NUTRITIONAL COMPOSITION COMPRISING 2'FUCOSYLLACTOSE AND 3'GALACTOSYLLACTOSE

Abstract

The invention pertains to a nutritional composition for infants or young children comprising 2′fucosyllactose, and 3′galactosyllactose, and preferably dietary butyric acid.

Claims

1. A nutritional composition for infants or young children, which is a formula feeding and which is not human milk, comprising: a. 2′fucosyllactose (2′-FL) in an amount of (i) 0.01 to 1 g per 100 ml nutritional composition; (ii) 0.075 to 7.5 wt. % based on dry weight; and/or (iii) 0.015 to 1.5 g per 100 kcal, and b. beta 3′galactosyllactose (beta3′-GL) in an amount of (i) 0.010 to 0.500 g per 100 ml; (ii) 0.075 to 3.75 wt. % based on dry weight and/or (iii) 0.015 to 0.75 g per 100 kcal.

2. The nutritional composition according to claim 1 further comprising dietary butyrate.

3. The nutritional composition according to claim 1, wherein the composition is at least partly fermented by lactic acid producing bacteria and comprises 0.1 to 1.5 wt. % of the sum of lactic acid and lactate based on dry weight of the nutritional composition, and wherein at least 90 wt. % of the sum of lactic acid and lactate is L-lactic acid and L-lactate.

4. The nutritional composition according to claim 1, wherein the composition further comprises LC-PUFA selected form the group of DHA, ARA, and EPA, preferably DHA and EPA, preferably DHA, EPA and ARA, more preferably comprising at least 1 wt. % of the sum of DHA, ARA and EPA based on total fatty acids.

5. The nutritional composition according to claim 1, wherein the formula further comprises galacto-oligosaccharides and/or fructo-oligosaccharides.

6. The nutritional composition according to claim 1, wherein the nutritional composition is selected from the group consisting of infant formula, a follow on formula or a young child formula.

7.-8. (canceled)

9. The nutritional composition according to claim 1, comprising (i) 0.3 to 5 wt. % dietary butyrate based on total fatty acids; (ii) 10 mg to 175 mg per 100 ml; (iii) 15 to 250 mg per 100 kcal; and/or (iv) 0.075 to 1.3 wt. % based on dry weight.

10. The nutritional composition according to claim 1, comprising (i) 0.2 to 5 g of the sum of galacto-oligosaccharides and fructo-oligosaccharides per 100 ml; (ii) 0.3 to 7.5 g per 100 kcal; and/or (iii) 1.5 to 35 wt. % based on dry weight.

11.-14. (canceled)

15. A method for improving the intestinal barrier function and/or for improving the immune system and/or for improving the intestinal microbiota and/or for the treatment or prevention of infections in particular intestinal infections, the method comprising administering to a subject in need thereof the nutritional composition according to claim 1

16. The method according to claim 15, wherein the method is for the treatment and/or prevention of allergy, for the induction of tolerance to allergens and/or for the prevention and/or treatment of atopic dermatitis.

17. The method according to claim 15, wherein the subject in need thereof is an infant or young child.

18. The method according to claim 17, wherein the subject in need thereof is an infant.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0113] FIG. 1: Effects of different galactosyllactoses (GLs) on the DON-induced impairment of the Caco-2 cell monolayer integrity. FIGS. 1A and 1B show the transepithelial electrical resistance (TEER) for different GLs. FIGS. 1C and 1D show the translocation of lucifer yellow (LYF) to the basolateral compartment. TEER was expressed as a percentage of the initial value and LYF was expressed in ng/cm.sup.2×h, i.e. in ng/ml. alpha3′-GL is Gal (alpha 1-3)-Gal (beta 1-4)-Glc; beta3′-GL is Gal (beta 1-3)-Gal (beta 1-4)-Glc; beta4′-GL is Gal (beta 1-4)-Gal (beta 1-4)-Glc′; beta6′-GL is Gal (beta 1-6)-Gal (beta 1-4)-Glc. Data are the mean±s.e. *: p<0.05 compared to control, **: p<0.01 compared to control, ***: p<0.001 compared to control, {circumflex over ( )}: p<0.05 compared to DON control, {circumflex over ( )}{circumflex over ( )} p<0.01 compared to DON Control, {circumflex over ( )}{circumflex over ( )}{circumflex over ( )} p<0.001 compared to DON Control.

[0114] FIG. 2: Different effects of GLs on the DON-induced increase in IL8 release by Caco-2 cells. IL-8 secretion is expressed as pg/mi as mean±s.e. alpha3′-GL is Gal (alpha 1-3)-Gal (beta 1-4)-Glc; beta3′-GL is Gal (beta 1-3)-Gal (beta 1-4)-Glc, beta4′-GL is Gal (beta 1-4)-Gal (beta 1-4)-Glc, beta6′-GL is Gal (beta 1-6)-Gal (beta 1-4)-Glc. Data are the mean±s.e. *: p<0.05 compared to control, **: p<0.01 compared to control, ***: p<0.001 compared to control, {circumflex over ( )}: p<0.05 compared to DON control, {circumflex over ( )}{circumflex over ( )} p<0.01 compared to DON Control, {circumflex over ( )}{circumflex over ( )}{circumflex over ( )} p<0.001 compared to DON Control.

EXAMPLES

Example 1: Infant Formula with 2′-FL and Dietary Butyrate Improve Intestinal Alkaline Phosphatase Expression

[0115] Two infant formulae were subjected to an in vitro digestion step and after the in vitro digestion step the effect on intestinal barrier maturation was examined, in particular the maturation of alkaline phosphate (AP). AP is an intestinal enzyme that is expressed and secreted by enterocytes and used as differentiation marker. AP plays a pivotal role in intestinal homeostasis and innate immune defense by dephosphorylating harmful substances such as microbial ligand lipopolysaccharide (endotoxin).

[0116] The control infant formula was a non-fermented infant formula supplemented with non-digestible oligosaccharides (scGos/lcFOS) in an amount of 0.8 mg/100 ml when in ready to drink form. The scGOS being derived from Vivinal GOS and the lcFOS being derived from RaftilineHP. The fat component being mainly vegetable oils, fish oil and microbial oil (source of arachidonic acid). The amount of butyric acid was below 0.05 wt. % based on total fat.

[0117] The active infant formula was the partly fermented infant formula similar to example 8, i.e. additionally containing 0.1 g 2′-FL, the lipid component comprising about 50 wt. % of bovine milk fat, and having about 1.5 wt. % of butyric acid based on total fatty acids, about 3.4 g fat per 100 ml about 0.28 wt. % lactic acid based on dry weight, and about 25 mg 3′-GL per 100 ml when in ready to drink form.

