Interaction of Fragment A from Diphtheria Toxin with Nicotinamide Adenine Dinucleotide

Abstract

Abstract The inhibition of protein synthesis by diphtheria toxin in extracts of mammalian cells results from the inactivation of elongation factor 2 (EF-2) by covalent attachment of the adenosine diphosphate ribose (ADPR) moiety of NAD+. NAD+ + EF-2 ⇌ ADPR-EF-2 + nicotinamide + H+ This reaction is catalyzed by Fragment A (mol wt 24,000) or other less common fragments generated by limited proteolysis and reduction of the toxin, but not by the toxin itself (mol wt 63,000). This report describes studies of the interaction of NAD+ with Fragment A. In addition to its major enzymic activity, Fragment A also catalyzes the slow hydrolysis of the nicotinamide-ribose linkage of NAD+. NAD+ + H2O ⇌ ADPR + nicotinamide + H+ This activity (NAD+-glycohydrolase; EC 3.2.2.5) is several orders of magnitude lower than the NAD+:EF-2-ADPR-transferase activity and probably does not contribute to the toxicity of diphtheria toxin. However, its existence implies a direct binding of NAD+ to Fragment A and is of interest with regard to the mechanism of transfer of ADPR to EF-2. Binding of NAD+ and related compounds to Fragment A was studied by dynamic dialysis, fluorescence quenching, and spectral absorbance techniques. The fragment contains a single binding site for NAD+ with a Kd of about 8 µm. Binding of NAD+ is rapidly reversible, and no evidence of a covalent ADPR-Fragment A intermediate was found. NAD+ strongly quenches the intrinsic fluorescence of the fragment (emission maximum 333 nm) and induces a broad peak of absorbance at 360 nm (e ∼ 500). Both effects are interpreted to result from formation of a charge transfer complex between the nicotinamide moiety of NAD+ and at least one of the tryptophan residues of the fragment. Binding of NAD+ also enhances the resistance of the fragment to trypsin or chymotrypsin. No significant change in either the sedimentation coefficient (2.19 S) or the rotational relaxation time was found upon addition of NAD+. The affinities of various NAD+ analogs and partial structures for the binding site on Fragment A closely paralleled their activities as inhibitors or substrates of ADPR transfer. These results, together with the fact that Fragment A alone hydrolyzes the same linkage which is ruptured during ADPR transfer, leave little doubt that it is this binding site which is involved in transfer of ADPR to EF-2. A model of the ADP-ribosylation of EF-2 is proposed based on these and other results.