Search

Structures of the N-linked oligosaccharides of a recombinant soluble form of human CD4 glycoprotein (sCD4) have been investigated by enzymic microsequencing. The glycoprotein has two N-glycosylation sites, Asn271 and Asn300, at both of which evidence for the presence of complex type biantennary sialo-oligosaccharides has been obtained previously by mass spectrometric analyses [Carr, S.A., Hemling, M.E., Folena-Wasserman, G., Sweet, R.W., Anumula, K., Barr, J.R., Huddleston, M.J. & Taylor, P. (1989) J. Biol. Chem. 264, 21,286-21,295]. Among oligosaccharides released from sCD4 by hydrazinolysis and labelled with NaB3H4, neutral (12.8%) and acidic (87.2%) oligosaccharides were detected by paper electrophoresis. The latter were rendered neutral following sialidase treatment indicating that acidity was due exclusively to the presence of sialic acid residues. By enzymic microsequencing of the sialidase-treated oligosaccharides (fractionated on affinity columns of Ricinis communis agglutinin 120 and concanavalin A) in conjunction with methylation data from the earlier study, 14 sequences were identified. These accounted for over 80% of the sialidase-treated oligosaccharides of sCD4 as follows: [formula: see text] where +/- indicates residues present on only a proportion of chains. The spectrum of oligosaccharide structures released from each glycosylation site was assessed as being similar to that of total oligosaccharides on the basis of their chromatographic profiles on the lectin columns and on Bio-Gel P-4.

The N-linked oligosaccharides were released from purified human intercellular adhesion molecule (ICAM)-3 by hydrazinolysis. Approximately 6 mol of oligosaccharides were released from 1 mol of ICAM-3. The oligosaccharides reduced with NaB[3H]4 were separated into neutral and acidic fractions by paper electrophoresis. Most of the acidic oligosaccharides were converted to neutral ones by digestion with sialidase, indicating that they are sialyl derivatives. The neutral and sialidase-treated acidic oligosaccharides were fractionated by serial lectin column chromatography followed by Bio-Gel P-4 column chromatography. Structural studies of each oligosaccharide by sequential exo- and endo-glycosidase digestion and by methylation analysis revealed that N-linked oligosaccharides of ICAM-3 are mainly of tri- and tetra-antennary complex-type, about 60% of which contain two to three poly N-acetyllactosamine chains terminated with the type 1 structure and those without the type 1 structure per oligosaccharide. In addition, a small amount of the high mannose-type oligosaccharide with six alpha-mannose residues, which could act as a ligand for the dendritic cell-specific ICAM-3 grabbing nonintegrin, was detected.

Cell-free extracts from the oleandomycin producer, Streptomyces antibioticus, possess an intracellular glycosyltransferase capable of inactivating oleandomycin by glycosylation of the 2'-hydroxyl group in the desosamine moiety of the molecule [Vilches, C., Hernandez, C., Mendez, C. & Salas, J. A. (1992) J. Bacteriol. 174, 161-165]. Using a four-step purification procedure, we have purified an enzyme activity from the culture supernatants from this organism which is able to release glucose from the inactive glycosylated molecule thus reactivating the antibiotic activity. This enzyme activity appeared in the culture supernatants immediately before oleandomycin is detected. The enzyme (molecular mass 87 kDa) showed a high degree of substrate specificity, not acting on other glycosylated macrolides such as methymycin, lankamycin and rosaramicin which are substrates for the glycosyltransferase. A second activity was detected corresponding to a 34-kDa polypeptide which probably originates from proteolytic cleavage of the larger polypeptide. The 87-kDa polypeptide possibly catalyses the last biosynthetic step in oleandomycin biosynthesis by S. antibioticus.

3-Phosphonoalanine has been made by the Strecker synthesis from phosphonoacetaldehyde, which is easily prepared from vinyl acetate. It gives phosphonopyruvate by transamination when treated with glyoxylate. Phosphonolactate, an analogue of phosphoglycerate, is prepared by reducing phosphonopyruvate. Diazotization of phosphonoalanine was investigated as a route for making phosphonolactate: addition of NaNO2 to the isoelectric form of phosphonoalanine gave much scission of the C-P bond with release of phosphate; addition of HBr prevented this release and gave largely the bromo acid. The supplement reports the synthesis of arsonolactate, a similar analogue, by treating chlorolactate with alkaline arsenite.

The complete primary structures of the Asn-linked oligosaccharides from the conserved glycosylation site of the type-I variant surface glycoproteins of Trypanosoma brucei MITat 1.4 and MITat 1.6 were determined using a combination of exoglycosidase digestions, permethylation analysis, acetolysis and 1H NMR. Both variants contained almost exclusively oligomannose-type oligosaccharides, identical in structure to those of mammalian glycoproteins. The oligosaccharides ranged in size from (Man)9(GlcNAc)2 to (Man)5(GlcNAc)2. The relative abundance of each component was similar in both variants. The major components were (Man)8(GlcNAc)2 and (Man)7(GlcNAc)2 with slightly less (Man)9(GlcNAc)2 and (Man)6(GlcNAc)2 and much less (Man)5(GlcNAc)2. Both variants also contained the same structural isomers. The close similarity of the oligomannose series indicates identical processing at the conserved site in both variants.

The disulphide linkages of the 16 half-cystine residues in the sweet-tasting protein thaumatin have been investigated by enzymatic hydrolysis of the intact molecule. The peptides obtained after proteolytic cleavage with trypsin and pepsin, and in one case with chymotrypsin have been purified by gel filtration, high-performance liquid chromatography and peptide mapping by paper high-voltage electrophoresis in one direction and paper chromatography in the second dimension. Disulphide bonds appeared to be formed by cysteine residues in positions 9-204, 56-66, 71-77, 121-193, 126-177, 134-149, 145-158 and 159-164. The labile disulphide bond responsible for the enzymatic properties of the sweet tasting protein thaumatin appeared to be between Cys-145 and Cys-158.

