Modulation of the hydrophobicity of glutamine synthetase by mixed-function oxidation

Oxidative modification of Escherichia coli glutamine synthetase renders the enzyme susceptible to proteolytic degradation by a specific protease purified from the bacterium; native enzyme is not a substrate for the protease. A model oxidizing system consisting of ascorbate, iron, and oxygen was used to generate a series of glutamine synthetases of increasing oxidative modification. We assessed the effect of oxidative modification on the surface hydrophobicity of the glutamine synthetases, utilizing hydrophobic chromatography on a phenyl matrix. Initial exposure to the oxidizing system caused inactivation of the enzyme and generated a protein that was more hydrophilic than the native form; it was not a substrate for the protease. Continued exposure to the oxidizing system yielded a protein with additional oxidative modification. This form was distinctly more hydrophobic than the native form and it was very susceptible to proteolytic attack by the purified protease. Thus, oxidative modification modulates the surface hydrophobicity of glutamine synthetase, and this modulation can control susceptibility to proteolysis.     

Protease So from Escherichia coli preferentially degrades oxidatively damaged glutamine synthetase

After oxidative damage (e.g. induced with iron, ascorbate, and oxygen), the inactivated glutamine synthetase is selectively hydrolyzed in extracts of Escherichia coli. We therefore tested if glutamine synthetase treated with this system is hydrolyzed preferentially by any of the known E. coli proteases. Protease So, a cytoplasmic serine protease, was found to degrade the oxidized form of glutamine synthetase to acid-soluble peptides 5-10 times faster than the native glutamine synthetase. Degradation of the oxidized glutamine synthetase was inhibited by EDTA and stimulated 5-10-fold by Mg2+, Ca2+, or Mn2+, even though casein hydrolysis by protease So is not affected by divalent cations. Apparently, these cations affect the conformation of this substrate, making it more susceptible to proteolytic attack. Protease Re, another cytoplasmic protease, also degrades preferentially the oxidized form of glutamine synthetase and seems to correspond to the glutamine synthetase-degrading activity recently described by Roseman and Levine [1987) J. Biol. Chem. 262, 2101-2110). However, it is much less active in this reaction than protease So. No other soluble E. coli protease, including Do, Ci, Mi, Fa, Pi, or the ATP-dependent proteases Ti and La (the lon product), appears to degrade this oxidized protein. These results suggest that protease So participates in the hydrolysis of oxidatively damaged proteins and that E. coli has multiple systems for degrading different types of aberrant proteins.     

Limited proteolysis of glutamine synthetase is inhibited by glutamate and by feedback inhibitors.

Limited proteolysis of glutamine synthetase from Escherichia coli has been studied under nondenaturing conditions (pH 7.6, 20 degrees C). Trypsin cleaves the polypeptide chain of glutamine synthetase into two principal fragments, Mr = about 32,000 and 18,000. The covalently bound AMP group is attached to the larger fragment and its presence does not affect cleavage. Although the cleaved polypeptide chain does not dissociate under nondenaturing conditions, catalytic activity is lost. Chymotrypsin and Staphylococcus aureus protease produce similar cleavages in glutamine synthetase. The substrate L-glutamate retards tryptic as well as chymotryptic digestion. Tryptic digestion is also retarded by some of the feedback inhibitors of glutamine synthetase including CTP, L-alanine, L-serine, L-histidine, and glucosamine 6-phosphate. An implication of these findings is that there is a region of the glutamine synthetase polypeptide chain that is particularly susceptible to proteolysis. Either the glutamate and inhibitor sites are formed partly by this suceptible peptide or the binding of glutamate and some inhibitors induces conformational changes within the E. coli glutamine synthetase molecule in the region of the susceptible peptide.     

Enhanced degradation of oxidized glutamine synthetase in vitro and after microinjection into hepatoma cells

Mixed-function oxidation of Escherichia coli glutamine synthetase has previously been suggested to mark the enzyme for intracellular degradation, and in vitro studies have demonstrated that oxidation renders the enzyme susceptible to proteolytic attack. In this study, the susceptibility of glutamine synthetase to degradation by purified proteases has been compared with the rate of degradation after microinjection into hepatoma cells. Upon exposure to an ascorbate mixed-function oxidation system the enzyme rapidly loses most of its activity, but further oxidation is required to cause susceptibility to extensive proteolytic attack either by a high-molecular-weight liver cysteine proteinase or by trypsin. The rate of degradation of biosynthetically 14C-labeled native and oxidized glutamine synthetase preparations after injection into hepatoma cells parallels their susceptibility to proteolysis in vitro. Native enzyme preparations and enzyme oxidatively inactivated, but not susceptible to extensive degradation by purified proteases, had similar intracellular half-lives; however, oxidized enzyme preparations that were susceptible to proteolytic breakdown in vitro were degraded almost ten times faster than the native enzyme within the growing hepatoma cells. These results suggest that the same features of the oxidized enzyme that render it susceptible to proteolysis in vitro are also recognized by the intracellular degradation system. In addition, they show that loss of enzyme activity does not necessarily imply decreased metabolic stability.     