In Vitro Digestion:

[0118] Infant formulae were prepared at 13.7% (w/v) in MiliQ water and 35 ml was transferred to bio-reactors in a computer controlled semi-dynamic gastrointestinal model simulating infant conditions. Each reactor was equipped with a pH electrode and four dosing lines. Each dosing line was connected to a pump adding either; a) hydrochloric acid 0.25M and b) Sodium bicarbonate 0.5 M for pH control or c) Simulated Gastric Fluid (SGF), d) Simulated Intestinal Fluid (SIF). The pH was controlled by standardizing to 6.8 at the start of digestion, then lowering the pH gradually during a 2-hour gastric phase to 4.3. In the intestinal phase of digestion, the pH is gradually raised from 6.5 to 7.2 over 2 hours. At t=0 (the start of digestion), 5.8 ml of Simulated Salivary Fluid (100 mM NaCl, 30 mM KCl, 1.4 mM CaCl.sub.2, 14 mM NaHCO.sub.3, 0.6 mg/ml α-amylase from Aspergillus oryzae (SIGMA, A9857)) was added as a bolus. From t=0 onwards 12.25 ml of SGF (100 mM NaCl, 30 mM KCl, 1.4 mM CaCl.sub.2, 50 mM Sodium acetate, 0.125 mg/ml pepsin from porcine gastric mucosa (SIGMA, P7012), and 0.05 mg/ml Lipase from Rhizopus oryzae, Amano) was gradually added until t=120 (the end of the gastric phase). The consecutive intestinal phase started with the pH being increased to 6.5, and the gradual addition of 31.5 ml SIF (100 mM NaCl, 10 mM KCl, 1.7 mM CaCl.sub.2, 0.17 mg/ml trypsin from bovine pancreas (SIGMA, T9201), 0.18 mg/ml chymotrypsin from bovine pancreas (SIGMA, C4129), 0.09 mg/ml pancreatic Lipase from porcine pancreas (SIGMA, L0382), 1.42 mg/ml Taurocholate (SIGMA, 86339) and 0.6 mg/ml Tauroursodeoxycholate (SIGMA, T0266)). At the end of simulated gastro-intestinal digestion a 5 ml sample was taken, mixed with 5 ml enzyme inhibitor buffer (0.1 M sodium phosphate, pH 5.5, 0.58 mg/ml trypsin-chymotrypsin inhibitor from Glycine max (SIGMA, T9777), 34.5 μg/ml Orlistat (SIGMA, 04139)) snap frozen and stored at −20° C. until further use.

Cell Differentiation

[0119] Cells from the enterocyte-like and brush border expressing human intestinal cell line C2BBe1 (ATCC® CRL-2102™) were seeded at 5000 cells/well in 96-wells Nunc™ Edge plates and grown to confluency in Dulbecco's Modified Eagle's Medium, (Catalog No. 30-2002) with 10% fetal calf serum, 1% penicillin/streptomycin and 0.01 mg/ml human transferrin. After reaching confluency, culture medium was replaced with predigested infant formula diluted in culture medium without fetal calf serum at final concentrations of 0.34%, 0.17% and 0.08 5% (w/v) in quadruplicates and incubated at 37° C., 5% CO.sub.2 for 96 hours, refreshing with the diluted predigested infant formula after 48 hours. At the end of the incubation period, 50 μl of culture medium was collected per well, the quadruplicates were pooled and stored at −20° C. until measurement of the AP activity. Then, all wells were washed with ice-cold Phosphate Buffered Saline and to each well, 100 μl of 50 mM Tris-HCL, 150 mM NaCl, 0.5% triton-100 at pH 7.0 was added. After 30 min incubation on ice, cell lysates were collected and protein content was determined using Thermo Fischer, Pierce BCA Protein Assay Kit according to the manufacturer's instructions. AP activity was determined by Biovision Alkaline Phosphatase Activity Colorimetric Assay Kit, according to the manufacturer's instructions. AP activity was expressed as Units/mg protein

Results

[0120] The AP activity was statistically significantly increased (p<0.05, t-test) in the enterocytes that were treated with the predigested infant formula of the invention, when compared to the enterocytes treated with predigested control formula. This effect was dose dependent and significantly different at all concentrations tested. The increase in extracellular AP activity compared to the control formula was 43%, 36% and 32% at infant formula concentrations of 0.34, 0.17 and 0.085% (w/v), respectively, see Table 1. This increase in extracellular AP activity is indicative for an improved intestinal barrier function maturation and an improved defense against intestinal pathogenic bacteria.

TABLE-US-00001 TABLE 1 AP activity of intestinal enterocytes exposed to predigested control or experimental formula in mU/mg protein. Control Test Dilution IF concentration formula formula (x) (g/100 ml) Mean SEM Mean SEM P* 40 0.34 0.84485 0.08992 1.20802 4.601E−02 0.023 80 0.17 1.16073 0.05346 1.57569 6.284E−02 0.007 160 0.085 1.49291 0.11494 1.96274  5736E−02 0.022 *p value determined by t-test, 2-tailed, two-sample equal variance.

Example 2 Infant Formula with 2′-FL and 3′-GL Improve Intestinal Lactase Expression and Cell Proliferation

[0121] The nutritional compositions of example 1 were tested in a similar experiment as example 1. Instead of 13.7, 13.6% (w/v) of the formula was used. Instead of lipase from Rhizopus oryzae, rabbit lipase was used at 16.6 mg/ml (Germ, REG.340) in the gastric phase. During the intestinal phase, 0.06 mg/ml pancreatic Lipase from porcine pancreas (SIGMA, L0382), and 3.5 mg/ml porcine pancreatic lipase (SIGMA L0126) was used instead of 0.09 mg/ml pancreatic Lipase from porcine pancreas (SIGMA, L0382).

[0122] Lactase activity was measured by mixing 30 μl of cell lysate with 30 μl assay buffer (maleic acid 0.625 M, lactose 0.12 M, pH 6.0) and incubated at 37° C. for 4 hours, the resulting glucose was quantified. Lactase activity was expressed as μmol glucose/min/mg.

[0123] It was found that the lactase activity was significantly increased when cells were treated with the predigested experimental test infant formula compared to predigested control formula, see Table 2.

TABLE-US-00002 TABLE 2 Lactase activity of intestinal enterocytes exposed to predigested control or experimental formula in mU/mg protein. Dilution IF concentration Control formula Test formula (x) (g/100 ml) Mean SEM Mean SEM P* 80 0.17 0.63 0.02 0.816 0.008 0.001 160 0.085 0.69 0.01 0.831 0.037 0.020 *p value determined by t-test, 2-tailed, two-sample equal variance.

