BIFIDOGENIC HYPOALLERGENIC GOS COMPOSITIONS AND METHODS FOR PROVIDING THE SAME INVOLVING BETA-GALACTOSIDASE FROM A STRAIN OF LACTOBACILLUS DELBRUECKII SSP BULGARICUS

Abstract

The invention relates to the field of hypoallergenic oligosaccharides for use in nutritional compositions, in particular to oligosaccharides having prebiotic properties. Provided is a hypoallergenic oligosaccharide composition comprising galacto-oligosaccharides (GOS), wherein (i) the galacto-oligosaccharides (GOS) content is at least 40% by weight of the total dry matter of the composition; (ii) the allolactose content is at least 10% by weight of the total dry matter of the composition; (iii) the 6′-GL content is at least 30% by weight of the total GOS in the composition; and (iv) at least 0.5% by weight of the total GOS has a polymerization degree (DP) of six or more. The GOS composition does not trigger GOS-allergy as assessed in a basophil activation test (BAT).

Claims

1. An oligosaccharide composition comprising: (i) at least 40% galacto-oligosaccharides (GOS) by weight of the total dry matter of the composition; (ii) at least 10% allolactose by weight of the total dry matter of the composition; (iii) at least 30% 6′-galactosyl-lactose (6′-GL) by weight of the total GOS in the composition; and (iv) at least 0.5% by weight of the total GOS has a polymerization degree (DP) of six or more.

2. The oligosaccharide composition of claim 1, wherein the GOS content is at least 42% by weight of the total dry matter of the composition.

3. The oligosaccharide composition of claim 1, wherein at least 1% by weight of the total GOS has a DP of six or more.

4. The oligosaccharide composition of claim 1, wherein the allolactose content is at least 12% by weight of the total dry matter of the composition.

5. The oligosaccharide composition of claim 1, wherein the 6′-GL content is at least 40% by weight of the total GOS in the composition.

6. The oligosaccharide composition of claim 1, wherein (i) the GOS content is at least 70% by weight of the total dry matter of the composition; (ii) the allolactose content is at least 12% by weight of the total dry matter of the composition; (iii) the 6′-GL content is at least 40% by weight of the total GOS in the composition; and (iv) at least 1% by weight of the total GOS has a polymerization degree (DP) of six or more.

7. A nutritional composition comprising the oligosaccharide composition of claim 1, wherein said nutritional composition is an infant formula, a follow-up formula, a growing-up milk, a dairy product, a cereal, or a medical nutritional product.

8. A nutritional composition comprising protein source, a lipid source, a carbohydrate source, and the oligosaccharide composition of claim 1.

9. (canceled)

10. A method of promoting gut microbiota balance and health in a subject, the method comprising administering to the subject an effective amount of the oligosaccharide composition of claim 1.

11. The method of claim 10, wherein said promoting gut microbiota and health comprises enhancing bifidogenic micro-organisms in the intestinal tract and/or improving patient tolerance to at least one medical treatment that leads to a gastrointestinal tract disorder.

12. The oligosaccharide composition of claim 1, wherein the oligosaccharide composition has prebiotic properties effective to promote gut microbiota balance and health in an intestinal tract of a human or animal.

13. A method of making a bifidogenic infant formula or dietetic food, the method comprising adding the oligosaccharide composition of claim 1 to a component selected from the group consisting of fats, carbohydrates, minerals, trace elements, vitamins, and combinations thereof.

14. A method of making the oligosaccharide composition of claim 1, the method comprising (i) contacting a lactose feed with a beta-galactosidase (EC 3.2.1.23), and (ii) allowing for oligosaccharide synthesis, wherein said beta-galactosidase has an amino acid sequence having at least 98% sequence identity to SEQ ID NO:1.

15. The method of claim 14, wherein said beta-galactosidase is contacted with said lactose feed while being present in a micro-organism endogenously expressing said beta-galactosidase, wherein said micro-organism is used as whole cells or an active part or fraction thereof, and the micro-organism is Lactobacillus delbrueckii subspecies bulgaricus strain DSM20080.

16. The oligosaccharide composition of claim 1, wherein the GOS content is at least 60% by weight of the total dry matter of the composition.

17. The oligosaccharide composition of claim 1, wherein at least 1.5% by weight of the total GOS has a DP of six or more.

18. The oligosaccharide composition of claim 1, wherein the 6′-GL content is at least 43% by weight of the total GOS in the composition.

