Affinity chromatography of glucose‐specific lectin using silica‐based support

Abstract

Four silica-based adsorbents were prepared from covalent attachment of four carbohydrates : i.e. maltose, cellobiose, N-acetyl-D-glucosamine and paminophenyl-#?-D-glucopyranoside, respectively. These adsorbents possess either terminal D-glucose or N-acetyl-D-glucosamine as the ligand on their surfaces with a ligand density ranging from 20 to 29.2 pmol g-'. The binding of the glucose-specific lectin, concanavalin A (Con A), to the immobilized ligand on the silica surface depended on the configuration of the immobilized glucose and the linkage of the glucose to the support. Con A showed strong affinity for maltoseimmobilized silica, which contains terminal a-D-glucose, and p-aminophenyl-#?-D-glucopyranoside-immobilized silica. On the other hand, Con A showed no affinity for cellobiose-immobilized silica, which contains terminal #?-D-glucose groups, and N-acetyl-D-glucosamine-immobilized silica. The binding constants for the interactions between Con A and immobilized ligands were determined. The columns packed with the resultant affinity adsorbents were then adopted for the purification of Con A from Jack bean meal. As the diluted NaCl extract of Jack bean meal was applied to the column packed with maltose-immobilized silica, a 13.2-fold purification was achieved by stepwise-elution.

Superparamagnetic adsorbents for high‐gradient magnetic fishing of lectins out of legume extracts

“…This is further supported by the results of Blomberg et al (1993) and Lee et al (1995) for sugars coupled via glycosyl amide linkage and reductive amination, respectively, and these two chemistries were tested for adsorbent preparation due to their chemoselective nature. The recorded sugar densities of support type I and III of approximately 50 Amol g À1 support (for the simple sugar ligands) were within the range of typical values observed using other activation chemistries (Hubbuch, 2001), suggesting good coupling efficiency for both reductive amination and glycosyl amide linkage.…”

“…The six atoms introduced with DVS and hydrazine in the preparation of supports II and IV was of sufficient length to achieve this, although the flexibility created by opening one pyranoside ring (support III) could yield a comparable effect. The combination of spacer length and high ligand densities (for supports II, III, and IV) resulted in significantly tighter interaction than found for both glucose -Con A in solution (K d c 1.8 mM; Schwarz et al, 1993) and glucose-linked silica supports described by Lee et al (1995), where a K d of 39 AM was found. The binding of Con A to monosaccharide-linked particles may possibly be improved further by attaching other sugars than glucose; e.g., Me-a-Man binds to Con A c 20 times more strongly than does glucose (Schwarz et al, 1993).…”

“…Schwarz et al (1993) determined dissociation constants in the millimolar range for the binding of these monosaccharides by Con A and LCA. Despite this relatively weak binding strength immobilized glucose and other simple sugar molecules have been employed effectively as affinity ligands in purification of lectins (Blomberg et al, 1993;Lee et al, 1995), and it is on this basis that glucose and maltose were examined as The concentration was calculated from the relative composition, as determined by SDS-PAGE, and then protein concentration of the extract. Values are given as mean F standard deviation (j nÀ1 ) based on analysis of three (i.e., n = 3) separate electrophoretograms.…”

“…In contrast, neither of the maltose-or glucose-linked materials could supply as much immobilized sugar to effect adsorption. The silica based glucose-functionalized supports prepared by Lee et al (1995) were packed in a column and used for recovery of Con A from a jack bean extract. Following application of a protein load of 0.63 g L À1 bed volume Lee et al (1995) recorded a Con A yield of just 16%.…”

“…The silica based glucose-functionalized supports prepared by Lee et al (1995) were packed in a column and used for recovery of Con A from a jack bean extract. Following application of a protein load of 0.63 g L À1 bed volume Lee et al (1995) recorded a Con A yield of just 16%. In contrast the HGMF processing resulted in a total yield of > 60% with a much higher load of 4.8 g L À1 filter canister volume, illustrating the advantage of batch adsorption to our dextran-linked adsorbents.…”