Affinity chromatography of nicotinamide–adenine dinucleotide-linked dehydrogenases on immobilized derivatives of the dinucleotide

Abstract

1. Three established methods for immobilization of ligands through primary amino groups promoted little or no attachment of NAD+through the 6-amino group of the adenine residue. Two of these methods (coupling to CNBr-activated agarose and to carbodi-imide-activated carboxylated agarose derivatives) resulted instead in attachment predominantly through the ribosyl residues. Other immobilized derivatives were prepared by azolinkage of NAD+(probably through the 8 position of the adenine residue) to a number of different spacer-arm–agarose derivatives. 2. The effectiveness of these derivatives in the affinity chromatography of a variety of NAD-linked dehydrogenases was investigated, applying rigorous criteria to distinguish general or non-specific adsorption effects from truly NAD-specific affinity (bio-affinity). The ribosyl-attached NAD+derivatives displayed negligible bio-affinity for any of the NAD-linked dehydrogenases tested. The most effective azo-linked derivative displayed strong bio-affinity for glycer-aldehyde 3-phosphate dehydrogenase, weaker bio-affinity for lactate dehydrogenase and none at all for malate dehydrogenase, although these three enzymes have very similar affinities for soluble NAD+. Alcohol dehydrogenase and xanthine dehydrogenase were subject to such strong non-specific interactions with the hydrocarbon spacer-arm assembly that any specific affinity was completely eclipsed. 3. It is concluded that, in practice, the general effectiveness of a general ligand may be considerably distorted and attenuated by the nature of the immobilization linkage. However, this attenuation can result in an increase in specific effectiveness, allowing dehydrogenases to be separated from one another in a manner unlikely to be feasible if the general effectiveness of the ligand remained intact. 4. The bio-affinity of the various derivatives for lactate dehydrogenase is correlated with the known structure of the NAD+-binding site of this enzyme. Problems associated with the use of immobilized derivatives for enzyme binding and mechanistic studies are briefly discussed.