The extent of gap junctions and dye coupling between insulin-producing B-cells was analyzed on islets of Langerhans isolated from adult rats treated for one day with glibenclamide, an insulin secretagogue, or diazoxide, a blocker of insulin release, or a combination of the two drugs. Glibenclamide treatment was associated with a marked depletion of the islet insulin content, an effect which was blocked by pretreatment of the rats with diazoxide. Diazoxide alone caused a marked increase in the plasma glucose level, and a decrease in the level of circulating insulin and in the hormone content of the B-cells. Quantitative analysis showed that (1) under control conditions, B-cells are connected by minute gap junctions (as evaluated on freeze-fracture replicas) and show a nonuniform and apparently restricted dye coupling (as determined by microinjection of the low-molecular-weight fluorescent probe Lucifer Yellow CH); (2) each of the 3 treatments tested significantly increased the relative and absolute gap junction area of the B-cells and the number of detectable, dye coupled B-cells per microinjection. After treatment with glibenclamide alone or with diazoxide plus glibenclamide, a 1.5-1.8-fold increase in gap junction area and a 2.7-3.7-fold increase in the number of dye-coupled B-cells were observed. In contrast, following treatment with diazoxide alone, gap junctions and dye coupling were found increased 1.8 and 8.7 times, respectively, as compared with control values. The results suggest that small territories of coupled B-cells may exist within adult islets of Langerhans, and show that such a pattern can be modulated in vivo via changes in the gap junctions and dye coupling of B-cells. The mechanism(s) regulating these changes is unknown but is clearly operative after both stimulation and inhibition of insulin release.