Search

A strategy has been developed for archaebacterial lipid analysis which provides three times the information to describe archaebacterial isolates and is compatible with simultaneous eubacterial/eukaryotic lipid analysis of environmental samples. Eubacterial and micro-eukaryotic biomass, community structure, and nutritional status have been routinely defined in environmental samples by lipid analysis. Lipid profiles are also useful in eubacterial identification and taxonomy. Polar lipid or whole cell ester-linked fatty acids are generally analyzed by gas chromatography-mass spectroscopy. Archaebacteria are characterized by their ether-linked membrane lipids. There is, however, less diversity in the side chains of archaebacterial membrane lipids as compared the eubacterial ester-linked membrane lipids. The information content of the archaebacterial lipid profile was increased by separately analyzing the polar lipid, glycolipid, and lipid-extracted residue fractions. Identification and quantification were performed by supercritical fluid chromatography. Results are presented for three species of methanogens and four thermoacidophile isolates, and compared with a literature review.

Fast atom bombardment mass spectrometry was used to analyse intact polar ether lipids present at microgram levels in crude lipid mixtures extracted from Halobacterium halobium, Natronococcus occultus and Halobacterium marismortui. Negative-ion spectra showed the intact deprotonated lipid molecules and in some instances their sodium salts. The simplicity of the mass spectra permits the rapid screening of polar lipid mixtures for the presence of novel lipids. Additional structural information of ions with selected masses was obtained after collisionally induced decomposition.

Ribosomal RNA (rRNA) operons of the archaea reflect both the unity and the diversity of this third primary taxon. They have proven to be a rich source of both molecular biological and phylogenetic information.

Archaeoglobus fulgidus is an extremely thermophilic, sulphate reducing archaebacterium thought to represent a biochemical missing-link between sulphur-metabolizing bacteria and methanogenic bacteria. Whereas the phylogenetic position of A.fulgidus is closer to the sulphur-metabolizing bacteria, there is a partial overlap in the biochemical machinery of A.fulgidus with both groups of bacteria. In particular, the presence of a number of aberrant cofactors up to now thought to be involved exclusively in the process of methanogenesis in methanogenic archaebacteria, i.e. coenzyme F420, methanofuran and methanopterin, has been indicated by previous studies. Here we present evidence for the structural identity of the methanopterin cofactor of A.fulgidus with the methanopterin isolated from Methanobacterium thermoautotrophicum and show that this non-methanogenic bacterium contains two as yet unknown analogues of coenzyme F420. The levels of the various cofactors were determined in cultures grown either on formate or lactate as the carbon source and sulphate or thiosulphate as the sulphur source.

The outer surface of the archaebacterium Haloferax volcanii (formerly named Halobacterium volcanii) is covered with a hexagonally packed surface (S) layer. The gene coding for the S-layer protein was cloned and sequenced. The mature polypeptide is composed of 794 amino acids and is preceded by a typical signal sequence of 34 amino acid residues. A highly hydrophobic stretch of 20 amino acids at the C-terminal end probably serves as a transmembrane domain. Clusters of threonine residues are located adjacent to this membrane anchor. The S-layer protein is a glycoprotein containing both N- and O-glycosidic bonds. Glucosyl-(1----2)-galactose disaccharides are linked to threonine residues. The primary structure and the glycosylation pattern of the S-layer glycoproteins from Haloferax volcanii and from Halobacterium halobium were compared and found to exhibit distinct differences, despite the fact that three-dimensional reconstructions from electron micrographs revealed no structural differences at least to the 2.5-nm level attained so far (M. Kessel, I. Wildhaber, S. Cohe, and W. Baumeister, EMBO J. 7:1549-1554, 1988).

Two-dimensional electrophoresis of total protein from 50 S ribosomal subunits of the archaebacterium Sulfolobus solfataricus demonstrated a complex between two proteins that was stable in 6 M urea, but dissociable in detergent or below pH 5.5. The proteins, numbered L1 and L10 according to their electrophoretic mobilities, corresponded to Escherichia coli ribosomal proteins L10 and L7/L12, respectively. The members of the complex were therefore designated Sso L10e and Sso L12e. Sso L12e had other properties in common with E. coli L7/L12: low molecular weight, relative acidity, selective release from the ribosome by high salt/ethanol, and dimeric structure. The Sso L12e.Sso L10e complex was isolated by gel filtration of total 50 S proteins in 4 M urea. The stoichiometry of the components was approximately four copies of Sso L12e to one copy of Sso L10e. The occurrence in an archaebacterium of a complex of acidic ribosomal proteins similar to E. coli (L7/L12)4.L10 and eukaryotic (P1)2/(P2)/.P0 strongly supports the concept that this element of quaternary structure is a major conserved feature of the ribosome and reaffirms its importance in the translocation step of protein synthesis.

