Primary structure and analysis of the location of the regulatory disulfide bond of pea chloroplast NADP-malate dehydrogenase.

Purified pea chloroplast NADP-malate dehydrogenase (S)-malate: NADP+ oxidoreductase, EC 1.1.1.82) was digested with trypsin and the resulting peptides were separated by HPLC and sequenced. Together with the information from earlier work (Fickenscher, K. et al. (1987) Eur. J. Biochem. 168, 653-658) the total sequence is not known to an extent of 78%. Comparison with the sequence of the corn NADP-malate dehydrogenase deduced from its cDNA (Metzler, M.C. et al. (1989) Plant Mol. Biol. 12, 713-722) showed 84% agreement; however, the 11 N-terminal residues exhibit only 27% similarity. The N- and C-terminal extrapeptides of the pea NADP-malate dehydrogenase when aligned with non-regulatory NAD-malate dehydrogenases from bacteria or mammals consist of 30 and 17 amino acids, respectively. Since all cysteine-containing peptides were sequenced, the number of eight cysteines per subunit of the pea enzyme was established. The native, oxidized enzyme is characterized by an extremely slow reactivity of two thiols. Titration of the thiols of the denatured, oxidized enzyme both with DTNB and with pCMB resulted in six thiols not involved in disulfide formation. Therefore, one disulfide bridge must be present per 38.9 kDa subunit. Analysis of disulfide bonds by urea gel electrophoresis confirmed this finding. Using digestion products of NADP-malate dehydrogenase with aminopeptidase K, the location of the single disulfide bridge was established to be on the N-terminal arm (Cys-12 and Cys-17) of the polypeptide chain.

Unexpected translation initiation within the coding region of eukaryotic genes expressed in Escherichia coli.

When expressing several eukaryotic genes in Escherichia coli, we observed N-terminally truncated proteins which were attributed to translation initiation at downstream AUG codons. These AUG codons are located between 4 and 20 nucleotides 3' from sequences resembling bacterial SD elements. Although the presence of such downstream SD sequences is not sufficient for downstream initiation to occur, in two cases their removal abolishes synthesis of the truncated proteins. In one construct, a potential hairpin-loop structure is likely to inhibit translation initiation at the correct site and favor downstream initiation.

Isolation and structure elucidation of an alpha-amylase inhibitor, AI-3688, from Streptomyces aureofaciens.

A novel polypeptide inhibitor, AI-3688, which acts upon human pancreatic alpha-amylase, was isolated from fermentation broth of Streptomyces aureofaciens. The purified peptide contains no unusual amino acids. Its Mr is 3936. The primary structure of AI-3688 was elucidated by automatic Edman degradation of the native or modified inhibitor. Two intramolecular cysteines form a disulphide bridge, thus creating a ring structure consisting of 17 amino acids. Strong sequence homology also exists to another microbial alpha-amylase inhibitor, tendamistat (HOE 467). This paper discusses the role of a common partial sequence, -Gln-Ser-Trp-Arg-Tyr-, present in the loop of both inhibitors as the active site of microbial peptide alpha-amylase inhibitors.

Primary structures of new 'iso-hirudins'.

Crude hirudin (12.7 U/micrograms), a complex mixture of polypeptides obtained from the leech, could be separated by microbore HPLC. A combination of amino acid analysis, N-terminal microsequencing and chemical as well as enzymatic fragmentation made the primary sequence of the new isohirudins Ia-IIIb' accessible. The biological activity determined in the thrombin inhibition test showed a comparable value for all of these compounds. The results presented address the question as to whether these isohirudins are true mutations from a family of genes or a family of leeches.

Identification of phosphorylation sites on murine nuclear lamin C by RP-HPLC and microsequencing.

