Dependence of the activity of colicin E1 in artificial membrane vesicles on pH, membrane potential, and vesicle size.

The dependence on pH and membrane potential of the ability of colicin E1 and a COOH-terminal tryptic fragment of the colicin to form membrane channels has been measured using a chloride-sensitive electrode to measure colicin-induced ion efflux from asolectin vesicles of two different size classes. This method allows measurement of ion efflux on a faster time scale, with half-times for efflux less than or equal to 3 s, than previously possible using labeled solutes. Activity measurements were also made through the use of potential-indicating fluorescence probes. The activities of both colicin E1 and the fragment increased with decreasing pH. The activity of the colicin was maximum at pH values near 4.0, with an apparent pK of 4.5-4.6, whereas that of the COOH-terminal fragment continued to increase to the lowest pH value, 3.4, that could be used, showing an apparent pK less than or equal to 3.8. Using relatively small vesicles (average diameter approximately equal to 0.1 micron) made by a freeze-thaw procedure, chloride efflux caused by addition of fragment or colicin was independent of the initial transmembrane K+-diffusion potential imposed upon the system. However, with larger (0.5-micron diameter) vesicles prepared by a fusion method, the chloride efflux showed a dependence upon membrane potential, with the activity decreasing as the membrane potential was made more positive. The average size of the different vesicle populations was determined by electron microscopy. It is proposed that the lack of potential dependence observed in the freeze-thaw vesicles and the small voltage dependence, relative to planar membranes, seen in the larger fused vesicles, results from rapid discharge of the membrane potential and internal ion content of the vesicles.