Correlation of sequence and spatial position in globular proteins

for 30min at 37OC and bound radioactivity was quantified by a standard centrifugation assay. Using concentrations of 0.15.0nM, the larger part of the observed binding was inhibited by an excess of unlabelled p-bungarotoxin; this saturable binding showed a maximum value of 160fmol/mg of protein and K, of 0 . 6 n ~ (Table I). The latter is in close agreement with that calculated from the measured association and dissociation rate constants and corresponded to the K, observed for unlabelled toxin; this indicates that tritiation did not alter appreciably the toxin’s binding affinity. The binding component is highly specific as other presynaptically-acting neurotoxins, including taipoxin, botulinum neurotoxin and a-latrotoxin were unable to inhibit 13H]/l-bungarotoxin binding, consistent with these proteins having distinct sites of action (Chang & Su, 1980; Othman et al., 1982). Likewise, pure phospholipase A, from the venoms of bee, Apis mellifera, and Naja melanoleuca displayed low (Table 1) and negligible affinities respectively. In contrast, toxin I, a single chain protein purified from Dendroaspis polylepis and devoid of phospholipase A, activity (Strydom, 1976), inhibited totally the observed binding with a K, similar to the K, value for [3Hl/?-bungarotoxin (Table I). Toxin I has been shown to increase the release of acetylcholine at synapses in skeletal muscle and to antagonize the inhibitory action of p-bungarotoxin thereon (Harvey, 1982). Hence, the effectiveness of this neurotoxin in preventing [3H1fibungarotoxin binding, together with the other binding and electrophysiological data cited above, establish that /I-bungarotoxin interacts specifically with a functional macromolecule on nerve membranes concerned with neurotransmitter release.