Isolated interphase lamin C, obtained from Ehrlich ascites tumor cells, was digested by Lys-C endoproteinase, the resulting peptides separated by reversed-phase HPLC and subjected to microsequencing in order to identify phosphorylation sites in interphase and following phosphorylation in vitro by cdc2-kinase, protein kinase C (PKC) and protein kinase A (PKA), respectively. Nuclear lamin C showed partial phosphorylation of Ser392 and Ser409, and possibly Ser407 in interphase. Phosphorylation was increased in response to cdc2-kinase at Ser390 and Ser392 and to PKC at Ser572. The N-terminal peptide (aa 1-32) containing consensus sequences for the 3 kinases was phosphorylated by cdc2-kinase, PKC and PKA. The sequence data suggests that multiple molecular switches via lamina modification control the dynamic behaviour of the nucleoskeleton during the cell cycle.

Cyclic somatostatin analogs bind specifically to pI 6.1 carboxylesterase of rat liver cells.

The hydrophobic cyclohexapeptide cyclo(Phe-Thr-Lys-Trp-Phe-DPro) (008), an analog of somatostatin with retro sequence, was previously shown to competitively inhibit the uptake of cholate and taurocholate into isolated rat liver cells. Conversely, the competitive uptake inhibition of 008 into isolated rat hepatocytes by bile acids confirmed the observation of common binding and transport sites by bile acids and cyclosomatostatin. Furthermore the transport characteristics of 008 uptake revealed a significant and rapid binding to cell membranes. In this context it was of special interest to investigate the specificity of the binding component since specific binding of the substrate to membrane proteins could be responsible for the low Km of 008-transport. Therefore, the cyclohexapeptide 008 could be used as the ligand in affinity chromatography in order to isolate such binding proteins. The gel matrix used did not interact non-specifically with octylglucoside-solubilized proteins from isolated rat liver plasma membranes. In affinity chromatography of octylglucoside-solubilized plasma membranes, two dominant proteins with apparent molecular masses of 60 and 58 kDa bound specifically to the 008 ligand. When used as ligands in affinity chromatography, these membrane-associated 60 and 58 kDa proteins bound exclusively to aromatic cyclopeptides, e.g. cyclosomatostatin 008, but not to linear peptides or taurocholate derivatives. The amino acid sequences of tryptic digests of the 008-affinity-purified 58 kDa protein were identical to the sequence of a microsomal pI6.1 carboxylesterase. Immunofluorescence of intact hepatocytes showed that this xenobiotic metabolizing enzyme is also located in sinusoidal rat liver plasma membranes and could therefore account for the extensive and specific binding of the cyclosomatostatin to sinusoidal plasma membranes of rat liver.

Binding proteins for cyclic and linear oligopeptides in plasma membranes and the cytosol of rat hepatocytes.

Using a cyclolinopeptide A analogue, the hydrophobic cyclic peptide c(-Ala-Lys-Pro-Phe-Phe-Ala-Lys-Pro-Phe-Phe-), termed CDP (cyclodecapeptide), as ligand in affinity chromatography, hepatocellular peptide binding proteins were isolated from the integral part of plasma membranes and the cytosol. The sequence of the isolated protein with MW of 50 kDa from the integral part of the plasma membrane fraction was identical to cytochrome P450 II C13 and cytochrome P450 II C22, whereas the sequence of the 54 kDa protein was identical to 3-hydroxyandrogen-UDP-glucuronosyltransferase. These proteins have also been described as binding proteins for bile acids. As shown in earlier studies, bile acids and CDP also compete for uptake into hepatocytes. In the cytosol, a further known bile acid binding protein, the glutathione-S-transferase (G-S-T) subunit Yb1, was isolated and sequenced as binding protein for CDP and also for a further cyclopeptide, the somatostatin analogue OO8, and a linear peptide with renin-inhibiting activity, EMD 55068. As shown in uptake studies using isolated basolateral plasma membrane vesicles, G-S-T was able to increase the uptake of EMD 51921, a linear peptide with renin-inhibiting potency, into the vesicles when the latter were preloaded with G-S-T. The binding of the substrate to the outside of the preloaded vesicles was not different than binding to unloaded vesicles. The maximal transport rate of the carrier-mediated/facilitated diffusion and the rate of permeation, however, were doubled in the presence of G-S-T, pointing to the involvement of intracellular binding proteins such as G-S-T in the unloading of the carrier protein and in the reduction of the free substrate concentration.

