The light-driven proton pump, cruxrhodopsin-2 in Haloarcula sp. arg-2 (bR+, hR-), and its coupled ATP formation.

Haloarcula sp. arg-2, a natural bacterial isolate from Andes heights, has a light-driven proton pump but not a light-driven anion pump. We have cloned and sequenced the gene encoding for the proton pump which has been named cruxrhodopsin-2. The gene consists of 768 bp encoding 255 amino acids with a molecular mass of 27,544 Da. The deduced amino acid sequence of cruxrhodopsin-2 is 77%, 50%, 48% and 48% identical to those of cruxrhodopsin-1, bacteriorhodopsin, archaerhodopsin-1 and archaerhodopsin-2, respectively. The charged amino acids important for the proton pump function were conserved among all these molecules. Cruxrhodopsin-2 accounted for 0.05 nmol/mg protein in arg-2, which was 20-30-fold less than the proportion of bacteriorhodopsin in Halobacterium salinarium R1M1. In contrast to R1M1, under anaerobic conditions, arg-2 showed light-induced proton extrusion concomitant with an increase in ATP level without transient proton uptake. Dicyclohexylcarbodiimide enhanced the rate and extent of proton extrusion and inhibited ATP formation in the light. The apparent stoichiometry of H+/ATP was estimated to be more than three in this natural bR+hR- strain.

A vacuolar ATPase and pyrophosphatase in Acetabularia acetabulum.

Vacuole-rich fractions were isolated from Acetabularia acetabulum by Ficoll step gradient centrifugation. The tonoplast-rich vesicles showed ATP-dependent and pyrophosphate-dependent H(+)-transport activities. ATP-dependent H(+)-transport and ATPase activity were both inhibited by the addition of a specific inhibitor of vacuolar ATPase, bafilomycin B1. A 66 kDa polypeptide present in the preparation cross-reacted with the anti-IgG fractions against the alpha and beta subunits of Halobacterium halobium ATPase and with the antibody against the A subunit (68 kDa subunit) of mung bean vacuolar ATPase. A 56 kDa polypeptide present in the vacuole preparation showed cross-reactivity with the antibody against the B subunit (57 kDa) of mung bean vacuolar ATPase but not with the anti-beta subunit of H. halobium ATPase. A 73 kDa polypeptide cross-reacted with the antibody against inorganic pyrophosphatase of mung bean vacuoles. These results suggest that vacuolar membrane of A. acetabulum equipped energy transducing systems similar to those found in other plant vacuoles.

Evolution of the archaeal rhodopsins: evolution rate changes by gene duplication and functional differentiation.

The amino acid sequences of 25 archaeal retinal proteins from 13 different strains of extreme halophiles were analyzed to establish their molecular phylogenetic relationship. On the basis of amino acid sequence similarity, these proteins apparently formed a distinct family designated as the archaeal rhodopsin family (ARF), which was not related to other known proteins, including G protein-coupled receptors. The archaeal rhodopsin family was further divided into four clusters with different functions; H+ pump (bacteriorhodopsin), Cl- pump (halorhodopsin), and two kinds of sensor (sensory rhodopsin and phoborhodopsin). These four rhodopsin clusters seemed to have occurred by gene duplication(s) before the generic speciation of halophilic archaea, based on phylogenetic analysis. Therefore, the degrees of differences in amino acid sequences within each cluster simply reflected the divergent evolution of halophilic archaea. By comparing the branch lengths after speciation points of the reconstituted tree, we calculated the relative evolution rates of the four archaeal rhodopsins bacteriorhodopsin:halorhodopsin:sensory rhodopsin: phoborhodopsin to be 5:4:3:10. From these values, the degrees of functional and structural restriction of each protein can be inferred. The branching topology of four clusters grouped bacteriorhodopsin and halorhodopsin versus sensory rhodopsin and phoborhodopsin by likelihood mapping. Using bacteriorhodopsin (and halorhodopsin) as an outgroup, the gene duplication point of sensory rhodopsin/phoborhodopsin was determined. By calculating the branch lengths between the gene duplication point and each halophilic archaea speciation point, we could speculate upon the relative evolution rate of pre-sensory rhodopsin and pre-phoborhodopsin. The evolution rate of pre-sensory rhodopsin was fivefold faster than that of pre-phoborhodopsin, which suggests that the original function of the ancestral sensor was similar to that of phoborhodopsin, and that sensory rhodopsin evolved from pre-sensory rhodopsin by the accumulation of mutations. The changes in evolution rate by gene duplication and functional differentiation were demonstrated in the archaeal rhodopsin family using the gene duplication date and halobacterial speciation date as common time stamps.

