Evaluation of the impact of a rat small intestinal extract on the digestion of four different functional fibers.

The degree of digestion, modulated by rat small intestinal extract on different functional fibers was investigated. In general, inulin-type fructans and fructooligosaccharides were the most resistant to the enzymatic digestion. Results evidenced the high-resistance of fructosyl-fructose bonds. This fits well with the concept of prebiotic carbohydrates. However, the mixture of melibiose, manninotriose and verbascotetraose (α-GOS) from peas, with a considerably lower molecular weight (0.6 kDa) than the fructans studied, were highly digested (61.2%). Interestingly, the Gal-(1 → 6)-Gal bonds present into the manninotriose and verbascotetraose were more prone to be hydrolyzed than Gal-(1 → 6)-Glc (melibiose). However, when melibiose was the only disaccharide present in the reaction mixture, the hydrolysis was also high (67.7%). The use of small intestinal enzymatic preparations is a realistic approximation to evaluate the digestion of different carbohydrates, thus, showing that recognized non-digestible carbohydrates can also be partially digested.


Introduction
The recommended daily allowances (RDAs) for total fiber consumption for healthy men and women (19-50 years old) are 38 g day −1 and 25 g day −1 , respectively, and these general needs can vary depending on the health status of the individual. Although fibers have revealed to possess numerous positive health effects on severe pathologies (obesity, diabetes, cardiovascular disease, among others), the mean daily intake for most people is much lower than the RDA. There is no superior acceptable amount for fiber consumption, although the tolerance depends mainly on the individual; bloating and abdominal pain being the most important consequences of excessive intake. 1,2 In general, the total fiber consumption is the quantity of dietary fiber and functional fiber. Technically, dietary fiber is a complex group of carbohydrates and lignin, which are not digested nor absorbed, in the human body. They can be divided in two major groups: soluble and insoluble fiber according to their water solubility, both being indigestible in the small intestine. [3][4][5] Functional fibers constitute a wide range of non-digestible carbohydrates that are either isolated or synthesized from natural sources mainly agro-food by-products. Once the functional fiber is produced is added to food during processing with the aim of providing beneficial effects on human health. Functional fibers include polysaccharides such as β-glucans, cellulose, chitins and chitosan, fructans, gums, pectin, polydextrose, polyols, resistant dextrins, resistant starches and oligosaccharides that are resistant to digestion. 2 According to previous studies, these carbohydrates can reach intact the large intestine, where they are hydrolyzed and fermented by the intestinal microbiota, thus causing the production of short chain fatty acids (SCFAs) that exert a beneficial effect not only in the colon but also systemically. [6][7][8] However, few studies have been conducted on the digestive resistance of carbohydrates (also refereed as non-digestible carbohydrates), probably due to the lack of reliable digestion methods specifically designed for carbohydrates.
To date, the focus of the most common approach to simulate the small intestinal digestion is dedicated to proteins, lipids and starch, by using pancreatic enzymes from porcine origin, salivary enzymes and microbial enzymes, which could not reflex most of the carbohydrase activities of the whole small intestine. 9 The Association of Official Analytical Chemist (AOAC) developed an integrated determination method for dietary fiber, including non-digestible oligosaccharides (NDOs) and resistant starch, which was modified later in 2015 (AOAC 2009.01). 10,11 Other methods, as well as this method, are based on the use of isolated digestive enzymes. Porcine pancreatic α-amylase and a fungal amyloglucosidase from Aspergillus niger are used to produce the complete hydrolysis of digestible saccharides, and therefore, to distinguish between digestible and non-digestible carbohydrates. However, similar to the InfoGest protocol, these enzymes cannot completely hydrolyze digestible saccharides since they do not represent the fully complex enzymatic environment of the small intestine, mainly because of the absence of the brush border enzymes of the enterocytes. As a result, digestible saccharides that are not fully degraded are detected as non-digestible carbohydrates, which lead to an inaccurate determination of the digestion resistance of these carbohydrates. 12,13 Recently, a promising in vitro digestibility method of dietary carbohydrates using rat small intestinal extract (RSIE) has questioned the belief that recognized prebiotics oligosaccharides derived from lactose (GOS) and fructooligosaccharides (FOS) reach the distal portions of colon without alterations. [14][15][16] Therefore, with the aim of gain more insight on the benefits of non-digestible carbohydrates, in this work the digestibility of commercial functional fiber, such as α-galactooligosaccharides derived from peas and different types of fructans have been tested using a rat small intestine extract.