Methanogen chromosomal protein MC1 is a polypeptide of 93 amino acid residues (Mr 10,757) which represents the major protein associated with the DNA of the archaebacterium Methanosarcina barkeri and can protect DNA against thermal denaturation. The conformation of protein MC1 has been investigated by means of predictive methods, infrared spectroscopy, circular dichroism and tryptophan fluorescence studies. Protein MC1 has a low amount of alpha-helix but contains antiparallel beta-sheet strands. The larger hydrophobic cluster which contains tryptophan at position 61 appears buried in the protein. Addition of salts induces the unfolding of the protein and makes the tryptophan indole ring more rigid. With respect to its primary structure and its conformation, protein MC1 appears radically different from the chromosomal DNA-binding protein II (also called HU-type protein) in eubacteria.

We cloned and sequenced the gene coding for the polypeptide of a halorhodopsin in Natronobacterium pharaonis (named here pharaonis halorhodopsin). Peptide sequencing of cyanogen bromide fragments, and immunoreactions of the protein and synthetic peptides derived from the COOH-terminal gene sequence, confirmed that the open reading frame is the structural gene for the pharaonis halorhodopsin polypeptide. The flanking DNA sequences, as well as those for other bacterial rhodopsins, were compared to previously proposed archaebacterial consensus sequences. In pairwise comparisons of the open reading frame with DNA sequences for bacterio-opsin and halo-opsin from Halobacterium halobium, silent divergences (mutations/nucleotide at codon positions which do not result in amino acid changes) were calculated. These indicate very considerable evolutionary distance between each pair of genes. In spite of this, the three protein sequences show extensive similarities, indicating strong selective pressures. Conserved and conservatively replaced amino acid residues in all three proteins identify general features essential for ion-motive bacterial rhodopsins, responsible for overall structure and chromophore properties. Comparison of the bacteriorhodopsin sequence with those of the two halorhodopsins, on the other hand, identifies features involved in their specific (proton and chloride ion) transport functions.

The chromosomal protein MC1 of Methanothrix soehngenii is a family of three variants a, b, and c. These are small basic polypeptides of 89, 87, and 90 residues, respectively. Their primary structures have been determined from automated sequence analyses of the intact proteins and from structural data provided by peptides derived from the variants by cleavage at aspartic acid, glutamic acid, arginine, and methionine residues. By comparison with variant b taken as reference, variants a and c present 18 and 24 differences, respectively. The extent of sequence homologies between protein MC1 from M. soehngenii and proteins MC1 from two other species of Methanosarcinaceae is only 60%. The sequences 17-35 and 45-58 of the protein MC1 appear well conserved. Deletions are observed in region 36-44. Many changes, most of them nonconservative, occur in the carboxyl-terminal third of the protein. However, proline residues at positions 68, 72, 76, and 82 remain strictly conserved. Predictive methods for secondary structures indicate a low content of alpha helix and beta sheet structures in proteins MC1.

The amino acid sequence of the protein HMb, a protein of 93 residues (Mr 10757) which represents the major acid-soluble component of the Methanosarcina barkeri nucleoprotein complex, has been established from automated sequence analysis of the protein and from structural data provided by peptides derived from cleavage of the protein at aspartic acid, arginine and methionine residues. The protein HMb is mainly characterized by a high amount of charged residues (15% of acidic residues and 26.8% of basic residues) which are distributed all along the polypeptide chain. The amino acid sequence of the protein HMb is not homologous to any eubacterial, archaebacterial or eukaryotic chromosomal proteins known up to now.

The stereostructure of the archaebacterial C40 diol has been established as (3R,7R,11R,15S,18S,22R,26R,30R)-3,7,11,15,18,22,26,30- octamethyldotriacontane-1,32-diol by stereorational total synthesis. This provides the final evidence necessary to establish the structure of an archaebacterial membrane substance that is a 72-membered-ring tetraether with 18 stereocenters.

Analysis of the fluorescent compounds extracted from six different species of halobacteria and one species each of Sulfolobus and Thermoplasma revealed the universal occurrence of coenzyme F420, (N-[N-[O-[5-(8-hydroxy-5-deazaisoalloxazin-10-yl)-2,3,4-trihydroxy -4-pentoxyhydroxyphosphinyl]-L-lactyl]-L-gamma-glutamyl]-L -glutamic acid), or its gamma-monoglutamyl derivative or both. The total amount (approximately 100 pmol/mg [dry weight]) of these compounds found in the halobacteria studied was approximately 5% of the amount previously reported for methanogenic bacteria. The amount of F420 found in the Sulfolobus and Thermoplasma strains was approximately 1% of that found in the halobacteria. The major compound in all but one of the examined strains was the gamma-monoglutamyl derivative of F420; one strain of halobacteria contained only F420. For the halobacterium-derived samples, the additional glutamic acid was shown to be linked by a gamma-glutamyl peptide bond to the terminal glutamic acid of the F420 core structure by enzymatic hydrolysis of the samples with three different gamma-glutamyltranspeptidases. The product of this enzymatic hydrolysis was F420 with one less glutamic acid in the side chain.