Hirudin: a family of iso-proteins. Isolation and sequence determination of new hirudins.

A recent NMR study of hirudin conformation in solution reveals a large extent of the tertiary structure of the inhibitor [2]. This model exhibits a highly packed core and two other extended wings-like domains. One of these domains centred at Gly (position 34) shows an accumulation of amino acid mutations, insertions, deletions located at an exposed position. This may represent evolutionary optimisation of the thrombin-binding site.

Antithrombotic effects of recombinant hirudin in different animal models.

The effect of recombinant hirudin (r-hirudin; HBW 023) was studied in four different models of thrombosis as well as on bleeding time. Arteriolar thrombus formation was induced in rats either by argon laser injury or by photochemical reaction. r-Hirudin, given by intravenous infusion, significantly inhibited arteriolar thrombus formation and arterial bleeding time at a dose of 40 micrograms/kg.min. In rabbit jugular veins and carotid arteries, occluding thrombi were produced by stenosis and endothelial damage. r-Hirudin reduced the incidence of thrombosis dose dependently (ED50 0.7 and 1.0 mg/kg i.v. in arteries and veins, respectively). Caval vein thrombosis was initiated in rats by insertion of a stainless steel coil. This thrombosis was dose dependently reduced by the injection of r-hirudin (ED50 0.16 mg/kg i.v.). The duration of the antithrombotic effect of 0.25 mg/kg of r-hirudin in caval veins was about 60 min, whereas the arterial bleeding time was significantly prolonged in these rats for only 20 min. The plasma level of r-hirudin had dropped to 0.12 micrograms/ml 60 min after injection. These results suggest that the duration of the effect on bleeding time may be shorter than the antithrombotic action of r-hirudin.

Cloning and expression of the genes of two fumarate reductase subunits from Wolinella succinogenes.

The fumarate reductase complex of the anaerobic bacterium Wolinella succinogenes catalyzes the electron transfer from menaquinol to fumarate. Two structural genes coding for subunits of the enzyme have been cloned in Escherichia coli. The genes were isolated from a lambda EMBL3 phage gene bank by immunological screening and subcloned in an expression vector. The genes frdA and frdB, which encode the FAD protein (Frd A, Mr 79,000) and the iron-sulfur protein (Frd B, Mr 31,000) of the fumarate reductase complex, were cloned together with a W. succinogenes promoter. The gene order was promoter-frdA-frdB. The FAD protein and the iron-sulfur protein were expressed in the correct molar mass in E. coli from the clones. The identity of the frdA gene and the suggested polarity were confirmed by comparing the amino-terminal sequence of the Frd A protein with that predicted from the 5'-terminal nucleotide sequence of frdA. The frdA and frdB genes are present only once in the genome. A region downstream of frdB, possibly a gene encoding cytochrome b of the fumarate reductase complex, hybridizes with a second site in the genome.

Stereospecific production of the herbicide phosphinothricin (glufosinate) by transamination: isolation and characterization of a phosphinothricin-specific transaminase from Escherichia coli.

