Regulation of the number of Ca(2+)-activated K(+) channels at the endothelial cell surface contributes to control of the endothelium-derived hyperpolarizing factor response, although this process is poorly understood. To address the fate of plasma membrane-localized KCa2.3, we utilized an extracellular epitope-tagged channel in combination with fluorescence and biotinylation techniques in both human embryonic kidney cells and the human microvascular endothelial cell line, HMEC-1. KCa2.3 was internalized from the plasma membrane and degraded with a time constant of 18 h. Cell surface biotinylation demonstrated that KCa2.3 was rapidly endocytosed and recycled back to the plasma membrane. Consistent with recycling... More
Regulation of the number of Ca(2+)-activated K(+) channels at the endothelial cell surface contributes to control of the endothelium-derived hyperpolarizing factor response, although this process is poorly understood. To address the fate of plasma membrane-localized KCa2.3, we utilized an extracellular epitope-tagged channel in combination with fluorescence and biotinylation techniques in both human embryonic kidney cells and the human microvascular endothelial cell line, HMEC-1. KCa2.3 was internalized from the plasma membrane and degraded with a time constant of 18 h. Cell surface biotinylation demonstrated that KCa2.3 was rapidly endocytosed and recycled back to the plasma membrane. Consistent with recycling, expression of a dominant negative (DN) RME-1 or Rab35 as well as wild type EPI64C, the Rab35 GTPase-activating protein, resulted in accumulation of KCa2.3 in an intracellular compartment. Expression of DN RME-1, DN Rab35, or wild type EPI64C resulted in a decrease in steady-state plasma membrane expression. Knockdown of EPI64C increased cell surface expression of KCa2.3. Furthermore, the effect of EPI64C was dependent upon its GTPase-activating proteins activity. Co-immunoprecipitation studies confirmed an association between KCa2.3 and both Rab35 and RME-1. In contrast to KCa2.3, KCa3.1 was rapidly endocytosed and degraded in an RME-1 and Rab35-independent manner. A series of N-terminal deletions identified a 12-amino acid region, Gly(206)-Pro(217), as being required for the rapid recycling of KCa2.3. Deletion of Gly(206)-Pro(217) had no effect on the association of KCa2.3 with Rab35 but significantly decreased the association with RME-1. These represent the first studies elucidating the mechanisms by which KCa2.3 is maintained at the plasma membrane.