Time-resolved, flash-induced difference absorbance spectra (300-700 am) at pH 10.5 and 5C for the bacteriorhodopsin photocycle fast and slow decaying forms of the M intermediate (M' and MS, respectively) MATERIALS AND METHODS Purple membranes were isolated from Halobacterium halobium, S9 strain, and suspended in 30 mM piperazine/30 mM glycylglycine, pH 10.5. All samples contained 40%o (vol/vol) glycerol except those used for the flash-induced absorbance changes in Fig. 4. All measurements were carried out on light-adapted BR.Two spectrophotometers were used to obtain the lightinduced difference spectra. For the faster measurements, the data were taken and analyzed as described (10,11). For the slower measurements a Hewlett-Packard model 8452A diode array spectrophotometer was used to take complete lightinduced difference spectra from 300 to 700 nm with 2-nm spectral resolution. (The absorption spectrum of the lightadapted sample was used as the baseline.) The actinic source was a commercial photoflash with Coming long-pass filter (CS 3-67, wavelength >550 nm). The half-duration time was 150 tUs. Since the measuring beam of the diode array spectrophotometer was relatively strong white light, special care was taken to check and avoid a possible actinic effect of the measuring beam. Sample OD570 was between 1.3 and 1.5. RESULTSSeparation of the Difference Spectra of Intermediates M', ME, and R. From the time-dependent series of difference spectra, we can determine the difference spectra of the late intermediates involved in the BR recovery at high pH (Mf, MS, and R). We have chosen conditions for which the differences in the lifetimes and yields of these three components are maximal. We assume that the intermediates before M can be neglected because of their much shorter lifetimes and that no significant amount of the late intermediate 0 is formed because of the low temperatures and high pH values (refs. 5, 6, and 10 and unpublished data). Fig. 1 6358The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

The permanent dipole moment, polarizability, and the retinal angle of Halobacterium halobium purple membranes were determined at different pH values. All of the parameters have a maximum between pH 5 and 6. There is a reversal in the direction of the permanent dipole moment near pH 5. The value of permanent dipole moment was determined to be 60 D/protein at pH 6.6, and the value obtained for polarizability was 3 X 10(-28) Fm2/membrane fragment. The retinal angle of all-trans retinal was 0.8 degrees smaller than that of the 13-cis conformation.

Kinetic curves for the bacteriorhodopsin (BR) photocycle were obtained both at 570 and at 412 nm at a series of increasing levels of intensity of the exciting laser. Singular value decomposition (SVD) of these curves showed two transitions in the kinetic profiles that occurred at specific levels of actinic light. This means that the photocycle was influenced by photon density in two ways. In a separate application of SVD, time-resolved optical spectra were analyzed at each of many levels of exciting laser intensities. The studies showed that the transition at the low level of laser intensity was due principally to an increase in the amount of BR that was turning over. The transition at the higher level of laser intensity showed a fundamental change in kinetics of the photocycle. At low intensity levels, the fast form of M (Mf) predominated, whereas at high levels the slow form of M (Ms) predominated. A distinction was found between Mf and Ms, in that the former decayed directly to the O intermediate whereas the latter decayed directly to BR.

The cell membrane of Halobacterium halobium (H. halobium) contains the proton-pump bacteriorhodopsin, which generates a light-driven transmembrane protonmotive force. The interaction of the bacteriorhodopsin photocycle with the electric potential component of the protonmotive force has been investigated. H. halobium cell envelope vesicles have been prepared by sonication and further purified by ultracentrifugation on Ficoll/NaCl/CsCl density gradients. Under continuous illumination (550 +/- 50 nm) varied from 0 to 40 mW cm-2, the vesicles maintain a membrane potential of 0 to -100 mV. The membrane potential was measured by flow dialysis of 3H-TPMP+ uptake and could be abolished by the uncoupler carbonylcyanide-m-chlorophenylhydrazone. Time-resolved absorption spectroscopy was used to measure the decay kinetics of the M photocycle intermediate, which was initiated by a weak laser flash (588 nm), while the vesicles were continuously illuminated as above. The M decay kinetics were fitted with two exponential decays by a computer deconvolution program. The faster decaying form decreases in amplitude (70 to 10% of the total) and the slower decaying form increases in amplitude and lifetime (23 to 42 ms) as the background light intensity increases. Although any correlation between the membrane potential and the bacteriorhodopsin photocycle M-forms is complex, the present data will allow specific tests of the physical mechanism for this interaction to be designed and conducted.

A new two step photobleaching process is observed under continuous illumination of bacteriorhodopsin. This photobleaching is considerable even at physiological temperatures and becomes large at 50^60³C. The photobleaching also increases with increasing pH from 7 to 10. We suggest that the bleaching at its final stage could be due to the dissociation of the retinal and a local thermal denaturation-like process. These facts may question the generally held belief that BR is a stable protein in vivo for a long period of time. Our results may have relevance also to practical applications of bacteriorhodopsin where the stability of bacteriorhodopsin is a key issue. In certain instances, the use of bacteriorhodopsin may require cooled conditions. Here, we defined the conditions under which bacteriorhodopsin is stable. The permanent photobleaching offers a new way of picture imaging and information input for bacteriorhodopsinbased optical devices.z 1999 Federation of European Biochemical Societies.