Formation of 9-cis- and 11-cis-retinal pigments from bacteriorhodopsin by irradiating purple membrane in acid

Unusual Photoisomerization Pathway in a Near-Infrared Light Absorbing Enzymerhodopsin

Microbial and animal rhodopsins possess retinal chromophores which capture light and normally photoisomerize from all-trans to 13-cis and from 11-cis to all-trans-retinal, respectively. Here, we show that a near-infrared light-absorbing enzymerhodopsin from Obelidium mucronatum (OmNeoR) contains the all-trans form in the dark but isomerizes into the 7-cis form upon illumination. The photoproduct (λ_max = 372 nm; P₃₇₂) possesses a deprotonated Schiff base, and the system exhibits a bistable nature. The photochemistry of OmNeoR was arrested at <270 K, indicating the presence of a potential barrier in the excited state. Formation of P₃₇₂ is accompanied by protonation changes of protonated carboxylic acids and peptide backbone changes of an α-helix. Photoisomerization from the all-trans to 7-cis retinal conformation rarely occurs in any solvent and protein environments; thus, the present study reports on a novel photochemistry mediated by a microbial rhodopsin, leading from the all-trans to 7-cis form selectively.

QuasAr Odyssey: the origin of fluorescence and its voltage sensitivity in microbial rhodopsins

Rhodopsins had long been considered non-fluorescent until a peculiar voltage-sensitive fluorescence was reported for archaerhodopsin-3 (Arch3) derivatives. These proteins named QuasArs have been used for imaging membrane voltage changes in cell cultures and small animals. However due to the low fluorescence intensity, these constructs require use of much higher light intensity than other optogenetic tools. To develop the next generation of sensors, it is indispensable to first understand the molecular basis of the fluorescence and its modulation by the membrane voltage. Based on spectroscopic studies of fluorescent Arch3 derivatives, we propose a unique photo-reaction scheme with extended excited-state lifetimes and inefficient photoisomerization. Molecular dynamics simulations of Arch3, of the Arch3 fluorescent derivative Archon1, and of several its mutants have revealed different voltage-dependent changes of the hydrogen-bonding networks including the protonated retinal Schiff-base and adjacent residues. Experimental observations suggest that under negative voltage, these changes modulate retinal Schiff base deprotonation and promote a decrease in the populations of fluorescent species. Finally, we identified molecular constraints that further improve fluorescence quantum yield and voltage sensitivity.

The authors present an in-depth investigation of excited state dynamics and molecular mechanism of the voltage sensing in microbial rhodopsins. Using a combination of spectroscopic investigations and molecular dynamics simulations, the study proposes the voltage-modulated deprotonation of the chromophore as the key event in the voltage sensing. Thus, molecular constraints that may further improve the fluorescence quantum yield and the voltage sensitivity are presented.

Protein catalysis of the retinal subpicosecond photoisomerization in the primary process of bacteriorhodopsin photosynthesis

The rate of retinal photoisomerization in wild-type bacteriorhodopsin (wt bR) is compared with that in a number of mutants in which a positively charged (Arg(82)), a negatively charged (Asp(85) or Asp(212)), or neutral hydrogen bonding (Asp(115) or Tyr(185)) amino acid residue known to be functionally important within the retinal cavity is replaced by a neutral, non-hydrogen bonding one. Only the replacements of the charged residues reduced the photoisomerization rate of the 13-cis and all-trans isomers present in these mutants by factors of approximately 1/4 and approximately 1/20, respectively. Retinal photo- and thermal isomerization catalysis and selectivity in wt bR by charged residues is discussed in terms of the known protein structure, the valence-bond wave functions of the ground and excited state of the retinal, and the electrostatic stabilization interactions within the retinal cavity.