[0124] Lactase activity increases in differentiating enterocytes, followed by an increase in sucrase activity after which brush border lactase activity starts dropping off. Since the cells did not show sucrase activity at the time of measurement (data not shown), an increased lactase activity is thus indicative for a more differentiated cell state.

Cell Proliferation Test

[0125] Crypt-like human colon carcinoma HT-29 cells were seeded at 5.Math.10.sup.4 in 96-wells Nunc™ Edge plates in DMEM with 10% FCS, 1% penicillin/streptomycin and 1 g/L galactose. Cells were allowed to adhere for 30 hours after which medium was replaced with digested IF diluted in culture medium without fetal calf serum at final concentrations of 0.23%, 0.17% and 0.085% (w/v) in triplicates. Different cell proliferation rates resulted in different cellular protein contents after 72 hours incubation, these were measured by lysing cell followed by protein content determination with Thermo Fischer, Pierce BCA Protein Assay Kit according to the manufacturer's instructions.

[0126] Cell proliferation was significantly increased as shown by an increased cellular protein content of cells treated with the predigested experimental, test infant formula compared to predigested control formula (Table 3).

TABLE-US-00003 TABLE 3 Proliferation (cellular protein ug/well) of intestinal enterocytes exposed to predigested control or experimental formula in mU/mg protein. Dilution IF concentration Control formula Test formula (x) (g/100 ml) Mean SEM Mean SEM P* 80 0.17 21.2 0.1 23.9 0.6 0.011 160 0.085 21.4 0.3 23.9 0.6 0.018 *p value determined by t-test, 2-tailed, two-sample equal variance.

[0127] To achieve its function as a barrier to the external environment, the gut epithelium must be continuously renewed. The growth and renewal of gut epithelial cells depends on proliferating cells in the intestinal crypts. Stimulation of the cell proliferation rate thus is expected to support the gut barrier function.

Example 3: Beta1,3′-Galactosyllactose and 2′Fucosyllactose Protects Against Intestinal Barrier Disruption and Prevents Permeability Increase

[0128] Beta1,3′-galactosyl-lactose (beta3′-GL), beta1,4′-galactosyllactose (beta4′-GL) and beta1,6′-galactosyl-lactose (beta6′-GL) were obtained from Carbosynth (Berkshire, UK). Alpha1,3′-galactosyl-lactose (alpha3′-GL) was obtained from Elicityl (Crolles, France). Purified deoxydivalenol (DON) (D0156; Sigma Aldrich, St Luis, Mo., USA) was dissolved in pure ethanol and stored at −20° C. Human epithelial colorectal adenocarcinoma (Caco-2) cells were obtained from American Type Tissue Collection (Code HTB-37) (Manasse, Va., USA, passage 90-102).

[0129] Caco-2 cells were used according to established methods. In brief: cells were cultured in Dulbecco's modified Eagle medium (DMEM) and seeded at a density of 0.3×10.sup.5 cells into 0.3 cm.sup.2 high pore density (0.4 μm) inserts with a polyethylene terephthalate membrane (BD Biosciences, Franklin Lakes, N.J., USA) placed in a 24-well plate. The Caco-2 cells were maintained in a humidified atmosphere of 95 air and 5% CO.sub.2 at 37° C. After 17-19 days of culturing, a confluent monolayer was obtained with a mean transepithelial electrical resistance (TEER) exceeding 400 Ωcm.sup.2 measured by a Millicell-Electrical Resistance System voltohm-meter (Millipore, Temecula, Calif., USA).

[0130] Caco-2 cell monolayers were thus grown in a transwell system, which is a model for intestinal barrier function. The monolayers were pretreated for 24 h with different GLs, including beta3′-GL, alpha3′-GL, beta4′-GL and beta6′-GL in a concentration of 0.75 wt. % of the GL, before being exposed to the fungal toxin deoxynivalenol (DON), which is a trigger and model compound to impair intestinal barrier. DON was diluted to a final concentration of 4.2 μM in complete cell medium and added to the apical side as well as to the basolateral side of the transwell inserts. This DON concentration did not affect the viability of the Caco-2 cells. Incubation with DON was 24 h.

[0131] Measurements of the transepithelial electrical resistance (TEER) and lucifer yellow (LY) permeability were conducted to investigate barrier integrity. For TEER measurements a Millicel-ERS voltohmmeter connected to a pair of chopsticks electrodes was used to measure the TEER values. Results are expressed as a percentage of the initial value. For paracellular tracer flux assay the membrane impermeable lucifer yellow (LY) (Sigma, St Luis, Mo., USA) was added in a concentration of 16 μg/ml to the apical compartment in the transwell plate for 4 h, and the paracellular flux was determined by measuring the fluorescence intensity in the basolateral compartment with a spectrophotofluorimeter (FLUOstar Optima, BMG Labtech, Offenburg, Germany) set at excitation and emission wavelengths of 410 and 520 nm, respectively. Release of interleukin-8 (IL-8 or CXCL8), which is a typical marker for inflammation, was quantified in the medium of the apical side and the basolateral side of the Caco-2 transwell inserts in response to the treatments. CXCL8 concentrations were measured by using the human IL-8 ELISA assay (BD Biosciences, Pharmingem, San Diego, Calif., USA) according to manufacturer's instructions. For more details on materials and methods see Akbari et al, 2016, Eur J Nutr. 56(5):1919-1930.

[0132] The results are shown in FIG. 1 A, B, C and D and in FIG. 2. FIG. 1 shows the effects of different galactosyllactoses (GLs) on the DON-induced impairment of the Caco-2 cell monolayer integrity. FIGS. 1A and 1B show the transepithelial electrical resistance (TEER) for different GLs. FIGS. 1C and 1D show the translocation of lucifer yellow (LYF) to the basolateral compartment. TEER was expressed as a percentage of the initial value and LYF was expressed in ng/cm.sup.2×h. alpha3′-GL is Gal (alpha 1-3)-Gal (beta 1-4)-Glc; beta3′-GL is Gal (beta 1-3)-Gal (beta 1-4)-Glc; beta4′-GL is Gal (beta 1-4)-Gal (beta 1-4)-Glc; beta6′-GL is Gal (beta 1-6)-Gal (beta 1-4)-Glc. Data are the mean±s.e. *: p<0.05 compared to control, **: p<0.01 compared to control, ***: p<0.001 compared to control, {circumflex over ( )}: p<0.05 compared to DON control, {circumflex over ( )}{circumflex over ( )} p<0.01 compared to DON Control, {circumflex over ( )}{circumflex over ( )}{circumflex over ( )} p<0.001 compared to DON Control.