Characterization of substrate by HPSEC-ELSD
The molecular weight (M w ) of each carbohydrate was obtained by High Performance Size Exclusion Chromatography (HPSEC) coupled to an Evaporative Light Scattering Detector (ELSD), following the method described by Muñoz-Almagro et al.
2.4. Enzymatic characterization of rat small intestine extract 2.4.1. Protein determination. Protein quantification was done through the Bradford method. 18 Bovine Serum Albumin (BSA) was used as a standard and absorbance was measured at 595 nm.
The carbohydrate hydrolysis was measured through GC-FID as described below. Specific enzymatic activities (U) of RSIE were calculated and expressed in µmol per (min per g per protein). Each unit of specific enzymatic activity was defined as the amount of enzyme which released 1 µmol of the corresponding monosaccharides in 1 min of incubation (n = 4).

In vitro digestion of polysaccharides with RSIE
The digestibility of three types of inulin, a mixture of α-GOS and the corresponding blanks (no added carbohydrate sample) were digested with RSIE following the method used by Ferreira-Lazarte et al. (2017) 14 with slight changes. Initially, a solution of 0.5 mg mL −1 of prebiotic carbohydrate in distilled water was prepared, then 40 mg of RSIE was mixed with 1 mL of prebiotic solution and the mixture was incubated to perform the reactions. Digestions were carried out in an orbital Thermomixer comfort (Eppendorf®) at 37°C during 3 h of reaction with continuous agitation (750 rpm). Duplicate of individual reactions were carried out for each time (0, 60, 120 and 180 min) in order to avoid any possible enzymes/substrate composition changes produced by taking aliquots, and reactions were stopped by heating in boiling water for 5 min.

Chromatographic analysis of carbohydrates
Gas chromatography, equipped with a flame ionization detector (GC-FID), was used to analyze the fraction of carbohydrates for each enzymatic characterization and digestion. Samples were derivatized, to obtain trimethylsilylated oximes (TSMO) of carbohydrates, according to the method of Brobst and Lott (1966). 19 Samples solutions were prepared with 500 µL of digestion samples (0.25 mg of carbohydrates) and 250 µL of phenyl-β-D-glucoside (internal standard, 0.5 mg mL −1 ), and were evaporated under vacuum. Three hundred µL of hydroxylamine chloride in pyridine (2.5%, w/v) were added to the samples and incubated at 70°C for 30 min with agitation. Then, 300 µL of hexamethyldisilazane (HDMS) and 30 µL of trifluoroacetic acid (TFA) were added and heated at 50°C for 30 min under continuous agitation. Finally, samples were centrifuged at 10 000 rpm for 3 min. Supernatants were injected in GC-FID.
GC-FID analysis was carried out in an Agilent Technologies 7820A gas chromatograph system. Separations of the compounds were achieved with a fused silica capillary column DB-5HT (5% phenyl methylpolysiloxane, 30 m × 0.25 mm × 0.1 µm, Agilent J&W Scientific, Folsom, CA, USA). The initial oven temperature was 150°C, then increased at a rate of 3°C min −1 to 380°C. The carrier gas used was nitrogen at a flow rate of 1 mL min −1 . Injector and detector temperatures were set at 280 and 385°C, respectively. Split mode 1 : 20 were used for the injections.
Interpretation and identification of the TMSO derivatives were performed using Agilent ChemStation software (Washington, DE, USA). Quantitative analysis was obtained through the internal standard method, thus calculating the response factors of standards solutions of carbohydrates (D-fructose, D-galactose, D-glucose, sucrose, lactose, kestose, nystose) at known concentrations (0.005 to 1 mg mL −1 ).

Statistical analysis
All digestions were made in duplicate and two GC-FID analysis (n = 2). For the statistical analysis, comparisons were made using analysis of variance (ANOVA) and Tukey's post hoc test with SPSS software for Windows (SPSS Inc., Chicago, II). Differences between content in carbohydrates were considered statistically significant when p < 0.05. Table 1 shows the sucrase, inulinase and melibiase activities of RSIE analyzed by GC-FID in the same conditions of digestion. Sucrase activity (42.07 U) was the highest, being 2-fold higher than melibiase activity (26.56 U) and eight times higher than inulinase (5.46 U). Sucrase activity obtained was moderately higher than the values obtained in previous reports in rats by Ferreira-Lazarte et al. (2017) 14 (23.5 U). These differences could be due to the variability between the batches of commercial intestinal acetone powders from rat. Regarding inulinase and melibiase activity, no previous studies have reported these activities in these enzymatic substrates.

