Membrane potential in a potassium transport-negative mutant of Escherichia coli K-12. The distribution of rubidium in the presence of valinomycin indicates a higher potential than that of the tetraphenylphosphonium cation.

The membrane potential across the cytoplasmic membrane of EDTA-treated cells of a K+ transport-negative mutant of Escherichia coli K-12 was estimated from the equilibrium distribution of different lipid-soluble cations. With glucose as a substrate and at low K+ out, the membrane potential calculated from the distribution ratio of 86Rb+ in the presence of valinomycin (delta psi rb+) was considerably higher than that indicated by the [3H]tetraphenylphosphonium cation (delta psi TPP+). The lipid-soluble anion phenyldicarbaundecaborane (PCB-) increased delta psi TPP+ close to delta psi Rb+. To investigate whether these results were due to different binding of the cations to cellular components, residual Rb+ and TPP+ uptake was measured in cells permeabilized with 5% n-butanol (by volume). In those cells the distribution ratios or Rb+, K+ and Na+ approached a value of 4, indicating that the uptake of all three ions was driven by a residual negative surface potential or transmembrane Donnan potential (internally negative). The distribution ratio of TPP+ was 3--4-times higher than that of other cations and up to 10 mM TPP+ out was almost independent of the added TPP+ concentration. This extra uptake presumably represents binding of TPP+ to the cellular membranes. Thus, at pH 7.5, delta psi Rb+ was about 180--200 mV, whereas after correction for binding delta psi TPP+ was 110--150 and 150--170 mV in the absence and presence of PCB-, respectively. It is proposed that TPP+ indicates too low a potential, because by its strong binding it decreases the negative surface potential of the cytoplasmic membrane, and thereby inhibits its own further uptake. This is taken to mean that TPP+ distribution can be used as a qualitative probe only for the bacterial membrane potential.