[0133] As can be seen from FIGS. 1A-D, the presence of DON disrupted the barrier function as shown by a decreased TEER value and an increased LY flux for the DON-control samples. Additionally, the presence of DON increased CXCL8 (IL-8) release, as shown in FIG. 2. FIGS. 1A-D further show that the presence of beta3′-GL prevented the DON-induced loss of epithelial barrier integrity as measured by increased TEER values and a reduction in the DON-affected LY flux across the intestinal epithelial monolayer.beta4′-GL and beta6′-GL did not show a significant effect on the intestinal epithelial barrier function. Interestingly, beta3′-GL, i.e. the galactosyl-lactose with a β1-3 glycosidic linkage, was effective in protecting the intestinal barrier function, whereas alpha3′-GL, i.e. the galactosyl-lactose with an α1-3 glycosidic linkage, did not prevent the DON-induced disrupted intestinal barrier. In contrast, all galactosyl-lactoses were able to decrease the DON-induced IL-8 release, as is shown in FIG. 2.

[0134] These results are indicative for the specific effect of beta3′-GL (herein also referred to as beta1,3′-galactosyllactose or Gal (beta 1-3)-Gal (beta 1-4)-Glc) on protecting the intestinal epithelial barrier function, in particular under conditions of challenges, which goes beyond and/or is independent of an effect on preventing an inflammatory response, and/or of an effect on or via the microbiota. These results are thus indicative of an effect that beta3′-GL has on increasing the intestinal barrier function and/or on the prevention and/or treatment of intestinal barrier disruption. In addition, these results are indicative of an effect of beta3′-GL on the treatment, prevention and/or alleviation of a toxin exposure associated condition in a subject, in particular when the toxin is a tricothecene toxin, and more in particular when the toxin is deoxynivalenol.

[0135] In a separate experiment the effect of 2′fucosyllactose on TEER and LYF flux was determined in the same model. 2′-FL was tested in a concentration of 1 mg/ml, and was found to statistically significantly prevent the DON induced reduction in TEER and increase in LYF, see Table 4. This is indicative for the advantageous effect 2′FL has on the intestinal barrier function. Therefore this example is indicative of a further improved effect on the intestinal barrier function in a composition, when combining 2′FL and beta3′-GL.

TABLE-US-00004 TABLE 4 Effect of 2′-FL on the DON-induced impairment of the Caco-2 cell monolayer integrity. TEER (% of Lucifer yellow initial value) flux in ng/ml Mean (s.e.) (cm.sup.2xh) Control 101.3 (0.379) 302.7 (7.325) Control with DON 34.33 (1.088)*** 530.8 (3.975)*** DON with 2′-FL (1 mg/ml) 42.79 (0.844).sup.∧∧ 446.4 (8.302).sup.∧ ***p <0.001 compared to control without DON. .sup.∧∧p <0.01 compared to control with DON, .sup.∧p <0.05 compared to control with DON

Example 4: Butyrate Improves Intestinal Barrier Function

[0136] The effect of butyrate on the intestinal barrier function was examined

Methods

[0137] T84 human intestinal epithelial cells are commonly used to study intestinal barrier integrity in vitro. T84 cells (ATCC, USA) were cultured on 12 mm transwell inserts (0.4 μm, Corning Costrar, USA) in DMEM-F12 glutamax with penicillin-streptomycin (100 IU/ml), supplemented with 5% FBS-Hl. T84 cells were used 14 days after reaching confluence. Monolayers of T84 cultured on transwell filters were pre-incubated for 48 h with or without butyrate. These samples were subsequently incubated for an additional 48 h in the presence of IL-4 (25 ng/ml). IL-4 was added to the basolateral compartment; medium and additives were changed every 24 h.

[0138] Epithelial barrier integrity was assessed by measuring transepithelial resistance (TEER; Ω×cm.sup.2) with the epithelial volt-ohm meter (EVOM; World Precision Instruments, Germany).

[0139] Results are shown in Table 5 where relative TEER values are presented. The 48 h and 96 h column is the TEER increase relative to t=0 value. IL-4 treatment disrupted the intestinal barrier function; however in the presence of butyrate this disruption was ameliorated.

TABLE-US-00005 TABLE 5 Effect of butyrate on the intestinal barrier function. Butyric acid % TEER at 96 h (IL- concentration (mM) % TEER at 48 h 4) — 17 (12)  0 (4) 4 79 (25) 26 (16)

[0140] Therefore this example is indicative of a further improved effect on the intestinal barrier function in a composition, when combining beta3′-GL, 2′-FL and optionally dietary butyrate.

Example 5: 2′-FL and (3′-GL and/or Butyrate) Effect the Immune System Differently

[0141] Immune cell activation and responses were determined by culturing human peripheral blood mononuclear cells (PBMCs) in the presence or absence of 2′-FL, 3′-GL and butyric acid with and without T cell specific stimulation.

Material and Methods

[0142] Isolation of PBMC from healthy donors: Human peripheral blood mononuclear cells (PBMC) from healthy donors were isolated from buffy coats (Sanquin, Amsterdam, the Netherlands). PBMC were obtained by centrifugation using Leucosep tubes (Greiner Bio-One). PBMC were collected and washed in PBS (Gibco, Thermo Fisher Technologies)+2% heat-inactivated FCS (Invitrogen), followed by hypotonic lysis of erythrocytes with sterile lysis buffer (0.15 M NH.sub.4Cl, 0.01 M KHCO.sub.3 and 0.1 mM EDTA, pH of 7.4 at 4° C., all from Merck, Darmstadt, Germany). After lysis, the PBMC were resuspended in freezing medium (70% RPMI 1640 medium (Gibco, Thermo Fisher Technologies) supplemented with 10% heat-inactivated FCS and 100 U/ml penicillin-streptomycin, 20% heat-inactivated FCS and 10% DMSO (Sigma)) and cryopreserved.

[0143] PBMC activation model: PBMC (0.2.Math.10.sup.6 cell/well) were cultured in 96-well flat bottom plates (Corning). For 24 hours the cells were pre-incubated with 2′-FL (Jennewein), 3′-GL (0-0.3% w/v; Carbosynth) or sodium butyrate (0.2 mM; Sigma Aldrich) and combinations thereof. Subsequently, the cells were CD3/CD28-activated (Pelicluster CD3 and Pelicluster CD28, Sanquin) for an additional 24 hours. After incubation, IFNγ was measured by ELISA in the supernatants (see below). To determine cell activity after stimulation, PBMC were incubated with cell proliferation reagent WST-1 (10 μl; Roche) and/or 10% Triton (5 μl; negative control). After 3 hours, absorbance was measured at OD450 nm and OD650 nm and cell activity was calculated according to manufacturer's instructions.

[0144] IFNγ production PBMC: PBMC were incubated with indicated reagents and after incubation the supernatants were collected and mediator levels were measured using human IFNγ ELISA kits (R&D Systems Europe Ltd.) according to manufacturers instructions.