19. The method of claim 11, wherein the at least one medical treatment is selected from the group consisting of radiotherapy, chemotherapy, gastrointestinal surgery, anesthesia, the administration of antibiotics, analgesic drugs, and treatment for diarrhea.

20. The method of claim 14, wherein the beta-galactosidase (EC 3.2.1.23) is obtained from Lactobacillus delbrueckii subspecies bulgaricus strain DSM20080.

Description

LEGEND TO THE FIGURES

[0058] FIG. 1: HPLC chromatogram of (panel A) a reference GOS+6′-GL and (panel B) a representative L-GOS composition of the invention. For peak identification see Table 1.

[0059] FIG. 2: Comparison of GOS Dionex pattern synthesized by whole cells of the strains RFC-219, RFC-227, RFC-302.

[0060] FIG. 3: Comparison of GOS profile by using cell-free extract and whole cells of Lactobacillus strain RFC227.

[0061] FIG. 4: Comparison of 5 feces bifidobacterial growth using the L-GOS of the invention, sugar control or reference GOS1 and 2 as the only carbon sources in MRS medium. Panel A: after 7 hours of fermentation. Panel B: after 24 hours of fermentation.

[0062] FIG. 5: Basophil activation in 4 test subjects (panel A-D) as measured by expression of the basophil activation marker CD203c (MF1=Mean Fluorescence). Different concentrations of test composition (L-GOS) and reference composition (vGOS) were included in the study. For details see Example 5.

[0063] FIG. 6: Basophil activation in 4 test subjects (panels A-D) as measured by expression of the basophil activation marker CD36 (MF1=Mean Fluorescence). Different concentrations of test composition (L-GOS) and reference composition (vGOS) were included in the study. For details see Example 5.

[0064] FIG. 7: (panel A) Amino acid sequence (SEQ ID NO:1) and (panel B) nucleotide sequence (SEQ ID NO:2) of an exemplary beta-galactosidase enzyme for use in the present invention.

EXPERIMENTAL SECTION

Example 1: Preparation of Beta-Galactosidase Enzyme from Lactobacillus

[0065] Three selected in house Lactobacillus test strains RFC 219, RFC 227 and RFC 302 were inoculated in MRS media and were grown to an optical density at 600 nm (O.D..sub.600) of ˜1.0, and subsequently the inoculates were diluted in fresh MRS medium to an O.D..sub.600 of 0.01-0.02 and to grow to OD of ˜1.0-1.5 at 37° C. after 16-32 hours fermentation under aerobic conditions.

[0066] Whole cells were harvested by centrifugation of the fermentation broth at 6000 rpm and 18° C. for 10 minutes. After decanting the fermentation broth, two washing steps were performed by repeatedly dispersing the whole cells in demineralized water and centrifugation, aiming to remove any insoluble residues.

[0067] The obtained wet whole cells were dispersed in 10 mM natrium citrate buffer, pH6.5 by a ratio of 10% (w/w). The whole cell dispersions were used directly for

[0068] GOS synthesis (whole cells) or were disrupted (cell free extract) by a min-bead beater (Biospec Product) using 0.1 mm glass beads at a maximal speed.

[0069] Due to the heat generation during the homogenization process, the homogenization process needs to be stopped after 60 seconds. Subsequently, the samples of whole cells were cooled down to 0° C. by immersing in the ice water bath before repeating the homogenization process for second round.

[0070] The cell debris after second round homogenization was removed by centrifugation and the cell-free extract (supernatant) was used for GOS synthesis directly without any further treatment.

Example 2: Enzymatic GOS Synthesis

[0071] GOS synthesis was performed under the following conditions:

[0072] 27 gram lactose crystals (28.42 gram lactochem®, pharma grade, containing 95% lactose) was added to 27 gram 10 mM citrate buffer, pH6.5, which contains beta-galactosidase as a whole cells dispersion (in 10 mM sodium citrate buffer) of Example 1 or the cell-free extract originating from same amount of whole cells dispersion of Example 1, in order to facilitate the comparison. The reaction mixture was stirred using a magnetic stir and the temperature was regulated by water bath to be 50° C. The reaction time was 65 hours.

[0073] The amount of enzyme activity needed is pre-determined in an assay by the clarification time of the reaction mixture under the same conditions as above but in ¼ of the above scale, starting from a lactose slurry. Subsequently, the activity of the enzyme preparations, was estimated using the following equation, which was prepared on the basis of a reference enzyme Biolacta N5 (Amano):

Enzyme dosage (Unit/gram lactose)=36.77*(clarification time (hour){circumflex over ( )}−0.549.

