Chemical modification of spinal cord membranes reveals [3H]strychnine binding sites that are not located on the 48 kDa subunit of the synaptic glycine receptor.

High-affinity binding sites for the convulsant alkaloid strychnine appear t o reside on a 48 kDa subunit of receptors for the inhibitory neurotransmitter glycine in the mammalian spinal cord [ I , 21. During a series of experiments to investigate the role of different amino acid residues in the recognition of strychnine at the glycine receptor, we found that modification o f arginine residues with 2,3-butanedione [ BD; 80 mM for 40 min in phosphate-buffered saline (100 mMNaCI, SO mM-Na2HP0,/NaH,P0, pH 7.4, 20°C)] increased [ ’Hlstrychnine binding to mouse spinal cord membranes [ 3. 41. This BD-induced binding (with the 3H-labelled ligand at 6 n M ) was specific in that it could be inhibited by unlabelled strychnine ( 1 0 -‘ M). However, BD treatment was found to abolish the ability of glycine ( l o ? M ) t o inhibit [ ZH]strychnine binding. Further experiments have now shown that BD treatment reveals additional [3H]strychnine binding sites in these membranes, sites which d o not appear t o reside on the previously characterized [ 5 I synaptic glycinc receptor complex. We have examined the distribution of the BD-induced increase in [ jH1strychnine binding by treating membranes prcparcd from different regions of the mouse central nervous system. A significant correlation ( I > < 0.05; Spearman rank correlation coefficient) was found between the amount o f specific, strychnine-displaceable, [jHIstrychnine binding in control untreated membranes and the magnitude of the increase in binding of this ligand after BD treatment; both parameters being highest, for example, in the spinal cord and lowest in the cerebellum. The distribution o f the strychnine binding sites revealed by BD treatment, therefore. appears t o parallel the distribution o f the high-affinity binding sites for this ligand in untreated membranes. In contrast, BD-induced [>H ]strychnine binding to spinal cord membranes prepared from the mutant mouse spastic was not significantly different ( 1 9 0.05; unpaired /-test) from that to spinal cord membranes prepared from unaffected littermate control mice. This occurred despite a reduction in high-affinity [ 3H]strychnine binding t o untreated membranes to less than 20% of control in the mutant. Further evidence that the BD-induced strychnine-binding sites arise from structures distinct from those identified by high-affinity [ ‘Hlstrychnine binding has come from experiments on synaptic glycine receptors solubilized [in 1 .S‘%, (w/v) Chaps] from mouse spinal membranes and affinity purified on a 2-aminostrychninc-agarose column [6] . When treated with BD, these purified receptors do not show an increase in [3H]strychnine binding. As found in whole membrane preparations, however, this BD-treatment of purified receptors abolishes the ability of glycine, but not unlabelled strychnine, to inhibit [ jH ]strychnine binding. Strychnine can be used as a photoaffinity label for the glycine receptor [ I ] . However, the strychnine-binding sites revealed by BD-treatment o f spinal-cord membranes failed to incorporate [ 3H]strychninc irreversibly upon illumination with ultraviolet light (nominal peak intensity at 254 nm o f 300 pW/cm?, for 30 min. 4°C). Indeed, the amount o f radioactivity incorporated into these BD-treated membranes was similar to that incorporated into untreated membranes. Solubilization of the photolabelled membranes [in 10%) (w/v) SDS] followed by SDS/PAGE and fluorography showed that thc [3H]strychnine was incorporated into a 48 kDa peptide in both BD-treated and untreated membranes. Preliminary kinetic studies suggest that the strychninebinding sites revealed by BD-treatment of spinal cord membranes are of a low affinity, with a K , , of approximately I p ~ . Although these binding sites appear to coexist in normal spinal cord membranes with the high affinity ( K , ) approximately 6 nM) strychnine-binding sites previously identified [7, 81, the present evidence suggests that they are not located on the same multi-subunit glycine receptor. Further characterization o f these low-affinity strychnine binding sites may indicate their physiological role.