Mechanism and Role of Divalent Cation Binding of Bacteriorhodopsin

Several observations have already suggested that the carboxyl groups are involved in the association of divalent cations with bacteriorhodopsin (Chang et al., 1985). Here we show that at least part of the protons released from deionized purple membrane (`blue membrane') samples when salt is added are from carboxyl groups. We find that the apparent pK of magnesium binding to purple membrane in the presence of 0.5 mM buffer is 5.85. We suggest this is the pK of the carboxyl groups shifted from their usual pK because of the proton concentrating effect of the large negative surface potential of the purple membrane. Divalent cations may interact with negatively charged sites on the surface of purple membrane through the surface potential and/or through binding either by individual ligands or by conformation-dependent chelation. We find that divalent cations can be released from purple membrane by raising the temperature. Moreover, purple membrane binds only about half as many divalent cations after bleaching. Neither of these operations is expected to decrease the surface potential and thus these experiments suggest that some specific conformation in purple membrane is essential for the binding of a substantial fraction of the divalent cations. Divalent cations in purple membrane can be replaced by monovalent, (Na⁺ and K⁺), or trivalent, (La⁺⁺⁺) cations. Flash photolysis measurements show that the amplitude of the photointermediate, O, is affected by the replacement of the divalent cations by other ions, especially by La⁺⁺⁺. The kinetics of the M photointermediate and light-induced H⁺ uptake are not affected by Na⁺ and K⁺, but they are drastically lengthened by La⁺⁺⁺ substitution, especially at alkaline pHs. We suggest that the surface charge density and thus the surface potential is controlled by divalent cation binding. Removal of the cations (to make deionized blue membrane) or replacement of them (e.g. La⁺⁺⁺-purple membrane) changes the surface potential and hence the proton concentration near the membrane surface. An increase in local proton concentration could cause the protonation of critical carboxyl groups, for example the counter-ion to the protonated Schiff's base, causing the red shift associated with the formation of both deionized and acid blue membrane. Similar explanations based on regulation of the surface proton concentration can explain many other effects associated with the association of different cations with bacteriorhodopsin.     

The concentrations and thermodynamic activities of cations in intact Bryopsis chloroplasts

The internal cation levels of chloroplasts isolated from a green sea alga, Bryopsis maxima, were studied. Atomic absorption spectroscopy, combined with the determination of the sorbitol-impermeable and water-permeable spaces, revealed that chloroplasts contain an extremely high concentration of K⁺ and high levels of Na⁺, Mg²⁺ and Ca²⁺. A method was developed to estimate the thermodynamic activities of monovalent and divalent cations present in chloroplasts. pH changes induced by the addition of an ionophore (plus an H⁺ carrier), which makes the outer limiting membranes of chloroplasts permeable to both a cation and H⁺, were determined. Provided that the external pH was set equal to the internal pH, the internal concentration of the cation was estimated by determining the external cation concentration which gave rise to no electrochemical potential difference of the cation and hence no pH change on addition of the ionophore. The internal pH was determined by measuring distributions of radioactive methylamine and 5,5-dimethyloxazolidine-2,4-dione between the chloroplast and medium (Heldt, H.W., Werdan, K., Milovancev, M. and Geller, G. (1973) Biochim. Biophys. Acta 314, 224–241). The internal pH was also estimated by measuring pH changes caused by the disruption of the outer limiting membrane with Triton X-100. The results indicate that a significant part of the monovalent cations and most of the divalent cations are attracted into a diffuse layer adjacent to the negatively charged surfaces of membranes and proteins, or form complexes with organic and inorganic compounds present in the intact chloroplasts.     

Protein-chromophore interactions in bacteriorhodopsin: the effects of a change in surface potential.