[0074] For whole cells, the enzyme dosage was calculated to be 2.95 LU/gram lactose for RFC227, 3.3 LU/gram for RFC219 and 4.4 LU/gram lactose for RFC302. As is known by a person skilled in the art, the reaction time can be shortened by adding more enzyme at any moment of the reaction, in order to boost the reaction.

Example 3: Characterization of GOS Compositions

[0075] The content of different oligosaccharide was analyzed by Dionex HPAEC-PAD chromatography (van Leeuwen et al., Carbohydrate Research 2014, 400:59-73) on a analytic CarboPac PA-1 column. The GOS content was estimated by the peak percentage. The validity of this method was confirmed by the reference composition of the commercial Vivinal ®GOS (Table 2).

[0076] 6′-GL component was identified by spiking a reference GOS with 6′-GL standard. As shown in FIG. 1A, 6′-GL is peak 6. In the same way, peak 6 in the L-GOS was also identified to be 6′-GL (See FIG. 1B).

[0077] The 6′-GL content in L-GOS was calculated by the peak percentage of 6′-GL of the total GOS (excluding the allolactose), as shown in Table 1.

TABLE-US-00001 TABLE 1 Composition of L-GOS and its 6′-GL content Peak percentage Peak no Peak Name (%) 1 Galactose 10.32 2 Glucose 19.4 3 Allolactose 14.07 4 Lactose 8.18 5 Lactulose 1.1 6 6′-galactosyl-lactose 17.12 GOS (excluding allolactose) 46.93 6′-GL (% on GOS excl. allolactose) 36.5

[0078] To determine the GOS Degree of polymerization (DP), oligosaccharides were separated using ion-exchange chromatography on a Rezex RSO column from Phenomenex, which has a high resolution for oligosaccharide till approximately DP18 (Degree of Polymerization). After the separation on the column the different components are measured with a RI detector. This detector is able to detect compounds on basis of refractive index. The individual DP percentage is calculated by the respective peak percentage. Table 2 shows the DP composition of an L-GOS composition according to the invention compared to reference composition Vivinal®GOS.

TABLE-US-00002 TABLE 2 DP composition of L-GOS and Vivinal ®GOS DP composition on Dry matter DP composition on GOS (%) DP Vivinal- DP composition L-GOS GOS composition L-GOS Vivinal-GOS DP7 0.32 0.73 DP7 0.67 1.25 DP6 0.26 1.37 DP6 0.54 2.35 DP5 1.06 4.19 DP5 2.21 7.17 DP4 5.57 9.99 DP4 11.62 17.11 DP3 22.24 21.49 DP3 46.38 36.80 DP2-GOS 18.5 20.63 DP2-GOS 38.58 35.33 sum of GOS* 47.95 58.4 DP2 26.68 36.81 DP2 Lactose 8.18 16.18 lactose Glucose 22.07 18.89 Glucose Galactose 21.8 0.33 Galactose

[0079] The GOS profiles obtained with the 3 whole cells are depicted in FIG. 2, which shows that the GOS profiles synthesized with each of the 3 enzyme preparations is actually identical, suggesting that the beta-galactosidases associated with these 3 Lactobacillus strains are functionally identical.

[0080] In view of this similarity, subsequent experiments were performed with whole cells and cell-free extract of strain RFC219 only, at an enzyme dosage of 3.3 LU/gram lactose used under the same conditions as mentioned above except with lactose concentration of 50% (w/w). Strain RFC219 is a Lactobacillus delbrueckii subsp. bulgaricus. The beta-galactosidase obtained from RFC219 has an amino acid sequence according to SEQ ID NO:1. An aliquot of 2.0 ml sample were taken at reaction time of 29 h, 36 h, 53 h and 65 hour and deactivated by addition of 1.5% 1.5 M HCl (v/v). The GOS composition and fingerprint profile were analyzed and summarized in Table 3 and FIG. 3. As expected, the GOS profiles by using whole cells or the isolated cell-free enzyme completely match, suggesting that the enzymatic kinetics are identical.