The Archaebacterium Thermoplasma acidophilum contains a basic chromosomal protein remarkably similar to the histones of eukaryotes. Therefore, it was of interest to examine a different Archaebacterium for similar proteins. We chose to examine Sulfolobus acidocaldarius because it is thermophilic, like T. acidophilum, but nevertheless the two organisms are not particularly closely related. Two major chromosomal proteins were found in S. acidocaldarius. The smaller of these was soluble in 0.2 M H2SO4 and had a molecular weight of 14500. The larger was acid-insoluble and had a molecular weight of about 36000. Together, the proteins protected about 5% of the DNA against nuclease digestion and stabilized about 50% against thermal denaturation. Overall, the properties of these proteins were intermediate between those of the Escherichia coli protein HU and T. acidophilum protein HTa.

Cell extracts of methanogens and the thermoacidophile Sulfolobus solfataricus contained little or no folic acid (pteroylglutamate) or pteroylpolyglutamate activity (less than 0.1 nmol/g [dry weight]). However, the halophile Halobacterium salinarum contained pteroylmono- or pteroyldiglutamates, and Halobacterium volcanii and Halobacterium halobium contained pteroyltriglutamates at levels equivalent to those in eubacteria (greater than 1 nmol/g [dry weight]).

We describe in this work the structure and polymorphism of a variety of lipids extracted from Sulfolobus solfataricus, an extreme thermoacidophilic archaebacterium growing at about 85 degrees C and pH 2. These lipids are quite different from the usual fatty acid lipids of eukaryotes and prokaryotes: each molecule consists of two C40 omega-omega' biphytanyl residues (with 0 to 4 cyclopentane groups per residue), ether linked at both ends to two (variably substituted) glycerol or nonitol groups. Four lipid preparations were studied; the total and the polar lipid extracts, and two hydrolytic fractions, the symmetric glycerol dialkyl glycerol tetraether and the asymmetric glycerol dialkyl nonitol tetraether, as a function of water content and temperature, using X-ray scattering techniques. The main conclusions from the study of the four lipid preparations can be summarized as follows. (1) As with other lipids, a remarkable number and variety of phases are observed over a temperature-concentration range close to "physiological" conditions. The possibility is discussed that this polymorphism reflects a fundamental property of lipids, closely related to their physiological rôle. (2) As in other lipids, two types of chain conformations are observed: a disordered one (type alpha) at high temperature; at lower temperature, a more ordered packing of stiff chains, all parallel to each other (type beta'). At temperatures and degrees of hydration approaching the conditions prevailing in the living cell, the conformation is of type alpha. (3) In all the phases with chains in the alpha conformation, the unsubstituted glycerol headgroups, whose concentration is high in these lipids, segregate in the hydrocarbon matrix, away from the other polar groups. This property may have interesting biological consequences: for example, the chains of a fraction of the bipolar lipid molecules can span hydrocarbon gaps as wide as 75 A. (4) Two cubic phases are observed in the total and the polar lipid extracts, which display a remarkable degree of metastability, most unusual in lipid phase transitions involving structures with chains in the alpha conformation. This phenomenon can be explained by the interplay of the physical structure of the cubic phases (the two contain two intertwined and unconnected three-dimensional networks of rods) and the chemical structure of the lipid molecules: the two headgroups of most molecules being anchored on each of the two networks of rods, the migration of the lipid molecules is hindered by the two independent diffusion processes and by the entanglement of the chains.(ABSTRACT TRUNCATED AT 400 WORDS)

The most commonly accepted secondary structure models for 5S RNA differ for molecules of eubacterial origin, where the four-helix model of Fox and Woese is generally cited, and those of eukaryotic origin, where a fifth helix is assumed to exist. We have carefully aligned all available sequences from eukaryotes, eubacteria, chloroplasts, archaebacteria and plant mitochondria. We could thus derive a unified secondary structure model applicable to all 5S RNA sequences known to-date. It contains the five helices already present in the eukaryotic model, extended by additional segments that were not previously assumed to be universally present. One of the helices can be written in two equilibrium forms, which could reflect the existence of a flexible, dynamic structure. For the derivation of the model and the estimation of the free energies we followed a set of rules optimized to predict the tRNA cloverleaf. The stability of the unified model is higher than that of nearly all previously proposed sequence-specific and general models.