The incorporation of radioactivity in vitro from UDP-[14C]glucose into endogenous acceptors in a crude enzymatic preparation from Acetobacter xylinum is described. A fraction of the total incorporated radioactivity could be extracted with organic solvents, like butanol. The remainder was mostly solubilized by hot alkali, leaving an insoluble fraction. No incorporation was observed in the absence of magnesium ions. Mn2+ could only partially replace Mg2+, and had an inhibitory effect when both ions were present. Analysis of the butanol-soluble fraction permitted the isolation of at least three components characterized as: lipid-diphosphate α-glucose, lipid-diphosphate α-cellobiose and lipid-monophosphate β-galactose. In the presence of [β-32P]UDP-glucose the system incorporated radioactivity only into the diphosphate derivatives, indicating the transfer of a α-glucose 1-phosphate group from UDP-glucose. Experiments with doubly labelled substrate confirmed these results. The role these compounds might play in polysaccharide biosynthesis is discussed.

The O antigen polysaccharide of Salmonella montevideo was isolated from a core-defective mutant by the phenol/water procedure, and was suspected to contain phosphomonoester and cyclic phosphodiester at its reducing end in anology to the O hapten from Salmonella typhimurium (Kent and Osborn, 1968). Therefore, it was chromatographed on a DEAE-cellulose column. Whereas one part eluted with water the other part of the polysaccharide could only be eluted with buffer. Both fractions were further purified on Sephadex G100 and contained mannose, glucose, N-acetylglucosamine and phosphate in a molar ratio of 4:1:1: < 0.1. In order to specifically label the reducing end phosphate was removed enzymatically, or the presumed cyclic diester was cleaved by mild hydrolysis, and the fractions were reduced with sodium boro[3H]hydride. Both fractions yield mainly [3H]glycerol after hydrolysis and paper chromatography. In addition, [3H]mannitol and [3H]monohydroxyacetone could be identified by paper chromatography and were concluded to be the result of phosphate migration and β-elimination reactions taking place during the isolation procedure and the various treatments prior to sodium boro[3H]-hydride reduction. These findings in addition to periodate oxidation studies indicated that the O antigen polysaccharide of Salmonella montevideo had glyceraldehyde phosphate at its reducing end. From the incorporation of 3H into the polysaccharide the O antigen was calculated to consist of about 19 repeating units of 6 sugar residues each.

Three rat alpha 1-fetoprotein fractions were obtained by chromatography on concanavalin-A--Sepharose: one non-reactive, one weakly reactive and one reactive to concanavalin A. The non-reactive and reactive variants were found to vary in the structure of their carbohydrate chains while the conformation of the weakly reactive form may modulate the accessibility of these chains to the lectin. N-Glycosidically linked glycans from unfractionated alpha 1-fetoprotein were isolated and chemically characterized. Particular attention was paid to develop sensitive methods based upon hydrazinolysis, quantitative re-N-acetylation of glycans with [14C]acetic anhydride and thin-layer chromatography of labeled compounds. With the aid of these methods two main kinds of glycans (1a and 2a) were obtained and fractionated on concanavalin-A--Sepharose into non-reactive (1a) and reactive (2a) molecules. Moreover it was demonstrated that each alpha 1-fetoprotein variant contained either two glycans 1a or two glycans 2a, not randomly, but a pair of the identical carbohydrate chains at the two glycosylation sites.

The isolation and amino-acid and carbohydrate composition of three tryptic glycopeptides, containing 16,17 and 20 amino-acid residues, respectively, from the neuraminidase-treated major form of human ceruloplasmin have been described earlier. It has also been suggested that the peptides possibly represent all of the carbohydrate-binding sites in the protein. The present paper describes the amino-acid sequences of these peptides as obtained by stepwise Edman degradation and carboxypeptidase digestions of the whole peptides and their fragments after cleavage by chymotrypsin, pepsin and thermolysin. The results confirm that the carbohydrate in all cases is attached via an asparagine (or aspartic acid) residue, that is followed by an X-Thr/Ser sequence. No sequence homologies are found when the peptides are compared to each other or to other glycopeptides of known sequence.

A partial sequence of the manganese-containing superoxide dismutase from Bacillus stearothermophilus has been determined by mass spectrometry. A new fragmentation reaction at N-terminal proline is described and an 'in-chain' fragmentation, also occurring at proline, has been observed. There is some evidence that the latter cleavage may also occur at other residues. Some differences from the classically determined sequence are reported, and the validity of these observations is discussed.

Incubation of calf lens cortex homogenate with [14C]putrescine or dansylcadaverine, followed by two-dimensional gel electrophoresis and fluorography, enabled the identification of three different beta-crystallin chains as the endogenous substrates of Ca2+-dependent lens transglutaminase (R-glutaminyl-peptide:amine-gamma-glutamylyltransferase, EC 2.3.2.13). One of these is beta Bp, the predominant subunit of beta-crystallin, of which the amino acid sequence is known. The site of amine-labeling in beta Bp could be located, by limited proteolysis, in the N-terminal domain of this chain. Tryptic digestion of the N-terminal domain and subdigestion with elastase of the N-terminal tryptic peptide identified glutamine-7 as the single residue to which the amines are bound. This is the first example of an endogenous substrate of intracellular transglutaminase in which the site of the acyl-donor glutamine residue has been established. Tryptic digestion of the putrescine-labeled beta-crystallin aggregate, followed by high-voltage paper electrophoresis, provided a preliminary characterization of the labeled peptides originating from the other two labeled beta subunits.

High-affinity penicillin binding sites from which the antibiotic could not be removed by washings at 4°C in 0.017 M K2HPO4 or 0.05 M Tris-HCl pH 7.5, were shown to occur in the isolated membranes of Streptomyces R61. These sites caused the attachment of 25 picomoles of [14C]benzylpenicillin per milligram membrane protein. Penicillins and cephalosporins competed for the same binding sites. The antibiotic concentrations which excluded [14C]benzylpenicillin from 50% of the binding sites were those which inhibited by 50% the membrane-bound transpeptidase. The same rate constant (about 1 × 10−4 s−1) for the dissociation of the benzylpenicillin membrane complex at 37°C and in 0.017 M K2HPO4, was calculated either from the release of the radioactivity (using [14C]benzylpenicillin) or from the recovery of the transpeptidase activity. These observations supported the conclusion that the high-affinity binding sites in the isolated membranes were the transpeptidase molecules. All the complexes formed between the membranes and the various penicillins and cephalosporins examined were reversible at 37°C and in 0.017 M K2HPO4 at least with regard to the transpeptidase. Depending upon the antibiotics, the rate constants for the dissociation of these complexes varied from 3.3 × 10−3 to 0.73 × 10−4 s−1. The radioactivity released through the dissociation of [14C]benzylpenicillin membrane complex occurred mainly in the form of a compound which behaved as [W]-benzylpenicilloic acid both by paper electrophoresis and thin-layer chromatography. It was impossible to choose between several possible mechanisms for the release of the antibiotic molecule.

