The insulin receptor in insulin-dependent diabetes mellitus: An in vivo and in vitro study

Abstract

Insulin binding to circulating monocytes was studied in eight nonobese patients with insulin-dependent diabetes mellitus (IDDM) in poor metabolic control, along with eleven normal subjects. Insulin secretory responses to glucose and tolbutamide were studied before treatment in the three newly diagnosed patients. Six of the patients underwent repeat insulin binding studies after improvement in metabolic control. Four metabolic abnormalities commonly seen in IDDM, hyperglycemia, ketonemia, acidemia, and hyperosmolarity, were simulated in vitro using cultured human lymphocytes (IM-9 line), and their effects on insulin binding examined. Insulin binding in poorly controlled insulin-dependent diabetics is heterogeneous. Mean tracer binding is slightly above the mean of normal subjects. The three newly diagnosed patients have minimal insulin reserve. They cannot be distinguished from previously treated diabetic subjects with poor metabolic control on the basis of insulin binding. There is no correlation of insulin binding, or R 0 (receptors/cell), or K e (affinity of the empty receptor) in the poorly controlled state with degree of ketonuria, fasting glucose concentrations, fasting growth hormone concentrations, or 8 a.m. cortisol concentrations. After treatment with insulin (5–78 days) mean tracer insulin binding fell significantly (9.05% ± 1.07% to 5.8% ± 1.08%, p < 0.05), associated with decreases of both R 0 and affinity. Insulin binding to cultured human lymphocytes was not altered by preincubation of cells for 18 hr at 37°C with varying concentrations of glucose (100–800 mg/dl), varying pH (6.6–7.6), or 40 mm β-OH butyric acid. There was a small but significant inverse correlation between tracer insulin binding and preincubation osmolality ( r = −0.42, p < 0.05). We conclude that (1) insulin binding in poorly controlled IDDM is normal or somewhat elevated similar to animal models of IDDM and distinctly different from non-IDDM, and (2) insulin binding decreases with insulin treatment and improvement in metabolic control, associated with both R 0 and affinity changes. These studies suggest that hyperglycemia, ketonemia, acidemia, and hyperosmolarity do not account for the changes in insulin binding observed in these patients. We have clearly demonstrated two types of regulation of the insulin receptor; type one requires the presence of the regulator in the binding assay, and is readily reversible so that the receptor has no “memory” for the effect, and type two which is readily demonstrable in the in vitro binding assay and for which the receptor has “memory.”