1. The calcium binding capacity (kappa(S)) of bovine chromaffin cells preloaded with fura-2 was measured during nystatin-perforated-patch recordings. 2. Subsequently, the perforated patch was ruptured to obtain a whole-cell recording situation, and the time course of kappa(S) was monitored during periods of up to one hour. 3. No rapid change (within 10-20 s) of kappa(S) was observed upon transition to whole-cell recording, as would be expected, if highly mobile organic anions contributed significantly to calcium buffering. However, approximately half of the cells investigated displayed a drop in kappa(S) within 2-5 min, indicative of the loss of soluble Ca2+ binding proteins in the range of 7-20 kDa. 4. The average Ca2+ binding capacity (differential ratio of bound calcium over free calcium) was 9+/-7 (mean+/-S.E.M.) for the poorly mobile component and 31+/-10 for the fixed component. It was concluded that a contribution of 7 from highly mobile buffer would have been detected, if present. Thus, this value can be considered as an upper bound to highly mobile Ca2+ buffer. 5. Both mobile and fixed calcium binding capacity appeared to have relatively low Ca2+ affinity, since kappa(S) did not change in the range of Ca2+ concentrations between 0.1 and 3 muM. 6. It was found that cellular autofluorescence and contributions to fluorescence of non-hydrolysed or compartmentalized dye contribute a serious error in estimation of kappa(S). 'Balanced loading', a degree of fura-2 loading such that the calcium binding capacity of fura-2 equals cellular calcium binding capacity, minimizes these errors. Also, changes in kappa(S) at the transition from perforated-patch to whole-cell recording can be most faithfully recorded for similar degrees of loading in both situations. 7. Nystatin was found unable to make pores from inside of the plasma membrane of chromaffin cells. With careful preparation and storage the diluted nystatin solution maintained its high activity of membrane perforation for more than one week. 8. An equation for the effective diffusion constant for total cytoplasmic calcium, D(Ca)', was derived, which takes into account fixed buffer and poorly mobile buffer as determined, as well as calcium bound to fura-2 and some highly mobile buffers. 9. It was concluded that even for fura-2 concentrations as small as 50 muM, the dye and possibly some immeasurable contribution of highly mobile buffer dominate the diffusion process, such that D' varies between approximately 10(-7) cm2 s-1 (in the absence of fura-2 and of highly mobile buffer) and 4 x 10(-7) cm2 s-1 in the presence of fura-2 and/or highly mobile buffer.