1. The ribosomal subunits of one thermoacidophilic archaebacterium (Caldariella acidophila) and of two reference eubacterial species (Bacillus acidocaldarius, Escherichia coli) were compared with respect to ribosome mass and protein composition by (i) equilibrium-density sedimentation of the particles in CsCl and (ii) gel-electrophoretic estimations of the molecular weights of the protein and the rRNA. 2. By either procedure, it is estimated that synthetically active archaebacterial 30S subunits (52% protein by wt.) are appreciably richer in protein than the corresponding eubacterial particles (31% protein by wt.) 3. The greater protein content of the archaebacterial 30S subunits is accounted for by both a larger number and a greater average molecular weight of the subunit proteins; specifically, C. acidophila 30S subunits yield 28 proteins whose combined mass is 0.6 X 10(6) Da, compared with 20 proteins totalling 0.35 X 10(6) Da mass for eubacterial 30S subunits. 4. No differences in protein number are detected among the large subunits, but C. acidophila 50S subunits exhibit a greater number-average molecular weight of their protein components than do eubacterial 50S particles. 5. Particle weights estimated by either buoyant-density data, or molecular weights of rRNA plus protein, agree to within less than 2%. By either procedure C. acidophila 30S subunits 1.15 X 10(6) Da mass) are estimated to be about 300 000 Da heavier than their eubacterial counterparts (0.87 X 10(6) Da mass); a smaller difference. 0.15 X 10(6) Da, exists between the archaebacterial and the eubacterial 50S subunits (respectively 1.8 X 10(6) and 1.65 X 10(6) Da). It is concluded that the heavier-than-eubacterial mass of the C. acidophila ribosomes resides principally in their smaller subunits.

DNA-binding proteins in eubacteria, such as Escherichia coli NS1 and NS2, are generally small basic molecules. In contrast, the archaebacterium Sulfolobus acidocaldarius contains three groups of DNA-binding proteins which have molecular masses of 7, 8, and 10 kDa. In the first group, five proteins (7a-7e) have been identified, while in the second and third group only two proteins each are present, denoted 8a and 8b and 10a and 10b, respectively. In this paper, we present the primary structures of proteins 7a, 7b, and 7e from the first group. All three proteins contain lysyl residues which are monomethylated to different extents. The modified lysines are found in the NH2-terminal regions of all 7-kDa proteins and in the COOH-terminal part of protein 7e. The sequences of the 7-kDa group are highly similar to each other. All of these macromolecules have been shown to interact specifically with DNA. Protein 7e of the 7-kDa group shows the tightest binding to DNA.

The spherical cells of the thermophilic, sulfur-dependent archaebacterium Desulfurococcus mobilis are completely covered with a relatively poorly ordered, tetragonally arrayed surface protein. The structure of this surface protein was examined by using three-dimensional electron microscopy. The protein lattice forms an open meshwork composed of cross-shaped morphological units, which are released when glycerol is added. These subunits make contact at the distal ends of their four arms. The p4 symmetry requires that each of these morphological subunits represents a tetramer. The strong interaction of the monomers within the crosses and the relatively weak interaction of the intersecting arms of the crosses within the lattice structure suggest that the tetramers are assembled before their incorporation into the lattice.

We examined the nature of the diphtheria toxin fragment A recognition site in the protein synthesis translocating factor present in cell-free preparations from the archaebacteria Thermoplasma acidophilum and Halobacterium halobium. In agreement with earlier work (M. Kessel and F. Klink, Nature (London) 287:250-251, 1980), we found that extracts from these organisms contain a protein factor which is a substrate for the ADP-ribosylation reaction catalyzed by diphtheria toxin fragment A. However, the rate of the reaction was approximately 1,000 times slower than that typically observed with eucaryotic elongation factor 2. We also demonstrated the presence of diphthine (the deamidated form of diphthamide, i.e., 2-[3-carboxyamide-3-(trimethylammonio)propyl]histidine) in acid hydrolysates of H. halobium protein in amounts comparable to those found in hydrolysates of similar preparations from eucaryotic cells (Saccharomyces cerevisiae and HeLa). Diphthine could not be detected in hydrolysates of protein from the eubacterium Escherichia coli. Whereas both archaebacterial and eucaryotic elongation factors contain diphthamide, they differ importantly in other respects.

