Hydrolysis of oligosaccharides in the gastrointestinal tract alters their prebiotic effects on probiotic strains

Abstract

Oligosaccharides have been widely used as prebiotics in the food industry, however their properties have been examined in vitro, without considering hydrolysis in the human digestive tract, especially in the small intestine. Here, we hypothesized that the prebiotic effects and utilization efficiency of ingested oligosaccharides would be altered in the colon, as their structures are partially hydrolyzed during digestion. Different types of oligosaccharides were partially degraded during simulated digestion, and digestible monosaccharides were released from the initial substrates. The growth of some probiotic strains responded to the presence of digestible/absorbable mono- and disaccharides (components of the prebiotic oligosaccharides), but not to that of the oligosaccharides themselves. These findings regarding oligosaccharide degradation in the gastrointestinal tract can be used to achieve greater experimental accuracy when examining the effects of prebiotics on gut flora via in vitro studies (e.g., on fecal fermentation or microbial growth rates).

Supplementary Information

The online version contains supplementary material available at 10.1007/s10068-023-01474-z.

Introduction

Oligosaccharides, consisting of 2–10 glycosyl units with varying monosaccharide compositions, have been widely applied as prebiotics in the food industry, as they enhance colon health after passing through the small intestine without undergoing structural degradation (Rajendran et al., 2017). Oligosaccharides (such as galacto-, fructo-, isomalto-, and xylo-oligosaccharide) act as carbon sources for colonic microbiotic organisms such as Bifidobacterium, which have specific carbohydrate utilization systems to hydrolyze the various types of linkages (Wang et al., 2020). The prebiotic properties of oligosaccharides have been determined by inoculating different microorganisms or human fecal samples into media containing oligosaccharides rather than a primary carbon source (e.g., glucose and sucrose), or by increasing the oligosaccharide ratio (Sanz et al., 2005; Saulnier et al., 2007). However, these methods, in which oligosaccharides were directly applied to probiotics, have not considered oligosaccharide degradation and absorption during digestion from the mouth to the gastrointestinal (GI) tract.

Carbohydrate digestion proceeds in three main stages, in the mouth, stomach, and small intestine (Jung et al., 2019). The function of the mouth is to decompose food into small particles via mastication; the endohydrolytic activity of α-amylase in the secreted saliva randomly hydrolyzes α-1,4 bonds in carbohydrates (Mulet-Cabero et al., 2020). Carbohydrates with weak specific bonds to acids are further hydrolyzed in the gastric fluid of the stomach. Finally, carbohydrates that reach the small intestine are broken down into monosaccharides by α-amylase from the pancreas and by mucosal glucosidases such as maltase-glucoamylase, sucrase-isomaltase, and lactase-phlorizin hydrolase (Kushwaha and Kumar, 2017). When oligosaccharides, which comprise several sugars linked by glycosidic bonds, pass along the GI tract, their structures will be partially degraded, especially by glucosidases in the small intestine that hydrolyze various α- and β-glycosidic bonds (Ferreira-Lazarte et al., 2017). Therefore, we hypothesized that prebiotic galacto-, fructo-, isomalto-, and xylo-oligosaccharides are digestible/degradable in the mammalian digestive tract, and that their prebiotic effects on the colon microbiota, and their utilization efficiency, are altered by partial absorption in the small intestine. Our in vitro findings, which improve the understanding of prebiotic digestibility, can be used to develop methods to accurately determine the effects of prebiotics on the activity of various target microorganisms, and to tailor oligosaccharides to specific target bacteria, for use in the food and pharmaceutical industries.