Oxidation of Neurospora crassa glutamine synthetase.

The glutamine synthetase of Neurospora crassa, either purified or in cell extracts, was inactivated by ascorbate plus FeCl3 and by H2O2 plus FeSO4. The inactivation reaction was oxygen dependent, inhibited by MnCl2 and EDTA, and stimulated in cell extracts by sodium azide. This inactivation could also be brought about by adding NADPH to the cell extract. The alpha and beta polypeptides of the active glutamine synthetase were modified by these inactivating reactions, giving rise to two novel acidic polypeptides. These modifications were observed with the purified enzyme, with cell extracts, and under in vivo conditions in which glutamine synthetase is degraded. The modified glutamine synthetase was more susceptible to endogenous phenylmethylsulfonyl fluoride-insensitive proteolytic activity, which was inhibited by MnCl2 and stimulated by EDTA. The possible physiological relevance of enzyme oxidation is discussed.     

Preferential degradation of the oxidatively modified form of glutamine synthetase by intracellular mammalian proteases

Four intracellular proteases partially purified from liver preferentially degraded the oxidatively modified (catalytically inactive) form of glutamine synthetase. One of the proteases was cathepsin D which is of lysosomal origin; the other three proteases were present in the cytosol. Two of these were calcium-dependent proteases with different calcium requirements. The low-calcium-requiring type (calpain I) accounted for most of the calcium-dependent activity of both mouse and rat liver. The calcium-independent cytosolic protease, referred to as the alkaline protease, has a molecular weight of 300,000 determined by gel filtration. Native glutamine synthetase was not significantly degraded by the cytosolic proteases at physiological pH, but oxidative modification of the enzyme caused a dramatic increase in its susceptibility to attack by these proteases. In contrast, trypsin and papain did degrade the native enzyme and the degradation of modified glutamine synthetase was only 2- to 4-fold more rapid. Adenylylation of glutamine synthetase had little effect on its susceptibility to proteolysis. Although major structural modifications such as dissociation, relaxation, and denaturation also increased the rate of degradation, the oxidative modification is a specific type of covalent modification which could occur in vivo. Oxidative modification can be catalyzed by a variety of mixed function oxidase systems present within cells and causes inactivation of a number of enzymes. Moreover, the presence of cytosolic proteases which recognize the oxidized form of glutamine synthetase suggests that oxidative modification may be involved in intracellular protein turnover.     

The effect of mixed-function oxidation of enzymes on their susceptibility to degradation by a nonlysosomal cysteine proteinase

Mixed-function oxidation of Escherichia coli glutamine synthetase by ascorbate, oxygen, and iron has previously been shown to cause inactivation of the enzyme and enhanced susceptibility to proteolytic attack by a variety of proteases. One of these proteases, from rat liver, is a high molecular weight cysteine proteinase which does not degrade native glutamine synthetase at neutral pH. Although inactive, the oxidized glutamine synthetase preparations used in this study were only partially degraded by this proteinase. Some of the subunits were degraded to acid soluble products with no detectable intermediates; the remaining subunits had not become susceptible to proteolytic attack during the limited exposure to the ascorbate mixed-function oxidation system. Several mammalian enzymes which are known to be inactivated by mixed-function oxidation were tested as substrates for the proteinase. Native rabbit muscle enolase and pyruvate kinase were resistant to degradation, but their oxidatively inactivated forms were degraded. Oxidized phosphoglycerate kinase and creatine kinase were also preferentially degraded. Moreover, trypsin degraded oxidized preparations of all of these enzymes faster than control preparations. Oxidative inactivation of superoxide dismutase by hydrogen peroxide caused a slight increase in susceptibility to proteolytic attack, but the enzyme was still relatively resistant to degradation both by the cysteine proteinase and by trypsin. Although oxidation conditions may not have been optimal for demonstrating enhanced proteolytic susceptibility, the results do indicate that mixed-function oxidation can render some mammalian enzymes, as well as bacterial glutamine synthetase, susceptible to degradation. Mixed-function oxidation of these proteins may be a mechanism of marking them for intracellular turnover.     