An aminotransferase capable of transaminating 2-oxo-4-[(hydroxy)(methyl)phosphinoyl]butyric acid to L-phosphinothricin [L-homoalanine-4-yl-(methyl)phosphinic acid], the active ingredient of the herbicide Basta (Hoechst AG), was purified to apparent homogeneity from Escherichia coli K-12. The enzyme catalyzes the transamination of L-phosphinothricin and various analogs with 2-ketoglutarate as the amino group acceptor. The transaminase has a molecular mass of 43 kilodaltons by sodium dodecyl sulfate-gel analysis and an isoelectric point of 4.35. The enzyme was most active in the high-pH region, with a maximum at pH 8.0 to 9.5, and had a temperature optimum of 55 degrees C. Heat stability was observed up to 70 degrees C. Substrate specificity studies suggested that the enzyme is identical with the 4-aminobutyrate:2-ketoglutarate transaminase (EC 2.6.1.19). The first 30 amino acids of the N terminus of the protein were determined by gas phase sequencing. The transaminase was immobilized by coupling to the epoxy-activated carrier VA-Biosynth (Riedel de Haen) and used in a column reactor for the continuous production of L-phosphinothricin. The enzyme reactor was operated for 7 weeks with only a slight loss of catalytic capacity. Production rates of more than 50 g of L-phosphinothricin per liter of column per h were obtained.

Purification and characterization of an elastase inhibitor derived from hyaline cartilage.

Articular cartilage has the ability to retard the invasiveness of inflammatory cells such as leucocytes. This resistance is in part due to the avascular nature, organization, and composition of the extracellular matrix. Extractable components (anti-invasion factors, [AIF]) within the matrix have been shown to be responsible for protecting the tissue against proteases produced by invading leucocytes. In the present study, AIF is examined for leucocyte elastase inhibitory activity using both analytical and preparative electrophoretic and chromatographic techniques. A specific leucocyte elastase inhibitor was isolated and identified as a 15,000 dalton protein on SDS-PAGE. The inhibitor is distinct from aprotinin (Trasylol) in size, electrophoretic pattern, and comparative peptide maps. The inhibitor exhibited a highly cationic (pI greater than 9.5) as well as a strong hydrophobic character. These properties of primary structure were confirmed by amino acid composition and sequence of the amino-terminal residues. The amino-terminal sequence showed no homologies to other proteins or known protease inhibitors. This matrix-derived protease inhibitor may be an important mediator in the regulation of and protection against cartilage destruction.

Quantitative determination of tiamenidine hydrochloride in human serum using gas chromatography mass spectrometry.

A method has been developed for the blood level determination of the antihypertensive agent tiamenidine hydrochloride. The serum samples are mixed with deuterium labelled tiamenidine hydrochloride as an internal standard and extracted with methylene chloride. The extracts are derivatized with heptafluorobutyric acid anhydride and analysed by means of gas chromatography mass spectrometry using the selected ion monitoring technique to measure the molecular ion intensities of the bis-heptafluorobutyryl derivatives of tiamenidine hydrochloride and of the internal standard. Using 5 ml serum, the limit of detection is 0.2 ng ml-1 with an accuracy of +/- 0.17 ng (Syx of the calibration curve).

Disulphide bridge formation of proinsulin fusion proteins during secretion in Streptomyces.

To study disulphide bridge formation by Streptomyces lividans TK 24 in secreted single chain precursors of insulin a fusion protein (PTF 1) was investigated consisting of monkey proinsulin and the aminoterminal sequence Asp1 to Gly43 of the alpha-amylase inhibitor tendamistat from Streptomyces tendae. The purified soluble protein PTF 1 has a molecular mass of 14.4 kDa. The primary structure was elucidated after digestion with lysyl endopeptidase and fragment analysis. In this system, disulphide bond formation occurs in a way that the first cysteine in proinsulin is linked to the next following cysteine in the amino-acid chain resulting in a non-natural folding of the insulin part of the fusion protein. Re-folding of PTF 1 by reduction and re-oxidation followed by proteolytic digestions led to insulins which are identical to authentic material. The ease of correct disulphide bond formation in solution and incorrect processing during secretion suggests involvement of yet unknown factors leading to an unfavourable folding of proinsulin.

Identification, purification, and characterization of a zyxin-related protein that binds the focal adhesion and microfilament protein VASP (vasodilator-stimulated phosphoprotein).

