A new trisaccharide sugar chain linked to a serine residue in bovine blood coagulation factors VII and IX

A new trisaccharide sugar chain was identified in bovine blood coagulation factors VII and IX. A pentapeptide isolated from factor VII contained Ser-52, which could not be identified with a gas-phase sequencer, suggesting an unknown substituent on the serine residue (Takeya, H. et al. (1988) J. Biol. Chem., in press). The same results were obtained for a pentapeptide containing Ser-53 of factor IX. Component sugar analysis revealed that the peptide contained 1 mol of glucose and 2 mol of xylose. This sugar component was also confirmed by high-resolution fast atom bombardment mass spectrometric analysis of the pentapeptide. The trisaccharide was released from the peptides by means of beta-elimination reaction and its reducing end was coupled with 2-aminopyridine. The fluorescent pyridylamino (PA-) derivative of the trisaccharide was purified by gel-filtration and reversed-phase HPLC. The sugar composition of the PA-trisaccharide was found to be 2 mol of xylose and 1 mol of PA-glucose. These results indicate the existence of a (Xyl2)Glc-Ser structure in factors VII and IX.     

Biosynthesis and processing of lysosomal cathepsin L in primary cultures of rat hepatocytes

The biosynthesis and proteolytic processing of lysosomal cathepsin L was studied using in vitro translation system and in vivo pulse-chase analysis with [35S]methionine and [32P]phosphate in primary cultures of rat hepatocytes. Messenger RNA prepared from membrane-bound but not free polysomes directed the synthesis of a primary translation product of an immunoprecipitable 37.5-kDa cathepsin L in vitro. The 37.5-kDa form was converted to the 39-kDa form when translated in the presence of dog pancreas microsomes. During pulse-chase experiments with [35S]methionine in cultured rat hepatocytes, cathepsin L was first synthesized as a 39-kDa protein, presumably the proform, after a short time of labeling, and was subsequently processed into the mature forms of 30 and 25 kDa in the cell. On the other hand, considerable amounts of the proenzyme were found to be secreted into the culture medium without further proteolytic processing during the chase. The precursor and mature enzymes were N-glycosylated with high-mannose-type oligosaccharides, and the proenzyme molecule contained phosphorylated oligosaccharides. The effects of tunicamycin and chloroquine were also investigated. In the presence of tunicamycin, a 36-kDa unglycosylated polypeptide appeared in the cell and this protein was exclusively secreted from the cells without undergoing proteolytic processing. These results suggest that cathepsin L is initially synthesized on membrane-bound polysomes as a 37.5-kDa prepropeptide and that the cotranslational cleavage of the 1.5-kDa signal peptide and the core glycosylation convert the precursor to the 39-kDa proform, which is subsequently processed to the mature form during biosynthesis. Thus, the biosynthesis and secretion of lysosomal cathepsin L in rat hepatocytes seem to be analogous to those of the major excreted protein of transformed mouse fibroblasts [S. Gal, M. C. Willingham, and M. M. Gottesman (1985) J. Cell Biol. 100, 535-544] and the mouse cysteine proteinase of activated macrophages [D.A. Portnoy, A. H. Erickson, J. Kochan, J. V. Ravetch, and J. C. Unkeless (1986) J. Biol. Chem. 261, 14697-14703].     

Identity of soluble thiamin-binding protein with thiamin-repressible acid phosphatase in Saccharomyces cerevisiae

Two secretory glycoproteins of Saccharomyces cerevisiae, a soluble thiamin-binding protein and a thiamin-repressible acid phosphatase, were shown to be repressed to a similar extent by excess thiamin in the growth medium. Thiamin-repressible acid phosphatase was co-purified throughout the purification of the soluble thiamin-binding protein. Purified and deglycosylated soluble thiamin-binding proteins exhibited both thiamin-binding and acid phosphatase activity on non-denaturing polyacrylamide gel electrophoresis. Heat treatment of the purified soluble thiamin-binding protein caused a decrease in both activities with a similar inactivation profile. Furthermore, two thiamin-repressible acid phosphatase-defective mutants isolated had no and decreased soluble thiamin-binding activity, respectively. From the results, it was concluded that the soluble thiamin-binding protein is identical to the thiamin-repressible acid phosphatase in S. cerevisiae.     