The chromophore retinal is bound to bacteriorhodopsin via a protonated Schiff base linkage. The retinal binding site is reported to be buried in the transmembrane portion of the protein, distant from the membrane surfaces. When bound to bacteriorhodopsin, the absorption maximum of retinal is red-shifted from 366 nm to 568 nm producing a purple color. This color persists across a wide pH range. However, when the pH is raised above 12.0, the membranes become pink in color, while at pH values of 3.0 or below, a blue color is produced. The blue color can also be obtained by removing the divalent cations bound to the surface of the protein. In this study, bacteriorhodopsin was examined by circular dichroism and absorption spectroscopy to determine if protein conformational changes were associated with the color shifts. It was found that although the retinal chromophore can be completely removed by bleaching with hydroxylamine with no significant influence on the secondary structure of the protein, a change in the surface charge of bacteriorhodopsin results in measurable conformational change in the protein, which apparently affects the nature of the retinal binding site.     

The surface potential on the purple membrane measured using a modified bacteriorhodopsin chromophore as the spectroscopic probe

The surface potential of the purple membrane was measured by a novel method by using an artificial bacteriorhodopsin whose chromophore was 13-CF₃ retinal instead of retinal. When attached to the apoprotein by a Schiff base, the intrinsic pK of the 13-CF₃ chromophore is around 7.3. The apparent p_K of this pigment depends on the surface potential and thus on the electrolyte concentration. This allowed us to determine the surface charge density using the Gouy-Chapman equation. The surface charge density was found to be −1.65 ± 0.15 × 10⁻³ electronic charges per Ų or about 2 negative charges/bacteriorhodopsin. This large value for the surface potential probably explains both part of the strong apparent association of divalent cations with the membrane and the effect of low salt concentrations on light-induced proton release from the purple membrane.     

Two possible roles of bacteriorhodopsin; a comparative study of strains of Halobacterium halobium differing in pigmentation.

The light-induced changes in pH and ATP level were compared for cell suspensions between strains of Halobacterium halobium differing in pigmentation after growth under the same conditions. Upon illumination, red cells which contained no detectable amount of bacteriorhodopsin showed only a pH increase, which, in the case of purple cells containing bacteriorhodopsin, was followed by a spontaneous pH decrease during illumination. Pre-incubation of cells at 75° for 5 min depressed the pH increase in both cells. Pre-illumination of cells with hydroxylamine depressed the pH decrease in purple cells. Whenever the pH increase was observed, the cellular ATP level increased. The presence of a bacteriorhodopsin different from that in the purple membrane is postulated.     

Measurement of chloroplast internal protons with 9-aminoacridine. Probe binding, dark proton gradient, and salt effects

A defined ratio, gamma, of the total proton uptake to the concentration change of free internal H+ is observed for illuminated envelope-free chloroplasts (Haraux, F. and de Kouchkovsky, Y. (1979) Biochim. Biophys. Acta, 546, 455-471). Proton uptake is measured by the external pH shift, free internal H+ by 9-aminoacridine fluorescence quenching. Extension of this work leads to the following conclusions, which, in the case of 9-aminoacridine behaviour, should apply to any kind of diffusible protonizable delta pH probe: 1. The gamma constancy is preserved when the internal volume (Vi) is modulated by chlorophyll and osmolarity changes: thus, 9-aminoacridine behaves as expected from the delta pH distribution of an amine of high pK; previous doubts on this point are attributed to the lack of control of the external proton uptake. 2. With variable 9-aminoacridine concentration, however, some variation of gamma confirms the existence of slight light-induced probe-membrane interactions. 3. According to the diffuse layer theory, salts decrease the negative potential at the 'plane of closest approach' of the thylakoids, thereby releasing the excess 9-aminoacridine in this diffuse layer, which increases its fluorescence. Although of equal valency, NH4+ is more potent than K+, suggesting competition between amines for specific anionic binding sites. 4. Two categories of membrane modifications are induced by salts: in addition to the above-mentioned electrical effect, mono- and divalent cations at high concentration increase the chloroplast proton binding capacity. La3+ is only able to release the excess dye in the diffuse layer and leaves gamma unchanged. Therefore the probe-membrane interactions should have limited importance for steady-state delta pH measurement. 5. A Donnan-type dark pH difference, which could seriously bias these delta pH estimates, is found experimentally to be less than 2 (no significant gamma change when Vi varies) and even theoretically less than 1 (on the basis of the concentration of the non-diffusible internal protonizable groups). Similarly, the predictable errors of Vi and its possible light-induced variations must have a small effect on delta pH under present experimental conditions.     