Proteoliposomes with right-side-out oriented purple membrane/bacteriorhodopsin require cations inside for proton pumping.

Proteoliposomes were prepared by sonication of phospholipids and blue membranes (cation-free purple membranes carrying little activity of light-driven proton pumping) in an acidic medium of very low ionic strength. The majority of the bacteriorhodopsin population in these proteoliposomes was in the right-side-out (as in living cells) orientation as judged from the resultant polypeptides after papain digestion. By raising the pH of sonication, the population of right-side-out oriented bacteriorhodopsin decreased, and consequently that of the inversely oriented one increased. In KCl and NaCl up to certain concentrations or in choline chloride even at high concentrations, in the light, the proteoliposomes with right-side-out bacteriorhodopsin did not pump protons, whereas those with inversely oriented bacteriorhodopsin did. The former began to pump only after cations were likely incorporated/permeated into the proteoliposome and reached the carboxyl terminal (cytosol) side of bacteriorhodopsin/purple membrane.

The gamma-subunit of ATP synthase from spinach chloroplasts. Primary structure deduced from the cloned cDNA sequence.

cDNA clones encoding the gamma-subunit of chloroplast ATP synthase were isolated from a spinach library using synthetic oligonucleotide probes. The predicted amino acid sequence indicated that the mature chloroplast gamma-subunit consists of 323 amino acid residues and is highly homologous (55% identical residues) with the sequence of the cyanobacterial subunit. The positions of the four cysteine residues were identified. The carboxyl-terminal region of the chloroplast gamma-subunit is highly homologous with those of the gamma-subunits from six other sources (bacteria and mitochondria) sequenced thus far.

Activation and inhibition of ATP synthesis in cell envelope vesicles of Halobacterium halobium.

The characteristics of ATP synthesis in cell envelope vesicles of Halobacterium halobium were further studied. The results confirmed the previous conclusion (Mukohata et al. (1986) J. Biochem. 99, 1-8) that the ATP synthase in this extremely halophilic archaebacterium can not be an ordinary type of F0F1-ATPase, which has been thought to be ubiquitous among all the aerobic organisms on our biosphere. The ATP synthesis was activated most in 1 M NaCl and/or KCl, and at 40 degrees C, and at 80 mM MgCl2 where F0F1-ATPase loses its activity completely. The synthesis was negligible at 10 degrees C, and at 5 mM MgCl2. The Km for ADP was about 0.3 mM in the presence of 20 mM Pi, 1 M NaCl, 80 mM MgCl2, and 10 mM PIPES at pH 6.8 and 20 degrees C. The ATP synthesis was not inhibited by NaN3 and quercetin (specific inhibitors for F0F1-ATPase) or vanadate (for E1E2-ATPase) or ouabain (for Na+,K+-ATPase) or P1,P5-di(adenosine-5')pentaphosphate (AP5A, for adenylate kinase). The ATP synthesis was not inhibited by modification (pretreatment) with NaN3 or 5'-p-fluorosulfonylbenzoyladenosine (FSBA). On the contrary, the ATP synthesis was rather non-specifically inhibited by N-ethylmaleimide (NEM), trinitrobenzenesulfonate (TNBS), phenylglyoxal, and pyridoxal phosphate. 7-Chloro-4-nitrobenz-2-oxa-1,3-diazole (NBD-Cl) as well as N,N'-dicyclohexylcarbodiimide (DCCD) was found to be a specific inhibitor at least partly, because the NBD-Cl inhibition was partly prevented by ADP added to the modification mixture.