The pink membrane: the stable photoproduct of deionized purple membrane

When cations are removed from the purple membrane of Halobacterium halobium it turns blue (lambda(max) = 603 nm); continuous irradiation with intense red light (lambda's >/= 630 nm) converts this deionized blue membrane into a pink membrane (lambda(max) approximately 491 nm). The rate and extent of the transformation from the blue to the pink membrane is facilitated by the removal of the last twenty COOH-terminal amino acids of bacteriorhodopsin. While the chromophore of the blue membrane is a 32:68 mixture of the 13-cis and all-trans isomers of retinal, the chromophore of the pink membrane is 9-cis rectinal. The quantum efficiency of the pink to blue membrane photoconversion is relatively high compared with that of the blue to pink membrane photoconversion. Proton release is observed when the pink membrane is converted to the blue form, and proton uptake occurs during the reverse transition. Unlike the blue membrane, the absorbance maximum of the pink membrane is only slightly affected by cation addition at low pH and ionic strength.

Combined kinetic and thermodynamic analysis of alpha-helical membrane protein unfolding

The analytical toolkit developed for investigations into water-soluble protein folding has yet to be applied in earnest to membrane proteins. A major problem is the difficulty in collecting kinetic data, which are crucial to understanding any reaction. Here, we combine kinetic and thermodynamic studies of the reversible unfolding of an alpha-helical membrane protein to provide a definitive value for the reaction free energy and a means to probe the transition state. Our analyses show that the major unfolding step in the SDS-induced denaturation of bacteriorhodopsin involves a reduction in alpha-helical structure and proceeds with a large free-energy change; both our equilibrium and kinetic measurements predict that the free energy of unfolding in the absence of denaturant is +20 kcal.mol(-1), with an associated m-value of 25 kcal.mol(-1). The rate of unfolding in the absence of denaturant, k(u)(H(2)O), is surprisingly very slow ( approximately 10(-15) s(-1)). The kinetics also give information on the transition state for this major unfolding step, with a value for beta (m(f)/[m(f) + m(u)]) of approximately 0.1, indicating that the transition state is close to the unfolded state. We thus present a basis for mapping the structural and energetic properties of membrane protein folding by mutagenesis and classical kinetics.

pH-dependent transitions in xanthorhodopsin

Xanthorhodopsin (XR), the light-driven proton pump of the halophilic eubacterium Salinibacter ruber, exhibits substantial homology to bacteriorhodopsin (BR) of archaea and proteorhodopsin (PR) of marine bacteria, but unlike them contains a light-harvesting carotenoid antenna, salinixanthin, as well as retinal. We report here the pH-dependent properties of XR. The pKa of the retinal Schiff base is as high as in BR, i.e. > or =12.4. Deprotonation of the Schiff base and the ensuing alkaline denaturation cause large changes in the absorption bands of the carotenoid antenna, which lose intensity and become broader, making the spectrum similar to that of salinixanthin not bound to XR. A small redshift of the retinal chromophore band and increase of its extinction, as well as the pH-dependent amplitude of the M intermediate indicate that in detergent-solubilized XR the pKa of the Schiff base counterion and proton acceptor is about 6 (compared to 2.6 in BR, and 7.5 in PR). The protonation of the counterion is accompanied by a small blueshift of the carotenoid absorption bands. The pigment is stable in the dark upon acidification to pH 2. At pH < 2 a transition to a blueshifted species absorbing around 440 nm occurs, accompanied by loss of resolution of the carotenoid absorption bands. At pH < 3 illumination of XR with continuous light causes accumulation of long-lived photoproduct(s) with an absorption maximum around 400 nm. The photocycle of XR was examined between pH 4 and 10 in solubilized samples. The pH dependence of recovery of the initial state slows at both acid and alkaline pH, with pKas of 6.0 and 9.3. The decrease in the rates with pKa 6.0 is apparently caused by protonation of the counterion and proton acceptor, and that at high pH reflects the pKa of the internal proton donor, Glu94, at the times in the photocycle when this group equilibrates with the bulk.

Closing in on bacteriorhodopsin: progress in understanding the molecule

Bacteriorhodopsin is the best understood ion transport protein and has become a paradigm for membrane proteins in general and transporters in particular. Models up to 2.5 A resolution of bacteriorhodopsin's structure have been published during the last three years and are basic for understanding its function. Thus one focus of this review is to summarize and to compare these models in detail. Another focus is to follow the protein through its catalytic cycle in summarizing more recent developments. We focus on literature published since 1995; a comprehensive series of reviews was published in 1995 (112).