Materials and methods

Materials and microbial strains

Galacto-, fructo-, isomalto-, and xylo-oligosaccharides (GOS, FOS, IMOS, and XOS, respectively) were obtained from NeoCremar Co. (Seoul, Republic of Korea). Colonic bacterial strains (Streptococcus thermophiles KCCM 35496, Lactobacillus helveticus KCTC 3545, Lactobacillus casei KCCM 12452, Lactobacillus paracasei KCTC 3510, Lactobacillus gasseri KCTC 3163, Lactobacillus plantarum KCTC 21024, Lactobacillus fermentum KCTC 3112, Lactobacillus reuteri KCTC 3594, Lactobacillus acidophilus KCCM 32820, Lactobacillus rhamnosus KCCM 32405, Lactobacillus. delbrueckii ssp. bulgaricus KCCM 35463, Lactobacillus salivarius KCCM 40210, Enterococcus faecium KCCM 12118, Enterococcus faecalis KCCM 40450, Lactococcus lactis KCTC 3769, Bifidobacterium animalis ssp. lactis KCTC 5854, Bifidobacterium bifidum KCTC 3440, Bifidobacterium breve KCCM 42255, and Bifidobacterium longum KCTC 3421) were purchased from the Korea Collection for Type Cultures (KCTC) and Korean Culture Center of Microorganisms (KCCM). All other chemicals were purchased from Sigma-Aldrich Co. (St. Louis, MO).

Oligosaccharide hydrolysis during simulated digestion

Simulated digestion of oligosaccharide samples was conducted for three independent phases: oral, gastric, and intestinal. Human saliva α-amylase (0.11 mg/mL; Sigma-Aldrich) was reacted with individual oligosaccharides (1%, w/v) for 30 s at 37 °C in 100 mM phosphate buffer (pH 7.0, 1 mL) to mimic salivary digestion. Gastric digestion was simulated for 10 min, 0.5, 1, 2, and 3 h at 37 °C using artificial gastric fluid adjusted to pH 2 using HCl. For the intestinal phase, porcine pancreatic α-amylase (2 mg/mL, Megazyme Co., Wicklow, Ireland) was reacted with each oligosaccharide solution (1%, w/v) in 100 mM phosphate buffer (pH 6.0, 1 mL) for 0.5, 1, 2, 3, and 6 h at 37 °C (Htoon et al., 2009). The amount of reducing sugar produced during each digestive phase was measured using a 3,5-dinitrosalicylic acid (DNS) assay (Miller, 1959), and the degree of oligosaccharide hydrolysis (%) was quantified as follows:

$$\text{The}\text{degree}\text{of}\text{total}\text{oligosaccharide}\text{hydrolysis}\left(\text{\%}\right)=\frac{W_r-W_t}{W_t}\times 100$$

where $W_r$ is the amount (mg) of reducing sugar released, and $W_t$ is the amount (mg) of initial reducing sugar in the individual oligosaccharides.

Intestinal digestion of oligosaccharides via mammalian glucosidase

Individual oligosaccharide samples (1%, w/v) were hydrolyzed using pre-washed precipitated enzyme from rat intestinal acetone powder (RIAP, Sigma-Aldrich; 1%, w/v) in pH 6.0 phosphate buffer for 24 h at 37 °C (Seo et al., 2020), then the reactant was boiled to inactivate for 10 min. Each sample was filtered through a 0.45 μm nylon syringe filter (GVS, IN), and then injected into a high-performance anion exchange chromatography (HPAEC) system (Thermo Fisher Scientific, Sunnyvale, CA) connected with an electrochemical detector (Thermo Fisher Scientific). Chromatographic separation was achieved using a CarboPac PA-1 pellicular anion-exchange column (Thermo Fisher Scientific) with solvent system eluents A (150 mM NaOH) and B (600 mM sodium acetate in 150 mM NaOH), using a gradient method (0–20 min, 100% eluent A; and 20–60 min, 0–40% eluent B).

Utilization of prebiotic oligosaccharides for various probiotics

Individual strains were inoculated in 15 mL of MRS broth (BD Difco, Detroit, MI) and incubated until the logarithmic phase. The strains were further inoculated under anaerobic conditions using the GasPak Container System (BD Difco) at 37 °C for 24 h in glucose-free MRS broth (MB Cell, Seoul, Republic of Korea) in which the glucose was replaced by oligosaccharides. The oligosaccharide samples were remanufactured to have the identical sugar compositions as GOS, FOS, IMOS, or XOS, and were analyzed using the HPAEC system. Absorbance at 600 nm was used to quantify the prebiotic effect on probiotic bacterial growth.