Isolation and characterization of halorhodopsin from Halobacterium halobium.

Chromoprotein of a light-driven chloride pump, halorhodopsin (HR), was isolated from Halobacterium halobium L-33, which contains HR and "slowly cycling rhodopsin-like pigment" (SR) but lacks bacteriorhodopsin (BR). The isolation was run in the presence of more than 2 M NaCl, which was required to preserve this halophilic retinal protein. Cell envelope vesicles were washed with Tween-20 to remove 80% of the proteins. The residual membranes were solubilized with 0.5% C12E9, which had little effect on the photochemical activities of HR and SR. HR was purified by passing it through a hydroxyapatite and then a phenyl-Sepharose column in 2 M NaCl and 0.5% C12E9. The absorption maximum of HR was 578 nm and the ratio of absorbance at 280 nm to 580 nm was 1.52. The apparent molecular weight of HR was 20,000 on polyacrylamide gel electrophoresis in the presence of SDS. The characteristic, bilobed CD spectrum of HR in the visible region suggested that HR exists as an oligomer in both its membrane-bound and isolated forms.

Magnesium ion-induced changes in the binding mode of adenylates to chloroplast coupling factor 1.

The effect of Mg2+ on the binding of adenylates to isolated chloroplast coupling factor 1 (CF1) was studied using CD spectrometry and ultrafiltration. At adenylate concentrations smaller than 100 muM, one mole of CF1 binds three moles of ATP (or ADP) regardless of the presence of Mg2+. In the presence of Mg2+, the first two ATP's bind to CF1 independently with the same binding constant of 2.5 X 10(-1) muM-1, then the third ATP binds with a much higher affinity of 10 muM-1. In the absence of Mg2+, the first ATP binds to CF1 with a binding constant of 2.5 X 10(-1) muM-1 then the other two ATP's bind less easily with the same binding constant of 4.0 X 10(-2) muM-1. The binding mode of ADP to CF1 is quite similar to that of ATP. In the presence of Mg2+, the binding constants of the first two ADP's are both 7.6 X 10(-2) muM-1, that of the third ADP being 4.0 muM-1. In the absence of Mg2+, the binding constant of the first ADP is 7.6 X 10(-2) muM-1, the constants of the other two ADP's both being 4.0 X 10(-2) muM-1. AMP caused a negligible change in CD.

Archae-opsin expressed in Escherichia coli and its conversion to purple pigment in vitro.

Archae-opsin-1 (aO-1) has been expressed efficiently as a fusion protein with 13 heterologous amino acids at the amino terminus of the mature aO-1 in Escherichia coli under the control of T7 promoter. The E. coli-expressed aO-1, designated as aO-1002, which was located in the membrane fraction, was extracted with 8 M urea and partially purified by gel filtration chromatography in the presence of SDS. When all-trans retinal was added, aO-1002 in dimyristoylphosphatidylcholine and detergent-mixed micelles was converted to a purple pigment with lambda max at 558 nm at 20 degrees C via a 435/460 nm intermediate. Conversion of the intermediate to purple pigment was the rate-limiting step and proceeded as a two-state transition, because an isosbestic point was seen at 485 nm. Similar spectral changes were also observed in the regeneration process of hydroxylamine-bleached claret membranes and aO-1 isolated from claret membranes. Thus, the polypeptide of aO-1002 is considered to fold and form a retinal binding pocket in phospholipid and detergent micelles similarly to aO-1 isolated from the halobacterial membranes. Purple pigment showed a light-driven proton-pumping activity when reconstituted into phosphatidylcholine liposomes.

