Average degree of polymerization (DP): The average number of monosaccharide building blocks per chain of glucose-based saccharides in each carbohydrate composition.

Maltose-alternan-oligosaccharide (MAOS): an example of a glucose based saccharide with maltose at the reducing end.

Mono-saccharides: any class of simple sugars that can act as building blocks of larger carbohydrate structures. Simple sugars cannot be hydrolysed to give a simpler sugar. Examples of a simple sugar include glucose and fructose. Di-saccharides: any sugar whose molecules contain two monosaccharide building blocks, for example sucrose, leucrose and maltose.

Added sugar: all mono and di-saccharides described above present in the composition

Glycemic response or postprandial glucose: the glucose concentration measured in the blood or interstitial tissue following the consumption of carbohydrates by a subject, preferably a human subject.

The terms “food,” “food product” and “food composition” mean a product or composition that is intended for ingestion by an individual such as a human and provides at least one nutrient to the individual. As used herein, these terms encompass food in any form, including both liquid (e.g., a beverage) and solid. The compositions of the present disclosure, including the many embodiments described herein, can comprise, consist of, or consist essentially of the elements disclosed herein, as well as any additional or optional ingredients, components, or elements described herein or otherwise useful in a diet.

A “beverage” is a substantially homogenous liquid that is at least 85 wt. % water, in some embodiments at least 90 wt. % water or at least 95 wt. % water. A “ready-to-drink” beverage is in a liquid form that can be consumed without further addition of liquid and preferably is aseptic. Reconstitution and dilution can comprise addition of water and/or milk to the powder or concentrate respectively, and in some embodiments the method comprises a reconstitution or dilution step.

Those skilled in the art will understand that they can freely combine all features of the present invention disclosed herein. In particular, features described for the product of the present invention may be combined with the method of the present invention and vice versa. Further, features described for different embodiments of the present invention may be combined. Where known equivalents exist to specific features, such equivalents are incorporated as if specifically referred to in this specification.

Further advantages and features of the present invention are apparent from the figures and non-limiting examples.

EXAMPLES
Example 1

Production of Digestible Carbohydrate Composition

The prior art process is known for example from WO 0047727 A2 and WO 2009095278 A2, and comprises steps P1-P4 of FIG. 1a.

In step P1 bioconversion of sucrose and maltose is done in in batch reactor for a duration of about 20 h at T=37° C. The sucrose: maltose ratio is chosen to 7:1 (w/w or mol/mol). Step P1 is done in the batch reactor for bioconversion which is shown in FIG. 2a.

P2 is a step of ultrafiltration for removal of alternan-polymer (alternan polysaccharide) and alternansucrase enzyme (AlSu). This step is done in the ultrafiltration device shown in FIG. 2a.

In P3, fructose is removed by nanofiltration. This step is done in the nanofiltration device shown in FIG. 2a. Here, water is added and a mixture of water and fructose is removed.

In the final step P4 the product from P3 is concentrated by evaporation. This is done in the evaporation device in FIG. 2a and maltose-alternan oligosaccharide (MAOS) is obtained.

The process of the invention, in a specific embodiment, comprises three steps S1-S3, shown in FIG. 1b.

In comparison to the prior art process of FIG. 1a, the bioconversion in step S1 comprises continuous feed of sucrose and continuous removal of fructose. Half-continuous feed and removal is possible. The mass ratio of sucrose (total amount added) to maltose is 19:1 (19 kg sucrose per 1 kg maltose) and the duration is about 72 h. Step P3 of the prior art can be omitted because fructose is removed in step S1 already. Removing fructose makes possible using a higher ratio of sucrose to maltose and reaching a higher degree of polymerization of the maltose-alternan-oligosaccharide.

Step S1, a bioconversion is done in the reactor in FIG. 2b, wherein maltose and alternansucrase enzyme (AlSu) are present in water. A stated above, bioconversion is done with continuous feed of sucrose, continuous removal of fructose, for a duration about 72 h (variable), T=37° C., and a sucrose: maltose ratio of 19:1 (w/w or mol/mol). Sucrose is added as feed (dissolved in water) and the reactor content is stirred. In the bioconversion in the reactor, alternan oligosaccharide comprising acceptor molecule maltose, also called maltose alternan oligosaccharide (MAOS) is formed as main product and alternan polymer, fructose and leucrose are formed as by-products. Content from the reactor is continuously circulated through the membrane cell (diafiltration cell), where water, fructose and leucrose are removed in membrane filtration, which in this example is a nanofiltration, done as a constant volume diafiltration. Leucrose content is reduced from about 30% to less than 10%, in comparison with the prior art. Removed water is replaced by the water feed stream.