Surface pH controls purple-to-blue transition of bacteriorhodopsin. A theoretical model of purple membrane surface.

We have developed a surface model of purple membrane and applied it in an analysis of the purple-to-blue color change of bacteriorhodopsin which is induced by acidification or deionization. The model is based on dissociation and double layer theory and the known membrane structure. We calculated surface pH, ion concentrations, charge density, and potential as a function of bulk pH and concentration of mono- and divalent cations. At low salt concentrations, the surface pH is significantly lower than the bulk pH and it becomes independent of bulk pH in the deionized membrane suspension. Using an experimental acid titration curve for neutral, lipid-depleted membrane, we converted surface pH into absorption values. The calculated bacteriohodopsin color changes for acidification of purple, and titrations of deionized blue membrane with cations or base agree well with experimental results. No chemical binding is required to reproduce the experimental curves. Surface charge and potential changes in acid, base and cation titrations are calculated and their relation to the color change is discussed. Consistent with structural data, 10 primary phosphate and two basic surface groups per bacteriorhodopsin are sufficient to obtain good agreement between all calculated and experimental curves. The results provide a theoretical basis for our earlier conclusion that the purple-to-blue transition must be attributed to surface phenomena and not to cation binding at specific sites in the protein.     

Transmembrane pH gradients and functional heterogeneity in reconstituted vesicle systems

The pH-sensitive, membrane impermeant fluorescence probes 8-hydroxy-1,3,6-pyrenetrisulfonate (pyranine; pK_a = 7.2) and 1-naphthol-3,6-disulfonate (Naps pK_a = 8.2) can be simultaneously entrapped within the intravesicular aqueous compartment of unilamellar vesicles and reconstituted proteoliposomes, where they function as reliable reporters of the intravesicular pH. Because the two probes are sensitive to pH over different but overlapping ranges, the useful monitoring range for the co-trapped probe pair extends from pH 6.5 to 9. In vesicles pre-equilibrated at a given pH and then subjected to a sudden change in external pH, the rate and extent of the subsequent change in internal pH are identical at all times during the re-equilibration, regardless of which probe is used to monitor the change. However, in reconstituted bacteriorhodopsin proteoliposomes, the size of the transmembrane pH gradient generated in the light always appears greater when pyranine is used to monitor internal pH. This discrepancy can most readily be understood in terms of heterogeneity in the vesicle suspension, with at least two populations of vesicles, one active in proton and one inactive. A simple algorithm was developed which generates, from the observed internal pH changes for two probes of different pK_a, the percentage of vesicles which are inactive, as well as the actual internal pH of the active fraction. The applicability of this algorithm was subsequently confirmed using a suspension of vesicles in which the level of heterogeneity was deliberately altered by the addition of various amounts of gramicidin. The apparent transmembrane pH gradient for the vesicle population as a whole decreased with increasing gramicidin, as did the calculated percentage of vesicles able to maintain a pH gradient, while the transmembrane gradient calculated for the active vesicle fraction only was essentially unaffected by gramicidin.     

Stimulation of ATP synthesis in Halobacterium halobium R1 by light-induced or artificially created proton electrochemical potential gradients across the cell membrane

The relationship between proton movement and phosphorylation in Halobacterium halobium R₁ has been investigated under anaerobic conditions. The light-induced changes in the bacteriorhodopsin are accompanied by proton movements across the cell membrane which result in pH changes in the suspending medium. The initial alkaline shift is shown to be closely paralleled by (and hence correlated with) ATP synthesis. Acidification of the medium in the presence of valinomycin, under conditions of low external potassium, brings about ATP synthesis in the dark.     

