Reconstitution of halorhodopsin in black lipid membranes.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to D Oesterhelt.
Explore the source record for details and available documents.
The native chromoprotein of the light-driven chloride pump halorhodopsin (HR) was isolated from Halobacterium halobium strain L-33 which lacks bacteriorhodopsin but contains 'slow cycling rhodopsin-like pigment' (SR). A membrane fraction was prepared in low salt and dissolved in a high salt medium by the detergents Lubrol PX or octylglucoside. These conditions destroyed the chromophore of SR but not the HR pigment. Chromatography on phenyl-Sepharose and hydroxylapatite produced, in 60% yield, a 230-fold enriched monomeric chromoprotein with an apparent mol. wt. of 20,000. The chromoprotein was stable in 1 M NaCl and 1% octylglucoside and remained stable upon removal of detergent. It reacted with borohydride in the dark and with hydroxylamine in the light. The absorption maximum of the light-adapted state is at 580 + 2 nm and its molar extinction approximately 50,000/M/cm. Upon illumination in the presence of detergent it was converted into a 410 nm absorbing species with concomitant release of protons. A thermal reconversion to the 580 nm species occurred with a half time of 76 s at -6 degrees C. Blue light absorbed by the photoproduct accelerated the re-conversion as well as the re-uptake of protons. Removal of the detergent prevented the light-induced formation of the 410 nm species. Under these conditions a photochemical behaviour similar to that in intact cells and cell vesicles, i.e., a photocycle in the 10-20 ms range was observed. These findings form the basis for functional reconstitution of HR.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Retinol-free fetal calf serum for cell cultures was obtained by UV irradiation and by removal of retinol-binding protein (RBP) by gel-filtration. RBP from bovine serum was purified in a simple two-step procedure. This made it possible to replenish the culture medium with retinol in its physiological i.e. RBP-bound form. Retinol-free and retinol-containing medium was used to assay growth of 3T3 and L929 fibroblast cells. None of these cell lines showed a significant change in growth rate or saturation density that could be attributed to free or RBP-bound retinol.
Five mol of lysine per mol of bacteriorhodopsin were modified with methylacetimidate. This treatment did not inactivate bacteriorhodopsin but prevented all lysines from subsequent reaction with diazotized sulfanilic acid. This reaction predominantly modified tyrosine 64 and light-induced proton translocation was abolished. Reduction of the mono(p-azobenzene sulfonic acid) tyrosine 64 to the corresponding 3-amino derivative with sodium dithionite led to complete reactivation of the proton translocation activity of bacteriorhodopsin. The relative location of tyrosines 26 and 64 and the COOH terminus on the two surfaces of the purple membrane was determined by incorporation into phospholipid vesicles, subsequent modification, and proteolytic treatment. The results obtained support the models proposed by Engelmann et al. (Engelman, D. M., Henderson, R. McLauchlan, A. D., and Wallace, B. A. (1980) Proc. Natl. Acad. Sci. U. S. A. 77, 2023-2027) and by Ovchinnikov et al. (Ovchinnikov, Yu. A., Abdulaev, N. G., Feigina M. Yu., Kiselev A. V., and Lobanov, N. A. (1979) FEBS Lett. 100, 219-224). Tyrosine 64 is located on the extracellular side of the membrane, whereas tyrosine 26 and the COOH terminus are located on the cytoplasmic side. Because specific nitration of tyrosine 26 also leads to inactivation of bacteriorhodopsin (Lemke, H. D., and Oesterhelt, D. (1981) Eur. J. Biochem. 115, 595-604), the results obtained demonstrate that amino acid residues located on both surfaces of the purple membrane are involved in proton translocation.
Cell envelope vesicles prepared from a retinal-deficient strain of Halobacterium halobium contained the apoprotein of halorhodopsin, but not the apoprotein of bacteriorhodopsin. Halorhodopsin was reconstituted in these membranes with tritium-labeled retinal and the preparation was reduced with sodium cyanoborohydride. The product was a bleached pigment in which the retinal-protein bond was resistant to hydroxylamine cleavage. Fluorography of sodium dodecyl sulfate/urea-polyacrylamide gels showed that one protein was radioactively labeled, almost exclusively. This protein migrated with an apparent molecular weight somewhat lower than that of bacteriorhodopsin, which was reconstituted in membranes from another strain and radioactively labeled under the same conditions. Thus, the retinal-binding component of halorhodopsin is a small protein, with an apparent molecular weight of approximately 25,000.
