Human red blood cells pretreated with low-ionic-strength solutions and resuspended in saline respond biphasically to extracellular Ca. At first, addition of Ca causes a large transient K efflux of as much as 600 mM . liter cell H2O-1 . h-1; this is followed by a decrease in K flux below control levels. The first phase (phase I) resembles the Gardos effect in several respects. It is inhibited by oligomycin, by external K, and by increased exposure time to Ca. Further, the K permeability of phase I is similar to that of the Gardos effect (5 X 10(-8)-9 X 10(-8) cm/s), and the cells hyperpolarize in a low-K medium when Ca2+ is added. However, phase I is not identical to the Gardos phenomenon. For example, La, which prevents the Gardos response, is ineffective on phase I. Moreover, external Ba prevents the development of phase I but not the Gardos response, whereas internal Ba prevents the Gardos response. Attempts to demonstrate a Ca leak or pump failure during phase I have failed; passive Ca movements of both treated and normal cells are similar. The results suggest that low-ionic-strength solution exposes Ca-sensitive sites to the external medium; these sites are maintained when the cells are returned to saline.