Reaction of the purple membrane with a carbodimide

Purple membrane was reacted with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide at pH 4.5 and 8.0. At pH 4.5, the reaction yields cross-linked bacteriorhodopsin. The cross-linking is inhibited by pretreatment of the membrane with papain, or by the presence of carbohydrazide or glycine ethyl ester in the reaction mixture. The product of the pH 8.0 reaction is not cross-linked, but it displays altered properties. Two measures of photochemical activity (light-induced change in proton binding (Δh?) and decay of photointermediate M) show changes indicative of slowed proton uptake. The Δh? is increased by ethyl dimethylaminopropylcarbodiimide. This increase is unaffected by pretreatment of the membrane with papain, and it is not reversed by NH₂OH. However, the reaction is inhibited by millimolar concentrations of CaCl₂. The altered Δh? is not apparent in detergent-solubilized membranes. Ethyl dimethylaminopropyl-carbodiimide does not appear to cause a large alteration in the membrane surface charge, as measured by Ca²⁺ binding.We conclude that (1) at acid pH, ethyl dimethylaminopropylcarbodiimide can be used for cross-linking or for attachment of specific probes to the C-terminal region of bacteriorhodopsin, and hence to the cytoplasmic side of the purple membrane, and (2) at alkaline pH, ethyl dimethylaminopropylcarbodiimide reacts at a different type of site and appears to inhibit the proton pump.     

Light-induced movement of magnesium ions in intact chloroplasts. Spectroscopic determination with Eriochrome Blue SE

The metallochromic indicator Eriochrome Blue SE was used to measure light-induced internal movement of Mg²⁺ in intact chloroplasts. By dual-wavelength spectroscopy (measuring wavelength 554 nm, reference 592 nm) a light-induced, dark-reversible absorbance increase of Eriochrome Blue in samples of isolated intact chloroplasts was observed. The light/dark difference spectrum of Eriochrome Blue between 550 and 590 nm (reference wavelength 562 nm) indicated that this absorbance increase was caused by an increased concentration of free Mg²⁺ in a neutral or slightly alkaline chloroplast compartment.

The signal was seen only with intact, but not with broken, envelope-free chloroplasts, which had lost most of their divalent cations. This is interpreted to show that the indicator responds to an increase of Mg²⁺ concentration in the chloroplast stroma, which represents an efflux of Mg²⁺ from the intra-thylakoid space caused by light-dependent proton pumping.

As calculated from corrected values of the absorbance increase of Eriochrome Blue, the light-induced internal release of Mg²⁺ was close to 100 nequiv per mg chlorophyll at pH 7.6 and 250 nequiv at pH 7.1. This corresponds to a light-dependent increase in the concentration of free Mg²⁺ in the stroma of about 2 and 5 mM, respectively.     

Thermal denaturation and gelation of rubisco: effects of pH and ions

In order to understand the mechanism of thermal gelation of rubisco, its native and heat denatured states were characterized by absorbance, fluorescence and circular dichroïsm spectroscopies as well as by differential scanning calorimetry in the presence of various salts. It appears that during the denaturation process, divalent anions are released while divalent cations are fixed by the protein, while it is disorganized and while the environment of its aromatic chromophores becomes more hydrophilic. The pH transition of gelation is shifted 1–2 pH units higher than the transition of denaturation temperature which occurs near the isoelectric point of the native molecule. This shift probably corresponds to the breaking of saline bridges within the protein molecule. Finally, a large effect of divalent cations on the phase diagram indicates that a particular denatured state is attained when these cations are in the denaturation medium.     

Functional reconstitution of a membrane protein in a diacetylenic polymerizable lecithin

It is shown that polymerized diacetylenic lecithins may be used for the functional reconstitution of a membrane protein. Purple membrane patches isolated from Halobacterium halobium and liposomes of the polymerizable diacetylenic lecithin 1,2-bis(10,12 tricosadiynoyl)-sn-glycero-3-phosphocholine were sonicated together to form mixed vesicles highly enriched in the polymerizable lipid. A net inward proton flow on illumination as determined by the change of pH of the external medium demonstrated the stability of the vesicular form in this mixed lipid system as well as vectorial orientation of the bacteriorhodopsin in the bilayer. When bacteriorhodopsin was incorporated in non-polymerizable lipids, irradiation with ultraviolet light resulted in complete loss of function. In the diacetylenic lipids, the loss of function was slower than the increase in polymer concentration. This demonstrates the utility of the diacetylenic lecithin system for study of interactions between membrane proteins and polymerizable lipids, as well as its potential in the development of biosensors based on membrane proteins.     

