An indirect method for obtaining a reliable measure of the rate of glucose transport into adipocytes is described. Evidence is presented that altered levels of 3':5'-cyclic AMP can influence the transport of glucose into adipocytes. When cyclic AMP levels were lowered with antilipolytic agents (insulin, nicotinic acid, or clofibrate), rates of glucose transport were increased. In contrast, when adipose tissue levels of cyclic AMP were elevated by lipolytic hormones or theophylline, glucose transport was inhibited. Insulin no longer increased glucose transport when cyclic AMP levels were elevated by lipolytic agents. Agents that can raise cyclic AMP but inhibit lipolysis (procaine, amitryptyline, and phenylethylbiguanide) reduced the rate of glucose transport. Other data are presented that are consistent with the conclusion that cyclic AMP inhibits glucose transport into adipocytes.The relatively small intracellular water volume in adipose tissue makes it technically difficult to measure rates of glucose transport into adipocytes (1). This rate is usually determined by measuring the volume of distribution of added glucose or derivatives in cells rapidly separated from the incubation medium or by measuring countertransport from prelabeled adipocytes (1-9). However, these methods suffer from the disadvantage that incubation is required at reduced temperature or for very short time intervals. The magnitude of the response of the glucose transport system in adipocytes to insulin as observable by previously used methods is small compared to the observed increases in rates of glucose metabolism. Consequently, a considerable portion of information relating to glucose transport in adipose tissue has been inferred from indirect techniques in which glucose utilization has been measured (10-12). Even though these indirect techniques do not clearly dissociate effects on glucose transport from effects on subsequent glucose metabolism, it is generally accepted that insulin stimulates glucose transport in adipose tissue (for review, see ref. 13).Addition of insulin to adipose tissue results in a fall in levels of 3':5'-cyclic AMP (cAMP) (for review, see refs. 13 and 14). Evidence is lacking, however, that the insulin-induced decrease in cAMP is directly associated with increased glucose transport. The apparent absence of an influence of cAMP on glucose transport is incompatible with the possibility of a unitary concept that would link the major metabolic effects of insulin to its actions on intracellular cAMP levels (15).In this report, we first describe a method to determine the rate of glucose transport in adipocytes. This technique offers several advantages over those previously described in that incubation is permitted at normal temperatures for longer periods. Using this method, we then report observations demonstrating an inverse correlation between adipocyte cAMP levels and the rate of glucose transport. On the basis of these data, we discuss the hypothesis that the major actions of insulin in adipose tissue may ...