Digestion of prebiotic carbohydrates using RSIE
The digestibility of three recognized prebiotics (Inulin Orafti GR, Raftiline HP and Raftilose P95) and two potential prebiotic carbohydrates (AlphaGOS P and melibiose) were tested using RSIE. Blank samples of digestion without carbohydrates were also carried out to measure possible matrix effects of this complex mixture. 31 Table 2 shows the individual composition of each carbohydrate incubated with RSIE, including mono-, di-, tri-and tetrasaccharide fractions. Before digestion treatments, Raftiline HP which is described as an inulin-type long-chain fructans, where DP below 10 is removed, 32 did not show any carbohydrate with DP < 4, apart from small amounts of fructose. On the other side, Inulin Orafti GR contains small amounts of FOS, detected as nystose, kestose and sucrose. Fructooligosaccharides (Raftilose P95) showed higher contents of small compounds (DP < 4) due to their oligosaccharide composition. Inulobiose (β-D-Fru(2 → 1)-β-D-Fru), inulotriose (β-D-Fru(2 → 1)-β-D-Fru(2 → 1)-β-D-Fru), inulotetraose (β-D-Fru(2 → 1)-β-D-Fru(2 → 1)-β-D-Fru(2 → 1)-β-D-Fru) and nystose were detected in Raftilose P95 samples, in agreement with the data reported by Montilla et al. (2006). 33 Interestingly, the analysis of AlphaGOS P showed that this commercial product contains melibiose (α-D-Gal(1 → 6)-D-Glc), manninotriose (α-D-Gal(1 →   35 previously, found that this enzyme, under appropriate conditions, is able to completely eliminate the fructose from stachyose, forming manninotriose. The evolution in the content of di-, tri-and tetrasaccharides in the samples was highly dependent on the structure during the RSIE digestion treatment. Inulin based samples such as Raftiline HP and Inulin Orafti GR exhibited the lowest changes in their composition due to the resistance of these substrates to intestinal enzymes. A minimum increase of fructose was observed in Raftiline HP possibly produced by the hydrolysis of high M w inulin species and no bigger structures (DP < 4) were detected. Regarding Inulin Orafti GR sample, slight degradations, but not significant, in the high M w structures such as nystose (24 to 19.5 mg nystose per g of sample) was registered after the digestion process (Table 2), and increases in the trisaccharide fraction were observed. Sucrose was the most digested structure causing important increases in fructose content. A similar trend was observed by Ferreira-Lazarte et al. (2017) 14 after the digestion of FOS with RSIE exhibiting the tetrasaccharide structure the highest degradation (40%). The trisaccharide also increased, most likely, due to degradation of the tetrasaccharide. These changes are in line with the low inulinase activity measured in this extract (Table 1).
Regarding Raftilose P95, both tetrasaccharides detected showed slight but not significant degradations, being inulotetraose more hydrolyzed than nystose. A consequent increase in the corresponding trisaccharide (inulotriose) was observed after 120 min digestion, probably due to the degradation of the former. Moreover, the presence of higher DP compounds in the sample could also produce tri-and tetrasaccharides during the digestion treatment. 29 Increases in smaller structures such as disaccharides (inulobiose) and monosaccharides (fructose) were also observed due to the hydrolysis of the biggest compounds.
These results underline the relevance of the higher M w composition in terms of digestibility. 16,36 Higher hydrolysis in Inulin Orafti GR was observed in the tetrasaccharide fraction after 180 min of digestion, supporting the data obtained in the in vivo and in vitro studies by Ferreira-Lazarte et al. (2017) 14 and Molis et al. (1996), 37 resulting in 12% and 11% of total digestion, respectively. In the case of Raftilose P95, a not significant decrease of tri-and tetrasaccharides was observed: hydrolysis being higher in the case of the linkage β(2 → 1) between fructose monomers (inulotriose and inulotetraose of Raftilose P95), when glucose is not present in the structure ending compared to sucrose oligosaccharides (kestose and nystose of Inulin Orafti GR) ( Table 2). This could suggest the lower resistance of the β(2 → 1) bonds to the action of the digestive enzymes. 14 With respect to oligosaccharides from galactose, chromatographic profiles by GC-FID of AlphaGOS P undigested and after digestion with RSIE are shown in Fig. 2. A decrease of verbascotetraose and manninotriose was observed ( peaks 4 and 5) and, consequently, melibiose and galactose contents were increased after 180 min of digestion due to the degradation of the structures with higher M w (Table 2). However, decreases in the melibiose levels were detected when a standard of only melibiose was digested, showing the highest hydrolysis values (67.7%) after 180 min of digestion. This degradation is in accordance with the considerable previous melibiase activity detected in the enzymatic extract, thus suggesting that the Total oligosaccharides was calculated by the sum of di-, tri-and tetrasaccharides of each carbohydrate.