Halorhodopsin (HR) was reconstituted in cell vesicles prepared from Halobacterium halobium strain L-07 by addition of tritium-labelled retinal and subsequently reduced with cyanoborohydride. Lysis of the labelled vesicles in water and dissolution of the cell membranes with 4% SDS allowed the purification of the retinyl protein (RP) by a 3-step procedure. Gel filtration on AcA-44 ultrogel was followed by chromatography on hydroxylapatite and preparative SDS-polyacrylamide gel electrophoresis. This procedure yielded material which migrated as a single band of an apparent mol. wt. of 25 000 on analytical SDS-polyacrylamide gels. The purification was 400-fold with an overall yield of 15%. Not only the mol. wts. but also the amino acid compositions of the RPs from bacteriorhodopsin (BR) and HR are very similar. Polyclonal antibodies against BR and HR did not, however, crossreact. When the two RPs were partially digested with staphylococcal V8 protease the proteolytic pattern of the retinyl peptides was similar, but not identical: two extra peptides are present in BR. The same kind of differences were found in the h.p.l.c. elution profiles of retinyl peptides produced by subtilisin digestion. Therefore, the two proteins must be different gene products and not modification products of one and the same protein.
Thermoplasma acidophilum and Sulfolobus acidocaldarius contain coenzyme A-acylating 2-oxoacid:ferredoxin oxidoreductases similar to those found in halophilic archaebacteria. A common feature of these enzymes is the formation of a free radical intermediate in the course of the catalytic cycle. The electron-accepting ferredoxins and a similar protein from Desulfurococcus mobilis have been purified and characterized. In contrast to the [2Fe-2S] ferredoxin of Halobacterium halobium, the ferredoxins of thermoacidophilic archaebacteria most likely contain two [4Fe-4S]2 + (2 + .1 +) clusters per molecule. Properties of these proteins are compared with respect to the evolution of archaebacteria.
Explore the source record for details and available documents.
Pyruvate:ferredoxin oxidoreductase and 2-oxoglutarate:ferredoxin oxidoreductase were obtained from cell-free extracts of Halobacterium halobium as homogeneous proteins after ammonium sulfate precipitation, salting-out chromatography with ammonium sulfate on unsubstituted agarose, gel filtration and chromatography on hydroxyapatite. The respective molecular weights are 256000 and 248000. Both enzymes consist of two sets of non-identical subunits of Mr 86000 and 42000 in the case of the pyruvate-degrading enzyme and of 88000 and 36000 in the case of the 20 -oxogluatarate-degrading enzyme. Analyses indicate that an intact enzyme molecule contains two [4 Fe-4S]2 + (2 + , 1+) clusters and two molecules of thiamin diphosphate. Flavin nucleotides, lipoic acid and pantetheine are absent. Thus the enzymes are very similar to the 2-oxoacid:ferredoxin oxidoreductases from fermentative and photosynthetic anaerobes described previously, but are clearly different from the 2-oxoacid dehydrogenase multienzyme complexes which commonly occur in anaerobic organisms.
The catalytic cycle of the 2-oxoacid:ferredoxin oxidoreductases from Halobacterium halobium was investigated. The first step is binding of the 2-oxoacid to the enzyme followed by decarboxylation and transfer of one electron to the [4Fe-4S] cluster of the functional unit. The cluster is then reoxidized by ferredoxin or, in the absence of the physiological electron acceptor, by oxygen. In the resulting stable enzyme-intermediate radical the decarboxylation product of the 2-oxoacid remains tightly bound until reaction with coenzyme A caused formation of acyl-CoA and concomitant transfer of the second electron to the cluster, which again is reoxidized by ferredoxin or oxygen. After purification, part of the enzyme molecules still contain the intermediate radical. Enzyme preparations either free of radical or containing enhanced amounts are obtained by treatment with coenzyme A or 2-oxoacid, respectively. Whenever the radical is present in an enzyme molecule the respective binding site for the 2-oxoacid is blocked.
Treatment of the purple membrane with tetranitromethane under controlled conditions leads to the nitration of 3 mol tyrosine/mol bacteriorhodopsin. The combination of subtilisin digestion and cyanogen bromide cleavage with subsequent analysis of the resulting peptide mixture by high-performance liquid chromatography, allows identification of the positions modified in the polypeptide chain. Tyrosines 26 and 64 are fully nitrated, whereas tyrosines 131 and 133 are nitrated to about 60% and 40%, respectively. Reduction of the nitrated membranes with the water-soluble ionic agent dithionite leaves only tyrosine 26 nitrated indicating that the residues 64, 131 and 133 are located on the membrane surface. As a result of nitration, the purple complex shifts its absorption maximum from 568 nm to 532 nm. Dithionite reduction of the nitrated membrane does not reverse this effect. Removal of the retinal and reconstitution maintains the blue-shifted absorption of the chromophore. A pH-dependent equilibrium of the chromophore with a further red-shifted form is observed. The pK of this transition is at about pH 9. Because nitration of tyrosine leads to a drastic decrease of its pK a participation of tyrosine 26 in the chromophoric structure via a hydrogen bridge is suggested. This finding is consistent with a model of chromophore structure published earlier [U. Fischer and D. Oesterhelt (1979) Biophys. J. 31, 139--146].