In the deoxyribonucleoprotein complex of Methanosarcina sp. CHTI 55, DNA is associated with two proteins, named MC1 (methanogen chromosomal protein 1) (Mr 10,760) and MC2 (Mr 17,000). Protein MC1, the most abundant of these proteins, is closely related to the Methanosarcina barkeri MS protein MC1. The effect of Methanosarcina sp. CHTI 55 protein MC1 on the thermal stability of DNA has been studied in native deoxyribonucleoprotein complex, as well as in reconstituted complexes, and it has been compared to the effect of E. coli DNA-binding protein II. Both proteins are able to protect DNA against thermal denaturation, but the differences observed in the melting profiles suggest that they interact by different mechanisms. Moreover, our studies indicate that one molecule of protein MC1 protects eight base pairs of DNA.

The thermoacidophilic archaebacterium Sulfolobus solfataricus possesses several DNA binding proteins which may have a histone-like function. Two particularly dominant species have molecular masses of 7 and 10 kDa, respectively. We have purified one of the small proteins which occurs in relatively large amount and have determined its amino acid sequence. The protein is characterized by a high lysine content; in the N-terminal region the lysine residues occur in an alternating order: X-K-X-K-X-K-X-K. The amino acid sequence does not indicate any obvious homology to those DNA binding proteins whose sequences have been determined.

The flagella of Methanococcus voltae were isolated by using three procedures. Initially, cells were sheared to release the filaments, which were purified by differential centrifugation and banding in KBr gradients. Flagella were also prepared by solubilization of cells with 1% (vol/vol) Triton X-100 and purified as described above. Both of these techniques resulted in variable recovery and poor yield of flagellar filaments. Purification of intact flagella (filament, hook, and basal body) was achieved by using phase transition separation with Triton X-114. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of purified flagella revealed two major proteins, with molecular weights of 33,000 and 31,000. This result indicates the likely presence of two flagellins. The filament had a diameter of 13 nm. The basal structure consisted of a small knob, while a slight thickening of the filament immediately adjacent to this area was the only evidence of a hook region. Flagella from three other Methanococcus species were isolated by this technique and found to have the same ultrastructure as flagella from M. voltae. Isolation of flagella from three eubacteria and another methanogen (Methanospirillum hungatei [M. hungatii]) by the phase separation technique indicated that the detergent treatment did not affect the structure of basal bodies. Intact ring structures and well-differentiated hook regions were apparent in each of these flagellar preparations.

The structure of several eubacterial and archaebacterial surface (glyco)proteins as determined by three-dimensional electron microscopy is described. Particular emphasis is placed on surface proteins which interact with membranes. Some structure-function relationships deduced from the structural information, such as shape maintenance and molecular recognition phenomena, are discussed.

The membrane of thermophilic archaebacteria is characterized by the presence of unusual isoprenoid bipolar lipids. The molecular organization of these lipids is still a matter of study. Important information could come from forming artificial black membranes. Black films can be formed from n-alkane or squalene dispersions of bipolar lipids extracted from the membrane of Caldariella acidophila. Membrane formation occurred only above a critical temperature (approximately 70 degrees C) corresponding to the physiological one. At lower temperatures, special solvent systems (n-alkanes or squalene, butanol and n-alkanes or squalene, butanol chloroform) were required. To characterize the physical parameters of these membranes, conductance and capacitance measurements were performed. Conductance was in the range of 10(-8) - 10(-7) omega -1 cm -2 , where specific capacitance at T = 72 degrees C was Cs = 0.685 +/- 0.004 microF/cm2 and Cs = 0.658 +/- 0.08 microF/cm2, corresponding to a dielectric thickness of 27 and 29 A for squalene and dodecane dispersions, respectively. Capacitance was shown to vary as the square of membrane potential, as usual in lipid bilayers. Values of the proportionality constant alpha have been compared to those of solvent-containing and solvent-free bilayers. The behavior of capacitance as a function of temperature is also shown by lowering temperature; the occurrence of complex structural changes was indicated. All the experimental data suggest that the presence of solvent is very low. Two possible molecular configurations of the films are discussed.

From the sulphur-dependent, anaerobically grown archaebacterium Sulfolobus ambivalens Caldariella quinone, CQ-6(12H) and the new Sulfolobus quinone SQ-6(12H), 6-(3,7,11,15,19,23-hexamethyltetracosyl)-5-methyl-benz[b]thioph en-4, 7-quinone have been isolated as main components. Lower homologues SQ-5-(10H), SQ-4(8H), SQ-3(6H), phylloquinone-like species CQ-6(10H), SQ-6(10H) and the menaquinone MK-6(12H) are present as minor components. The results are compared with those from Sulfolobus acidocaldarius. Thermococcus celer, Desulfurococcus mucosus and Desulfurococcus mobilis do not contain quinones in comparable amounts.