Semliki-Forest-virus 42-S virion RNA is translated in a mouse ascites cell-free system. One of the major polypeptides made in vitro has been identified as Semliki-Forest-virus capsid protein by co-electrophoresis on dodecylsulfate–polyacrylamide gels and by tryptic peptide analysis on paper. Another major product of the cell-free system has been characterised as envelope protein by peptide analysis but it does not co-electrophorese on gels with envelope proteins from purified virions. It is possible that the nascent, non-glycosylated polypeptide is partially degraded by proleolytic enzymes present in the cell-free extract, or alternatively that fragments of envelope protein are produced by premature termination on the viral RNA. Several other polypeptides are synthesised in the cell-free system in variable amounts but these have not been identified. The genetic content and the mechanism of translation of Semliki-Forest-virus 42-S RNA are discussed.

The positions of the disulfide bridges in toxin II of Androctonus australis Hector were investigated using proteolytic enzymes. The resulting peptides were separated by column and paper chromatography and high-voltage electrophoresis. Determination of the amino acid compositions of the cystine-containing peptides or their oxidized derivatives and, when necessary, dansyl Edman degradation allowed to locate the disulfide bonds in the toxin. The four disulfide bridges were found to link the half-cystine residues number 12 and 63, 16 and 36, 22 and 46, 26 and 48. These positions are probably the same in all scorpion neurotoxins.

The primary structure of sheep para-xA-casein, the N-terminal portion of x-casein, was established. The reduced alkylated protein was subjected to digestion with trypsin. The resulting soluble peptides were purified by a combination of Dowex 1 X2 chromatography and electrophoresis and chromatography on paper; the core peptides were solubilized in 50% formic acid and filtered on Sephadex G-50. The amino-acid sequence of these peptides was determined chiefly with a Sequencer. Alignment of the tryptic peptides into a single chain containing 105 amino acids was determined from basic overlap peptides (tryptic core peptides with several basic amino acids; chymotryptic peptides). Some comparisons were achieved with cow para-xa-casein. %-Casein is the principal casein fraction affected by chymosin (or rennin) in the primary phase of the milk clotting process [ 11. The group of Jollbs established already in 1965 that during this enzymic reaction, a Phe -+ Met linkage was specifically split in cow as well as in sheep x-caseins [2-41. An insoluble part, para-x-casein (N-terminal moiety of x-casein), was from Merck or Prolabo except those employed for the Sequencer which were purchased from Socosi (94100Saint Maur, France) and 4-sulfophenylisothiocyanate from Pierce.

A purification procedure applied to H-2 antigens (controlled by the H-2 locus) was used for the characterization of HL-A substances (HL-A2 and HL-A7) (controlled by the HL-A locus) solubilized from lymphoid cell membranes by delipidation. Upon Sephadex chromatography the serologically active material had an apparent Mr of 32000, and displayed, in relation to H-2 antigens, a faster electrophoretic mobility and higher pI values. One sulfhydryl group and one (intrachain) disulfide bond were present in partially purified HL-A preparations. Reduction and aminoethylation in 4 M urea did not influence the electrophoretic migration of HL-A active substances, suggesting that, as already demonstrated for H-2 alloantigens, in HL-A molecules the genetic determinants are present on single polypeptide chains. Highly purified HL-A alloantigens, like H-2 substances, revealed a high degree of microheterogeneity. These findings further support the concept of the genetic and structural homology of the main histocompatibility systems in mouse and man.

The substrate specificity of the glucose-proton symport system was studied to gain information about the spatial relationship between the binding sites for glucose and proton. Charged glucose analogues such as amino sugars or sugar acids were not transported by the uptake system, with the exception of 2-amino-2-deoxy-d-glucose. This glucosamine was taken up in the charged form in uniport mechanism, i.e. without symport of proton. This result was interpreted to mean that the proton-binding site of the symport system is close to the hydroxyl at carbon 2 of glucose. This interpretation was strengthened by the following facts: The steric position of hydroxyl groups at carbons 1, 2 or 3 of glucose were especially important for efficient transport. O-Methylation was not tolerated at carbon 1, but it was tolerated at carbons 3,4 or 6. The stoichiometric flow of proton and sugar could be disturbed by removal of hydroxyl group at carbon 1 of glucose. The pH-dependence of sugar transport is sugar-specific, e.g. the amino group at carbon 2 of glucose improves transport at higher pH. The configuration at carbon 2 of glucose influences the specificity for the symported ion. It is concluded that the coupled flow of proton and glucose occurs by simultaneous coordinate movement of both in a transmembrane channel.

An inhibitor for S-adenosyl-L-methionine (AdoMet)-dependent methyltransferases has been purified from rat liver particulate fraction to apparent homogeneity, as judged by high-performance liquid chromatography, two-dimensional paper electrophoresis and isoelectric focusing chromatography. This inhibitor molecule, which is composed of 27 amino acid residues with an additional fluorescent chromophore, is rich in glycine, contains no basic amino acid, and has an isoelectric point (pI) of 3.70. A single absorption peak was observed at 248 nm in acidic as well as in neutral media, while two peaks were detected in alkaline medium at 206 nm and 248 nm. The former peak was found to be quite labile. The fluorescent spectra with excitation peak at 285 nm and emission peak at 358 nm are greatly influenced by the pH, being the highest in alkaline medium. The purified inhibitor inhibits all the AdoMet-dependent methyltransferases examined.

