it gave rise to a disappearance of bands greater than 37 kDa, although the effect was less marked respect to L. acidophilus [25]. While we don’t know if and in which manner such proteins could be involved in the resistance of microorganism to those stress conditions, the degradation of the protein profile might be ascribable to the increase in the mortality of the cell population and can qualitatively be related to the data obtained by the enumeration of the colony forming units.

Fermentation of carbohydrates in the colon by the intestinal microbiota and probiotics leads to production of short-chain fatty acids (SCFA) such as acetate and butyrate and gases (mainly CO₂ and H₂) [26]. SCFAs are quickly used by gut microbiota or absorbed (through the mucosal surface of the caecum and ascending colon) and utilized as energy sources by the host. Mucosal cells efficiently use butyrate to preserve and maintain their structure. Acetate goes across the intestinal wall without being metabolized much and is carried to the systemic circulatory system, reaching the blood and entering the liver [27]. The capacity of certain intestinal microbes to utilize carbohydrates not digested by the host plays a key part in their effectiveness to adequately colonize the mammalian gastrointestinal tract. The adaptation of L. plantarum subsp. plantarum to a variety of environmental niches can be associated with its diverse catabolic potential, demonstrating the competitive benefits of utilizing complex sugars [28]. The amount of lactate, acetate and butyrate produced by L. plantarum subsp. plantarum grown in the presence of glucose, pectin or inulin, was evaluated by HPLC and is shown in Table 1.

The presence of glucose gave rise to higher amounts of acetate respect to inulin. Probably, the presence of inulin promoted the production of butyrate, which amount was higher (about 30%) respect to glucose, used as control. Conversely to L. acidophilus grown in presence of the same substrates [25],pectin, although it stimulated the production of acetate and lactate, did not induce the strain to produce detectable levels of butyrate, confirming the hypothesis that different carbohydrates might affect the metabolic pathways in lactobacilli in diverse ways. In any case, the production of butyrate may be considered a result of biological relevance, based on its known healthy effects: in fact, it is the preferred energy source for colonic epithelial cells, and has anti-carcinogenic and anti-inflammatory effects [29]; butyrate has been found to slow proliferation and promote the expression of phenotypic markers of differentiation in different cancer cell lines and in vivo [30,31,32]. Our data suggest that the presence of inulin in the growth medium of L. plantarum subsp. plantarum might induce an up-regulation of butyrate production; however, conversely to L. acidophilus grown under the same conditions [25], such a capability cannot be correlated to the higher resistance exhibited by the strain through its passage the gastrointestinal juice, as demonstrated by the absence of butyrate production after growth of L. plantarum subsp. plantarum in the presence of pectin.

Reactive oxygen species (ROS), are formed in the process of incomplete reduction of oxygen. Physiological concentrations of ROS are needed for a regular functioning of the cell and for the production of energy; indeed, they are involved in phagocytosis and in the regulation of the mechanisms involved in intercellular signaling [33]; however, high levels of ROS may lead to damage of the host’s DNA, proteins, carbohydrates and lipids [34]. Different studies on the antioxidant activity of lactic acid bacteria have demonstrated that some strains are capable of decreasing the risk of ROS accumulation and degrading superoxide anion and hydrogen peroxide. Kaizu et al. first reported the antioxidant activity of Lactobacillus sp SBT 2028[35]; Lin and Yen indicated the intracellular cell-free extracts of B. longum ATCC 15708 and L. acidophilus ATCC 4356 have antioxidant activity [36]; other LAB strains, such as L. fermentum E3, L. fermentum E18 and L. acidophilus ATCC 4356, could overcome the oxidative stress [6,37,38]. In our study, antioxidant activity was exhibited by the HKC of the L. plantarum subsp. plantarum grown with the three carbohydrates as energy source. The activity (16.45% ± 1.06) was generally in the range observed for other Lactobacillus strains [39]. The presence of inulin and pectin increased in an impressive manner such activity compared to glucose, and ten times higher (24.5 % ± 1.04, and 31.5% ± 2.75, respectively) antioxidant activity values were observed. This aspect can be of particular relevance, taking into account that probiotic strains with high antioxidative activity might not only contribute to fighting all diseases related to ROS, but also they can exhibit, in vivo, a greater survival in a particular ecological niche and, probably also through the production of organic acids, a high rank of anti-infectious potential, as demonstrated by Hutt et al. [39] and Truusalu et al. [40] in different probiotics.

Cell adhesion is a complex process implicating contact between the bacterial cell membrane and the interacting surfaces. The capacity of adhesion to epithelial cells and mucosal surfaces has been suggested to be an important property of many bacterial strains used as probiotics. Several studies have investigated the composition, structure and forces involved in bacterial adhesion to intestinal epithelial cells [41] and mucus [42]. The microbial adhesion to hydrocarbons (MATH) test has been extensively used for measuring cell surface hydrophobicity/hydrophilicity in lactic acid bacteria [43]. The test was used to evaluate the hydrophobic/hydrophilic cell surface properties of L. plantarum subsp. plantarum, deriving from the presence of inulin, pectin or glucose used as source of energy. After growth in the presence of inulin and pectin, microbial cells adhered to xylene, with percentages of 34.81 % (± 3.14) and 35.28% (± 2.75), respectively, resulting 10 times higher than that observed after growth with glucose (3.39 % ± 0.012). In probiotics, a good correlation between surface hydrophobicity and the capacity to adhere to the intestinal mucosa has been ascertained [44,45]: in fact, these bacteria, also through such a structural property, are capable of interacting more effectively with the intestinal mucosa. The fact that a much higher percentage of L. plantarum subs plantarum cells adhered to the apolar solvent xylene after growth in presence of the two prebiotics, compared to the microbial growth with glucose, demonstrated that these two prebiotics could positively affect the hydrophobicity of the cell surface of this strain; thus, pre-incubation of the strain in the presence of the two prebiotics might render its cell surface potentially more capable of colonizing the intestine. Indeed, our results confirm that hydrophobicity index might be dependent on the carbon source used as energy substrate [46].