A reproducible high performance liquid chromatography (HPLC) method for the separation of diethers and tetraethers isolated from archaebacterial phospholipids is reported. Fourier transform infrared spectroscopy was used for structural confirmation of these signature lipids. A mixture of tetraethers from a thermoacidophilic archaebacteria was resolved into three major components by the normal phase separation. These components were differentiated by Fourier self-deconvolution of infrared spectra. The application of the HPLC technique to environmental samples may provide an accurate assessment of archaebacterial biomass in various microbial communities.

Bacillus stearothermophilus large ribosomal subunits were reconstituted in the presence of 5S rRNAs from different origins and tested for their biological activities. The results obtained have shown that eubacterial and archaebacterial 5S rRNAs can easily substitute for B. stearothermophilus 5S rRNA in the reconstitution, while eukaryotic 5S rRNAs yield ribosomal subunits with reduced biological activities. From our results we propose an interaction between nucleotides 42-47 of 5S rRNA and nucleotides 2603-2608 of 23S rRNA during the assembly of the 50S ribosomal subunit. Other experiments with eukaryotic 5.8S rRNAs reveal, if at all, a very low incorporation of these RNA species into the reconstituted ribosomes.

The isoelectric points of most proteins from the extremely halophilic archaebacteria are between 4.0 and 4.65 which agrees with the generally high content of glutamic and aspartic acid in proteins from halobacteria. The subunits from two purified halobacterial membrane enzymes (ATPase and nitrate reductase) behaved differently with respect to isoelectric focusing, silver staining and interaction with ampholytes. Differential behavior was also observed in whole cell proteins from Halobacterium saccharovorum regarding resolution in two-dimensional gels and silver staining. We propose that these differences reflect the existence of two classes of halobacterial proteins, one resembling non-halophilic proteins, and the other possessing unique properties that may be related to salt dependence.

The microorganisms inhabiting a 91 degrees C hot spring in Yellowstone National Park were characterized by sequencing 5S rRNAs isolated from the mixed, natural microflora without cultivation. By comparisons of these sequences with reference sequences, the phylogenetic relationships of the hot spring organisms to better characterized ones were established. Quantitation of the total 5S-sized rRNAs revealed a complex microbial community of three dominant members, a predominant archaebacterium affiliated with the sulfur-metabolizing (dependent) branch of the archaebacteria, and two eubacteria distantly related to Thermus spp. The archaebacterial and the eubacterial 5S rRNAs each constituted about half the examined population.

From Sulfolobus solfataricus, a sulfur-oxidizing thermophilic member of archaebacteria, three unusual benzothiophenquinones were isolated. Detailed NMR studies on these quinones, including multipulse mono-dimensional and two-dimensional techniques, were performed to obtain carbon and proton assignments, one-bond, geminal and vicinal coupling constants and T1 relaxation times. This report extends the known quinone composition of thermophilic archaebacteria and further supports the concept that these biomolecules can serve as a useful chemotaxonomic tool.

The amino acid sequence of a ferredoxin from a thermoacidophilic archaebacterium, Sulfolobus acidocaldarius, was determined by a combination of various conventional methods to be as follows: Gly-Ile-Asp-Pro-Tyr-Arg-Thr-His-Lys-Pro-Val-Val-Gly-Asp-Ser-Ser-Gly-His- Lys-Ile -Tyr-Gly-Pro-Val-Glu-Ser-Pro-Lys(Me)-Val-Leu-Gly-Val-His-Gly-Thr-Ile-Val -Gly-Va l-Asp-Phe-Asp-Leu-Cys-Ile-Ala-Asp-Gly-Ser-Cys-Ile-Thr-Ala-Cys-Pro-Val-As n-Val-P he-Gln-Trp-Tyr-Glu-Thr-Pro-Gly-His-Pro-Ala-Ser-Glu-Lys-Lys-Ala-Asp-Pro-V al-Asn- Glu-Gln-Ala-Cys-Ile-Phe-Cys-Met-Ala-Cys-Val-Asn-Val-Cys-Pro-Val-Ala-Ala- Ile-Asp -Val-Lys-Pro-Pro. It was composed of 103 amino acid residues giving a molecular weight of 10,908 excluding Fe and S atoms. This ferredoxin contained an N6-monomethyllysine residue at position 29 which was determined by a comparison of the elution profile of the acid hydrolysates of the protein and peptides on an amino acid analyzer with three methyl derivatives of lysine and also by field desorption mass spectrometry of a purified peptide. The ferredoxin has only 7 cysteine residues, which probably participate in constructing the Fe-S clusters of this ferredoxin, indicating the presence of a unique chelate structure. Comparison of this ferredoxin with other archaebacterial ferredoxins indicated that the archaebacteria might have multiple origins in an evolutionary tree.