Photochemistry in dried polymer films incorporating the deionized blue membrane form of bacteriorhodopsin

The preparation and photochemical properties of dried deionized blue membrane (dIbR600; lambdamax approximately 600 nm, epsilon approximately 54, 760 cm-1 M-1, f approximately 1.1) in polyvinyl alcohol films are studied. Reversible photoconversion from dIbR600 to the pink membrane (dIbR485; lambdamax approximately 485 nm) is shown to occur in these films under conditions of strong 647-nm laser irradiation. The pink membrane analog, dIbR485, has a molar extinction coefficient of approximately 39,000 cm-1 M-1 (f approximately 1.2). The ratio of pink --> blue and blue --> pink quantum efficiencies is 33 +/- 5. We observe an additional blue-shifted species (dIbR455, lambdamax approximately 455 nm) with a very low oscillator strength (f approximately 0.6, epsilon approximately 26,000 cm-1 M-1). This species is the product of fast thermal decay of dIbR485. Molecular modeling indicates that charge/charge and charge/dipole interactions introduced by the protonation of ASP85 are responsible for lowering the excited-state all-trans --> 9-cis barrier to approximately 6 kcal mol-1 while increasing the corresponding all-trans --> 13-cis barrier to approximately 4 kcal mol-1. Photochemical formation of both 9-cis and 13-cis photoproducts are now competitive, as is observed experimentally. We suggest that dIbR455 may be a 9-cis, 10-s-distorted species that partially divides the chromophore into two localized conjugated segments with a concomitant blue shift and decreased oscillator strength of the lambdamax absorption band.

Evidence that aspartate-85 has a higher pK(a) in all-trans than in 13-cisbacteriorhodopsin

Three experimental observations indicate that the pK(a) of the purple-to-blue transition (the pK(a) of Asp-85) is higher for all-trans-bR(1) than for 13-cis-bR. First, light adaptation of bacteriorhodopsin (bR) at pHs near the pK(a) of Asp-85 causes an increase in the fraction of the blue membrane present. This transformation is reversible in the dark. Second, the pK(a) of the purple-to-blue transition in the dark is lower than that in the light-adapted bR (pK(a)(DA) = 3.5, pK(a)(LA) = 3.8 in 10 microM K(2)SO(4)). Third, the equilibrium fractions of 13-cis and all-trans isomers are pH dependent; the fraction of all-trans-bR increases upon formation of the blue membrane. Based on the conclusion that thermal all-trans <=> 13-cis isomerization occurs in the blue membrane rather than in the purple, we have developed a simple model that accounts for all three observations. From the fit of experimental data we estimate that the pK(a) of Asp-85 in 13-cis-bR is 0.5 +/- 0.1 pK(a) unit less than the pK(a) of all-trans-bR. Thus in 10 microM K(2)SO(4), pK(a)(c) = 3.3, whereas pK(a)(t) = 3.8.

Effective photochromic nonlinearity of dried blue-membrane bacteriorhodopsin films

We report the effective nonlinearity for photochromic conversion in a blue-membrane bacteriorhodopsin film hosted in a dry polyvinyl alcohol matrix. The shift in absorption maximum on photoconversion in this film is larger than that of the same material in hydrated form, thus offering a larger modulation of the refractive index. The photoexcited index modulation is stable for several months, which provides for holographic data recording and long-term photochromic data storage. The effective index modulation is experimentally measured and is in good agreement with the theoretical predictions based on the Kramers-Kronig transformation.