Some properties of a Saccharomyces cerevisiae mutant resistant to 2-amino-4-methyl-5-beta-hydroxyethylthiazole

A mutant of Saccharomyces cerevisiae highly resistant to 2-amino-4-methyl-5-beta-hydroxyethylthiazole (2-aminohydroxyethylthiazole), an antimetabolite of 4-methyl-5-beta-hydroxyethylthiazole (hydroxyethylthiazole), has been isolated. Its resistance to 2-aminohydroxyethylthiazole was about 10(4) times that of the sensitive parent strain. The amount of thiamin synthesized in the cells of the resistant strain grown in minimal medium was less than half of that of the sensitive strain. The ability to synthesize thiamin from 2-methyl-4-amino-5-hydroxymethylpyrimidine (hydroxymethylpyrimidine) and hydroxyethylthiazole in the resistant strain was low compared with that of the sensitive strain. These results were found to be due to a deficiency of hydroxyethylthiazole kinase in the resistant strain: in sonic extracts of cells the enzyme activity was only 0.67% of that of the sensitive strain. Although the cells of the sensitive strain could accumulate exogenous hydroxyethylthiazole in the form of hydroxyethylthiazole monophosphate, no significant uptake of hydroxyethylthiazole by the cells of the resistant strain was observed. The possibilities that 2-aminohydroxyethylthiazole monophosphate may be the actual inhibitor of the growth of Saccharomyces cerevisiae, and that hydroxyethylthiazole may not be involved in the pathway of de novo synthesis of thiamin via hydroxyethylthiazole monophosphate, are discussed.     

Coordination structures and reactivities of compound II in iron and manganese horseradish peroxidases. A resonance Raman study

Resonance Raman investigations on compound II of native, diacetyldeuteroheme-, and manganese-substituted horseradish peroxidase (isozyme C) revealed that the metal-oxygen linkage in the compound differed from one another in its bond strength and/or structure. Fe(IV) = O stretching frequency for compound II of native enzyme was pH sensitive, giving the Raman line at 772 and 789 cm-1 at pH 7 and 10, respectively. The results confirmed the presence of a hydrogen bond between the oxo-ligand and a nearby amino acid residue (Sitter, A. J., Reczek, C. M., and Terner, J. (1985) J. Biol. Chem. 260, 7515-7522). The Fe(IV) = O stretch for compound II of diacetylheme-enzyme was located at 781 cm-1 at pH 7 which was 9 cm-1 higher than that of native enzyme compound II. At pH 10, however, the Fe(IV) = O stretch was found at 790 cm-1, essentially the same frequency as that of native enzyme compound II. The pK value for the pH transition, 8.5, was also the same as that of native compound II. Unlike in native enzyme, D2O-H2O exchange did not cause a shift of the Fe(IV) = O frequency of diacetylheme-enzyme. Thus, the metal-oxygen bond at pH 7 was stronger in diacetylheme-enzyme due to a weaker hydrogen bonding to the oxo-ligand, while the Fe(IV) = O bond strength became essentially the same between both enzymes at alkaline pH upon disruption of the hydrogen bond. A much lower reactivity of the diacetylheme-enzyme compound II was accounted to be due to the weaker hydrogen bond. Compound II of manganese-substituted enzyme exhibited Mn(IV)-oxygen stretch about 630 cm-1, which was pH insensitive but down-shifted by 18 cm-1 upon the D2O-H2O exchange. The finding indicates that its structure is in Mn(IV)-OH, where the proton is exchangeable with a water proton. These results establish that the structure of native enzyme compound II is Fe(IV) = O but not Fe(IV)-OH.     

Volatile anesthetics cause conformational changes of bacteriorhodopsin in purple membrane

We examined the effects of volatile anesthetics on the structure of the bacteriorhodopsin in the purple membrane by measurements of the absorption spectrum and the visible circular dichroism (CD) spectrum and assay of the retinal composition. As the concentrations of halothane, enflurane and methoxyflurane were increased, the absorption at 560 nm decreased but that at 480 nm increased with an isosbestic point around 510 nm. These anesthetic-induced spectroscopic changes were reversible. The CD spectrum showed the biphasic pattern with a positive and a negative band. As the concentration of halothane was increased from 4 mM to 8mM, the negative band reversibly diminished more drastically than the positive band, and at 8 mM of halothane the positive band shifted to around 480 nm. These results show that halothane disturbed the exciton coupling among bacteriorhodopsin molecules. The retinal isomer composition was analyzed using high performance liquid chromatography. The ratio of 13-cis- to all-trans-retinal was 47:53, 34:66 and 19:81 at control, 7.4 mM and 14.9 mM enflurane, respectively. After elimination of enflurane, the ratio returned to the control value. These findings indicate that volatile anesthetic directly affect a bacteriorhodopsin in the purple membrane and induce conformational changes in it.     

