ANALYSIS OF THE DC PHOTOELECTRIC SIGNAL FROM MODEL BACTERIORHODOPSIN MEMBRANES - DC PHOTOCONDUCTIVITY DETERMINATION BY THE NULL CURRENT METHOD AND THE EFFECT OF PROTON IONOPHORES

On illumination by continuous light, a reconstituted bacteriorhodopsin membrane exhibits a stationary photocurrent under short-circuit conditions. It has been widely reported that this photocurrent is linearly dependent on the applied transmembrane potential, and that the photocurrent reverses its polarity at a critical potential. It is also well known that the stationary photosignal of a bacteriorhodopsin membrane is linearly dependent on the light intensity and eventually reaches saturation. In this paper, the null current method (F.T. Hong and D. Mauzerall, Biochim. Biophys. Acta, 275 (1972) 479) is applied to decompose the photocurrent into a photovoltaic part (photoemf) and a photoconductive part (photoconductance). It is found that the photoconductance is zero in the dark, and is activated by illumination to reach a fixed magnitude which is independent of a further increase in the light intensity (''step-function'' photoswitching). Furthermore, the photoconductance is ohmic (i.e. independent of the applied potential). The linear voltage dependence of the photocurrent can be explained in terms of the photoswitching by assuming that the light-activated proton conductance channel is also available for a transmembrane potential to drive a proton current through in either direction. With this assumption, the photoemf is shown to be voltage independent. The photoemf is initially linearly light dependent at low light intensities, but eventually reaches a saturation level. We confirm the reported enhancement effect of the proton ionophores carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP) and carbonyl cyanide-m-chlorophenylhydrazone (CCCP), which is caused by increases in both the photoconductance and the ionic conductance. The action of the Cl- ionophore, nystatin, is quite different. Nystatin inhibits the photocurrent and increases the ionic conductance, but does not affect the photoconductance. The enhancement effect of proton ionophores cannot be explained by the shunting effect alone, even if the sandwich model postulated by Bamberg et al. (Biophys. Struct. Mech., 5 (1979) 277) is invoked. We suspect that the incorporation of bacteriorhodopsin into the artificial black lipid membrane may be more complete than initially believed.