“Specific binding” of 3H-phencyclidine: Artifacts of the rapid filtration method

Abstract

Attempting to understand the nature of the specific binding of [ 3H]-phencyclidine ( 3H-PCP) to rat brain membranes, demonstrated recently with the rapid filtration method (I: Vincent et al., PNAS 76: 4678, 1979; II: Zukin and Zukin, ibid., 5372), we found that 3H-PCP binds to GF/B glass fiber filters in a manner indistinguishable from “specific binding”. The apparent Kd for this binding is 0.24±0.12 μM, not significantly different from the values in I and II. Binding of 3H-PCP to the P 2 fraction of rat brain membranes is not affected by denaturing the tissues in boiling water. The apparent Kd and Bmax values are nearly the same for the native and denatured P 2 fractions. The definition of pharmacologically meaningful binding sites for 3H-PCP by the rapid filtration method is further hampered by the fast dissociation rate of bound 3H-PCP, as expected from the relatively large Kd. The amount bound is sensitive to the separation (washing) procedure: two volumes of 5ml reduce specific binding to native tissue by 60%, four volumes reduce it by 80%. We show that these characteristics of 3H-PCP binding to both GF/B filters and denatured tissue are not shared by other radioligands, such as 3H-D-lysergic acid diethylamide ( 3H-D-LSD) and 3H-(±)3-quinuclidinyl benzylate ( 3H-QNB). The very large filter binding of 3H-PCP is shown to be the reason for spurious differences in the density of PCP binding sites in various brain regions: we find that the amount of 3H-PCP bound per assay is equal in each of the ten brain regions studied but that, due to differing amounts of protein in assays from different re-regions, an apparent regional distribution of binding sites is obtained when the results are normalized to binding per mg protein. The competition of other ligands for 3H-PCP binding to all three materials-— glass fiber filters, native and denatured P 2 fraction — yields nearly identical IC50 values and potency rankings. These values are in good agreement with I and II indicating in that this case the pharmacological identity criterion does not discriminate between receptor and nonreceptor interactions. We conclude that the artifacts created in this case by the rapid filtration method render it unsuitable for labeling the pharmacologically relevant binding sites of 3H-PCP. If specific PCP receptors exist, as claimed in I and II, methods other than rapid filtration must be used to reveal them.