The reactor and the membrane cell form a combined bioconversion and nanofiltration device, also called reactor system.

Process step S2 of this embodiment corresponds to step P2 of the prior art. Here, alternan polysaccharide (alternan polymer), as by product, and alternansucrase enzyme (AlSu) are removed by ultrafiltration. The step is beneficial in case that more alternan polymer as desired has been formed, or in order to steer desired DPw of the alternan species remaining.

Process step S3 of this embodiment corresponds to step P4 of the prior art. Here the product is concentrated by evaporation.

The composition of the reaction solution used during the production is shown in Table 1. The 2.1 L solution gives a total of about 5.6 L with the water in the system (dead volume).

The process is run without depletion for the first hour to minimize potential loss of maltose across the membrane. Subsequently, fructose is constantly depleted via a nanofiltration membrane Filmtec NF270-2540 (DOW). Upon completion of the chain extension, the nanofiltration module was replaced with a TRISEP 2540-UE50-QXF ultrafiltration module (Microdyn Nadir). This separated the maltose-alternan oligosaccharide (MAOS) fraction from the longer aging chains and the enzyme. The membranes used during the process and process parameters used are summarized in Table 2. The filtrate was finally concentrated to a dry matter content of >72%.

TABLE 1

Composition of the reaction solution

Component
Amount
feed rate

Maltose
451
g
Batch

Sucrose
8500
g
^~100 g/h

sodium acetate
57
g
Batch

Alternansucrase
1900
U
Batch

Water
ad 2.1
L
Batch

TABLE 2

membranes and parameters

process step
filter
pressure
Temperature

nanofiltration
Filmtec NF270-
5-30 bar, actually
30-40° C., actually

2540 (DOW)
used 15 bar
used 37° C.

ultrafiltration
TRISEP 2540-
2-15 bar, actually
30-60° C., actually

UE50-QXF
used 10 bar
used 40° C.

(Microdyn

Nadir)

FIG. 3 shows the increase in chain length over the course of the process. The values were recorded by GPC-RI measurements. The relationship for calculation of DPw from the Mw values of FIG. 3 is as follows: DPw=Mw/(162 Da). So it can be seen that at the end of the process, a DPw of about 15.4 is reached (2500/162).

In order to reach an average chain length DPw of about 15, 19 kg of sucrose (in total) were used per 1 kg of maltose.

Comparative example: Alternan-oligosaccharide was made according to a process as shown in FIG. 1a. A ratio of 21:1 (kg Sucrose: kg Maltose) was used and DPw of 9.9 was obtained. For the measurement of DPw, GPC-RI was used, but not exactly according to the method-protocol as mentioned above. Nevertheless, in comparison with the results of FIG. 3 it is shown that by the present invention alternan with higher DPw can be obtained.

DPw values for two samples obtained by the method of the invention that were analyzed with HPAEC-PAD method are summarized in table 4.

Example 2

Structural Characterization of Digestible Carbohydrate Composition Prepared According to Example 1

Simple sugars (glucose, fructose, leucrose, sucrose, maltose) were quantified using a Dionex ICS-3000 DC apparatus equipped with an HPLC carbohydrate column (CarboPac PA1 column, 4×250 mm, no guard at 30° C.), inert styrene divinyl benzene polymer (Dionex Corporation, 2010) and gold triple potential pulsed amperometric detection (PAD). Eluent A (300 mM aqueous NaOH), eluent B (MiliQ water) and eluent C (CH3COONa 500 mM in NaOH 150 mM) were used as the mobile phase in a gradient mode, with a total run of 35 mins.