The amino acid sequence of the porcine vasoactive intestinal octacosapeptide is His-Ser-Asp-Ala-Val-Phe-Thr-Asp-Asn-Tyr-Thr-Arg-Leu-Arg-Lys-Gln-Met-Ala-Val-L ys-Lys-Tyr-Leu-Asn-Ser-Ile-Leu-Asn-NH2. Its amino acid residues 1, 2, 6 and 7, counted from the N-terminus, are identical to those in the corresponding positions in both porcine glucagon and secretin. The residues in positions 3, 12, 13, 14 and 23 are identical to those in secretin, but not in glucagon, and position 10 is occupied by a tyrosyl and position 28 by an asparaginyl residue, like in glucagon but not in secretin. If in addition to identical amino acid residues also chemically similar residues, such as isoleucine in position 26 of the octacosapeptide, as compared to leucine in secretin and glucagon, are taken into consideration then the similarity between these three polypeptides is still more evident. At a more remote level there is some structural resemblance also between these peptides and the other four peptides from the intestinal wall, cholecystokinin-pancreozymin, motilin, the gastric inhibitory peptide and substance P, the structures of which are known. The elucidation of the structure of the octacosapeptide was facilitated by the finding that pancreatic kallikrein preferentially cleaved only one of the three bonds in its N-terminal cyanogen bromide heptadecapeptide that are susceptible to cleavage with trypsin.

The kinetic and regulatory properties of enzymes involved in the biosynthesis of UDP-D-galactose, UDP-N-acetylglucosamine. GDP-alpha-D-mannose and GDP-beta-L-fucose from D-glucose 6-phosphate in various reticulo-endothelial tissues was studied. The tissues examined include bovine liver, thyroid, spleen, salivary gland, lung, intestine and mesenteric; pulmonary, portal and sub-maxillary lymphnodes. The maximum rates of specific enzymes in these pathways which were slow enough to be rate-limiting in the formation of glycoproteins in these tissues was determined. UDP-D-galactose 4-epimerase was consistently the rate-limiting reaction in the conversion of -d-glucose 6-phosphate to UDP-D-galactose in all of the tissues examined. The series of reactions leading to the formation of GDP-alpha-D-mannose and GDP-beta-L-fucose were limited by the activity of GDP-alpha-D-mannose pyrophosphorylase and GDP-alpha-D-mannose oxidoreductase, respectively. The formation of UDP-N-acetylglucosamine was limited by the rate of the amination reaction which converts -d-fructose 6-hosphate to D-glucosamine 6-phosphate in the presence of glutamine. Several of these rate-limiting enzymes were partially purified from mesenteric lymph node extracts, and their regulatory properties were examined. GDP-alpha-D-mannose was found to be a competitive inhibitor of GDP-alpha-D-mannose pyrophosphorylase. The apparent Km for GTP was 0.06 mM and the Ki for GDP-alpha-D-mannose was 0.03 mM. The concentrations of GTP and GDP-alpha-D-mannose in lymph node extracts were determined to be 0.095 and 0.012 mumol per g, respectively. UDP-N-acetylglucosamine and UDP-D-glucose inhibited D-fructose 6-phosphate amidotransferase in a manner competitive with D-fructose 6-phosphate. The Km for fructose 6-phosphate was 0.3 mM, while the Ki for UDP-D-glucose and UDP-N-acetyglucosamine were determined to be 0.4 mM and 0.045 mM, respectively. The concentrations of these metabolites in lymph node tissue were: UDP-D-glucose, 0.42; UDP-N-acetylglucosamine 0.095; and D-fructose 6-phosphate, 0.073 mumol per g wet weight of tissue. The results obtained in these studies show that specific rate-limiting enzymes in the pathways for the biosynthesis of nucleoside diphosphate sugars in reticulo-endothelial tissues may be subject to cumulative feedback inhibition by the nucleoside diphosphate sugars which are the final products of these systems and the initial precursors of the oligosaccharide units of glycoproteins in these tissues.

A hitherto unknown amino sugar was discovered in lipopolysaccharides isolated from strains of chemotype III of the photosynthetic bacterium Rhodopseudomonas palustris. The sugar (2–3% of the lipopolysaccharides dry weight) is detected as a shoulder on the descending part of the glucosamine peak obtained when acid hydrolysates of the lipopolysaccharides are passed through an amino acid analyzer. It was separated from the other amino sugars of the lipopolysaccharide (i.e. glucosamine, galactosamine, quinovosamine) by high-voltage paper electrophoresis and subsequent paper chromatography. The mass spectrometric analysis of the isolated sugar as deuterium-reduced alditol acetate revealed the presence of a 6-O-methyl-2-amino-2-deoxyhexose. Demethylation of the isolated sugar afforded glucosamine and the demethylation of its ninhydrin degradation product yielded arabinose, which showed that the parental amino sugar is glucosamine. Authentic 6-O-methyl-d-glucosamine exhibited identical properties in paper chromatography and high-voltage electrophoresis as did its alditol acetate derivative in gas-liquid chromatography and mass spectrometry. These data prove that the unknown amino sugar in lipopolysaccharides of Rhodopseudomonas palustris is 6-O-methyl-glucosamine. The strong positive optical rotation of the isolated sugar indicates its d-configuration. 6-O-Methyl-d-glucosamine is part of the O-specific chain and not of thelipid A moiety of the lipopolysaccharide. Periodate-oxidation studies revealed that the sugar, like the other amino sugars of this lipopolysaccharide, does not occupy non-reducing terminal positions to any appreciable extent.

The complete primary structure of mouse prolactin has been established on the basis of tryptic peptides from cyanogen-bromide-treated, S-carboxymethylated mouse prolactin and Staphylococcus-aureus-protease-cleaved overlaps, which were sequenced by manual liquid-phase and solid-phase Edman degradation. Three disulfide bonds were assigned to Cys-4-Cys-9, Cys-56-Cys-172, and Cys-189-Cys-197 by digestion of intact prolactin with S. aureus protease. One of the characteristics to date is replacement of Trp-89, which is commonly present among prolactin, growth hormone and choriomammotropin, by serine. It was suggested, by comparison with five other prolactins, five growth hormones and human choriomammotropin, that Asp-18, His-25, Ser-60 and Thr-63 are essential to lactogenic activity.