From the archaebacterium Thermoproteus tenax, strain Kra-1 a mixture of 6 quinones of menaquinone and phylloquinone type with isopentyl side chains, MK-6(12H), MK-6-(10H), MK-5(10H), MK-5(8H), MK-4(8H), MK-4(6H) and two analogous quinones, containing in addition a methyl group in the naphthoquinone system, were isolated and characterized.

The structure of the major fluorescent pterin present in thermophilic archaebacterium Sulfolobus solfataricus has been assigned, by analysis of the intact molecule and its hydrolytic and periodate cleavage products, as erythro-neopterin-3'-D-2-deoxy-2-aminoglucopyranoside. The trivial name solfapterin is proposed for this compound.

A modified procedure for extraction and purification of hydrolyzed archaebacterial lipids is described. Lipids were extracted from Sulfolobus acidocaldarius using a Soxhlet extraction procedure followed by trichloroacetic acid solvent-extraction of the residue. The yield of total extractable material by this protocol was 14% which, after a two-phase wash, yielded 10% lipid. Modifications to the published steps for purifying the subsequently hydrolyzed lipids were developed to purify glycerol dialkyl nonitol tetraether (GDNT). The nearly colorless final macrocyclic product was characterized by TLC, IR, NMR, and mass spectrometry.

Cyclic AMP was found in species representative of the three major groups of the archaebacteria. In Methanobacterium thermoautotrophicum starvation for H2 led to a significant increase in cellular cAMP. The findings suggest that the occurrence of cAMP antedates the divergence of the major kingdoms of biology; the observations also imply that cAMP constitutes a very early regulatory molecule.

DNA-binding proteins have been extracted from the thermoacidophilic archaebacterium Sulfolobus solfataricus strain P1, grown at 86 degrees C and pH 4.5. These proteins, which may have a histone-like function, were isolated and purified under standard, non-denaturing conditions, and can be grouped into three molecular mass classes of 7, 8 and 10 kDa. We have purified to homogenity the main 7 kDa protein and determined its DNA-binding affinity by filter binding assays and electron microscopy. The Stokes radius of gyration indicates that the protein occurs as a monomer. The complete amino-acid sequence of this protein contains 14 lysine residues out of 63 amino acids and the calculated Mr is 7149. Five of the lysine residues are partially monomethylated to varying extents and the methylated residues are located exclusively in the N-terminal (positions 4 and 6) and the C-terminal (positions 60, 62 and 63) regions only. The protein is strongly homologous to the 7 kDa proteins of Sulfolobus acidocaldarius with the highest homology to protein 7d. Accordingly, the name of this protein from S. solfataricus was assigned as DNA-binding protein Sso7d.

A detailed analysis of the folate coenzymes in the nonmethanogenic archaebacteria has been performed. By using the Lactobacillus casei microbiological assay for folates, the levels of folates in Sulfolobus solfataricus and Sulfolobus acidocaldarius were found to be 3.7 and 8.3 ng/g (dry weight) of cells, respectively, compared with 88,000 and 28,000 ng/g (dry weight) of cells in Halobacterium halobium and Halobacterium strain GN-1, respectively. The levels of folates found in the Sulfolobus spp. were approximately 100 times less than those found in the typical eubacterium, whereas the levels in the halobacteria were approximately 10 times higher. The folate in Sulfolobus solfataricus was shown to consist of only 5-formyltetrahydropteroylglutamate, and the folate in Halobacterium strain GN-1 was shown to consist of only pteroyldiglutamate. The low folate levels in the Sulfolobus spp. are the same as those found in the methanogenic bacteria, suggesting that another C1 carrier may function in these cells.

Four DNA binding histone-like proteins have been purified from the nucleoid of the acidothermophilic archaebacterium Sulfolobus acidocaldarius to homogeneity employing DNA-cellulose chromatography and carboxymethylcellulose chromatography. The molecular weights of these proteins are in the range 8,000-12,500. Immunoblotting results suggest that a few antigenic determinants are common among these proteins which could not be detected by immunodiffusion. Spectroscopic properties of the proteins have been studied. The amino acid compositions of these proteins show both similarities and differences with histones and prokaryotic histone-like proteins. All of the four proteins bind native and denatured DNAs and single stranded RNA with differing affinities. Three of the proteins, denoted by HSNP (helix stabilizing nucleoid protein)-A, HSNP-C, and HSNP-C', show physiologically significant, strong, and synergistic effects in stabilizing duplex DNA against thermal denaturation with Tm increases in the range of 15-30 +/- degrees C.