The simple sugars present in the tested products on dry basis are summarized in the Table 3 below:

Tested

product
Glucose
Fructose
Leucrose
Maltose
Sucrose

DCC-1
0.0
1.2
12.2
1.6
0.3

DCC-2
0.0
0.1
12.1
0.0
0.0

P1

DCC-3
1.5
2.8
11.2
0.0
0.0

P2

Values for DPw were measured by HPAEC-PAD after reducing and hydrolyzing glucose-based saccharides. Two milliliters of a solution containing 6 g/mL of digestible carbohydrate composition were treated with 0.2 mL of a NaBH₄solution (40 mg/mL) in 0.5 M ammonia at 40° C. for 30 min. Reduced samples were subsequently hydrolyzed with 0.5 mL 2 M Trifluoroacetic acid heated at 121° C. for 1 h to release monomers. The released monomers were quantified by injecting sample solutions on a Thermo Scientific™ Dionex™ ICS-6000 ion chromatograph system equipped with a CarboPac™ MA1 and fed with eluents (water and NaOH 1000 mM) at 0.4 mL/min. DP values are calculated with the following formula:

DP
=

(

sugar
⁢

alcohols

1
⁢
8
⁢
2

)

+

(

glucose
⁢

content

1
⁢
8
⁢
0

)

(

sugar
⁢

alcohols

1
⁢
8
⁢
2

)

DP values for samples are summarized in the table 4 below:

Tested Product
Avg DP (DPw)

Comparative Sample DCC-1
6.9 ± 0.02

DCC-2
12.4 ± 0.24

P1

DCC-3
17.3 ± 0.53

P2

The MW of DCC-1, DCC-2 P1 and DCC-3 P2 were also analysed quantitatively using Thin Layer Chromatography, using a silica TLC plate, a mixture of chloroform:acetic acid:water:ethanol (30:35:25:20) as mobile phase and the diphenylamine-aniline reagent as a visualization agent (FIG. 4).

Glycosidic linkage profiles for glucose-based oligosaccharides were measured with partially-methylated alditol acetates by GC-MS. Briefly, samples were dissolved in anhydrous DMSO, deprotonated by an addition n-Butyl Lithium (Sigma 230707) and methylated with Methyl iodide (Sigma 289566). The methylated samples were subsequently hydrolyzed with 2 N TFA (60 min at 121° C.). The hydrolyzed samples were evaporated under a nitrogen air draft, re-dissolved in 1 M ammonium hydroxide and aldehyde groups were reduced with a DMSO solution containing sodium borodeuteride (20 mg/ml). Glacial acetic acid was added drop wise to stop reaction and acetylation was done by addition of 1-methylimidazole and acetic anhydride. Partially methylated alditol acetates in acetone were quantified by GC-MS (7890A-5975C MSD, Agilent Technologies, Inc., Santa Clara, CA, USA) using a Supelco 24111-U SP-2380 capillary column (injector volume, 0.5 μl; injector temperature, 250° C.; detector temperature, 250° C.; carrier gas, helium: 30 mL/min; split ratio, 40:1; temperature program, 100° C. for 3 min, 4° C./min to 270° C. for 20 min. Electron impact spectra were acquired at 69.9 eV over 50-550 Da mass range.

TABLE 5

Glycosydic-linkage
DCC-1
DCC-2
DCC-3

Terminal-Glc
36
31.5
29.3

1,3-D-Glc
13
15.2
16.8

1,6-D-Glc
39
44.1
44.3

1,4-D-Glc
11
6.8
5.0

1,3,6-D-Glc
1.4
2.4
4.7

Higher values of 1,6-glycosidic linkages are explained by the leucrose content in the digestible carbohydrate compositions that contributes to the amounts of 1,5,6-Tri-O-acetyl-1-deuterio-2,3,4-tri-O-methyl-D-glucitol observed. In addition, leucrose, as well as monomeric glucose, contributes to the amounts of terminal-Glc in the digestible carbohydrate compositions.

Example 3

Production of Digestible Carbohydrate Composition

In this example, parameters of the process described in Example 1 for DCC-3 were varied as shown in the following table and alternansucrase was given to the reactor in 4 equal portions, the first portion being present before sucrose was fed to the reactor.