[0145] Cytokine production of PBMC: PBMC were incubated with indicated reagents. After incubation, the supernatants were collected and IL2, IL6, IL10, IL13, IL21, TNFα, IFNγ, MIF, CCL1, CCL13, CCL17, CCL20, CCL22 and CXCL8-11 levels were measured by conducting a validated multiplex immunoassay based on Luminex technology (xMAP, Luminex Austin Tex. USA). Acquisition was conducted with the Biorad FlexMAP3D (Biorad laboratories, Hercules USA) in combination with xPONENT software version 4.2 (Luminex). Data was analyzed by 5-parametric curve fitting using Bio-Plex Manager software, version 6.1.1 (Biorad).

[0146] After PBMC stimulation, cell culture supernatants were collected, after which cytokine responses were measured in order to test immune responsiveness of the cells. The levels of cytokines measured in stimulated conditions were corrected for the (low) levels of the cytokines measured in non-stimulated conditions. In addition, since each donor is reacting in its own efficiency onto the T cell stimulus, we calculated the individual index of cytokine response by dividing the intervention induced response by the basal stimulated response.

[0147] Generally IL2, IL6, TNF-alpha, CCL1, CCL17, and CCL20 are considered to be associated with inflammation and/or proliferation. IFN-gamma, CXCL9, CXCL10, and CXCL11 are considered to be associated with a Th1 response. IL13, CCL13, and CCL22 are considered to be associated with a Th2 response. IL10 and Galectin-9 are considered to be associated with a Treg effect and IL21 is associated with a B-cell effect.

[0148] Statistical analysis: Comparison between CD3/CD28 stimulated and controls were made using paired one-tailed (Wilcoxon) t test, p<0.05 was considered significantly different.

[0149] Relative mean±SEM from the measured and calculated values in stimulated condition were statistically tested using paired two-tailed (Wilcoxon) t test p<0.05 was considered significantly different. The calculation values of the combined effect of the single ingredients was based on the per donor measured values.

[0150] Immune cell activity as measured by WST was significantly increased after 24 h by the addition of 2′-FL, whereas a decrease in activation was detected upon addition of 3′-GL. The addition of butyrate, did not influence immune cell metabolic activity neither in no-stimulated conditions, nor under T cell stimulated conditions (CD3/CD28).

[0151] The addition of 2′-FL altered the cytokine response, whereas the addition of 3′-GL did not result in the same changes. Interestingly the addition of 3′-GL with the 2′-FL seemed to boost the performance of 2′-FL significantly. Moreover, the difference that was found between the response derived from 3′-GL and 2′-FL on the metabolic activity of the cells vs the IFN-gamma production, is indicative for other immune responses to be indicted.

[0152] Overall it is concluded that the total pool of isolated human PBMCs is a diverse pool of immune cells, which respond directly and differently to provided HMOs. Although cells become more metabolic active, the cytokine production in the presence of 2′-FL is not equal to the cytokine production in the presence of 3′-GL, suggesting differential immune reactive responses.

Results on 2′-FL, 3′-GL and their Combination

[0153] The effect of coculturing with 2′FL and 3′GL and their combination on PBMC cultures from 10 human donors was studied. First the effect of T-cell specific stimulation via CD3/CD28 was determined. After stimulation of the human PBMCs, cell culture supernatants were collected, after which cytokine responses were measured in order to test immune responsiveness of the cells. Upon T-cell specific stimulation with CD3/CD28 several cytokines were detected within the cell supernatants using Luminex technology. The Th2 type of cytokines IL-4 and IL-13, chemokines CCL17 were significantly increased showing a robust T cell stimulation (Table 6).

TABLE-US-00006 TABLE 6 Cytokine IL-4, IL-13 and chemokine CCL17 levels (pg/ml) as measured in cell culture supernatants of PBMCs after stimulation with CD3/CD28 as compared to unstimulated conditions. Unstimulated Stimulated CD3/CD28 Mean (s.e.) Mean (s.e.) IL-4 1.117 (0.062) 25.83 (5.45)*** IL-13    6 (0) 50.25 (9.13)*** CCL17  1.80 (0.267) 83.97 (19.89)*** Paired one-tailed (Wilcoxon) t test *p <0.05, **p <0.01, ***p <0.001, *p <0.0001 stimulated vs unstimulated

[0154] Subsequently, in order to test the direct effect of specific compounds on PBMC activity, the cells were activated with CD3/CD28 for 24 hours after a pre-incubated for 24 h with either 2′-FL, 3′-GL and combinations thereof. In addition, since each donor is reacting in its own efficiency onto the T cell stimulus, the individual index of cytokine response was calculated by dividing the intervention induced response by the basal stimulated response (the blanc is set at 1). In this way the intervention within 10 different donors has been studied.

[0155] IL-4 and IL-13 are closely related cytokines, known to regulate many aspects of allergic inflammation. They play important roles in regulating the responses of lymphocytes, myeloid cells, and non-hematopoietic cells. For example; in T-cells, IL-4 induces the differentiation of naïve CD4 T cells into Th2 type of T cells, in B cells, IL-4 drives the immunoglobulin (Ig) class switch to IgG1 and IgE, and in macrophages, IL-4 and IL-13 induce alternative macrophage activation.

TABLE-US-00007 TABLE 7 Relative level of IL-4, IL-13 and CCL17 in CD3/CD28 stimulated condition with 2′-FL and 3′- GL, or the combination. IL-4 IL-13 CCL17 Mean (s.e.) Means (s.e.) Mean (s.e) blanc    1 (0)    1 (0)    1 (0) 0.2 wt % 2′-FL 0.9887 (0.0517)  1.039 (0.080)  1.061 (0.117) 0.1 wt % 3′-GL 0.5825 (0.0413) 0.5472 (0.0723) 0.6212 (0.0531) 0.2 wt % 2′-FL + 0.1 wt % 3′-GL 0.4290 (0.0280)* 0.4822 (0.0451)* 0.4647 (0.614)* Observed effect 0.2 wt % 2′-FL + 0.1 wt % 3′-GL 0.5712 (0.062)  0.687 (90.090) 0.6822 (0.123) Calculated effect *paired one-tailed (Wilcoxon) t test p <0.05 when compared with the 2′-FL + 3′-GL calculated effect.

[0156] T cell stimulation of the human PBMCs resulted in the significant increase of IL-4 and IL-13. Pre-incubation of the cells with 2′-FL had no effect on the levels of IL-4 and IL-13 as compared to controls. However, a decrease was detected in the presence of 3′-GL as compared to control. Moreover, the combination of 2′-FL and 3′-GL induced significantly lower levels of IL-4 and IL-13 as compared to control and 2′-FL. Interestingly, these reduced IL-4 and IL-13 levels were significantly lower than could be expected based on calculations of the individual effects of 2′-FL and 3′-GL, see Table 7.

