Dynamic Control of IKs Current Amplitude by the ‘Late-Assembly’ Strategy of Channel Subunit Association

Background: Ion channels composed of α (pore-forming) and β (auxiliary) subunits use an ‘early-assembly' strategy (Kir6.x/SUR assembly in ER) to control cell surface expression, or a ‘late-assembly' strategy (α/β2 of rat brain Nav channels, both subunits independently traffic to cell surface) to allow dynamic control of current amplitude/gating kinetics. Slow delayed rectifier (IKs) channel is composed of KCNQ1 (Q1, α) and KCNE1 (E1, β) subunits, and functions as ‘repolarization-reserve' in human heart. It is not clear which of the 2 assembly strategies Q1 and E1 use in forming IKs. Methods: We express Q1 and E1 tagged with fluorescent-protein (Q1-GFP and E1-dsR) or extracellular epitope (AU5-Q1 and HA-E1) in COS-7 and neonatal rat ventricular myocytes (NRVM), and use confocal imaging to track Q1 & E1 movements in cells. Results: In COS-7 cells 3 hr after transfection, Q1 and E1 travel in separate transport intermediates without mixing/fusion. E1 reaches the cell surface before Q1. By 24 hr, Q1 and E1 are colocalized on cell surface. Brefeldin-A (blocking protein export from ER) prevents surface expression of Q1-147C/E1-40C and reduces disulfide formation between the two (as a measure of functional Q1/E1 assembly). In NRVM during in vitro development, Q1-GFP and E1-dsR travel in distinctly different transport intermediates. As NRVM develops into mature-like phenotype, Q1-GFP and E1-dsR are colocalized on the cell surface, with a separate cytosolic Q1-GFP pool colocalized with α-actinin and calnexin (z-line and ER/SR markers, respectively). Conclusions: Q1 and E1 use the ‘late-assembly' strategy to afford dynamic control of IKs current amplitude. This explains why native Q1 and E1 in adult ventricular myocytes are often not well-colocalized. We propose that cardiac myocytes regulate IKs amplitude dynamically by adjusting the degree of Q1/E1 colocalization on cell surface.