- Sucrose amount: 16.5 kg
- Sucrose feeding rate: 775 g/h
- Enzyme activity: 79.2 kU
- Temperature: 43° C.
- Time: 27 h

Enzyme activity means the total units of alternansucrase enzyme used in the process.

Reaction time of the process of the invention (not including here further process steps like removing alternan-polysaccharide and alternansucrase enzyme by a further membrane filtration, or concentrating a retentate which is obtained in the further membrane filtration) could be reduced by increasing the sucrose amount, increasing the sucrose feeding rate, increasing the enzyme activity and increasing the temperature.

Glycosidic linkage profiles for glucose-based oligosaccharides were measured as per Example 2. Results are shown in Table 6 below:

TABLE 6

Glycosydic-linkage
%

Terminal-Glc
23.1

1,3-D-Glc
17.4

1,6-D-Glc
46.0

1,4-D-Glc
5.5

1,3,6-D-Glc
7.0

Example 4

Preliminary Crossover Study Results

In a previous randomized, controlled, crossover study, a digestible carbohydrate composition with a degree of polymerization of 7 (DCC-1) was tested in 16 healthy volunteers. 25 g of the DCC-1 dissolved in 300 mL of water were ingested and post-prandial glucose response was measured during two hours (FIG. 6 and FIG. 7). As a control, 25 g of glucose syrup were consumed.

In this study, the consumption of DCC-1 led to a lower glycemic response as compared to glucose syrup, however, the differences for tmax, iCmax and iAUC were not significant. Moreover, no major gastrointestinal discomfort was reported by the study participants. These results suggest that the α-glucan consumed is largely digested but the modest increase of degree of polymerization (DP of 7), as compared to glucose syrup, does not allow to obtain significant benefits towards glycemic response.

Example 5

Clinical Trial Methodology

This was a monocentric, controlled, randomized, double-blind, crossover study where 16 participants consumed different test products containing different types of carbohydrates differing in their structure and composition. Subjects' participation in this research project was voluntary and they could end their participation at any time without having to justify the withdrawal of their consent or any termination of their participation. If they withdrew, coded data that was already collected up to that point was used and was anonymized after analyses.

The aim of this study was to test the effect of different α-glucans on post-prandial glucose response (PPGR) and hydrogen (H2) production to provide insight into their digestibility.

The primary objective was to determine if the α-glucans tested induce a lower glucose response as compared with maltodextrin.

The secondary objective was to determine indirectly if the α-glucans tested are fully absorbed in the small intestine and induce gastrointestinal discomfort.

Glucose response was assessed during 3 hours after intake of the experimental product. The 3 h incremental area under the curve (3 h-iAUC) was the primary endpoint to address the primary objective.

As secondary endpoints, the following parameters were analyzed:

- Other parameters derived from PPGR (i.e. cross-sectional values, iCmax, Cmax, tmax, partial iAUCs, AUCs);
- Digestibility by measuring breath H2 during the 4 hours following test product ingestion. Parameters derived from these H2-curves were 4 h-iAUC, partial iAUC, partial AUC, Cmax and cross-sectional values. All these values were derived from H2-data corrected for CO2. Similar parameters were derived from H2-data not corrected for CO2, as well as from CH4-data (both with and without CO2 correction);
- Gastro-intestinal tolerance 3 h after the product intake with a visual analogue scale for each symptom of interest: 1) Diarrhea, 2) Abdominal cramping, 3) Vomiting, 4) Audible bowel sounds, 5) Flatulence or Gas.

The glycaemia was measured using a flash glucose monitoring (FGM, Free Style Libre®, Abbott) device, which is a sensor with minimal invasiveness for real-time monitoring of body glucose levels in interstitial fluids. The FGM has been developed and validated for use in adults with type 1 or type 2 diabetes [1-4] and measures interstitial glucose every 15 minutes for 14 days.

Hydrogen breath testing is widely used for detection of carbohydrate malabsorption. Its principle relies on the detection of hydrogen in the exhaled air resulting from bacterial fermentation of carbohydrates mainly in the colon. To obtain rapid and reliable indirect information on carbohydrate absorption, a breath analyser (Lactotest 202, M.E.C Belgium) measuring H2 was used in this study.