VASP (vasodilator-stimulated phosphoprotein), an established substrate of cAMP- and cGMP-dependent protein kinases in vitro and in living cells, is associated with focal adhesions, microfilaments, and membrane regions of high dynamic activity. Here, the identification of an 83-kDa protein (p83) that specifically binds VASP in blot overlays of different cell homogenates is reported. With VASP overlays as a detection tool, p83 was purified from porcine platelets and used to generate monospecific polyclonal antibodies. VASP binding to purified p83 in solid-phase binding assays and the closely matching subcellular localization in double-label immunofluorescence analyses demonstrated that both proteins also directly interact as native proteins in vitro and possibly in living cells. The subcellular distribution, the biochemical properties, as well as microsequencing data revealed that porcine platelet p83 is related to chicken gizzard zyxin and most likely represents the mammalian equivalent of the chicken protein. The VASP-p83 interaction may contribute to the targeting of VASP to focal adhesions, microfilaments, and dynamic membrane regions. Together with our recent identification of VASP as a natural ligand of the profilin poly-(L-proline) binding site, our present results suggest that, by linking profilin to zyxin/p83, VASP may participate in spatially confined profilin-regulated F-actin formation.

Charge heterogeneity of insulin fusion proteins expressed in Escherichia coli is not due to proteolytic degradation.

Evaluation of the yield of expression of exogeneous protein in transformed Escherichia coli cells by means of one-dimensional SDS-PAGE often leads to overestimation and miscalculation. For example, it is possible that proteins of similar size comigrate and thus mask the overexpressed product band. Therefore, two-dimensional electrophoresis was used to analyze two types of recombinant fusion proteins, i.e., a beta-galactosidase insulin fusion protein and a interleukin II insulin fusion protein, directly after fermentation. We found that production scale expression products show charge and size heterogeneity. The heterogeneous protein spots were characterized by subsequent blotting onto Immobilon membrane and by N-terminal sequencing. Some of the separated spots were either N-terminally blocked or already degraded to some extent. The integrity of the actual product component of the fusion protein was examined with a C-terminus-specific antibody and by Western blot analysis of the 2D gels.

Heat-resistant inhibitors of protein kinase C from bovine brain.

Bovine brain cytosol is shown to contain two heat-resistant inhibitors of protein kinase C, with the following characteristics: 1. One protein kinase C inhibitor can be easily purified to homogeneity. Evidence is presented that this polypeptide of Mr 19,000 is calmodulin. It inhibits protein kinase C with an EC50 of about 2.5 microM and the inhibition is Ca2+-independent. It inhibits only intact protein kinase C. Removal of the regulatory domain of protein kinase C, by limited proteolysis with trypsin, abolishes the inhibition. 2. Another protein kinase C inhibitory activity has been partially purified. Its Mr is low (Mr 600-700, as estimated by gel chromatography). It is not digested by proteases, is hydrophilic, acid- and alkali-resistant, acts Ca2+-independently, and, in contrast to calmodulin, inhibits even the catalytic fragment of protein kinase C after removal of the regulatory domain by limited proteolysis. This inhibition is, at least partially, due to a competition with ATP. Besides protein kinase C, calcium/calmodulin-dependent protein kinase II is inhibited to a similar extent. cAMP-dependent protein kinase is not affected.

Wolinella succinogenes fumarate reductase contains a dihaem cytochrome b.

The fumarate reductase operon of Wolinella succinogenes is made up of three structural genes (frd-CAB). The frdC gene was located next to the promoter region and identified as the cytochrome b structural gene encoding 256 amino acid residues. The N-terminal amino acid sequences of seven fragments derived from the cytochrome b moiety of the enzyme all mapped within the frdC gene. This suggested that the enzyme contained only one species of cytochrome b. Re-evaluation of earlier measurements of subunit composition, haem B content and molecular weight led to the conclusion that the enzyme contained one molecule of cytochrome b with two haem B groups. The hydropathy plot of the amino acid sequence predicted five membrane-spanning hydrophobic segments, the first four of which contained a single histidine residue each. These residues could form the axial ligands to the two haem B groups. FrdC was found to be homologous with the cytochrome b (SdhC) of the Bacillus subtilis succinate dehydrogenase, but not with the hydrophobic subunits of the fumarate reductase or succinate dehydrogenase of Escherichia coli.