Degradation studies performed on the original or dephosphorylated core oligosaccharides derived from lipopolysaccharides of Salmonella Ra, Rb, and RcP+ mutants revealed the presence of a branched heptose trisaccharide, −3(HepIII-1,7-)-HepII-1,3-HepI–. The branching heptose III is linked to position 7 of heptose II. Heptose II is also substituted (at either position 2, 4, or 6) by a phosphate group. Other previously identified substituents of the heptose units are a pyrophosphoryl-ethanolamine group linked to position 4 of heptose I and the hexose oligosaccharide of the core linked to position 3 of heptose II. The degree of substitution of the heptose main chain with heptose III varies in different lipopolysaccharides from 20 to 90%. A lipopolysaccharide of chemotype RcP− was analyzed and found not to contain the third heptose unit.

Lipopolysaccharide of a chloridazon-degrading bacterium was obtained by a two-stage extraction procedure with phenol/EDTA in a yield of 0.3% of dried bacteria. The carbohydrate moiety consisted of heptose, 3-deoxyoctulosonic acid and D-glucose in a molar ratio of 1:2:2 X 3. Lipid A was composed of 1 mol 2,3-diamino-2,3-dideoxy-D-glucose, 2 mol amide-bound and 2.6 mol ester-bound fatty acids/mol. Amide-bound fatty acids were 3-hydroxydodecanoic acid and 3-hydroxyhexadecanoic acid; dodecanoic acid and R-(-)-3-hydroxydodec-5-cis-enoic acid were found to be present in ester linkage. Under conditions of acidic hydrolysis, the latter was converted into the cis and trans isomers of 5-hexyltetrahydrofuran-2-acetic acid. Dodecanoic acid was demonstrated to be linked with the hydroxy groups of the amide-bound fatty acids. The taxonomic significance of these results, especially the demonstration of 2,3-diamino-2, 3-dideoxy-D-glucose, is discussed.

The chemical structure of Bacteroides fragilis NCTC 9343 lipid A was characterized by using conventional chemical procedures, methylation analysis, and laser desorption mass spectrometry. It was found that B. fragilis lipid A consists of a beta-D-glucosaminyl-(1-6)-D-glucosaminyl-1-O-phosphate backbone whose hydroxyl groups in positions 4, 4' and 6' are free, the latter serving as the attachment site for the polysaccharide component in lipopolysaccharide. This backbone molecule carries up to of five molecules of ester- and amide-bound long chain non-hydroxylated and (R)-3-hydroxy fatty acids. With regard to the distribution on the fatty acids on the lipid A backbone, a considerable heterogeneity was revealed by laser desorption mass spectrometry. Despite this heterogeneity, a major species of B. fragilis lipid A could be defined in which the hydroxyl group at position 3' of the distal GlcN carries (R)-3-hydroxyhexadecanoic acid and the hydroxyl group at position 3 of the reducing GlcN is acylated by (R)-3-hydroxypentadecanoic acid. The amino group of the distal GlcN residue carries (R)-3-(13-methyltetradecanoyloxy)-15-methylhexadecanoic acid and that of the reducing GlcN group (R)-3-hydroxyhexadecanoic acid. The absence of ester-bound phosphate and ester-linked 3-acyloxyacyl groups, the presence of not more than five acyl residues and the predominance of fatty acids possessing 15-17 carbon atoms are unique features of B. fragilis lipid A which differentiate it from enterobacterial and other lipids A and which are likely to be related to its low endotoxic activity.

By using the glutamate-amidated tetrapeptide l-alanyl-d-isoglutaminyl-(l)-meso-diamino-pimelyl-(l)-d-alanine as a probe, there appears to exist in the membranes of Escherichia coli K12 strain 44 a dd-carboxypoptidase-transpeptidase system which does not recognize this peptide and a dd-carboxypoptidase-transpeptidase system which recognizes it. The dd-carboxypeptidase-endopeptidase system is essentially hydrolytic. It catalyzes the hydrolysis of UDP-N-acetyl-muramyl-pentapeptide into UDP-N-acetylmuramyl-tetrapeptide and the hydrolysis of the wall peptidoglycan peptide dimer into monomers. These activities are not inhibited by the glutamate-amidated tetrapeptide. The system may consist either of two enzyme proteins having predominantly carboxypeptidase activity and endopeptidase activity, respectively, or of one enzyme protein of which the functioning would depend upon the environmental conditions. The dd-carboxypeptidase-transpeptidase system (a) catalyzes concomitant hydrolysis (carboxypeptidase activity) and transfer (natural model transpeptidase activity) reactions with the pentapeptide l-alanyl-γ-d-glutamyl-(l)-meso-diaminopimelyl-(l)-d-alanyl-d-alanine. The transfer reaction leads to the synthesis of a dimer that is identical to the one which occurs in the E. coli wall peptidoglycan; (b) utilizes the glutamate-amidated tetrapeptide as an acceptor. Simultaneous exposure of the pentapeptide and the glutamate-amidated tetrapeptide to the enzyme system leads to the formation of an hybrid monoamidated peptide dimer and causes a decreased hydrolysis of the pentapeptide; (c) by virtue of its own carboxypeptidase activity, it appears to exert some endopeptidase activity. Both carboxypeptidase and endopeptidase activities of this system are inhibited by the glutamate-amidated tetrapeptide, but this represents only a small fraction of the total hydrolytic activity of the membrane Brij-36T extract. (d) The system catalyzes an unnatural model transpeptidation reaction in which glycine replaces d-alanine at the C-terminal position of the nucleotide UDP-N-acetylmuramyl-pentapeptide. This system may also consist either of two enzyme proteins having predominantly natural model transpeptidase activity and unnatural model transpeptidase activity, respectively, or of one enzyme protein of which the functioning would depend upon the environmental conditions. Whatever the exact situation, the E. colidd-carboxypeptidase-transpeptidase system is in many respects, similar to the dd-carboxy-peptidase-transpeptidase single polypeptide enzymes isolated from Streptomyces strains R39 and R61.

The precursor proteins for winter flounder antifreeze polypeptide (AFP) were isolated from liver using gel filtration chromatography and reverse-phase high-performance liquid chromatography. Two major pro-antifreezes (Mr 5000), corresponding to the precursors for AFP-6 and AFP-8, were characterized by amino acid analyses and automated Edman degradation. These precursors showed significant antifreeze activity. The pro-antifreezes were synthesized in the liver seasonally as demonstrated by immunoblotting and in vitro liver incorporation studies. No mature AFP were detected in liver, thus indicating that the processing of pro-antifreezes, including amidation of the C-termini, occurred mainly in the serum. The function(s) of the prosequences, if any, remain unclear.

