In this study, the relationship between the concentration of extracellular insulin, insulin binding and insulin action was evaluated in skeletal muscle. Initially we investigated the dose-response relationship of insulin action using three different experimental models that are responsive to insulin, i.e. the isolated perfused rat hindquarter, incubated strips of soleus muscle, and insulin receptors partially affinity-purified from skeletal muscle. We selected as insulin-sensitive parameters glucose uptake in the perfused hindquarter, lactate production in the incubated muscle preparation, and tyrosine receptor kinase activity in the purified receptor preparation. Our results showed that the dose-response curves obtained in the perfused hindquarter and in the incubated muscle were superimposable. In contrast, the dose-response curve for insulin-stimulated receptor tyrosine kinase activity in partially purified receptors was displaced to the left compared with the curves obtained in the perfused hindquarter and in the incubated muscle. The differences between the dose-response curve for receptor tyrosine kinase and those for glucose uptake and lactate production were not explained by a substantial insulin concentration gradient between medium and interstitial space. Thus the medium/interstitial insulin concentration ratio, when assayed in the incubated intact muscle at 5 degrees C, was close to 1. We also compared the dose-response curve of insulin-stimulated receptor tyrosine kinase with the pattern of insulin-binding-site occupancy. The curve of insulin-stimulated receptor kinase activity fitted closely with the occupancy of high-affinity binding sites. In summary, assuming that the estimation of the medium/interstitial insulin concentration ratio obtained at 5 degrees C reflects the actual ratio under more physiological conditions, our results suggest that maximal insulin action is obtained in skeletal muscle at insulin concentrations which do allow full occupancy of high-affinity binding sites. Therefore our data provide evidence for a lack of spare high-affinity insulin receptors in skeletal muscle.