The test products were α-glucans with varying degrees of polymerization (DP) and glycemic linkages, and therefore different degrees of digestibility. Digestible carbohydrate composition 2 (DCC-2) has a DP of 12.4 and digestible carbohydrate composition 3 (DCC-3) has a DP of 17.3. As a reference, a fully digestible and fully caloric maltodextrin (glucose syrup) was administered. 33 g of total carbohydrates were contained in the test products, either in the form of powder (glucose syrup) or syrup (digestible carbohydrate composition). On the day before the test visits, the test products were dissolved in 300 mL of water and stored overnight at 4° C. On the morning before consumption, the beverages were heated up to room temperature and served to the participants

The target population was completely healthy males and females (based on anamnesis) aged between 18 and 45 years with a BMI between 20 and 29.9 kg/m2. The subjects were not eligible for participation if they presented one or several of the following exclusion criteria:

- Pregnant or lactating women;
- Any concomitant medication potentially interfering with study procedures and assessment, such as antibiotics, antiacids, or other medications impacting transit time, colonoscopy, irrigoscopy or other bowel cleansing procedures four weeks prior the test;
- Major medical/surgical event in the last 3 months potentially interfering with study procedures and assessments;
- Abnormal bowel transit and history of chronic constipation with passage of fewer than 3 spontaneous bowel movements per week on average or chronic or recurrent diarrhea with spontaneous bowel movements more often than 3 times daily;
- Known food allergy and intolerance to test product;
- Medically known cutaneous hypersensitivity to adhesives and plasters;
- Alcohol intake higher than 2 servings per day. A serving is 0.4 dl of strong alcohols, 1 dl of red or white wine, or 3 dl of beer;
- Smokers;
- Volunteers who cannot be expected to comply with the protocol;
- Subject having a hierarchical link with the research team members.

Potential candidates were contacted using adverts sent out within Nestlé entities. Interested and eligible volunteers were given a full oral information session (V0) by the research team on the research objectives, methodology, as well as its possible risks/inconveniences. Candidates were invited for a medical screening visit to ensure their eligibility and for signature of the informed consent.

At least 24 hours before the intake of the first test product, one FGM device was inserted into the non-dominant arm of each volunteer. Then, the participants attended a different test visit for each of the beverage, separated by at least one day to allow a low fibre diet on the day preceding each test visit. The sensor was removed on the last day of test, after completion of all measurements.

On the day before each test visit, the subjects were required to refrain from consuming alcohol. There were asked not to take any medication like aspirin or supplement containing Vitamin C that may affect FGM measurements, and promotility drugs, laxatives and antibiotics four weeks prior the test and during the whole study. Fermentable foods such as complex carbohydrates had to be avoided on the day prior to breath testing. The last meal on the day preceding the test had to be not too ample and not contain any dietary fibres. Participants were also asked to avoid intake of chewing gums from Bpm the day before each test visit.

The general study scheme is described in FIG. 5. At each test visit, the subjects arrived at the metabolic unit at 8 am, fasting since 8 pm the previous day. To minimize interferences with breath testing, the subjects were required to avoid chewing gums and wearing perfume and were asked to brush their teeth thoroughly at home before each visit at the metabolic unit.

Readings from the FGM device were taken before and after product intake. Mean of the two measures were considered as baseline value. Readings were also performed at 30, 60, 90, 120, 150, 180, 210 and 240 minutes. Breath samples to measure the production of hydrogen were taken before product intake, then at 30, 60, 90, 120, 150, 180, 210 and 240 minutes, just after glucose reading. 180 minutes after product intake and once the corresponding glucose reading and breath test were completed, the subjects were asked to fill in a questionnaire about gastrointestinal symptoms. Following these procedures, a breakfast without fibres composed of white bread, honey and coffee or tea or water was served. Any other food, including chewing gums, were prohibited during the test period and only water was allowed.

A minimal sample size of N=10 subjects completing a study is required by 150-26642 for determining the glycemic index (GI) of food products and to classify them [5]. In the case of α-glucans, the study 17.08.BIO has shown that N=16 is required to detect a reduction of 35% in PPGR 3 h-iAUC vs. maltodextrin control (SD=50%) with α=5% (two-sided) and power=80%. Since this previous study used a smaller serving size (i.e. 25 g), the calculated sample size is conservative (i.e. variability in healthy subjects is generally reduced with higher servings). This sample size is also appropriate for the H2-related endpoints as shown by the fact that N=16 allows to differentiate a difference of 0.35 log ppm with α=5%, power=80% [6]. As a conclusion, the sample size for this study was N=16.