Peptide studies on two acidic proteins, A1 and A2, from 50-S ribosomes of Escherichia coli indicate a closely related primary structure. 1 Amino acid compositions of the tryptic peptides of the two proteins are identical, with exception of the amino terminal peptide. 2 The N-terminal amino acid sequence of A2-protein is Ser-Ile-Thr-Lys, while that of A1-protein is N-acetyl-Ser-Ile-Thr-Lys. 3 The estimated number of 120 amino acid residues in each polypeptide chain is consistent with the physical molecular weight estimate. 4 In 50% of the polpeptide chains, in both A1 or A2-protein, a specific lysine residue is replaced by ɛ-N-monomethyl-lysine.

Cleavage with cyanogen bromide was employed to investigate the N-terminal amino acid sequence of the λ-chains of normal porcine immunoglobulin. The N-terminal homoserine-containing nonapeptide was isolated from all fractions of the λ-chains in a yield of 60% of theory. Its amino acid sequence was determined by mass spectrometry and was found to be uniform. Other homoserine-containing fragments were obtained from the cyanogen bromide hydrolyzate of the λ-chains with split disulfide bonds. Partial amino acid sequence of these fragments provided evidence that methionine in position 9 is present in more than 60% of variants and obviously has an invariant character. Variants of the λ-chains exist (by estimate not more than 15% of the total λ-chains) which, in addition to the invariant methionine residue in position 9, possess a variable methionine residue in the section between positions 46 and 51. The data obtained indicate that the sequence of 23 amino acid residues from the N-terminus of porcine λ-chains is uniform with the exception of a single replacement (in position 13). Moreover, in comparison with the sequence of λ-chains of man, mouse and birds, the sequence displays several species-specific replacements and a deletion of threonine which occupies position 5 in the chains of the other species. In the pig, the λ-chains represent the predominant type of light chains (some 70%). Hence it follows that some 50% of the light chains in this species possess a uniform amino acid sequence in the N-terminal section. These facts are difficult to reconcile with the germline theories of genetic control so that mechanisms of somatic diversification of immunoglobulins must be envisaged.

1. Succinate dehydrogenase flavocoenzyme ("SD-flavin"), previously shown to be an 8α-substituted riboflavin derivative containing a tertiary nitrogen homoconjugated to the flavin nucleus, was subjected to further hydrolysis and to reduction under acid conditions. Both conditions resulted in the liberation of 1 mole of histidine per mole of flavin. This proves histidine to be the covalent link between flavin and peptide backbone in succinate dehydrogenase and imidazole to be the tertiary nitrogen function homoconjugated to the flavin.2. 8α-Histidyl-riboflavin has been synthesized starting from riboflavin chemically and shown to be completely identical with the natural product in optical, ESR and NMR spectra, pH-fluorescence curve and behavior on thin-layer and paper chromatography, as well as paper electrophoresis.3. Both the natural compound isolated by acid hydrolysis of flavin peptide and the synthetic one contain two isomers, which may be separated by high voltage electrophoresis. The isomers appear to be the N(1)-and N(3)-imidazole substituted compounds. Digestion of the flavin peptide with aminopeptidase M yields only one isomer but on treatment with 6-N HC1 this is gradually converted to a mixture of the two isomers. The absolute assignment of the natural isomer is suggested as 8α-[N(3)-histidyl]-riboflavino n the basis of imidazole quaternization with CH3I, reductive cleavage of the flavin-imidazole bond and identification of the methyl-histidine liberated as 1-methyl-histidine.

A comparison has been made of the sequence of the tyrosine-containing tryptic peptide derived from the purified, unfractionated, lysine-rich histone complement of the calf thymus, rat thymus and calf liver. Two tyrosine-containing peptides, one of which is a fragment of the other, have been obtained from each tissue. The major tyrosine-containing peptide was isolated by a single electrophoretic step in about a 50%, yield. The results indicate that the sequence of amino acids around the single tyrosine present in the lysine-rich molecule is constant from tissue to tissue. However, this sequence varies among the molecular species comprising the lysine-rich histone complement of a single tissue. Within a tissue, a conservative amino acid replacement in the residue penultimate to tyrosine (from the amino terminus) has been detected. The sequence of the tyrosine-containing region was found to be Lys-Ala-Leu-Ala-AlaGly-Gly-Tyr-Asp-Val-Glu-Lys.

The addition of mRNA preparations to a number of heterologous cell-free systems has been shown to result in the synthesis of immunoglobulin light chains. This paper reports the results of experiments in which an mRNA fraction prepared from MOPC 21 mouse myeloma cells has been used to stimulate the synthesis of immunoglobulin heavy chains in rabbit reticulocyte lysates. The synthesis of heavy chains in a homologous cell-free system containing either microsomes or membrane-derived polysomes is also described. The synthetic product has been characterised by finger-printing in each case. The results reveal the presence of a unique methionine-containing peptide in digests of the product of the rabbit reticulocyte cell-free system. This peptide is not observed in corresponding digests of heavy chains synthesized in vivo or in the homologous cell-free system in vitro.

The primary structures of the ribosomal proteins A1 (= L7) and A2 (= L12) have been elucidated. Comparison of the two amino acid sequences confirms earlier studies by us (1972) which indicated that the two proteins are identical except that A1 possesses an N-terminal acetyl group. Sequencing of tryptic and chymotryptic peptides was accomplished primarily by automatic solid-phase Edman degradation of 50- to 70-nanomole peptide samples. A large tryptic peptide T1Phe, which could not be sequenced by this method, was fragmented with elastase and its sequence mainly derived by conventional methods. The sequence of the first 51 residues of the nonacetylated A2-protein was obtained by Edman degradation in the Beckman sequencer. A-protein comprises three distinct regions: I (residues 1—55), which is negatively-charged and hydrophobic, II (56—81), which is positively charged, and III (82—120), which is negatively charged and hydrophilic. α-Helix promoting residues are located primarily in regions I and III. ɛ-N-Monomethyllysine, which occurs in 50% of the A1 and A2 chains is located in the more flexible region II at position 81. The distinctive clustering of hydrophobic and charged residues is quite remarkable and suggests that in situ certain of these regions may contain a binding site for components involved in peptide chain elongation.

