Prebiotic galacto-oligosaccharides (GOS) were shown to support mucosal immune development by enhancing regulatory-type Th1 immune polarization induced by synthetic CpG oligodeoxynucleotides (TLR9 agonist mimicking a bacterial DNA trigger). Epithelial-derived galectin-9 was associated with these immunomodulatory effects. We aimed to identify the most active fractions within GOS based on the degree of polymerization (DP), and to study the immunomodulatory capacities of DP3-sized β-3′galactosyllactose (β-3′GL) using a transwell co-culture model of human intestinal epithelial cells (IEC) and activated peripheral blood mononuclear cells (PBMC). IEC were apically exposed to different DP fractions of GOS or β-3′GL in the presence of CpG, and basolaterally co-cultured with αCD3/CD28-activated PBMC, washed, and incubated in fresh medium for IEC-derived galectin analysis. Only DP3-5 in the presence of CpG enhanced galectin-9 secretion. DP3-sized β-3′GL promoted a regulatory-type Th1 response by increasing IFNγ and IL-10 or galectin-9 concentrations as compared to CpG alone. In addition, IEC-derived galectin-3, -4, and -9 secretion was increased by β-3′GL when combined with CpG. Therefore, the GOS DP3-5 and most effectively DP3-sized β-3′GL supported the immunomodulatory properties induced by CpG by enhancing epithelial-derived galectin secretion, which, in turn, could support mucosal immunity.

Human milk oligosaccharides (HMOs) and their most abundant component, 2′-Fucosyllactose (2′-FL), are known to be immunomodulatory. Previously, it was shown that HMOs and 2′-FL bind to the C-type lectin receptor DC-SIGN. Here we show, using a ligand-receptor competition assay, that a whole mixture of HMOs from pooled human milk (HMOS) and 2′-FL inhibit the binding of the carbohydrate-binding receptor DC-SIGN to its prototypical ligands, fucose and the oligosaccharide Lewis-B, (Le^b) in a dose-dependent way. Interestingly, such inhibition by HMOS and 2′-FL was not detected for another C-type lectin, langerin, which is evolutionarily similar to DC-SIGN. The cell-ligand competition assay using DC-SIGN expressing cells confirmed that 2′-FL inhibits the binding of DC-SIGN to Le^b. Molecular dynamic (MD) simulations show that 2′-FL exists in a preorganized bioactive conformation before binding to DC-SIGN and this conformation is retained after binding to DC-SIGN. Le^b has more flexible conformations and utilizes two binding modes, which operate one at a time via its two fucoses to bind to DC-SIGN. Our hypothesis is that 2′-FL may have a reduced entropic penalty due to its preorganized state, compared to Le^b, and it has a lower binding enthalpy, suggesting a better binding to DC-SIGN. Thus, due to the better binding to DC-SIGN, 2′-FL may replace Le^b from its binding pocket in DC-SIGN. The MD simulations also showed that 2′-FL does not bind to langerin. Our studies confirm 2′-FL as a specific ligand for DC-SIGN and suggest that 2′-FL can replace other DC-SIGN ligands from its binding pocket during the ligand-receptor interactions in possible immunomodulatory processes.

Apart from being a prebiotic and laxative, lactulose has direct immunomodulatory effects.

Human milk oligosaccharides (HMOs) and its most abundant component, 2’-Fucosyllactose (2’-FL), are known to be immunomodulatory. Previously, it was shown that HMOs and 2’-FL bind to the C-type lectin receptor DC-SIGN. Here we show, using a ligand-receptor competition assay, that a whole mixture of HMOs from pooled human milk (HMOS) and 2’-FL inhibit the binding of the carbohydrate-binding receptor DC-SIGN to its prototypical ligands, fucose and the oligosaccharide Lewis-B, (Le ^b ) in a dose-dependent way. Interestingly, such inhibition by HMOS and 2’-FL was not detected for another C-type lectin, Langerin, evolutionary similar to DC-SIGN. The cell-ligand competition assay using DC-SIGN expressing cells confirmed that 2’-FL inhibits the binding of DC-SIGN to Le ^b . Molecular dynamics (MD) simulations show that 2’-FL exists in a preorganized bioactive conformation before binding to DC-SIGN and this conformation is retained after binding to DC-SIGN. Le ^b has more flexible conformations and utilizes two binding modes, which operate one at a time via its two fucoses to bind to DC-SIGN. 2’-FL may have a reduced entropic penalty due to its preorganized state compared to Le ^b , and it has lower binding enthalpy, suggesting better binding to DC-SIGN. Thus, due to the better binding to DC-SIGN, 2’-FL may replace Le ^b from its binding pocket in DC-SIGN. MD simulations also showed that 2’-FL does not bind to Langerin. Our studies confirm 2’-FL as a specific ligand for DC-SIGN and suggest that 2’-FL can replace other DC-SIGN ligands from its binding pocket during ligand-receptor interactions in possible immunomodulatory processes.