Dual roles of DMPC and CHAPS in the refolding of bacterial opsins in vitro.

The bacterial opsins can be refolded to regenerate the chromophore by transfer from SDS to DMPC/CHAPS/SDS mixed micelles in the presence of retinal. A sequential refolding model has been proposed for bacterioopsin [Booth et al. (1995) Nature Struct. Biol. 2, 139-143]. However, the roles of DMPC and CHAPS in the refolding process are not clear. In this study we measured the effects of DMPC and CHAPS on the refolding of bacterial opsins in vitro by CD and fluorescence spectroscopy. In contrast to in experiments in the presence of large amounts of DMPC, the process of retinal binding pocket formation was a rate-determining step in overall chromophore regeneration with relatively low concentrations of DMPC. CHAPS triggered alpha-helix formation and long-range interactions between the helices within 1 s by providing a suitable hydrophobic environment for bacterial opsins. This CHAPS-induced transient molten globule-like structure would be identical to I1 postulated by Booth et al., to which DMPC bound and induced the proper packing of the side chains to form a retinal binding pocket. If DMPC was not present, CHAPS induced another conformation change in bacterial opsins, which led to denaturation. DMPC dependence of chromophore regeneration and the maintenance of the retinal binding pocket suggested that retinal binding pocket formation was part of the large structure changes during stable apoprotein formation.

Effect of digitonin on photophosphorylation and light-induced H+ uptake in isolated spinach chloroplasts.

The effects of the presence of digitonin in the reaction mixture on photophosphorylation, light-induced H+ uptake, and the size of isolated spinach chloroplasts were studied. Digitonin inactivated photosystem II and dissipated light-induced pH increase without affecting the photophosphorylation driven by photosystem I. Digitonin increased the concentration of NH4Cl required for uncoupling. This effect of digitonin was diminished by valinomycin which barely affects phosphorylation by itself. Although such properties have all been observed with digitonin subchloroplast particles, digitonin at the concentration required for acquiring such properties had little effect on the size of chloroplasts. These results suggest that the characteristic effects of digitonin subchloroplast particles are caused by thylakoid membranes penetrated by digitonin but not by the size of membrane vesicles. Both in chloroplasts with digitonin in the reaction mixture and in digitonin subchloroplast particles, N,N'-dicyclohexylcarbodiimide, purine nucleoside di- and tri-phosphates, and organic bases with low pKa values (pyridine and aniline) made the light-induced pH increase detectable, and this was lost upon the addition of proton ionophores. These results suggest that the lack of the light-induced pH increase by chloroplasts with digitonin in the reaction mixture and by digitonin subchloroplast particles is caused both by a membrane leaky to protons and by the loss of internal buffering capacity of thylakoids. Furthermore, the phosphorylation of chloroplasts with digitonin in the reaction mixture and of digitonin subchloroplast particles is considered to be driven by the proton motive force as is that of chloroplasts without digitonin.

A membrane-bound ATPase from Halobacterium halobium: purification and characterization.

An ATPase was newly identified on the inner face of the plasma membrane of the extremely halophilic archaebacterium Halobacterium halobium. The enzyme was released into an alkaline EDTA solution and purified by several chromatographic steps in the presence of sulfate at 1 M or over. The molecular weight of the native enzyme was around 320,000; it is most likely composed of two pairs (alpha 2 beta 2) of 86,000 (alpha) and 64,000 (beta) subunits. The enzyme hydrolyzed ATP and other nucleoside triphosphates but neither ADP nor AMP. The enzyme required divalent cations, among which Mn2+ was most effective (Mg2+ activated 35% of Mn2+). The ATPase activity was optimum at pH between 5.5 and 6, particularly in a nearly saturated Na2SO4 (or Na2SO3) solution, while it was very low in a chloride salt solution even at 4 M at any pH. The Km value for ATP was 1.4 mM and the K1 value for ADP (competitive to ATP) was 0.08 mM. Neither azide (a specific inhibitor for F0F1-and F1-ATPase) nor vanadate (for E1E2-ATPase) inhibited the enzyme. The ATPase was stable at high concentrations of sulfate. At low concentrations of salts, or at low temperatures even in high NaCl concentrations, the enzyme was inactivated. Although the ATPase isolated here from halobacterial membrane has such unusual characteristics, it is the most probable candidate for the (catalytic part of) halobacterial ATP synthase, which differs from F0F1-ATPase/synthase (Mukohata et al. (1986) J. Biochem. 99, 1-8; Mukohata and Yoshida (1987) J. Biochem. 101, 311-318).