Directly probing rapid membrane protein dynamics with an atomic force microscope: a study of light-induced conformational alterations in bacteriorhodopsin.

This paper demonstrates that an atomic force microscope can be used to directly monitor rapid membrane protein dynamics. For this demonstration the membrane-bound proton pump, bacteriorhodopsin, has been investigated. It has been unequivocally shown that the light-induced dynamic alterations that have been observed do not arise from external artifacts such as heating of the sample by the incident light, but that these changes can be directly linked to the light-induced protein conformational alterations in this membrane. In essence, it has been shown that the light energy absorbed by bacteriorhodopsin is converted not only to chemical energy but also to mechanical energy. In summary a new ultrasensitive tool is described for monitoring the molecular dynamics of materials with wide applicability to fundamental and applied science.     

H+ translocation in lysozyme-treated cells of the blue-green alga Plectonema boryanum. Difference between isotonically and hypotonically treated preparations and effects of divalent cations.

Lysozyme-treated cells of a blue-green alga, Plectonema boryanum, had an internal pH of 7.3+/-0.2 under isotonic and hypotonic conditions. This value was similar to that of untreated cells. The CCCP-induced biphasic H+ change seen in the isotonic cells was not observed in the hypotonically treated cells. The biphasic time course remained in the hypotonic preparation if CaCl2 or MgCl2 was added prior to the osmotic shock. It is suggested that the cells have two compartments of H+ concentration. The outer region may be more acidic than the inner region. A light-induced H+ efflux was observed under isotonic conditions and an influx of H+ under hypotonic conditions. The H+ influx was not observed when lysozyme-treated cells were incubated with CaCl2 or MgCl2 prior to the hypotonic treatment. Two types of effects of divalent cations, one on the rigidity of the outer membrane and another on the permeability characteristics of the inner photosynthetic membrane, are indicated. Rearrangement of the photosynthetic membranes and an apparent inversion of the H+ pump by hypotonic shock are also suggested.     

Reaction of the purple membrane with a carbodiimide.

Purple membrane was reacted with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide at pH 4.5 and 8.0. At pH 4.5, the reaction yields cross-linked bacteriorhodopsin. The cross-linking is inhibited by pretreatment of the membrane with papain, or by the presence of carbohydrazide or glycine ethyl ester in the reaction mixture. The product of the pH 8.0 reaction is not cross-linked, but it displays altered properties. Two measures of photochemical activity (light-induced change in proton binding (delta h) and decay of photointermediate M) show changes indicative of slowed proton uptake. The delta h is increased by ethyl dimethylaminopropylcarbodiimide. This increase is unaffected by pretreatment of the membrane with papain, and it is not reversed by NH2OH. However, the reaction is inhibited by millimolar concentrations of CaCl2. The altered delta h is not apparent in detergent-solubilized membranes. Ethyl dimethylaminopropylcarbodiimide does not appear to cause a large alteration in the membrane surface charge, as measured by Ca2+ binding. We conclude that (1) at acid pH, ethyl dimethylaminopropylcarbodiimide can be used for cross-linking or for attachment of specific probes to the C-terminal region of bacteriorhodopsin, and hence to the cytoplasmic side of the purple membrane, and (2) at alkaline pH, ethyl dimethylaminopropylcarbodiimide reacts at a diffent type of site and appears to inhibit the proton pump.     

