Migration plays an important role in the formation of tumor metastases. Nonetheless, little is known about electrophysiological phenomena accompanying or underlying migration. Previously, we had shown that in migrating alkali-transformed Madin-Darby canine kidney focus (MDCK-F) cells a Ca+-sensitive 53-pS K+ channel underlies oscillations of the cell membrane potential. The present study defines the role this channel plays in migration of MDCK-F cells. We monitored migration of individual MDCK-F cells by video imaging techniques. Under control conditions, MDCK-F cells migrated at a rate of 0.90±0.03 Mim/min (n = 201). Application of K+ channel blockers ( 1 and 5 mmol / liter Ba2+, 5 mmol /liter tetraethylammonium, 100 ,mol/liter 4-aminopyridine, 5 nmol/liter charybdotoxin) caused marked inhibition of migration, pointing to the importance of K+ channels in migration. Using patchclamp techniques, we demonstrated the sensitivity of the Ca2+-sensitive 53-pS K+ channel to these blockers. Blockade of this K+ channel and inhibition of migration were closely correlated, indicating the necessity of oscillating K+ channel activity for migration. Migration of MDCK-F cells was also inhibited by furosemide or bumetanide, blockers of the Na+/K+/2Cl-cotransporter. We present a model for migration in which oscillations of cell volume play a central role. Whenever they are impaired, migration is inhibited. (J. Clin. Invest. 1994.

Epithelial cells lose their usual polarization during carcinogenesis. Although most malignant tumours are of epithelial origin little is known about ion channels in carcinoma cells. Previously, we observed that migration of transformed Madin-Darby canine kidney (MDCK-F) cells depended on oscillating K+ channel activity. In the present study we examined whether periodic K+ channel activity may cause changes of cell volume, and whether K+ channel activity is distributed in a uniform way in MDCK-F cells. After determining the average volume of MDCK-F cells (2013+/-270 microm3; n=8) by means of atomic force microscopy we deduced volume changes by calculating the K+ efflux during bursts of K+ channel activity. Therefore, we measured the membrane conductance of MDCK-F cells which periodically rose by 22.3+/-2.5 nS from a resting level of 6.5+/-1.4 nS (n=12), and we measured the membrane potential which hyperpolarized in parallel from -35.4+/-1.2 mV to -71.6+/-1.8 mV (n=11). The distribution of K+ channel activity was assessed by locally superfusing the front or rear end of migrating MDCK-F cells with the K+ channel blocker charybdotoxin (CTX). Only exposure of the rear end to CTX inhibited migration providing evidence for "horizontal" polarization of K+ channel activity in transformed MDCK-F cells. This is in contrast to the "vertical" polarization in parent MDCK cells. We propose that the asymmetrical distribution of K+ channel activity is a prerequisite for migration of MDCK-F cells.