Fluorescence spectra of bacteriorhodopsin and the intermediates O and Q at room temperature

An unequivocal answer is given to the question of why the reported fluorescence spectra of bacteriorhodopsin (bR568) have been different from one another. The inconsistency is shown to arise from the accumulation of the fluorescent intermediates O and Q (KN) by cw excitation light. Their fractions in the photo-stationary states depend on the excitation power and the suspension pH. We report the intermediate-free fluorescence spectrum of bR568 obtained with a weak excitation source (632.8 nm, 5.3 x 10(15)-1.9 x 10(16) photons cm-2.s-1) and a near-IR sensitive intensified photodiode array system. The fluorescence maxima of the spectra, F(lambda) and f(nu), are located at 755 +/- 10 nm and 12700 +/- 200 cm-1, respectively. The spectrum of O is identical to that of the deionized purple membrane bR605 (Fmax = 750 +/- 5 nm, fmax = 13,000 +/- 100 cm-1). Q (KN) exhibits a blue-shifted spectrum more than that of bR568 (Fmax < 720 nm, fmax > 13,400 cm-1).

Interaction of aspartate-85 with a water molecule and the protonated Schiff base in the L intermediate of bacteriorhodopsin: a Fourier-transform infrared spectroscopic study

Fourier-transform infrared spectra were recorded at 170 K before and after irradiating the Asp85-->Asn mutant of bacteriorhodopsin. The difference spectrum exhibits protein bands such as those due to the perturbations of Asp96 and Asp115 and the N-H stretching vibration of tryptophan, characteristic of the L minus all-trans-bacteriorhodopsin spectrum of the wild-type protein. However, some vibrational bands of the peptide backbone and the chromophore are different from L and more characteristic of N of the wild-type protein. Remarkably, the shift observed for the vibrational band due to an internal water molecule upon L formation [Maeda, Sasaki, Shichida, and Yoshizawa (1992) Biochemistry 31, 462-467] is absent. These changes in the spectrum of the mutant could originate from the destruction of a hydrogen-bonding system consisting of Asp85, the water molecule, and the Schiff base, upon replacement of Asp85 with asparagine. These observations constitute direct evidence for the interaction of water with Asp85 at the time when it is protonated by the Schiff base.

Photochemical conversion of the O-intermediate to 9-cis-retinal-containing products in bacteriorhodopsin films

The photochemical activity of the O-state was investigated in bacteriorhodopsin (BR) films containing wildtype BR at pH 6.5 in the presence of glycerol. The formation of a photoproduct of O with an absorption maximum at 490 nm and 9-cis-retinal configuration was found. This 490-nm product was named P and shows a slow thermal reaction into a compound with a maximal absorption at 380 nm which was named Q and contains free 9-cis-retinal in the proteins binding site. The photoproducts of O, i.e., P and Q, are very similar, or even identical, to those previously observed in blue membranes. Common to the O-state and blue membrane forms of bacteriorhodopsin is a protonated aspartic acid 85, and we suggest that it is the reduced negative charge around the Schiff base which is responsible for the 9-cis photoisomerization. The release of a proton from aspartic acid 85 is linked to the conversion of the O-state back to the initial state of BR. Therefore the conditions of low proton mobility in BR films containing glycerol favor the accumulation of the O-state. For optical and holographic applications such BR films are very attractive. It is possible to create photoproducts with red light which are thermally stable at room temperature and that can be photochemically erased. Dependent on the light composition both properties can be realized in the same sample material. This feature may bridge the gap between information processing and short-term and long-term storage of information with BR.

Bacteriorhodopsin: a biological material for information processing

Technology which makes use of biological materials has advanced dramatically in the last few decades. Production of specific biochemicals by selected microbial strains, the use of enzymes for stereospecific biosynthesis of materials and gene technological production of biologically important macromolecules are a few examples of these developments.

Effect of a light-induced pH gradient on purple-to-blue and purple-to-red transitions of bacteriorhodopsin