The H+-translocating ATP synthase in Halobacterium halobium differs from F0F1-ATPase/synthase.

Cell envelope vesicles of Halobacterium halobium synthesize ATP by utilizing base-acid transition (an outside acidic pH jump) under optimal conditions (1 M NaCl, 80 mM MgCl2, pH 6.8) even in the presence of azide (a specific inhibitor of F0F1-ATPase) (Mukohata & Yoshida (1987) J. Biochem. 101, 311-318). An azide-insensitive ATPase was isolated from the inner face of the vesicle membrane, and shown to hydrolyze ATP under very specific conditions (1.5 M Na2SO4, 10 mM MnCl2, pH 5.8) (Nanba & Mukohata (1987) J. Biochem. 102, 591-598). This ATPase activity could also be detected when the vesicle components were solubilized by detergent. The relationship between ATP synthesis and the membrane-bound ATPase was investigated by modification of the vesicles with 7-chloro-4-nitrobenzo-2-oxa-1,3-diazole (NBD-Cl) or N-ethylmaleimide (NEM). The inhibition pattern of ATP synthesis in the modified vesicles and that of ATP hydrolysis of the solubilized modified vesicles were compared under the individual optimum conditions. The inhibition patterns were almost identical, suggesting that the ATP synthesis and hydrolysis are catalyzed by a single enzyme complex. The ATP synthase includes the above ATPase (300-320 kDa), which is composed of two pairs of 86 and 64 kDa subunits. This is a novel H+-translocating ATP synthase functioning in the extremely halophilic archaebacterium. This "archae-ATP-synthase" differs from F0F1-ATPase/synthase, which had been thought to be ubiquitous among all respiring organisms on our biosphere.

Acceleration of the ATPase activity of glycerol-treated muscle fibers by repeated stretch-release cycles.

Glycerol-treated muscle fiber bundles were fixed at their rest length in 50 mM KC1, 2 mM MgC1(2), and 10 micron CaC1(2) at pH 7.8 and 0 degrees C in the presence of sufficient amounts of ATP, creatine kinase, and creatine phosphate. The fiber bundles were stretched linearly with time for 0.3 s at a constant amplitude, suddenly released, then fixed at the rest length for a constant time interval (alpha seconds). The stretch-release cycle was repeated, and the ATPase activity (the rate of ADP liberation) [EC 3.6.1.3] was measured. It was found that: 1. ATPase was activated by repeated stretch-release. As repetitive stretch-release of 1--2% of the rest length caused maximum activation, we usually selected a value of 2.5% of the rest length. The activation of ATPase was found to be a function of the duration, alpha, of the isometric phase after sudden release from stretching. The ATPase activity of fiber bundles was almost unaffected when they were oscillated by a simple stretch-release without an isometric phase after the sudden release (alpha=0). 2. The ATPase activity of oscillated muscle fibers increased with increase in the value of alpha, reached a maximal level, then decreased gradually with further increase of alpha to a value slightly larger than that of static fibers. At 0 degrees C, the value of alpha for the maximum activation was observed at about 2 s, and the maximum activity was about 2.5 times that of static fibers. At 20 degrees C, the alpha value for maximum activation was about 0.5 s, and the maximum activity was about 1.8 times that of static fibers. 3. The time course of ADP liberation after one stretch-release cycle could be easily calculated from the ATPase activity of the summed durations of the isometric phase, alpha, assuming that the ATPase activation was turned off and on by the stretching and release, respectively, and that the state of cross-bridges immediately after the stretch-release was independent of alpha of the cycle. The rate of ADP liberation after stretch-release thus obtained showed a short lag phase, a sigmoidal increase, a decrease to almost zero, then a return to nearly the original level (the rate of static fibers). About 1.3 mol of ATP per mol of myosin was hydrolyzed at both 0 degrees C and 20 degrees C during one cycle of the changes in the rate of ADP liberation.