Resonance Raman study of cytochrome b562-o complex, a terminal oxidase of Escherichia coli in its ferric, ferrous, and CO-ligated states

Cytochrome b562-o complex, a terminal oxidase in the respiratory chain of aerobically grown Escherichia coli, has been studied by resonance Raman spectroscopy in its air-oxidized, dithionite-reduced, and reduced and CO-ligated states. In the reduced state, with a 406.7-nm excitation, there appeared 1494 and 1473 cm-1 lines, indicating that low spin and high spin components are included in the cytochrome b562-o complex. For the air-oxidized protein, resonance Raman lines were observed at 1372, 1503, and 1580 cm-1 with a 413.1-nm excitation, indicating that there is a ferric low spin heme. In addition, a weak but appreciable Raman line was observed at 1480 cm-1 assignable to a ferric high spin heme. Accordingly, it was concluded that low spin and high spin components are included in the cytochrome b562-o complex in the reduced and the air-oxidized states. In the CO-ligated state, with a defocused laser beam of 413.1 nm, two Raman bands assignable to the Fe-CO stretching mode have been observed at 489 and 523 cm-1, as a major and a minor component, respectively. When the laser beam was focused upon the sample to cause a photodissociation of CO from the heme moiety, the intensity of the major band at 489 cm-1 was reduced as expected. On the other hand, the minor band at 523 cm-1 remained still obvious. It was suggested that the cytochrome b562-o complex may have an additional anomalous site for CO that is resistant to photodissociation.     

Inactivation of rice bran thiamine-binding protein by N,N'-dicyclohexylcarbodiimide

The addition of a carboxyl-modifying reagent N,N'-dicyclohexylcarbodiimide (DCCD) to thiamine-binding protein isolated from rice bran resulted in a remarkable loss of its binding activity with [14C]thiamine. Thiamine and chloroethylthiamine substantially protected the protein against inactivation by DCCD, whereas thiamine phosphates did not. Another carboxyl reagent N-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinoline (EEDQ) also inactivated rice bran thiamine-binding protein. Inactivation of the thiamine-binding protein was accompanied by covalent binding of DCCD to the protein as shown by the use of [14C]DCCD. The binding of [14C]DCCD to the thiamine-binding protein was specific, and significantly inhibited by the addition of thiamine. The loss of thiamine-binding activity was proportional to the specific binding of [14C]DCCD. For complete inactivation of the thiamine-binding activity, the binding of 2.46 mol of [14C]DCCD per mol of thiamine-binding protein was required. Furthermore, limited proteolysis of the binding protein by trypsin yielded two polypeptides with molecular weights of 35,000 (large polypeptide) and 12,500 (small polypeptide) which were separated by SDS-polyacrylamide gel electrophoresis. The binding sites of [14C]DCCD were found to be located on the large polypeptide. These results suggest that a specific carboxyl residue in the large polypeptide releasable from rice bran thiamine-binding protein by trypsin digestion when modified by DCCD is involved in the binding of thiamine.     

Improvement of the dideoxy chain termination method of DNA sequencing by use of deoxy-7-deazaguanosine triphosphate in place of dGTP.

The dideoxy chain termination method using deoxy-7-deazaguanosine triphosphate (dc7GTP) in place of dGTP was found to be very useful. Sequencing of a part of the human N-myc gene having 85% GC content is impossible by the original method using dGTP, because of compression of bands. However, the nucleotide sequence of this part was unambiguously determined by analysis of both strands by the modified method. Use of dc7GTP is concluded to improve the dideoxy chain termination method for DNA sequencing.     

Modification of batroxobin with activated polyethylene glycol: reduction of binding ability towards anti-batroxobin antibody and retention of defibrinogenation activity in circulation of preimmunized dogs

Amino groups of batroxobin (Bothrops atrox thrombic protease) were modified with 2,4-bis(O-methoxypolyethylene glycol)-6-chloro-s-triazine (activated PEG2). The modified batroxobin had the reduced binding ability towards anti-batroxobin antibody but retained its enzymic activity in vitro and in vivo. Administration of modified batroxobin in which 29% of the total amino groups in the molecule had been modified, to beagle dogs preimmunized with native batroxobin gave rise to a marked reduction of the fibrinogen level in plasma, accompanied with an increased level of fibrinogen (fibrin) degradation products, FDP. On the other hand, no reduction of fibrinogen level was observed when native batroxobin instead of modified batroxobin was injected to immunized dogs.