Evidence supporting a two-receptor model for insulin binding by cultured embryonic heart cells

Curvilinear Scatchard plots for insulin binding by cultured embryonic heart cells at 15 C were interpreted assuming a two-receptor model (Santora II, A. C., F.B. Wheeler, R.L. DeHaan, and L.J. Elsas II, Endocrinology 104:1059, 1979). The present studies test this hypothesis further. Both 125I-labeled hormone binding and 2-methylamino-[1-14C]isobutyric acid ([14C]meAIB) transport were assayed at 24 C in identical buffers. Equilibrium chicken insulin-binding experiments yielded curvilinear Scatchard functions, consistent with binding to both high affinity, low capacity and low affinity, high capacity receptor sites at 24 C. Insulin-stimulated transport of the model amino acid [14C]meAIB (a specific alanine-preferring system substrate) paralleled the expected occupancy of the low affinity receptors. In competitive binding assays, the abilities of multiplication-stimulating activity (MSA), proinsulin, and insulin to compete for primarily high affinity insulin binding were compared with their abilities to stimulate transport. The relative potencies for binding competition with 34 pM chicken [125I]insulin were insulin greater than proinsulin greater than or equal to MSA (1:0,05:0.03). In contrast, the relative potencies for stimulating [14C]meAIB transport were MSA greater than insulin greater than proinsulin (3:1:0.28). Maximal stimulation of transport by insulin was not additive to that by MSA or proinsulin. The relative potency profile for binding competition with [125I]MSA was the same as that for stimulation of transport: MSA greater than insulin greater than proinsulin (2-3:1:0.2-0.3). Proinsulin, added in excess of insulin at a constant molar ratio, reduced low affinity insulin binding to a greater extent than it reduced high affinity insulin binding. Moreover, the extent of binding to each receptor was predicted from, and thus directly supported, our biological potency data, in which proinsulin was nearly one third as potent as insulin at the low affinity receptor yet possessed only 5% of insulin's potency at the high affinity receptor. The different relative affinity profiles suggest that these two receptors had binding site specificities for different portions of the insulin molecule. Each experiment with proinsulin and insulin in a constant molar ratio was carried out in parallel with an experiment in which insulin was the only unlabeled hormone. Analysis of the results from weighted nonlinear least squares regression fits of data to Michaelis-Menten equations showed that a two-receptor model was necessary to fit the data obtained with insulin as the only unlabeled hormone. Low affinity site binding parameters were also needed when data within each parallel experiment were pooled. The following kinetic experiments indicated the absence of enhanced dissociation of bound [125I]insulin by unlabeled insulin...