Food & Function Paper
This journal is © The Royal Society of Chemistry 2020 intestinal enzymes probably are more prone to hydrolyze higher M w structures present in the sample (manninotriose and verbascotetraose) rather than the disaccharide. Increases in galactose levels also suggest that linkages between galactose monomers are being broken by digestive enzymes, similar to the degradation of β-GOS derived from lactose and lactulose observed during small intestinal digestion in previous works. 14,16,36,38,39 Beneficial effect of the α-GOS on the modulation of the intestinal microbiota is wellknown. 40,41 Nevertheless, there is not enough information about their digestibility, even the expert scientific panel from the European Food Safety Authority (EFSA) claims that the α-GOS are non-digestible carbohydrates. 42 Considering the sum of total oligosaccharides of AlphaGOS P (manninotriose and verbascotetraose) the hydrolysis was considerably high (61.2%) ( Table 2), despite the lack of pancreatic α-galactosidase in mammals. 43 However, Ferreira-Lazarte et al. 16 showed meaningful hydrolysis degree of prebiotic β-GOS (53% for linkages β(1 → 3)) using brush border membrane vesicles of small intestine of pig. A possible explanation for this fact could be the promiscuous multienzymatic complexes in the rat small intestine extract. 12 Carbohydrases of the small intestine, such as α-amylase, hydrolyze large starch structures, whereas complete digestion is done by the mucosal α-glucosidases. 44,45 In this sense, mucosal maltase-glucoamylase and sucrase-isomaltase complexes could hydrolyze to α(1 → 6) bonds between the monomers of both AlphaGOS P and melibiose. 20,21,46 These enzymatic structures could also have more versatility in terms of hydrolytic activity, as was showed elsewhere. 16,47 To the best of our knowledge, these data are the first evidence about digestibility of α-GOS with small intestine enzymes.
Some in vivo studies with rats have demonstrated a partial digestibility of prebiotic carbohydrates, showing a considerable high hydrolysis rate of β-GOS obtained from lactose and lactulose. 36,48 In the same way, an in vitro study with RSIE reported a hydrolysis degree of 12% of a mixture of FOS after 120 min of digestion. 14 These reported data are consistent with this work, highlighting the key role of the mammalian intestinal enzymes on the digestibility of carbohydrates.

Conclusions
Limitations of traditional digestibility methods of carbohydrates have been shown in several works. 12,13 Therefore, results obtained in this work confirmed the usefulness and effectiveness of the use of a RSIE to evaluate the digestion of polysaccharides. Moreover, similarities between small intestinal enzymes of rat and human emphasized the viability of this extract. 49 Raftiline HP, which is mainly constituted by inulin, showed the highest resistance to the gastrointestinal enzymes, with only a slight increase of fructose. Inulin Orafti GR also showed high resistance, with a small hydrolysis of tetrasaccharides, followed by Raftilose P95, thus supporting the role of these substrates as prebiotic compounds. Finally, AlphaGOS P and melibiose showed a considerable high hydrolysis degree (61.2 and 67.7%, respectively), remarking the effect of the chemical structure (M w and type of linkage) of prebiotic oligosaccharides with respect to their resistance to digestibility. According to the obtained results, mucosal enzymes complexes have versatile hydrolytic activities and contribute to the digestion of different types of functional fiber which is belief to reach intact the distal colon to be fermented by the microbiota exerting its beneficial effects. Therefore, although more studies are required, including in vivo analysis, the results obtained underline the need to use specific methods for carbohydrates based on small intestinal extract of mammals, to test the resistance of these compounds to digestion. In general, the well-known prebiotic activity of these fibers is aligned with their partial digestibility, since not all the carbohydrate fraction is digested. In consequence, it is important to highlight that not all prebiotic carbohydrates are nondigestible and can be partially digested, still exerting the beneficial effect in the large intestine, which, therefore, warrants a revision of the current assumption of non-digestibility of prebiotic carbohydrates, as recently was suggested by Hernandez-Hernandez (2019). 50 Abbreviations used

Conflicts of interest
There are no conflicts to declare.