For the primary endpoint (i.e. 3 h-iAUC derived from PPGR), the test products were compared to the maltodextrin control using paired t-tests (following the logic of ISO-26642). In order to control the False Discovery Rate (FDR) at α=5%, the Benjamini-Hochberg procedure [7] was applied. A sensitivity analysis was performed by using a mixed model to take into account potential systematic position or carry-over effects [8]. For all other pairwise comparisons and all other endpoints, the analyses were the same but without any correction for multiplicity.

Example 6

Results of Clinical Trial

16 healthy volunteers, 6 females and 10 males, with a mean age of 31.4±5.9 years, a mean BMI of 23.0±1.6 kg/m2 and a mean fasting glycaemia of 4.8±0.5 mmol/L were recruited.

FIG. 8 and FIG. 9 illustrate the 4 h-post-prandial glucose and incremental glucose responses, respectively, of all test products. As compared to glucose syrup, both glucose-based saccharides led to significantly lower iCmax but no significant difference is observed for tmax. The time required for the glycaemia to return to baseline values is not significantly longer compared to the reference. While 1 h-iAUC is significantly lower for both test products, only DCC-3 led to significantly lower 2 h- and 3 h-iAUC, as compared to the glucose syrup.

When comparing the glucose curve shape of the α-glucans to the reference, it is observed that for both α-glucans the peak of glucose is lower and the return to baseline slower. The decrease of glycaemia after the peak is less rapid than the reference and the glycaemia remains slightly higher for a longer period of time, suggesting that the α-glucans might be digested and utilized slightly more slowly than the maltodextrin. Moreover, the consumption of α-glucans leads to less hypoglycemia after return to baseline compared to the control.

In FIG. 10 are represented the mean relative 1 h-, 2 h- and 3 h-iAUC of the test products compared to the glucose syrup. DCC-2 lead to 27% (p<0.05), 16% and 13% reduction of 1 h-, 2 h- and 3 h-iAUC, respectively, compared to the reference. DCC-3, with a higher degree of polymerization than DCC-2, led to greater glycemic reductions as compared to the reference: 45% for 1 h-iAUC (p<0.05), 38% for 2 h-iAUC (p<0.05) and 33% for 3 h-iAUC (p<0.05).

DCC-2 and -3 contain 2.1 and 2.9 g of fructose equivalent, respectively, and are therefore not fully glucose-based, unlike the glucose syrup. The dark grey bars of FIG. 10 are the adjusted data for the predicted relative iAUC if the product was pure glucose-based. Once adjusted the reduction of iAUC for DCC-2 remains significant only for the 1 h-iAUC (24%), while for DCC-3 1 h-, 2 h- and 3 h-iAUC reductions are significant (42%, 34% and 28% respectively). Even with the adjusted predictions, the reductions of glycaemia remain significant and it can be concluded that this is not the fructose being the main driver of PPGR reduction.

Post-prandial breath H2 was used as an indirect measurement of products digestibility. If non-digested carbohydrates reach the colon they are be fermented by colonic bacteria. This fermentation produces gases, such as H2 and CH4, which can be measured in the breath. Following consumption of the test products, no significant difference between each other was observed and, as well as no significant difference relative to the fully digestible control product (FIG. 11 and FIG. 12). The common range for fasting breath hydrogen is 7±3 ppm and most values during 240 minutes are found within this range. Only DCC-2 has higher values than the other products, but they remain under the carbohydrate malabsorption threshold of 20 ppm and decrease throughout the test to reach levels similar to the others. A small increase of hydrogen production is observed at 180 min, after breakfast consumption, suggesting that the lactotest is detecting even small amounts of H2, which can come from the presence of very small amounts of complex carbohydrates or a slightly incomplete absorption of simple carbohydrates (e.g. fructose). Therefore, all α-glucans seem to be largely digestible and do not induce carbohydrate fermentation like fibres.