Statistical analysis

All statistical analyses were performed using SAS 9.4 (SAS Institute, Cary, NC). Statistical differences among groups were determined using one-way analysis of variance (ANOVA), and additional analysis was conducted on the means by applying Tukey’s multiple comparison test. Differences were considered statistically significant at p < 0.05. All measurements were performed in triplicate.

Results and discussion

Hydrolysis of oligosaccharides during simulated oral, gastric, and intestinal digestion

During oral digestion, a negligible proportion (< 3.0%) of the oligosaccharides was digested by salivary α-amylase, whereas ca. 11.6% of the WCS, the digestible control for α-hydrolytic enzymes, was broken down (Fig. 1(A)). However, IMOS showed a significantly higher digestibility (p < 0.05) among oligosaccharide samples, which may be due to the remaining starch-based substrates during enzymatic production or small amounts of α-1,4 linkages in commercial IMOS (Oku and Nakamura, 2003). During the gastric phase (pH 2), the oligosaccharides were partially hydrolyzed (by 5.7–12.1%), with FOS degrading gradually, by up to 53.6% after 3 h (Fig. 1(B)). The poor stability of FOS at low pH can be explained by the weakness of the glycosidic bonds on the furanosyl group, whereas the other tested oligosaccharides have pyranosyl groups with stronger glycosidic bonds (Courtin et al., 2009). Oligosaccharide degradation patterns were similar under simulated intestinal degradation (by porcine pancreatic α-amylase) and oral digestion, although more substrates were hydrolyzed by pancreatic α-amylase owing to the relatively longer transit time for simulated intestinal digestion (Fig. 1(C)). These findings suggest that these oligosaccharides can be broken down in the GI tract without the assistance of mucosal α-glucosidases, thereby releasing monosaccharides into the small intestine. These results indicate that oligosaccharides may reach the large intestine partially digested, and that they will have different effects on colonic microbiota in vitro, owing to structural changes.

Fig. 1.

Hydrolysis of galacto-, fructo-, isomalto-, and xylo-oligosaccharides (GOS, FOS, IMOS, and XOS, respectively) at different phases of simulated digestion. (A) Oral phase: hydrolysis by α-amylase from human saliva; (B) Gastric phase: exposure to acidic conditions (pH 2); and (C) Intestinal phase: degradation by porcine pancreatic α-amylase. Waxy corn starch (WCS) was used as the digestible control for the α-amylase reaction. Groups with different lowercase letters (a–c) are significant differences at p < 0.05

Hydrolysis of oligosaccharides to monosaccharides by mammalian small intestinal glucosidases

HPAEC analysis revealed that all of the oligosaccharides tested were partially degraded to monosaccharides by the reaction with RIAP solution as a source of mucosal glucosidases, and the remaining structures were altered (Fig. 2). GOS hydrolysis by the rat intestinal extract increased the galactose, glucose, and lactose fractions and reduced the oligosaccharide fraction (Fig. 2(A)). This is because the β-galactosyl groups of both GOS and lactose can be hydrolyzed by the β-galactosidase activity of lactase-phlorizin hydrolase in the small intestine, although β-galactosidase expression and enzyme activity differ between mammalian species (Ferreira-Lazarte et al., 2019; Hernandez-Hernandez et al., 2019). The released monosaccharides (glucose and galactose) from GOS and lactose will be absorbed via the glucose transporters in the epithelial cells of the small intestine, and the remaining fractions reach the colon with deceased sizes compared to the initial structures. Similarly, the simulated intestinal digestion of FOS increased the fructose and glucose fractions and reduced the oligosaccharide fraction (Fig. 2(B)). FOS hydrolysis by rat intestinal extract is consistent with prior findings (Ferreira-Lazarte et al., 2017), and the structural weakness under the acidic condition of the gastric phase produces digestible monosaccharides in the small intestine (Fig. 1(B)). Unlike the other oligosaccharides, IMOS was almost entirely broken down into glucose by α-glucosidase (Fig. 2(C)). This is because its α-1,6 linkages are slowly hydrolyzed by mammalian digestive enzymes, especially isomaltase, and are not resistant to digestion (Song et al., 2022). A recent in vivo study proposed redefining IMOS as a glycemic carbohydrate, contrary to the traditional paradigm that considers IMOS a prebiotic that regulates beneficial bacteria in the colon (Oku and Nakamura, 2003). A recent study proposed that XOS is resistant to pancreatic and intestinal mucosal enzymes (de Figueiredo et al., 2020), however these structures were also partially hydrolyzed to xylose monomers by the rat intestinal extract (Fig. 2(D)). This is because the rat intestinal crude extract contains microorganism-derived hydrolytic enzymes (Seo et al., 2020). Carbohydrates hydrolysis is affected both by mammalian α-glucosidases and the intestinal microbiota. These findings confirm that GOS, FOX, XOS, and IMOS were partially degraded by mammalian α-glucosidases or microorganisms in the GI tract, and that the portions reaching the colon were altered in terms of structure and composition.