TABLE-US-00003 TABLE 3 GOS content and sugar analysis by Dionex HPAEC-PAD Allo- Sample Galactose Glucose lactose Lactose Lactulose GOS* GOS reference 1.85 17.04 4.52 16.25 1.18 59.16 Cell Free 7.67 17.31 13.09 20.6 1.31 40.02 extract -29 h Cell Free 7.84 17.14 13.03 19.18 1.17 41.64 extract -36 h Cell Free 8.74 18.12 13.78 17.49 1.02 40.85 extract -53 h Cell Free 10.32 19.4 14.07 8.18 1.1 46.93 extract -65 h Whole cell 7.69 17.09 12.98 19.27 1.26 41.71 synthesis-29 h Whole cell 8.37 17.52 13.34 18.3 0.88 41.59 synthesis-36 h Whole cell 9.43 18.77 13.77 15.79 0.93 41.31 synthesis-53 h GOS = 100 − galactose % − glucose % − allolactose % − lactose % − lactulose % (AOAC method)

Example 4: Bifidogenic Effect of GOS Compositions

[0081] Partial purification of GOS by removing the mono-sugars was performed using a published method to Rodriguez-Colinas et al. (2013, Appl. Microbiol. And Biotechn. Vol. 97, pp 5743-5752).

[0082] A partially purified GOS preparation (“L-GOS test composition”) with the composition shown in Table 2, was tested for its bifidogenic effect using baby feces in an established in vitro model.

TABLE-US-00004 TABLE 4 Composition of partially purified L-GOS Component Concentration (%) galactose 1.64 Glucose 4.07 allo-lactose 14.82 Lactose 6.86 lactulose 0.48 GOS 72.13 GOS + allolactose 86.95

[0083] The bifidogenic effect of GOS was evaluated using TIM-2 model (TNO in vitro model of the colon (TIM-2), Venema K. (2015), The TNO In Vitro Model of the Colon (TIM-2). In: Verhoeckx K. et al. (eds), The Impact of Food Bioactives on Health. Springer, Cham), which is able to simulate material passing the ileo-caecal valve in humans. The microbiota was fed into the system through a food syringe, which contains a simulated ileal efflux medium (SIEM).

[0084] The microbiota used in this model for the current invention was established via fecal donations from 6 healthy infants (between 1-6 months old, bottle fed and no use of antibiotics for at least one month prior donation). Moreover, all babies were predominately bottle-fed. Since the feces of baby 4 did not contain any detectable bifidogenic activity, this sample was withdrawn from the assay.

[0085] The standard medium used contained the following components (g): pectin (9.4), xylan (9.4), arabinogalactan (9.4), amylopectin (9.4), casein (47.0), starch (78.4), Tween 80 (34.0), Bacto Peptone (47.0) and ox bile (0.8). Dialysis liquid contained (per litre): 2.5 g K.sub.2HPO.sub.4.3H.sub.2O, 4.5 g NaCl, 0.005 g FeSO.sub.4.7H.sub.2O, 0.5 g MgSO.sub.4.7H.sub.2O, 0.45 g CaCl.sub.2.2H.sub.2O, 0.05 g bile and 0.4 g cysteine.HCl, plus 1 ml of a vitamin mixture containing (per litre): 1 mg menadione, 2 mg D-biotin, 0.5 mg vitamin B12, 10 mg pantothenate, 5 mg nicotinamide, 5 mg p-aminobenzoic acid and 4 mg thiamine.

[0086] To determine the bifidogenic effect, the total carbohydrate was equivalently substituted by either a sugar control, the L-GOS test composition of the invention or Reference compositions GOS1 and GOS2. The sugar control is a composition equivalent to the sugar composition of the mono sugars (galactose and glucose plus lactose present in the corresponding purified GOS preparation). GOS1 and GOS2 refer, respectively, to GOS prepared with a beta-galactosidase from Bifidobacteria longum and the commercial product Vivinal®GOS.

[0087] The Bifidobacterium growth rate was analyzed after 7 hours (FIG. 4A) and 24 hours (FIG. 4B).

[0088] As shown in FIG. 4, the L-GOS composition of the invention is able to stimulate the growth of the 5 baby's feces most effectively when compared to either the sugar control or the reference compositions GOS1 and 2.

Example 5: Hypoallergenicity of an Oligosaccharide Compositions of the Invention

[0089] This example demonstrates the reduced allergenicity of an oligosaccharide composition of the invention in four human subjects with known galacto-oligosaccharide allergy. L-GOS obtained by transgalactosylation using beta-galactosidase from strain RFC227 cell-free extract and a commercial GOS reference preparation obtained using B. circulans enzyme (vGOS) were included in the test.