[0157] These data show an unexpected reduced Th2 type of responsiveness upon T cell stimulation within the total PBMC population when the cells are in the presence of a combination of 2′-FL and 3′-GL, compared to 3′-GL or 2′FL alone.

[0158] The cytokines regulate cellular responses on transcriptional level, while chemokines play a role in recruiting inflammatory cells to the sites on inflammation. The chemokine CCL17 (thymus and activation-regulated chemokine) is a potent chemoattractant for Th2 lymphocytes and is thought to play an important role in inflammatory diseases like allergy. In example, serum CCL17 levels sharply reflect the disease activity of atopic dermatitis, which is considered to be a Th2-dominant inflammatory skin disease, especially in the acute phase.

[0159] Human T cell stimulation resulted in significant increase in CCL17, see Table 7. Although pre-incubation with either 2′-FL or 3′-GL had no significant effect on CCL17 levels within T cell stimulated PBMCs, a significant decrease was detected in the levels of CCL17 when the activated PBMCs were preincubated with both 2′-FL and 3′-GL as compared to single 2′-FL and 3′-GL. Based on changes induced by individual components as compared to control levels, one can calculate the expected change when combining the interventions. Interestingly when T cell stimulated PBMCs were cultured in the presence of the combination of 2′-FL and 3′-GL lower CCL17 levels were induced than expected. The changes in CCL17 levels indicate an unexpected further reduction of Th2 type responsiveness upon T cell stimulation within the total PBMC population when the cells are in the presence of a combination of 2′-FL and 3′-GL. These CCL17 data are in line with the IL-4 and IL-13 data.

[0160] The total pool of isolated human PBMCs is a diverse pool of immune cells, which can respond directly and differently to provided HMOs. Although cells become more metabolically active, Th2 mediators IL-4, IL-13 and CCL17 levels were not significantly affected by single 2′-FL exposure, while single 3′-GL exposure resulted in a reduction of these mediator levels. Interestingly, the simultaneous exposure of 2′-FL and 3′-GL statistically significantly reduced IL-4, IL-13 and CCL17 levels, thereby reducing the Th2 type of responses. These data indicate that the addition of 3′-GL to 2′-FL has the potential to reduce allergy development.

Results on 2′-FL and Butyric Acid

[0161] Interleukin-10 (IL10) is not a cell type-specific cytokine but is broadly expressed by many immune cells. The induction of IL10 often occurs together with other pro-inflammatory cytokines, although the pathways that induce IL10 may negatively regulate these pro-inflammatory cytokines. IL10 has a central role in infection by limiting/regulating the immune response to pathogens and thereby preventing damage to the host. Therefore, IL10 is generally regarded as a regulatory cytokine. IL 10 levels were measured in peripheral blood mononuclear cell (PBMC) cultures from 10 human donors.

TABLE-US-00008 TABLE 8 IL10 levels in PBMC under unstimulated condition, depicted as relative (normalized) values compared to blanc, thereby correcting for donor variation Relative IL10 level, mean (se) blanc    1 (0) 0.2% w/v 2′-FL 3.361 (1.867) 0.2 mM butyrate 1.911 (0.203) 0.2% w/v 2′-FL + 0.2 mM butyrate 14.77 (5.074)* Observed value 0.2% w/v 2′-FL + 0.2 mM butyrate 4.272 (0.643) Calculated value *paired two-tailed (Wilcoxon) t test, p <0.05 when compared to the calculated value.

[0162] In human PBMCs co-cultured with 0.2% 2′-FL a significantly (p<0.05) increased level of IL10 was detected as compared to blanc control, whereas the addition of butyrate did not have an effect on IL10 levels. The combination 0.2% 2′-FL and 0.2 mM butyrate significantly increased IL10 levels as compared to the blanc control and to 0.2 mM butyrate. Interestingly the combination of 2′-FL and butyrate increased the IL10 to higher levels than theoretically can be expected based on individual components, see Table 8. This is indicative for an unexpected, beneficial increased regulatory capacity of the human PBMC in the presence of a combination of 2′-FL and butyrate when compared to the single ingredients.

[0163] In general, CCL20 and CCR6 play a role in the recruitment of immature DCs and their precursors to sites of potential antigen-entry. Depending on the tissue microenvironment (e.g. local presence of TGF-beta, IL10 or IL15), immune cells may acquire functional CCR6 and hence migrate to sites of CCL20 production. CCL20 is shown to rapidly induce firm adhesion of subsets of freshly isolated T-lymphocytes to intercellular adhesion molecule-1. Regulation can therefore be obtained through modulation of CCL20 in un-stimulated conditions.

TABLE-US-00009 TABLE 9 CCL-20 levels in unstimulated condition depicted as relative values, thereby correcting for donor variation Relative CCL 20 mean (s.e.) blanc    1 (0) 0.2% w/v 2′-FL 2.586 (0.281) 0.1 w/v % 3′-GL 1.164 (0.212) 0.2 mM butyrate 1.053 (0.127) 0.2 w/v % 2′-FL + 0.2 mM butyrate 4.955 (1.206)** Observed value 0.2 w/v % 2′-FL + 0.1 w/v % 3′-GL + 8.127 (2.264)* 0.2 mM butyrate Observed value 0.2 w/v % 2-FL + 0.2 mM butyrate 2.639 (0.347) Calculated value 0.2 w/v % 2′-FL + 0.1 w/v % 3′-GL + 2.803 (0.524) 0.2 mM butyrate Calculated value paired two-tailed (Wilcoxon) t test: * p <0.05, ** p <0.01 when compared with the calculated value.

[0164] In human PBMCs exposed to 2′-FL an increased level of CCL20 was detected as compared to blanc control, whereas the addition of butyrate or 3′-GL alone did not have a statistically significant effect on CCL20 levels. Incubation of human PBMCs with the combination of 2′-FL and butyrate induced significantly higher levels of CCL20 compared to the blanc and butyrate alone. The further presence of 3′-GL in this combination of 2′-FL and butyrate further enhanced the CCL20 levels. Unexpectedly, the observed levels of the combination of 2′-FL and butyrate was significantly higher than can be calculated based on the single ingredients. This was also the case when the observed value of the combination of 2′-FL, butyrate and 3′-GL was compared with the theoretically calculated value based on the single ingredients. see Table 9.