Bacteriorhodopsin-containing vesicles that were able to alkalize the extravesicular medium by greater than 1.5 pH units under illumination, i.e., inside-out vesicles, were reconstituted by reverse-phase evaporation with Halobacterium halobium polar lipids or exogenous phospholipids. Acid titration of a dark-adapted sample was accompanied by a color change from purple to blue (pKa = 2.5-4.5 in 0.15 M K2SO4), and alkali titration resulted in the formation of a red species absorbing maximally at 480 nm (pKa = 7 to greater than 9), the pKa values and the extents of these color changes being dependent on the nature of lipid. When a vesicle suspension at neutral or weakly acidic pH was irradiated by continuous light so that a large pH gradient was generated across the membrane, either a purple-to-blue or a purple-to-red transition took place. The light-induced purple-to-red transition was significant in an unbuffered vesicle suspension and correlated with the pH change in the extravesicular medium. The result suggests that the purple-to-red transition is driven from the extravesicular side, i.e., from the C-terminal membrane surface. In the presence of buffer molecules outside, the dominant color change induced in the light was the purple-to-blue transition, which seemed to be due to a large decrease in the intravesicular pH. But an apparently inconsistent result was obtained when the extravesicular medium was acidified by a HCl pulse, which was accompanied by a rapid color change to blue. We arrived at the following explanation: The two bR isomers, one containing all-trans-retinal and the other 13-cis-retinal, respond differently to pH changes in the extravesicular and the intravesicular medium. In this relation, full light adaptation was not achieved when the light-induced purple-to-blue transition was significant; i.e., only the 13-cis isomer is likely to respond to a pH change at the N-terminal membrane surface.

Photoreactions of bacteriorhodopsin at acid pH

It has been known that bacteriorhodopsin, the retinal protein in purple membrane which functions as a light-driven proton pump, undergoes reversible spectroscopic changes at acid pH. The absorption spectra of various bacteriorhodopsin species were estimated from measured spectra of the mixtures that form at low pH, in the presence of sulfate and chloride. The dependency of these on pH and the concentration of Cl- fit a model in which progressive protonation of purple membrane produces "blue membrane", which will bind, with increasing affinity as the pH is lowered, chloride ions to produce "acid purple membrane." Transient spectroscopy with a multichannel analyzer identified the intermediates of the photocycles of these altered pigments, and described their kinetics. Blue membrane produced red-shifted KL-like and L-like products, but no other photointermediates, consistent with earlier suggestions. Unlike others, however, we found that acid purple membrane exhibited a very different photocycle: its first detected intermediate was not like KL in that it was much more red-shifted, and the only other intermediate detectable resembled the O species of the bacteriorhodopsin photocycle. An M-like intermediate, with a deprotonated Schiff base, was not found in either of these photocycles. There are remarkable similarities between the photoreactions of the acid forms of bacteriorhodopsin and the chloride transport system halorhodopsin, where the Schiff base deprotonation seems to be prevented by lack of suitable aspartate residues, rather than by low pH.

Iso-halorhodopsin: a stable, 9-cis retinal containing photoproduct of halorhodopsin

Dark-adapted halorhodopsin is a mixture of 13-cis and all-trans retinal chromophoric species. It is known that illumination with blue light increases the all-trans content, and this is reversed partially by brief red illumination. We now find that extended red-light illumination produces a third spectroscopic form. Analysis of composite absorption spectra recorded during various illumination regimes yielded the spectrum for the new species, whose absorption is shifted approximately 100 nm to the blue. The isomeric composition of retinal extracted from the illuminated pigment indicates that this form contains 9-cis retinal. This species, which we name iso-halorhodopsin, is stable in the dark at room temperature for at least a day, but can be quantitatively reconverted into a mixture of all-trans and 13-cis halorhodopsin by blue-light illumination. A kinetic scheme for the isomeric interconversions was drawn up, where iso-halorhodopsin is produced from either all-trans halorhodopsin only, or both 13-cis and all-trans forms. This kind of scheme is supported by the finding that red illumination of halo-opsin reconstituted with 13-trans-locked retinal will generate iso-halorhodopsin. A similar experiment with 13-cis-locked retinal could not be done because reconstitution with this retinal analogue was not possible. The photoreaction that leads to iso-halorhodopsin can be readily demonstrated in detergent-solubilized halorhodopsin or in halorhodopsin in liposomes made from phosphatidylcholine plus phosphatidyl-ethanolamine, but only to much reduced extent in cell envelope vesicles and in halorhodopsin incorporated into liposomes made from halobacterial polar lipids.