ATP synthesis in cell envelope vesicles of Halobacterium halobium driven by membrane potential and/or base-acid transition.

Cell envelope vesicles active in ATP synthesis were prepared from Halobacterium halobium cells, which genetically lack bacteriorhodopsin, by sonication in the presence of substrates. ATP was synthesized when vesicles were illuminated to build up membrane potential through the action of halorhodopsin. The threshold value of membrane potential for ATP synthesis was about -100 mV relative to the external medium, i.e., inside-negative. ATP synthesis also occurred in the dark upon acidification of the external medium of a suspension of cell envelope vesicles. This base-acid transition ATP synthesis took place when the pH difference was greater than 1.6 units. The threshold pH difference was lowered when the base-acid transition was carried out under dim light which induced a membrane potential of about -100 mV. Regardless of the sort of driving force, ATP synthesis was optimum at the intravesicular pH of around 6.5 and almost nil at 8, where ATP syntheses by F0F1 type ATPases in other organisms are most active. The synthesis could be inhibited by N,N'-dicyclohexylcarbodiimide (DCCD) with a half-maximum inhibition at around 25 microM/2 mg protein/ml. These results strongly suggest that in halobacteria a DCCD-sensitive H+-translocating ATP synthase is in operation which is driven by membrane potential and/or pH gradient, and obeys chemiosmotic energetics. The results also suggest that the ATP synthase may not be identical to F0F1 type H+-translocating ATPases found in mitochondria, chloroplasts and eubacteria.

A mechanism of respiratory control: studies with proteoliposomes containing cytochrome oxidase and bacteriorhodopsin.

Both beef heart cytochrome oxidase and bacteriorhodopsin of Halobacterium halobium were reconstituted into liposomes by the sonication-cholate dialysis method. The proteoliposomes showed the respiratory control ratio of 4.2, and steady-state illumination of the vesicles lead to the 2.7-fold stimulation of the oxidase activity in the absence of uncouplers. The light-stimulated state 4 respiration increased with light intensity, but light had no effect on the oxidase activity that had been relieved by addition of uncouplers. Proteoliposomes with the photosensitive oxidase activity were also obtained when cytochrome oxidase vesicles were fused with bacteriorhodopsin vesicles in the presence of calcium chloride, and the extent of photoactivation was maximally 1.4-fold. The light-induced respiratory release was observed even in the presence of valinomycin or nigericin, indicating that the oxidase activity was sensitive to both the membrane potential and the pH gradient. We propose as a mechanism of the respiratory control that the process of proton transport to the reaction center for water formation is the rate limiting step for the cytochrome oxidase activity.

The effect of carboxyl group modification on the chromophore regeneration of archaeopsin-1 and bacterioopsin.

Carboxyl group modification with DCCD and NCD-4 was employed to investigate the chemical environment of the side chains of archaeopsin-1 (aO-1) and bacterioopsin (bO). Some differences were observed between aO-1 and bO. Although DCCD or NCD-4 did not modify aO-1 in bleached membrane, they modified bO in bleached membrane and in mixed DMPC/CHAPS/SDS micelles at neutral pH, thereby affecting the opsin shift and the photocycle of the regenerated chromophore. On the contrary, after solubilization with SDS, aO-1 and bO were modified by DCCD and NCD-4, which decreased the chromophore regeneration. In particular, the reaction of aO-1 in SDS with NCD-4 proceeded in a 1:1 ratio at neutral pH. The fluorescence and CD spectra indicated that the modified site was located in the hydrophobic, asymmetrical region. Lysyl-endopeptidase digestion of NCD-4 modified aO-1 produced a fluorescent fragment and amino acid sequence analysis showed that Asp85 or Asp96 in helix C is a probable candidate for the modified residue at present. Kinetic CD measurements revealed that the introduction of N-acylurea at an Asp residue in helix C did not affect the formation of the transient intermediate but inhibited the side chain packing during refolding.