[0090] Eligible subjects were selected from the cohort previously studied for the prevalence of GOS-allergy in a Singapore atopic population, as described by Soh et al., (Allergy 2015, 70, 1020-3).

[0091] A Basophil Activation Test (BAT) was performed on patient blood samples. To that end, heparinized peripheral blood aliquots (100 μL) were pre-incubated at 37° C. for 5 minutes and then incubated with 100 μL of PBS (negative control), anti-IgE antibody (positive control, G7-18; BD Biosciences, San Jose, Calif) or diluted GOS samples for 15 minutes (37° C.). After incubation, cells were washed in PBS-EDTA (20 mmol/L) and then incubated with phycoerythrin-labeled anti-human IgE (Ige21; eBioscience, San Jose, Calif), biotin-labeled anti-human CD203c (NP4D6; BioLegend, San Jose, Calif), and fluorescein isothiocyanate-labeled anti-human CD63 (MEM-259, BioLegend) mAbs for 20 minutes at 48° C. Expression of CD203c and CD63 are both markers for basophil activation.

[0092] After washing the cells with 1% BSA/PBS, allophycocyanin-conjugated streptavidin (BD Biosciences) was added and incubated for 15 minutes at 48° C. Thereafter, samples were subjected to erythrocyte lysis with 2 mL of FACS Lysing Solution (BD Biosciences). Cells were then washed, resuspended in 1% BSA/PBS, and analysed by means of FACSCalibur (BD Biosciences). Basophils were detected on the basis of side-scatter characteristics and expression of IgE (IgEhigh).

[0093] In contrast to the reference GOS composition, L-GOS prepared with Lactobacillus enzyme of the strains used in the present invention elicits no positive reaction in BAT test, as evidenced by the very low or virtually no expression of the activation markers CD203c (FIG. 5) and CD63 (FIG. 6).

Example 6: Determination of the Gene and Protein Sequence of the Lactobacillus Enzymes

[0094] An attempt was made to determine the gene sequence encoding the ß-galactosidase produced by the three Lactobacillus delbreuckii strains.

[0095] Bacteria were cultured in MRS liquid media as mentioned above to reach a OD of ˜1.0. The DNA extraction was performed directly with this biomass. The DNA extraction protocol used is mainly based on the use of Zymo Research Bacterial/Fungal DNA microPrep kit D6007. The obtained DNA sample was analysed by the Illumina HiSeq2500.

[0096] Quality Analysis of FASTQ Sequence Reads

[0097] Paired-end sequence reads were generated using the Illumina HiSeq2500 system. FASTQ sequence files were generated using bcl2fastq2 version 2.18. Initial quality assessment was based on data passing the Illumina Chastity filtering. Subsequently, reads containing PhiX control signal were removed using an in-house filtering protocol. In addition, reads containing (partial) adapters were clipped (up to minimum read length of 50 bp. The second quality assessment was based on the remaining reads using the FASTQC quality control tool version 0.11.5.

[0098] De Novo Assembly

[0099] Assembly

[0100] The quality of the FASTQ sequences was enhanced using the read error correction module BayesHammer in the SPAdes version 3.10 genome assembly toolkit (Bankevich A et. Al. (2012)J Comput Biol. 19:455-477) The high-quality reads were assembled into contigs using SPAdes. Misassemblies and nucleotide disagreement between the Illumina data and the contig sequences are corrected with Pilon (Walker B J et. al. (2014) PLOS ONE 9(11): e112963) version 1.21.

[0101] Scaffolding

[0102] The contigs were linked and placed into scaffolds, where the orientation, order and distance between them were estimated using the insert size between the paired-end and/or matepair reads. The analysis has been performed using the SSPACE Premium Scaffolder version 2.3 (Boetzer et. al, 2011).

[0103] Automated Gap Closure

[0104] The gapped regions within the scaffolds are (partially) closed in an automated manner using GapFiller version 1.10 (Boetzer and Pirovano, 2012). The method takes advantage of the insert size between the paired-end and/or matepair reads.

[0105] The obtained genome sequences were annotated using DNA annotation tool: “ClustalW”(https://www.genome.jp/tools-bin/clustalw), the conversion of DNA to protein was done in an offline package (“sms2”, downloaded from http://www.bioinformatics.org/sms2/).

[0106] The amino acid sequence and the nucleotide sequences of a representative ß-galactosidase obtained are shown in FIGS. 7A and 7B, respectively.