Resonance Raman study of the pink membrane photochemically prepared from the deionized blue membrane of H. halobium

We report here the Resonance Raman spectrum of a 'pink' membrane (lambda max approximately 495 nm) photochemically generated from the deionized 'blue' membrane (Chang et al., 1985). Comparison of the Raman spectrum of the pink membrane with that of the model compounds, as well as the chromophore extraction data, indicate that the chromophore in the pink membrane is in the 9-cis configuration. The Schiff base peak at approximately 1,652 cm-1 shifts to approximately 1,622 cm-1 upon deuteration of the pink membrane, showing that the chromophore is bound to the bacterio-opsin by a protonated Schiff base linkage. The location of the Schiff base peak, as well as the 30 cm-1 shift that it undergoes upon deuteration, are quite different from the corresponding values for the native bacteriorhodopsin, suggesting differences in the local environment for the Schiff base in these pigments.

On the protein (tyrosine)-chromophore (protonated Schiff base) coupling in bacteriorhodopsin

The kinetics of formation of both the tyrosinate ion (from its absorption at 296 nm) and the deprotonated Schiff base (M412) (from its absorption at 404 nm) are studied simultaneously at different pH values (7-11) and temperatures (5-25 degrees C). Two formation rates are observed for M412 in agreement with previous observations. The slow one is dominant under physiological conditions and is found to be slightly faster than that for the tyrosinate formation. This is in disagreement with the proposal that the tyrosinate formation is a prerequisite to the deprotonation of the Schiff base (M412). The ratio of the amplitudes of the fast and slow components is found to be sensitive to pH and, at any pH, it can be used to calculate an amino acid pKa value of 9.6. This is explained by proposing the existence of two sites for the protonated Schiff base within the protein. In one site, the Schiff base is near the neutral form of an amino acid residue with a pKa value of 9.6 (giving rise to the slow component), while in the other, it is near its conjugate base. The formation of the tyrosinate ion as well as the formation of the slow and fast components of M412 all have activation energies that are comparable to H-bond energies. A model is suggested to account for this and the comparable deprotonation rates of tyrosine and the slow component of the protonated Schiff base. It involves the reduction of their pKa by their exposure to a positively charged species.

Absorption spectral properties of acetylated bacteriorhodopsin in purple membrane depending on pH

The dark-adapted form of bacteriorhodopsin in the purple membrane of Halobacterium halobium changes its absorption maximum from 560 to 600 nm if the pH is lowered to about 2 [Oesterhelt, D., & Stoeckenius, W. (1971) Nature (London), New Biol. 233, 149; Moore, T. A., Edgerton, M. E., Parr, G., Greenwood, C., & Perham, R. N. (1978) Biochem. J. 171, 469; Mowery, P. C., Lozier, R. H., Chae, Q., Tseng, T.-W., Taylor, M., & Stoeckenius, W. (1979) Biochemistry 18, 4100; Fischer, U., & Oesterhelt, D. (1979) Biophys. J. 28, 211; Muccio, D. D., & Cassim, J. Y. (1979) J. Mol. Biol. 135, 595]. We compared the pH dependence of the absorption spectra of acetylated membrane with that of unacetylated native membrane. The completely acetylated membrane showed a midpoint of pH 4.8 for the conversion to the acidic form; that of the native membrane was 3.4. On acetylation, the absorption maximum at neutral pH moved from 560 to 555 nm with about 20% decreases in extinction coefficients as compared with that of the native membrane, whereas the spectrum in acid was not affected. The chloride-dependent blue shift from the acidic form of the acetylated membrane was largely suppressed. The CD spectrum of the acetylated membrane was composed of two bands of an opposite sign with slightly decreased amplitudes. The chromophore of the acetylated membrane was sensitive to hydroxylamine, and the spectrum before bleaching was restored on addition of all-trans-retinal to the bleached membrane followed by dark incubation. Blue light irradiation accelerated the conversion to the acidic form in the native membrane but not in the acetylated membrane. Reductive ethylation did not affect the pH dependence of the absorption spectra.