Cholera toxin inhibitors studied with high-performance liquid affinity chromatography: a robust method to evaluate receptor-ligand interactions

Abstract

Anti-adhesion drugs may be an alternative to antibiotics to control infection of micro-organisms. The well-characterized interaction between cholera toxin and the cellular glycolipid GM1 makes it an attractive model for inhibition studies in general. In this report, we demonstrate a high-performance liquid affinity chromatography approach called weak affinity chromatography to evaluate cholera toxin inhibitors. The cholera toxin B-subunit was covalently coupled to porous silica and a (weak) affinity column was produced. The K(D) values of galactose and meta-nitrophenyl alpha-D-galactoside were determined with weak affinity chromatography to be 52 and 1 mM, respectively, which agree well with IC(50) values previously reported. To increase inhibition potency multivalent inhibitors have been developed and the interaction with multivalent glycopolypeptides was also evaluated. The affinity of these compounds was found to correlate with the galactoside content but K(D) values were not obtained because of the inhomogeneous response and slow off-rate from multivalent interactions. Despite the limitations in obtaining direct K(D) values of the multivalent galactopolypeptides, weak affinity chromatography represents an additional and valuable tool in the evaluation of monovalent as well as multivalent cholera toxin inhibitors. It offers multiple advantages, such as a low sample consumption, high reproducibility and short analysis time, which are often not observed in other methods of analysis.

Figure 1

Structures of monovalent inhibitors as well as the natural ligand GM1os. (A) 3-nitrophenyl α-D-galactopyranoside (MNPG). (B) N-(ε-aminocaproyl)-β-galactosylamine (CapGal). (C) 2-nitrophenyl β-D-galactopyranoside (ONPG). (D) Galβ1-3GalNAcβ1-4[NeuAcα2-3]Galβ1-4Glc (GM1os).

Figure 2

The schemes of multivalent CTB inhibitors with different density of pendant saccharide ligands.

Figure 3

The reaction scheme of the coating of silica with (3-glycidyloxypropyl)-trimethoxysilane and the immobilization with reductive amination.

Figure 4

Evaluation of the MNPG–CTB interaction with frontal chromatography at pH 7.0 and 22 °C. (A) Break-through curves (fronts) from the injection of 2.82, 1.41 and 0.35 mM MNPG (0.70 and 0.18 mM fronts are not shown). (B) First derivative of the fronts; the peak apex was used to determine the midpoint of each front. (C) One-site binding isotherm based on the frontal chromatography results; K_D (1.3 mM) and B_max (260 nmol) were determined from the non-linear regression. The corresponding Scatchard plot is shown as an inset.

Figure 5

Evaluation of the galactose–CTB interaction with inhibition chromatography at pH 7.0 and 22 °C. (A) Retardation of MNPG (injection volume: 5 μL, concentration 33 μM) with increasing galactose concentration in mobile phase (25, 50, 100 and 200 mM; 300 and 400 mM are not shown). (B) One-site binding isotherm based on the inhibition chromatography results; K_D (52 mM) was determined from the non-linear regression. (C) Scatchard plot of the inhibition chromatography experiment.

Figure 6

Zonal chromatography under linear conditions of MNPG, CapGal and ONPG at pH 7.0 and 22 °C. Injection volume was 5 μL and sample concentrations were 10 μg/mL (about 30 μM for all compounds).

Figure 7

Zonal chromatography of the four glycopolypeptides (solid lines) at pH 7.0 and 22 °C. The interaction with CTB was partly inhibited with 56 mM galactose in the mobile phase (dotted lines). (A) 17-H-6/CapGal₁₂; (B) 17-H-6/CapGal₅; (C) 35-H-6/CapGal₆; (D) 17-H-6/Glc₁₂.