The rate of regeneration of rhodopsin, from 11-cis-retinal and opsin, and bacteriorhodopsin from all-trans-retinal and bacterio-opsin, in the presence or absence of compounds whose structures partially resemble retinal were measured. Some of these compounds severely slowed down the regeneration process, but did not influence the extent of regeneration. In the case of compounds with a carbonyl functional group they were not joined to the active site of the apo-protein via a Schiff's base linkage since after treatment with NaBH4 an active apo-protein remained. The most effective inhibitors of rhodopsin regeneration were molecules whose structure could be superimposed on 9-cis or 11-cis retinal up to carbon atom 11. These C13 and C15 molecules were not distinguished between aldehyde, ketone or alcohol functional groups. The regeneration of bacteriorhodopsin was not inhibited by retinal analogues with short side chains. The most effective inhibitors were the all-trans C17-aldehyde (beta-ionylideneacetaldehyde) or C18-ketone (beta-ionylidenepent-3-ene-2-one) which, compared to retinal, lack two or three carbon atoms from the end of the poylene chain. The inhibition was very dependent upon the presence of the all-trans isomer and required aldehyde or ketone as functional group nitriles and alcohols were less effective. However, similarly to retinol, the all-trans C17 and C18 alcohols underwent a bathochromic shift and showed fine-structured spectra when mixed with bacterio-opsin.
The proton motive force across the cell membrane of halobacterial cells has been estimated and compared to intracellular values of ATP, ADP, and inorganic phosphate concentrations with respect to the chemiosmotic hypothesis. The accumulation of 14C-labeled indicator substances, triphenylmethylphosphonium for the membrane potential and 5,5-dimethyloxazolidine-2,4-dione for the pH difference between the cell interior and the medium, has been measured in the cells. Values up to 270 mV for the proton motive force have been found in cells pretreated with N,N'-dicyclohexylcarbodiimide (DCCD, 10(-4) M, 30 degrees C, 12 h). Upon illumination a high membrane potential is generated, which is then gradually replaced by a large pH difference. Cells treated with lower DCCD concentrations show only an enhancement of membrane potential upon illumination; the pH difference remains at a low level. Under anaerobic dark conditions, untreated cells maintain a proton motive force of 120-140 mV, which is equilibrated with the intracellular levels of ATP, ADP, and inorganic phosphate. The pH gradient is 1 unit at pH 6 but 0 at pH 8. The membrane potential is low (60-80 mV) at pH 6 and high (120-130 mV) at pH 8. We propose that the proton translocating ATPase compensates for the lowered pH difference at high external pH values by enhancing the membrane potential. The concentration difference of the potassium ions influences the proton motive force and the intracellular ATP levels, apparently via its action on the membrane potential. When the difference of the chemical potential of the potassium ion, expressed in millivolts, exceeds the preexisting membrane potential, the intracellular ATP level is enhanced. When the difference of the chemical potential of the potassium ion (millivolts) is smaller than the membrane potential, the ATP level is decreased.
The increase of the proton motive force mediated by bacteriorhodopsin is compared to the intracellular ATP concentration under steady-state illumination. The membrane potential was measured via the accumulation of the lipophilic ion [14C]triphenylmethylphosphonium and the pH gradient via the accumulation of the weak acid 5,5-dimethyloxazolidine-2,4-dione. Light causes a parallel increase of ATP level and membrane potential at an external pH of 8. In contrast, at pH 6 an increase of the intracellular ATP concentration occurs without a corresponding increase of the proton motive force. If the extracellular NaCl concentration is reduced and not replaced by other ions, no membrane potential at all can be measured at pH 6-7; nonetheless, light-induced ATP synthesis occurs. A significant enhancement of the pH gradient occurs only at irradiances higher than those required for the attainment of the maximal intracellular ATP concentration. In the presence of the uncoupler carbonyl cyanide m-chlorophenylhydrazone it is possible to obtain light-induced ATP synthesis without a measurable proton motive force in basal salt at pH 8. It is concluded that if bacteriorhodopsin acts as a proton pump, then the pumped protons can be used for ATP synthesis before they equilibrate with the protons in the extracellular bulk phase.
Explore the source record for details and available documents.
Protonation changes of the protein occur during the reconstitution of bacteriorhodopsin from bacterio-opsin and all-trans retinal in the purple membrane of Halobacterium halobium. The protonation changes are conveniently determined from measures of the pH changes after photoisomerisation of 9-cis retinal in apomembrane preparations, which induces the reconstitution. In addition, to the omega-amino group of the lysine which is involved in the condensation of retinal and bacterio-opsin, the dissociation equilibria of at least two other amino acid residues are changed during the reconstitution. The results are consistent with a proposed model of chromophore structure in which an interaction of the Schiff's base occurs with two protonable amino acid residues.