1 Acidic capsular polysaccharide of Vibrio parahaemolyticus A55 (05: K15) was esterified with ethylene oxide and then reduced with sodium borohydride. Hydrolysis of the reduced material gave two hexosamines identical with mannosamine (2-amino-2-deoxy-mannose) and gulosamine (2-amino-2-deoxy-gulose) on paper and ion-exchange chromatograms. This fact suggested that the K15 antigen contained mannosaminuronic acid (2-amino-2-deoxy-mannuronic acid) and gulosaminuronic acid (2-amino-2-deoxy-guluronic acid). 2 Direct acid hydrolysis of the antigen gave two 2-amino-2-deoxy-hexuronic acids. These two were identified as mannosaminuronic and gulosaminouronic acids by paper chromatography and paper electrophoresis, being compared with synthesized standard samples. 3 Titration of acid form of the polysaccharide indicated that about 91.5% of the polysaccharide was composed of the N-acetyl-hexosaminuronic acids.

The amino acid sequences of two rat C-peptides corresponding to two rat proinsulins have been elucidated. Both C-peptides have 31 amino acids. There are differences in two positions, namely, in position 8 where rat-I has a proline and rat-II an alanine residue, and in position 17 where rat-I has a glutamic acid, rat-II a glycine residue. The additional acidic residue in rat-I C-peptide is counteracted by the additional basic residue in the insulin moiety of rat-I proinsulin (lysine B 29), which provides the two proinsulins with the same net charge at neutral pH. The amino acid sequences are: The rat C-peptides contain threonine and arginine, neither of which is present in the C-peptides of man, pig and ox. A comparison of the 5 sequences shows only 10 identical positions out of 31.

2-Amino-2-deoxyhexuronic acid was found to constitute the main sugar component of the specific part of Shigella sonnei phase I lipopolysaccharide. The sugar was isolated from the lipopolysaccharide by acid hydrolysis and purified on a Dowex 50×8 column. From the chemical reactivity, gas chromatography-mass spectrometry, paper chromatography, high-voltage electrophoresis and deamination experiments it has been concluded that the isolated compound is a 2-amino-2-deoxyhexuronic acid.

The peptide side chains of the peptidoglycan of the Pseudomonas aeruginosa cell envelope have the sequence H2N-l-Ala-d-Glu-A2pm-d-Ala-COOH. The d-glutamate is linked to the meso-2, 5-diaminopimelate (A2pm) by a peptide bond formed by the C-5 carboxyl group of d-glutamate. The C-1 carboxyl group of the glutamate does not occur in the amidated state in the cell wall. The configurations of alanine, glutamate, and 2, 5-diaminopimelate have been confirmed.

Polynucleotide phosphorylase from Escherichia coli cells catalyzes the transfer of a single nucleotidyl residue from each of the 2′(3′)-O-isovalerylesters of ADP, GDP, CDP and UDP to the initiator oligonucleotide, A-A-A. The products of these reactions are identified as the 2′(3′)-O-isovaleryl esters of the corresponding tetranucleotides A-A-A-A, A-A-A-G, A-A-A-C and A-A-A-U. The 2′(3′)-O-isovaleryl residue can be removed from the product by treatment with aqueous methanolic ammonia under conditions that do not significantly damage the oligonucleotide chains. The monoaddition of 2′(3′)-O-acyl esters of nucleoside diphosphates to an oligonucleotide acceptor is proposed as a method for the stepwise synthesis of oligoribonucleotides of defined sequence.

1 32P-labelled ribosomal sub-particles were hydrolysed with ribonuclease T1, under conditions which have previously been shown to yield a specific fragment of RNA · protein (“band III”), containing only four of the 21 30-S ribosomal proteins. The hydrolysate was fractionated on a polyacrylamide gel using the methods already published, and gel fractions containing the specific RNA · protein fragment were subjected to electrophoresis on a second polyacrylamide gel, in the presence of sodium dodecylsulphate to dissociate RNA and protein. Two bands of [32P]RNA were reproducibly observed, containing about 320 and 360 nucleotides, respectively. 2 Two-dimensional fingerprints of ribonuclease T1, hydrolysates from both these RNA species showed that the two fragments were almost identical. Oligonucleotides from the fingerprints were digested with pancreatic ribonuclease, and the secondary digestion products analysed and compared with the known sequence of 16-S ribosomal RNA. Several characteristic oligonucleotides from regions D, E and P of the RNA were identified unambiguously. Other oligonucleotides were found which could have arisen from more than one part of the 16-S RNA, but in each case one of the possibilities was in regions D, E, E′ or P. 3 It was concluded that the band III proteins are associated with these contiguous regions of the 16-S RNA, i.e. near to its 3′-end.

The fate of the hydrogen atoms at C-3 and C-4 of uridine diphospho-d-glucuronic acid (UDP-GlcUA) in the course of its conversion to uridine diphospho-apiose (UDP-api) was investigated. The specifically labeled UDP-GlcUA substrates were prepared enzymatically from d-[3-3H]glucose and d-[4-3H]glucose by the combined action of hexokinase, phosphoglucomutase, UDPG-pyrophosphorylase, and UDPG dehydrogenase. UDP-[U-14C, 3-3H]GlcUA, UDP-[U14C, 4-3H]-GlcUA, and UDP-[14C, 3,4-3H2]GlcUA were separatelly converted to UDP-Api by an enzyme preparation from Lemna minor. In order to obtain an apiose (3-C-hydroxymethyl-d-erythro-furanose) derivative which was reasonably stable and which could be produced without passing through the aldehydo form, the apiosyl radical of UDP-Api was enzymatically transferred to 7-glucosylapigenin by an enzyme preparation from parsley. The product of this reaction, apiin (7-O-[β-d-apiofuranosyl-(12)-β-d-glucosyl-5,7,4′-trihydroxyflavone), was subjected to mild periodate oxidation to convert the branch hydroxymethyl group of the apiosyl residue to formaldehyde. Approximately 14–16% of each of the substrates was converted to UDP-Api. The transfer of the apiosyl moiety to 7-glucosylapigenin was essentially quantitative in all cases. The isolated apiin was shown to have retained 53.5% of the tritium from the substrate labeled at C-3, 99% from the substrate labeled at C-4, and 73% from the substrate labeled at C-3,4. All of the tritium and one-fifth of the 14C of apiin produced from the variously labeled substrate was accounted for in the formaldehyde-dimedone derivative. These results establish the occurrence of a hydride shift from C-4 of UDP-GlcUA to C-3′ of apiose in the biosynthesis of UDP-Api from UDP-GlcUA.