Regulation of rhizosphere acidification by photosynthetic activity in cowpea (Vigna unguiculata L. walp.) seedlings

In contrast to cereals or other crops, legumes are known to acidify the rhizosphere even when supplied with nitrates. This phenomenon has been attributed to N2 fixation allowing excess uptake of cations over anions; however, as we have found previously, the exposure of the shoot to illumination can cause rhizosphere acidification in the absence of N2 fixation in cowpea (Vigna unguiculata L. Walp). In this study, we examined whether the light-induced acidification can relate to photosynthetic activity and corresponding alterations in cation-anion uptake ratios. The changes of rhizosphere pH along the root axis were visualized using a pH indicator agar gel. The intensity of pH changes (alkalization/acidification) in the rhizosphere was expressed in proton fluxes, which were obtained by processing the images of the pH indicator agar gel. The uptake of cations and anions was measured in nutrient solution. The rhizosphere was alkalinized in the dark but acidified with exposure of the shoots to light. The extent of light-induced acidification was increased with leaf size and intensity of illumination on the shoot, and completely stopped with the application of photosynthesis inhibitor. Although the uptake of cations was significantly lower than that of anions, the rhizosphere was acidified by light exposure. Proton pump inhibitors N,N'-dicyclohexyl carbodimide and vanadate could not stop the light-induced acidification. The results indicate that light-induced acidification in cowpea seedlings is regulated by photosynthetic activity, but is not due to excess uptake of cations.     

Properties of the Isolated Intact Chloroplast at Cytoplasmic K Concentrations : I. Light-Induced Cation Uptake into Intact Chloroplasts is Driven by an Electrical Potential Difference

Photosynthesis, stroma-pH, and internal K⁺ and Cl⁻ concentrations of isolated intact chloroplasts from Spinacia oleracea, as well as ion (K⁺, H⁺, Cl⁻) movements across the envelope, were measured over a wide range of external KCl concentrations (1-100 millimolar).

Isolated intact chloroplasts are a Donnan system which accumulates cations (K⁺ or added Tetraphenylphosphonium⁺) and excludes anions (Cl⁻) at low ionic strength of the medium. The internally negative dark potential becomes still more negative in the light as estimated by Tetraphenylphosphonium⁺ distribution. At 100 millimolar external KCl, potentials both in the light and in the dark and also the light-induced uptake of K⁺ or Na⁺ and the release of protons all become very small. Light-induced K⁺ uptake is not abolished by valinomycin suggesting that the K⁺ uptake is not primarily active. Intact chloroplasts contain higher K⁺ concentrations (112-157 millimolar) than chloroplasts isolated in standard media. Photosynthetic activity of intact chloroplasts is higher at 100 millimolar external KCl than at 5 to 25 millimolar. The pH optimum of CO₂ fixation at high K⁺ concentrations is broadened towards low pH values. This can be correlated with the observation that high external KCl concentrations at a constant pH of the suspending medium produce an increase of stroma-pH both in the light and in the dark. These results demonstrate a requirement of high external concentrations of monovalent cations for CO₂ fixation in intact chloroplasts.

     

Paradoxical digestion of myo-inositol phosphates by alkaline phosphatase at low pH

The ability of bovine intestinal alkaline phosphatase (0.1-10 units/ml) to cleave myo-inositol bound phosphate moieties was examined. Paradoxically the digestion was optimal for a number of isomers at pH 5-7. It is possible that digestion at higher pH (9-10) does not proceed at maximal rates due to a conformation of the myo-inositol phosphate molecule which stabilizes the molecule against enzymatic attack. Alkaline phosphatase activity did not require the addition of added divalent cations. Moreover, several divalent cations, particularly zinc, were found to have a marked inhibitory effect. Further studies into this phenomenon suggested that some divalent cations can form insoluble complexes with myo-inositol phosphates, particularly those possessing a number of phosphate moieties, preventing the action of degradative enzymes. On the basis of these experiments we conclude that phosphate moieties can be removed from myo-inositol using relatively low concentrations of alkaline phosphatase as long as optimal incubation conditions are selected. This includes the use of a slightly acidic incubation media without the addition of divalent cations, particularly zinc.