Oxidative modification of glutamine synthetase. II. Characterization of the ascorbate model system

The first step in the proteolytic degradation of bacterial glutamine synthetase is a mixed function oxidation of one of the 16 histidine residues in the glutamine synthetase subunit (Levine, R.L. (1983) J. Biol. Chem. 258, 11823-11827). A model system, consisting of oxygen, a metal ion, and ascorbic acid, mimics the bacterial system in mediating the oxidative modification of glutamine synthetase. This model system was studied to gain an understanding of the mechanism of oxidation and of factors which control the susceptibility of the enzyme to oxidation. Availability of substrates and the extent of covalent modification of the enzyme (adenylylation) interact to modulate susceptibility of the enzyme to oxidation. This interaction provides the biochemical basis for physiologic regulation of intracellular proteolysis of glutamine synthetase. The oxidative modification requires hydrogen peroxide. While the reaction may involve Fenton chemistry, the participation of free radicals, superoxide anion, and singlet oxygen could not be demonstrated.     

Role of glutamine synthetase in the uptake and metabolism of methylammonium by Azotobacter vinelandii.

Methylammonium is a substrate for the ammonium transport system of Azotobacter vinelandii. During cellular uptake methylammonium is rapidly converted to a less polar metabolite (E. M. Barnes, Jr., and P. Zimniak, J. Bacteriol. 146:512-516, 1981). This metabolite has been isolated from A. vinelandii and identified as gamma-glutamylmethylamide by mass spectroscopy, 1H nuclear magnetic resonance spectroscopy, and cochromatography with the authentic compound. Escherichia coli also accumulated gamma-glutamylmethylamide during methylammonium uptake. The biosynthesis of gamma-glutamylmethylamide in vitro required methylammonium, ATP, L-glutamate, and a soluble cell extract from A. vinelandii. The enzyme responsible for gamma-glutamylmethylamide synthesis was glutamine synthetase. In a crude extract, L-methionine-DL-sulfoximine was equipotent in inhibiting the activities for gamma-glutamyltransferase and for the synthesis of glutamine and gamma-glutamylmethylamide. Likewise, an antiserum against the glutamine synthetase of E. coli precipitated the transferase and both synthetic activities at similar titers. During repression by growth of cells on ammonium medium, the synthesis of glutamine and gamma-glutamylmethylamide in vitro was also inhibited coordinately. A partially purified preparation of glutamine synthetase from A. vinelandii utilized methylammonium as substrate (Km = 78 mM, Vmax = 0.30 mumol/min per mg), although less efficiently than ammonium (Km = 0.089 mM, Vmax = 1.1 mumol/min per mg). The kinetic properties of glutamine synthetase with methylammonium as substrate as well as the insensitivity of this activity to inhibition by T1+ were strikingly different from methylammonium translocation. Thus, methylammonium (ammonium) translocation and intracellular trapping as glutamylamides are experimentally distinguishable processes.     

Purification of PII and PII-UMP and In Vitro Studies of Regulation of Glutamine Synthetase in Rhodospirillum rubrum

The P(II) protein from Rhodospirillum rubrum was fused with a histidine tag, overexpressed in Escherichia coli, and purified by Ni(2+)-chelating chromatography. The uridylylated form of the P(II) protein could be generated in E. coli. The effects on the regulation of glutamine synthetase by P(II), P(II)-UMP, glutamine, and alpha-ketoglutarate were studied in extracts from R. rubrum grown under different conditions. P(II) and glutamine were shown to stimulate the ATP-dependent inactivation (adenylylation) of glutamine synthetase, which could be totally inhibited by alpha-ketoglutarate. Deadenylylation (activation) of glutamine synthetase required phosphate, but none of the effectors studied had any major effect, which is different from their role in the E. coli system. In addition, deadenylylation was found to be much slower than adenylylation under the conditions investigated.     

Purification of a liver alkaline protease which degrades oxidatively modified glutamine synthetase. Characterization as a high molecular weight cysteine proteinase

A nonlysosomal alkaline protease which degrades the oxidatively modified form of Escherichia coli glutamine synthetase has been purified to apparent homogeneity from rat and mouse liver acetone powders. Its molecular weight was determined to be 300,000 by Sephacryl S-300 gel filtration but results of further studies using high pressure liquid chromatography gel filtration suggest a value of 650,000. Examination of the subunit structure by sodium dodecyl sulfate-polyacrylamide gel electrophoresis showed multiple bands of molecular weights between 22,000 and 34,000. The alkaline protease was inhibited by thiol reagents. Phenylmethylsulfonyl fluoride, aprotinin, leupeptin, antipain, and chymostatin partially inhibited the protease. The inhibition by phenylmethylsulfonyl fluoride was prevented by dithiothreitol, and alpha 1-antitrypsin and soybean trypsin inhibitor did not inhibit. No inhibition was observed with metalloprotease inhibitors. The alkaline protease is active over a broad range of pH with optimum activity for the degradation of oxidized glutamine synthetase around pH 9.0. Its activity is not stimulated by MgATP. A study of the products of insulin B chain degradation demonstrated major cleavage sites at Gln13-Ala14, Leu15-Tyr16, Cys(SO3H)19-Gly20, Gln4-His5, and Leu17-Val18. Based on its endopeptidase activity and its inhibitor specificity, the alkaline protease should be classified as a cysteine proteinase. It appears to be distinct from previously described proteinases and is likely involved in nonlysosomal mechanisms of intracellular protein turnover.     