For the elucidation of the amino-acid sequence of the carboxyl-terminal region of soybean trypsin inhibitor (Kunitz), fragments C and D were digested with trypsin, and the resulting peptides were separated by ion-exchange chromatography on Dowex 50X2 or by gel nitration on Bio-Gel P-4. Further fractionation and purification of the peptides were performed by ion-exchange chromatography on Dowex 1X2, by gel filtration on Bio-Gel P-2 or by high-voltage paper electrophoresis at pH 1.9 and 3.6. Three peptides were obtained in pure form from fragment C and ten peptides from fragment D, and their amino-acid sequences were determined by the direct Edman method and by carboxy-peptidase digestion technique. Overlapping peptides necessary for the alignment of the tryptic peptides from fragment D were obtained from a chymotryptic hydrolysate of fragment D. Nine main peptides and nine minor peptides were obtained. The amino acid composition and the partial amino-acid sequence of the 18 chymotryptic peptides of fragment D made it possible to establish the amino-acid sequence of the carboxyl-terminal region of the inhibitor. The complete amino-acid sequence of soybean trypsin inhibitor (Kunitz) deduced here was compared with that of Bowman-Birk inhibitor, another well-known soybean proteinase inhibitor.

The fingerprinting technique has been applied to double-stranded viral RNA. Under appropriate conditions pancreatic and T1 ribonuclease digests were obtained which upon two-dimensional paper electrophoresis yielded specific patterns, similar to those obtained from single-stranded RNAs. The structure of some simpler nucleotides on the fingerprints have been determined. The method seems suitable to characterize and sequence all kinds of double-stranded RNA molecules. Labelling of the virus in vivo yielded RNA preparations of rather low radioactivity. Polynucleotide kinase was used therefore to label the digest of non-radioactive RNA for fingerprinting. The possibilities and limitations of this technique for sequencing non-radioactive nucleic acids is discussed.

A short glycopeptide was isolated from bovine χ-casein by enzymic digestions, filtrations on Sephadex G-25 and was electrophoretically homogeneous. An O-glycosidic linkage was characterized between a residue of threonine and N-acetylgalactosamine. By chemical and enzymic proce- dures, the following formula was established: where NeuNAc =N-acetylneuraminic acid. The phosphoserine residue of χ-caseinoglycopeptide was characterized in the short glycopeptide which was localized in the C-terminal part of χ-casein. Thr-X-Pro might be suggested as “code sequence” for glycopeptides containing an O-glycosidic linkage.

Histone mRNA from HeLa cells synchronized by a double thymidine block has been identified and characterized. Its synthesis starts immediately after the cells are released from the thymidine block. Pulse labelling for 60 min with [5-3H]uridine at different times of the S-phase showed that radioactively labelled histone mRNA enters polyribosomes with a maximum during the first hour of S-phase, i.e., before the maximum of DNA synthesis. Histone mRNA sediments on sucrose gradients in the 8–10-S region and separates on polyacrylamide gels into three RNA species with molecular weights of about 1.5 × 105, 1.8 × 106, and 2.1 × 105, respectively. Polyribosomal 8–10-S RNA from synchronized cells in S-phase directs the synthesis of histones in a rabbit reticulocyte cell-free system. Histones synthesized in vitro were identified by polyacrylamide gel electrophoresis and by high-voltage paper electrophoresis of tryptic peptides. Polyribosomal 8–10-S RNA from cells in which DNA synthesis was blocked by thymidine or hydroxyurea had only a marginal histone mRNA activity.

Salmonella typhimurium rfa-657 strains, including both mutants and transductants, were shown to contain in their lipopolysaccharides both the (usual) l and the d-isomer of glycero-d-manno-heptose. By the application, in sequence, of two different extraction procedures S and R form lipopolysaccharides could be isolated separately from a representative rfa-657 strain. The R lipopolysaccharide was predominantly of the Re chemotype, only a few 2-keto-3-deoxyoctonate (KDO) residues being substituted by heptose (Rd2 chemotype) or by longer core chains. The d-heptose was enriched in the R lipopolysaccharide, while the S lipopolysaccharide contained mainly the natural l-isomer. It was found that d-heptose can replace l-heptoses I to III of the wild-type core. However, an elongation to completeness of the core stubs containing d-heptose seldom occurs. The heterogeneity in structure of the lipopolysaccharide produced by this class of (leaky) mutants is assumed to be due to an altered 6-epimerase which, in the wild form, catalyses the conversion of NDP-d-glycero-d-manno-heptose into NDP-l-glycero-d-manno-heptose and which is determined by the gene rfa-657 situated between cysE and pyrE. The symbol rfaD is proposed for this gene.

A protein kinase specifically associated with the ribosomal fraction has been isolated from rainbow trout testis. The enzyme was extracted from isolated trout-testis ribosomes in 0.6 M KCl, and was separated from trout testis protamine kinase by hydroxylapatite chromatography. The ribosomal protein kinase is a Mg2+-dependent enzyme that will transfer the terminal phosphoryl group from ATP into O-phosphoseryl linkages of the substrate. The enzyme will catalyze the phosphorylation of several basic proteins, and the slightly lysine-rich histone IIb2 is most readily phosphorylated followed by protamine and ribosomal proteins. Histones I, IIb1 and III are phosphorylated at an intermediate rate, while the arginine-rich histone IV and an acidic protein such as phosvitin are poor substrates. The enzyme is stimulated by the addition of dithiothreitol, and cyclic AMP enhances enzyme activity between 10 and 25%.