Since the beginning of the seventies, studies of the cellular mechanisms behind insulin resistance in man have included studies of insulin receptor binding and insulin action in isolated cells. In the first studies, only measurements of insulin binding to circulating blood cells (mononuclear cells and erythrocytes) were possible. In these studies it was thus necessary to anticipate that insulin binding to these cells was representative for binding to target cells for insulin (adipocytes, hepatocytes, muscle cells). Later, studies of the human adipocyte became available. In the isolated human adipocyte it was possible to measure both insulin binding and the action of insulin on glucose transport and on the intracellular glucose processing. Immediately, it was observed that receptor binding to the different cell types was not always comparable. Moreover, the relationship between fat cell insulin binding and action was not always straightforward. Because fat tissue is only responsible for a small fraction of total glucose uptake, it is not possible to know whether changes in insulin binding and action in this tissue is representative for changes in the total organism. In the present review these problems have been elucidated by studies of patients with insulin-dependent and non-insulin-dependent diabetes mellitus. In chapter one, the methods used in the clinical studies are reviewed. The precision (intraassay variability) and reproducibility (intraperson variability) has been measured for all insulin receptor assays. It was found that the earlier used assay for mononuclear cells was improved by using a pure monocyte assay, because precision as well as reproducibility was improved. On the other hand, these values were considerably poorer than those found for the other cell types. The precision was 0.09, 0.04, and 0.04 for monocytes, erythrocytes and adipocytes, respectively. The reproducibility was 0.19, 0.06 and 0.11. In order to be able to measure comparability between insulin binding to the above mentioned cell types and hepatocytes, methods for measurement of insulin binding to these cell types from swine have been developed. These studies showed that insulin binding to swine cells have many similarities to that of human cells whereas several dissimilarities were seen between insulin binding to rat and human cells. Thus, it is surmised that swine cells are more suitable than rat cells concerning insulin receptor binding and action studies.(ABSTRACT TRUNCATED AT 400 WORDS)