[0165] These data indicate that the addition of 2′-FL, and butyrate influence immune responsiveness of human PBMCs. The further presence of 3′-GL further improves the immune response. The total pool of isolated human PBMCs is a diverse pool of immune cells, which respond directly and differently to provided HMOs. Changes as detected both in IL-10 as well as with CCL20 levels are suggestive for an unexpected improved modulation in responsiveness of the human PBMC in the presence of a combination of 2′FL and butyrate, which is even further improved when 3′GL is present.

Example 6: 2′-FL Increases the Butyrate Formation by the Microbiota, in Particular if Also GOS is Present

[0166] A faecal sample from a 3 months old healthy infant born via C-section, exclusively breastfed with no history of antibiotic usage, was used as inoculum to simulate the infant intestinal microbiota in the colon compartments of a quad-SHIME®—a dynamic model of the human gastrointestinal tract comprises 4 SHIME® units running in parallel and each SHIME® unit is composed of 3 reactors simulating the stomach and small intestine, proximal and distal colons.

[0167] SCFA profiles showed that acetate is the most abundant in the distal colon, followed by propionate (Table 10). The concentrations of acetate and propionate were higher in the presence of scGOS/lcFOS and scGOS/lcFOS/2′-FL than in the control and 2′-FL-supplemented units. Similar observations were also seen in the proximal colons (data not shown). Interestingly, butyrate was generated earlier in the distal colon and at a higher concentration in the presence of 2′-FL and scGOS/lcFOS/2′-FL relative to the control and the scGOS/lcFOS groups. The level of iso-butyrate, a branched SCFA resulting from the proteolytic fermentation, was reduced in the distal colon in the presence of scGOS/lcFOS/2′-FL and scGOS/lcFOS.

TABLE-US-00010 TABLE 10 Short-chain fatty acids produced in the distal colons of the un-supplemented (control) and supplemented SHIME.sup. ® units at D1 to Day 3 (D1-D3), Day 4 to Day 11 (D4-D11), and Day 12 to D15 (D12-D15). Control scG/IcF 2′-FL D1-D3 D4-D11 D12-D15 D1-D3 D4-D11 D12-D15 D1-D3 D4-D11 D12-D15 Acetate   37 ± 3.77 27.77 ± 2.32  27.31 ± 5.43  58.83 ± 17.67 84.3 ± 9.8  82.13 ± 3.35  38.67 ± 10.25 26.22 ± 0.97  22.96 ± 1.04  (Mean ± SD) Propionate 10 ± 1  6.58 ± 0.58  6.5 ± 1.08 13.83 ± 3.4  18.75 ± 3.5  18.63 ± 1.03  10.17 ± 3.01  6.58 ± 0.49 6.13 ± 0.63 (Mean ± SD) Butyrate 0 ± 0 0.05 ± 0.12 0.71 ± 0.15 0 ± 0 0.04 ± 0.1  0.29 ± 0.08 0 ± 0 0.74 ± 0.49 0.99 ± 0.09 (Mean ± SD) Iso- 0 ± 0  0.6 ± 0.93 1.29 ± 0.28 0 ± 0 0.43 ± 0.49 0.20 ± 0.23 0 ± 0 0.68 ± 0.75 0.93 ± 0.21 Butyrate (Mean ± SD) 2′-FL + scG/IcF D1-D3 D4-D11 D12-D15 Acetate 58.5 ± 9.26 83.94 ± 6.8  84.43 ± 9.20  (Mean ± SD) Propionate 14.17 ± 1.04  20.17 ± 2.36  21.50 ± 2.55  (Mean ± SD) Butyrate 0 ± 0 0.64 ± 0.32 1.06 ± 0.15 (Mean ± SD) Iso- 0 ± 0 0.16 ± 0.19 0 ± 0 Butyrate (Mean ± SD)

[0168] The glycoprofile data revealed that 2′-FL was not metabolized when supplemented alone, but only utilised in the presence of scGOS/lcFOS where it was slowly metabolised across the proximal and distal colon. All other carbohydrates including scGOS were depleted within the first hour in the proximal colon. It was shown that 2′-FL was only fermented in the presence of other GOS, in particular GOS/lcFOS resulting in a microbial eco-system that is suggested to confer health benefits.

[0169] scGOS/lcFOS/2′-FL enhanced the production of butyrate, an important SCFA for the gut barrier function. scGOS/lcFOS/2′-FL resulted in a surprisingly lower level of iso-butyrate, which is an indication of a less proteolytic activity in the colon.

Example 7: Inhibition of Pathogens in the Microbiota by 2′-FL, 3′-GL and Butyrate

[0170] Anaerobic fermentation of fecal slurry samples was tested in a BioLector Pro microfluidics mini multifermentor. Faecal slurry samples were collected from breast fed infants and from formula fed infants. These faecal slurry samples were processed by adding 0.6 grams feces in 40 ml Baby Reichardt V.6 medium+mucus+Ammonium sulfate+lactate and acetate. The resulting solutions were inserted to the BioLector Pro microfluidics mini multifermentor. The test legs were supplemented with 3-GL, 2′-FL, 3′-GL+2′-FL, and GOS/FOS. The control leg was supplemented with sterile water.

[0171] In addition the test legs were supplemented with Clostridium difficile C153 (difficile agar), Salmonella enteriditis S29 (XLD agar), Cronobacter sakasakii E71 (chromogenic agar) or Klebsiella pneumonia K2 (Simons citrate inositol agar). For every NDO and for the control also a pathogen free culture was prepared.

[0172] After fermentation the fermented solutions were tested for SCFA content (in particular acetic acid, propionic acid, butyric acid and isobutyric acid), ammonia content, lactate content and pathogen concentration. Also DNA-isolation+identification and 16s sequencing was performed to determine the composition of the microbiota.

[0173] A 32 well plate that can handle low pH was used. The wells of this plate were filled with fecal solution. 2.5% (w/v) of the different sterile carbohydrate solutions (3′-GL, 2′-FL, 2′-FL/3′-GL (2.0+0.5%). and Glucose were added to the fecal slurry according to a template.

[0174] The experiment was started, and pH setpoint was either 5.5 (facial inoculum from infant 1, vaginally born, breast fed, 5 months of age) or pH 6.0 (Inoculum from infant 2, vaginally born, breast fed, 5 months of age) with continuous pH regulation, and temperature 37° C., humidity 85%, OD control. At 4, 8 and 24 hours a sample is taken for CFU determination on TOS-propionate MUP agar (total Bifidobacteria), XLD agar for Salmonella enteriditis S29, and Simons citrate inositol agar for Klebsiella pneumonia K2 and for SCFA, D- and L Lactate and ammonia analysis. Fecal pellet was used for 16s DNA sequencing.