Fig. 2.

Hydrolysis of galacto-, fructo-, isomalto-, and xylo-oligosaccharides (GOS, FOS, IMOS, and XOS; (A–D), respectively) by mammalian α-glucosidases (24 h at 37 °C). Gray and black lines: chromatograms for 0 and 24 h, respectively, of enzymatic digestion

Effects of mono- and di-saccharides in oligosaccharide samples for the growth of probiotic strains

The utilization of GOS, FOS, IMOS, and XOS by the prebiotic strains was analyzed by measuring their growth in media containing these oligosaccharides relative to their growth in the blank control (no carbon source) and in the sugar control (containing only mono- and di-saccharides and prepared based on their proportions in the oligosaccharides) (Fig. 3; Supplementary Table 1). The bacterial strains grew well without the carbon sources (blank), and the growth ratios of most microorganisms exhibited significantly better growth in the sugar control, as the mono- and di-saccharides were well-utilized as carbon sources by the prebiotic strains. These sugars, including glucose, fructose, sucrose, and lactose, are critical factors for microbial growth (Wang et al., 2019). Oligo-, mono-, and di-saccharides utilization varied between the different bacterial strains (Fig. 3), suggesting that some prebiotic bacteria utilize oligosaccharides as carbon sources for growth. This indicates that oligosaccharides are utilized by colonic bacteria as nutrients. Their utilization rate may be an indicator of their prebiotic effects (Gibson and Roberfroid, 1995). Nevertheless, the growth ratios of some strains using oligosaccharides were not significantly different (p > 0.05) compared to the control, suggesting that these ingredients have no prebiotic properties to affect the growth of certain microorganisms as we hypothesized. Several studies have suggested that oligosaccharides have strong prebiotic effects, without considering small intestinal digestion of digestible mono- and di-saccharides (Kim et al., 2021; Yang et al., 2020). Here, we examined the effects on bacterial growth of known quantities of digestible carbohydrates from oligosaccharides clearly demonstrate that the prebiotic effects of the oligosaccharides were altered during passage through the digestive tract. As a next step, the purified oligosaccharides without digestible mono- and di-saccharides will be applied to fecal fermentation to evaluate the accurate effects on altering the colon microbiota.

Fig. 3.

Prebiotic effects of the digestible/absorbable mono- and di-saccharides in galacto-, fructo-, isomalto-, and xylo-oligosaccharides (GOS, FOS, IMOS, and XOS; (A–D), respectively), quantified as bacterial growth relative to the control. ‘Blank’: medium without a carbon source. ‘Control’: medium with only digestible/absorbable mono- and di-saccharides. ‘Sample’: medium containing the oligosaccharide sample. (A–D): Groups with different lowercase letters (a–c) have significant differences at p < 0.05 between the blank, control, and sample

Supplementary Information

Below is the link to the electronic supplementary material.

Declarations

Conflict of interest

The authors declare no conflicts of interest.