Undulin is a novel member of the fibronectin-tenascin family of extracellular matrix glycoproteins.

We characterized cDNA clones specific for the extracellular matrix glycoprotein undulin. Two sets of cDNA clones were isolated from a human placental lambda gt11 expression library and from a rhabdomyosarcoma cell line encoding two partially identical carboxyl-terminal polypeptides of 843 (Un1) and 443 (Un2) amino acids suggesting differential splicing of a single gene transcript. Northern blot analysis of human rhabdomyosarcoma cell poly (A) RNA with cDNA specific for Un1 identified transcripts of approximately 4.2, 6.5, and 8.5 kilobases, whereas a probe specific for Un2 detected a single mRNA of approximately 5 kilobases. Since a monoclonal antibody that is reactive with a sequence encoded by Un1 and not by Un2 detects the bands considered characteristic for undulin in Western blots, the mRNAs related to Un1 may code for the major part of the undulin molecule. The protein sequences deduced from Un1 and Un2 reveal an amino-terminal differentially spliced von Willebrand factor A domain, characteristic of proteins that interact with interstitial collagens, which is linked to fibronectin-like type III homology units by a unique sequence of 57 amino acids. Whereas Un2 encodes two complete and one incomplete type III homologies followed by a unique acidic carboxyl-terminal domain of 118 amino acids, Un1 codes for seven complete and one truncated type III homologies, followed by a short proline-rich carboxyl-terminal segment of 23 amino acids. Considering the 298 amino acids occurring in identical segments, the 989 different amino acid positions deduced from clones Un1 and Un2 represent an estimated 40% of the overall undulin sequence. In the context of 1) rotary shadowing electron microscopy data showing undulin as a structure composed of nodules that are interconnected by flexible rods of varying size, 2) the presence of three major bands of Mr 270,000, 190,000, and 180,000 in sodium dodecyl sulfate-polyacrylamide gel electrophoresis, with 3) common antigenic epitopes and similar peptide maps (Schuppan, D., Cantaluppi, M.C., Becker, J., Veit, A., Bunte, T., Troyer, D., Schuppan, F., Schmid, M., Ackermann, R., and Hahn, E.G. (1990) J. Biol. Chem. 265, 8823-8832), our finding of differentially spliced type III homology units, as found in tenascin and fibronectin, suggests that undulin is another member of the fibronectin-tenascin family of extracellular matrix glycoproteins. Furthermore, as in fibronectin and tenascin, undulin bears an additional subset of interactive domains tailored to specific structural and functional roles in development and differentiation.

Enzymatic amidation of recombinant (Leu27) growth hormone releasing hormone-Gly45.

By chemoenzymatic synthesis the gene for a (Leu27) analogue of human growth hormone releasing hormone-Gly45 [(Leu27)GHRH-Gly45] was constructed, cloned and expressed in Escherichia coli as a fusion protein with beta-galactosidase under the control of the lac promoter and operator. Upon induction with isopropyl-D-thio-beta-galactopyranoside the fusion protein accumulated to a yield of 15-20% of the total cellular protein. After cyanogen bromide cleavage of the fusion protein the precursor peptide (Leu27)hGHRH-Gly45 was separated by extraction and purified by ion exchange and h.p.l.c.-RP18 chromatography. The purified peptide was analysed by sequencing, isoelectric focusing, amino acid analysis and amino acid analysis after V8 protease digestion. The carboxy-terminal glycine was subsequently amidated by PAM (peptidylglycine-alpha-amidating-monooxygenase), an enzyme which was isolated and characterized from fresh bovine pituitaries. Correct amidation of the penultimate amino acid, leucine, was verified by peptide sequencing with an authentic leucine amide reference.