Archaebacterial glycerol-bisdiphytanyl-glycerol tetraether core lipids containing from one to eight cyclopentane rings could be resolved from each other and from the parent uncyclized C40, C40 lipid by TLC. The core lipids of examples from the genera Methanobacterium, Methanobrevibacter, Methanogenium and Methanoplanus did not contain cyclized forms of glycerol-bisdiphytanyl-glycerol tetraethers, whereas the core lipids of Methanosarcina barkeri contained glycerol-bisdiphytanyl-glycerol tetraethers with from one to three cyclopentane rings in each C40 isopranoid chain.

As a continuation of an X-ray scattering study of the tetraether lipids extracted from the thermophilic archaebacterium Sulfolobus solfataricus, the phase behaviour of four fractions of the complex polar lipid extract (PLE) is described. Each molecule of two of these fractions (P1 and GL) carries an unsubstituted glycerol headgroup, those of another (P2) no such group; the fourth fraction (WPLE) is obtained by water-washing PLE, thus reducing its P2 content from approximately 48% to approximately 24% and increasing the average number of molecules bearing an unsubstituted glycerol headgroup from approximately 0.4 to approximately 0.6. The main result is a striking correlation between the phase behaviour and the average ratio of unsubstituted glycerol headgroups to the total number of headgroups: the fractions P1, GL and WPLE, in which that number is respectively 0.5, 0.5 and 0.3, form rod-containing phases; the fraction P2, in which that number is zero, yields a lamellar phase throughout the phase diagram. An analysis of the dimensions of the structure elements confirms our previous conclusion that, in the presence of a sufficient amount of water, the unsubstituted glycerol headgroups partition preferentially in the hydrocarbon regions rather than at the polar/apolar interfaces. These results, moreover, corroborate our previous conjectures regarding the correlations between the structure of the plasma membrane, the phase behaviour of the lipid extract and life at high temperature.

A method for preparing glycerophosphoesters from ether phospholipids by dealkylation with boron trichloride (BCl3) is described. Treatment of ether phospholipids in chloroform with BCl3 for 30 min at room temperature yielded almost quantitatively the corresponding glycerophosphoesters retaining the intact polar head group of the ether phospholipids. Thus, glycerophosphocholine, glycerophosphoinositol, glycerophosphoglycerol, glycerophosphoserine, glycerophosphate, and glycerophosphoethanolamine were prepared from the diether analogs of phosphatidylcholine, phosphatidylinositol, phosphatidylglycerol, phosphatidylserine, and phosphatidic acid, and the tetraether analog of phosphatidylethanolamine, respectively. BCl3 also cleaved diacyl, alkyl-acyl, and alk-1-enyl-acyl forms of phospholipids to yield corresponding glycerophosphoesters. The glycerophosphoesters were separated more rapidly by cellulose thin-layer chromatography with the same solvent system as in paper chromatography. This method is of great use for structure determination of glycerophosphoester backbones of ether phospholipids, analogous to the mild alkaline methanolysis of diacyl form of phospholipids, as well as for the analysis of alkyl chains. It is, however, not applicable to glycolipids because of cleavage of glycosidic bonds by BCl3.

The glycerol diether phospholipids of 25 monocultures of methanogenic bacteria were isolated and degraded with hydriodic acid. The resulting alkyl iodides were converted to acetate esters and alcohols which were examined using capillary gas-liquid chromatography. The presence of C20 phytanol was observed in accordance with previous studies. Soft fragmentation by chemical ionization mass spectrometry combined with selected ion monitoring enabled the detection, for the first time, of C15 and C25 isoprenologues as components of the diether phospholipids in several strains.

Indole-3-acetic acid (IAA) and 2-(indol-3-ylmethyl)indol-3-yl acetic acid were identified in lipid extracts of Sulfolobus acidocaldarius; they occurred at concentrations of 0.57 and 0.59 mumol/g (dry weight), respectively. The amount of IAA found in these cells is more than a thousand times greater than that found in a typical extract of a plant in which IAA serves as a plant growth hormone. Neither of these compounds was detected in the other archaebacteria that were analyzed; these included Sulfolobus sulfataricus, Halobacterium salinarium, and several strains of methanogenic bacteria. This is the first report of the natural occurrence of 2-(indol-3-ylmethyl)indol-3-yl acetic acid.

The basic assumption in this paper is that the secondary structure of a 5-S ribosomal RNA cannot be represented by a single model. We propose that the molecule can adopt, at least within the ribosome, a series of slightly different structures of nearly equal stability. The different structures arise from the existence of ambiguous base-pairing opportunities in bulged helices and the adjacent interior loops. In eubacterial 5-S RNAs there is one such an area, in eukaryotic 5-S RNAs two such areas that can give rise to structural switches. We explain how a change in secondary structure in these areas may influence the relative orientation of the surrounding helices, in other words how bulges and interior loops may serve as articulations and give rise to a flexible tertiary structure.