[0175] For both inocula the extent of fermentation, as measured by NaOH consumption, i.e. acid production, was highest with the combination of 3′-GL/2′-FL, when compared to 3′-GL or 2′-FL alone. The rate of initial acidification was high for 3′-GL and for 3′-GL/2′-FL. In general 2′-FL alone resulted in slower and lower acidification. As the amount of carbohydrates that can be fermented is the same in the reaction vessels, the higher total acidification with the combination is indicative for an unexpected, synergistic effect of the combination of 2′-FL and 3′-GL. The SCFA that were produced was for the largest part acetic acid. Also L-lactic acid was produced.

[0176] Growth of bifidobacteria was observed with 3′-GL, 2′-FL and with the mixture 2′-FL/3′-GL and growth stimulation was in general very similar. However, at 24 h the highest level was observed with the 3′-GL/2′-FL mixture for baby 1. Growth of Enterobacteriaceae was also observed, and was very similar under the conditions tested, but was lowest at 8 h for the combination of 2′-FL/3′-GL for the inoculum of baby 1. 16s Microbiota sequencing data at this time point showed a relative decrease was seen of the phylum of Proteobacteria (main contributor being the genus Escherichia). At the end of fermentation, when carbohydrates were depleted the 2′-FL/3′-GL fed microbiota was able to retain a more positive microbiota composition than the controls (glucose and blanc). For the inoculum of baby 2 the effect on bifidobacteria was highest in the presence of 3′-GL, and the growth reducing effect on enterobacteriacea was best when a combination of 3′-GL/2′-FL was used.

[0177] Under conditions where the vessels were spiked with the mixture of pathogens, in general a slightly reduced acidification was observed when compared to the conditions where there was no spiking with pathogens. However, the effects of 2′-FL, 3′-GL and 2′-FL/3′-GL on acidification, as determined by NaOH consumption, was not affected, and again was highest with 3′-GL/2′-FL for both inocula. For the inoculum of baby 1 Salmonella growth was most restricted by 2′-FL, whereas Klebsiella growth was most inhibited by the combination of 3′-GL/2-′FL. For the inoculum of baby 2 Salmonella growth was most restricted by 2′-FL, whereas Klebsiella growth was most inhibited by 3′-GL or the combination of 3′-GL/2′-FL. For both inocula the outgrowth of C difficile was restricted under all the conditions

[0178] These results are indicative for an improved effect on the intestinal microbiota function and composition combination of 2′-FL and 3′-GL going beyond the effects of 2′-FL alone or 3′-GL alone.

Example 8: Infant Formula

[0179] Infant formula, intended for infants of 0 to 6 months of age, comprising per 100 ml, after reconstituting 13.7 g powder to an end volume of 100 ml: [0180] 66 kcal, [0181] 1.3 g protein (whey protein/casein wt ratio 1/1), [0182] 7.3 g digestible carbohydrates (mainly being lactose), [0183] 3.4 gram fat (of which about 50 wt. % bovine milk fat, the remainder being vegetable oils, fish oil and microbial oil). Based on total fatty acids the amount of butyric acid is 1.48 wt. %, the amount of arachidonic acid is 0.52 wt. %, the amount of eicosapentaenoic acid is 0.11 wt. %, the amount of docosahexaenoic acid is 0.52 wt. %, [0184] 0.9 g non-digestible oligosaccharides, of which 0.1 g 2′-FL (source Jennewein), 0.08 g long chain fructo-oligosaccharides (source RaftilineHP), 0.72 g galacto-oligosaccharides (of which about 25 mg 3′galactosyllactose obtained by fermentation, the remainder being galacto-oligosaccharides from Vivinal GOS), [0185] Minerals, vitamins, trace elements and other micronutrients as according to directives for infant formula, [0186] Part of the formula about 26 wt. % based on dry weight, is derived from the Lactofidus product fermented by S. thermophilus and B. breve strains, resulting in about (about 0.28 wt. % lactic acid based on dry weight of the composition, of which more than 95 wt. % is in the L-form.

Example 9: Follow on Formula

[0187] Follow on formula, intended for infants over 6 months of age, comprising per 100 ml, after reconstituting 14.55 g powder to an end volume of 100 ml: [0188] 68 kcal, [0189] 1.36 g protein (whey protein/casein wt ratio 4/6), [0190] 8.1 g digestible carbohydrates (mainly being lactose), [0191] 3.2 gram fat (of which about 50 wt. % bovine milk fat, the remainder being vegetable oils, fish oil and microbial oil). Based on total fatty acids the amount of butyric acid is 1.47 wt. %, the amount of arachidonic acid is 0.29 wt. %, the amount of eicosapentaenoic acid is 0.12 wt. %, the amount of docosahexaenoic acid is 0.56 wt. %, [0192] 0.85 g non-digestible oligosaccharides, of which 0.05 g 2′-FL (source Jennewein, name?), 0.08 g long chain fructo-oligosaccharides (source RaftilineHP), 0.72 g galacto-oligosaccharides (of which about 25 mg 3′galactosyllactose obtained by fermentation, the remainder being galacto-oligosaccharides from Vivinal GOS), [0193] Minerals, vitamins, trace elements and other micronutrients as according to directives for infant formula, [0194] Part of the formula, about 26 wt. % based on dry weight, is derived from the Lactofidus product fermented by S. thermophilus and B. breve strains, resulting in about 0.28 wt. % lactic acid based on dry weight of the composition, of which more than 95 wt. % is in the L-form.

Example 10: Young Child Formula

[0195] Follow on formula, intended for young children over 12 months of age up to 36 months of age, comprising per 100 ml, after reconstituting 15.07 g powder to an end volume of 100 ml: [0196] 65 kcal, [0197] 1.3 g protein (whey protein/casein wt ratio 4/6), [0198] 8.7 g digestible carbohydrates (mainly being lactose), 2.6 gram fat (of which about 10 wt. % bovine milk fat, the remainder being vegetable oils, fish oil). Based on total fatty acids the amount of butyric acid is about 0.35 wt. %, the amount of eicosapentaenoic acid is 0.42 wt. %, the amount of docosahexaenoic acid is 0.63 wt. %, [0199] 1,22 g non-digestible oligosaccharides, of which 0.02 g 2′-FL (source Jennewein, name?), 0.12 g long chain fructo-oligosaccharides (source RaftilineHP), 1,08 g galacto-oligosaccharides (of which about 17 mg 3′galactosyllactose obtained by fermentation, the remainder being galacto-oligosaccharides from Vivinal GOS), [0200] Minerals, vitamins, trace elements and other micronutrients as according to directives for infant formula, [0201] Part of the formula, about 18 wt. % based on dry weight, is derived from the Lactofidus product fermented by S. thermophilus and B. breve strains, resulting in about (0.2 wt. % lactic acid based on dry weight of the composition, of which more than 95 wt. % is in the L-form.