Halobacterial rhodopsins.

Following the discovery of the bacteriorhodopsin proton pump in Halobacterium halobium (salinarum), not only the halorhodopsin halide pump and two photosensor rhodopsins (sensory rhodopsin and phoborhodopsin) in the same species, but also homologs of these four rhodopsins in strains of other genera of Halobacteriaceae have been reported. Twenty-eight full (and partial) sequences of the genomic DNA of these rhodopsins have been analyzed. The deduced amino acid sequences have led to new strategies and tactics for understanding bacterial rhodopsins on a comparative basis, as summarized briefly in this article. The data discussed include (i) alignment of the sequences to qualify/characterize the conserved residues; (ii) assignment of residues that cause differences in function(s)/properties; and (iii) phylogeny of the halobacterial rhodopsins to suggest their evolutionary paths. The four kinds of rhodopsin in each strain are assumed, on the basis of their genera-specific distributions, to have arisen by at least two gene-duplication processes during evolution prior to generic speciation. The first duplication of the rhodopsin ancestor gene yielded two genes, each of which was duplicated again to give four genes in the ancestor halobacterium. The bacterium carrying four rhodopsin genes, after accumulating mutations, became ready for generic speciation and the delivery of four rhodopsins to each species. The original rhodopsin ancestor is speculated to be closest to the proton pump (bacteriorhodopsin).

Purification and properties of a peroxidase from Halobacterium halobium L-33.

A peroxidase was purified from Halobacterium halobium L-33 to an electrophoretically homogeneous state and some of its properties were studied. The enzyme showed an absorption peak at 406 nm in the oxidized form and peaks at 440, 558, and 591 nm in the reduced form. The difference spectrum, reduced + CO minus reduced, of the enzyme showed peaks at 425, 538, and 577 nm and troughs at 444, 562, and 596 nm. These spectral properties were apparently similar to those of "cytochrome a1" except for the occurrence of the peak at 558 nm in the reduced form. The molecular weight of the enzyme was 110,000 and the enzyme possessed one unit of protoheme in the molecule. The activity to oxidize guaiacol in the presence of H2O2 of the peroxidase was about one-twentieth of that of horseradish peroxidase. The enzyme also showed a catalase-activity one-fourth as active as that of liver catalase. The reactions catalyzed by the enzyme were strongly inhibited by KCN.

Archaebacterial ATPases: relationship to other ion-translocating ATPase families examined in terms of immunological cross-reactivity.

Immunological cross-reactivity among three types of H(+)-ATPases, that is, three archaebacterial ATPases, the F1-ATPase from thermophilic bacterium PS3 (TF1) and the vacuolar membrane ATPase from Saccharomyces cerevisiae, was examined by means of immunoblot analyses. The three archaebacterial ATPases were very similar in immunological cross-reactivity, suggesting that they belong to the same family of ATPases. Cross-reaction was also observed between the ATPase from Sulfolobus acidocaldarius, one of the three archaebacteria, and TF1. S. cerevisiae vacuolar ATPase reacted with the antibodies prepared against each of the three archaebacterial ATPases, but did not react with the antibody against TF1. Electron microscopic examination revealed that the oligomeric structure of Sulfolobus ATPase was very similar to that of F1-ATPase. These results, taken together, suggest that the archaebacterial ATPases share close structural similarities with the vacuolar ATPases, and, to a lesser degree, with the F0F1-ATPases.