Bacteriorhodopsin photoreaction: identification of a long-lived intermediate N (P, R350) at high pH and its M-like photoproduct

Abstract
An alkaline suspension of light-adapted purple membrane exposed to continuous light showed a large absorption depletion at 580 nm and a small increase around 350 nm. We attribute this absorption change to an efficient photoconversion of bR570 into a photoproduct N (P, R350), which has a major absorption maximum between 550 and 560 nm but has lower absorbance than bR570. N was barely detectable at low pH, low ionic strength, and physiological temperature. However, when the thermal relaxation of N to bR570 was inhibited by increasing pH, increasing ionic strength, and decreasing temperature, its relaxation time could be as long as 10 s at room temperature. N is also photoactive; when it is present in significant concentrations, e.g., accumulated by background light, the flash-induced absorption changes of purple membrane suspensions were affected. Double-excitation experiments showed an M-like photoproduct of N, NM, with an absorption maximum near 410 nm and a much longer lifetime than M412. It may be in equilibrium with an L-like precursor NL. We suggest that N occurs after M412 in the photoreaction cycle and that its photoproduct NM decays into bR570. Thus, at high pH and high light intensity, the overall photoreaction of bR may be approximated by the two-photon cycle bR570 .fwdarw. M412 .fwdarw. N .fwdarw. (NL .tautm. NM) .fwdarw. bR570, whereas at neutral pH and low light intensity it can be described by the one-photon cycle bR570 .fwdarw. M412 .fwdarw. N .fwdarw. O640 .fwdarw. bR570. The result of light-induced pH changes in purple membrane suspensions suggested that one proton is taken up from the medium during the thermal relaxation N .fwdarw. bR570 or the light reaction N .fwdarw. bR570 (not during the reaction M412 .fwdarw. N). At high pH and high ionic strength, a small amount of N appears to also be present in the dark, which implies that a thermal backreaction from bR570 to N also exists. The proposed modification of the photoreaction cycle model will require confirmation and possibly corrections by other techniques, e.g., vibrational spectroscopy. However, as it stands, it offers a satisfactory explanation of a variety of earlier observations which are inconsistent with the simple, generally used model bR .fwdarw. K .fwdarw. L .fwdarw. M .fwdarw. O .fwdarw. bR.