Regulation and properties of the glutamine synthetase purified from Photobacterium phosphoreum

Glutamine synthetase from a marine enterobacterium, Photobacterium phosphoreum, was purified to homogeneity from cells grown in glycerol-yeast extract medium. The purified enzyme had a molecular weight of approximately 670,000 and a subunit size of 56,000, i.e. larger than that of the enzyme from E. coli. Regulation of the glutamine synthetase activity by adenylylation/deadenylylation was demonstrated on snake venom phosphodiesterase treatment. The state of adenylylation appeared to influence both the biosynthetic and gamma-glutamyltransferase activities of P. phosphoreum glutamine synthetase similar to in the case of the E. coli enzyme. The enzyme activity was controlled by adenylylation and possibly in combination with feedback inhibition by alanine, serine, and glycine, metabolites which are especially effective in inhibiting P. phosphoreum glutamine synthetase. When either Mn2+ or Mg2+ was added to the relaxed (divalent cation-free) enzyme, similar UV-difference spectra were obtained for the enzyme, indicating that the conformational states induced by these cations were also similar. The profile of these spectra varied from those published for E. coli, and three peaks were four 1 at 282.5, 288.5, and 298 nm.     

Expression of the Bacillus subtilis glutamine synthetase gene in Escherichia coli.

The structural gene for glutamine synthetase (glnA) in Bacillus subtilis ( glnAB ) cloned in the lambda vector phage Charon 4A was used to transduce a lysogenic glutamine auxotrophic Escherichia coli strain to prototrophy. The defective E. coli gene ( glnAE ) was still present in the transductant since it could be transduced. In addition, curing of the prototroph resulted in the restoration of glutamine auxotrophy. Proteins in crude extracts of the transductant were examined by a "Western blotting" procedure for the presence of B. subtilis or E. coli glutamine synthetase antigen; only the former was detected. Growth of the strain in media without glutamine was not curtailed even when the bacteriophage lambda pL and pRM promoters were hyperrepressed . The specific activities and patterns of derepression of glutamine synthetase in the transductant were similar to those of B. subtilis, with no evidence for adenylylation. The information necessary for regulation of glnAB must be closely linked to the gene and appears to function in E. coli.     

glnA mutations conferring resistance to methylammonium in Escherichia coli K12

Cells of Escherichia coli K12 were sensitive to 100 mM-methylammonium when cultured under nitrogen limitation, and resistant when grown with an excess of either NH4Cl or glutamine. Glutamine synthetase activity was required for expression of the methylammonium-sensitive phenotype. Mutants were isolated which were resistant to 100 mM-methylammonium, even when grown under nitrogen limitation. P1 bacteriophage transduction and F' complementation analysis revealed that the resistance-conferring mutations mapped either inside the glnA structural gene and/or elsewhere in the E. coli chromosome. Glutamine synthetase was purified from the wild-type and from some of the mutant strains. Strains carrying glnA-linked mutations that were solely responsible for the methylammonium-resistant phenotype yielded an altered enzyme, which was less active biosynthetically with either ammonium or methylammonium as substrate. Sensitivity to methylammonium appeared to be due to synthesis of gamma-glutamylmethylamide by glutamine synthetase, which was synthesized poorly, if at all, by mutants carrying an altered glutamine synthetase enzyme.     

Serine proteases from two cell types target different components of a complex that governs regulated intramembrane proteolysis of pro-sigmaK during Bacillus subtilis development

Upon starvation Bacillus subtilis undergoes a developmental process involving creation of two cell types, the mother cell and forespore. A signal in the form of a serine protease, SpoIVB, is secreted from the forespore and leads to regulated intramembrane proteolysis (RIP) of pro-sigmaK, releasing active sigmaK into the mother cell. RIP of pro-sigmaK is carried out by a membrane-embedded metalloprotease, SpoIVFB, which is inactive when bound by BofA and SpoIVFA. We have investigated the mechanism by which this complex is activated. By expressing components of the signalling pathway in Escherichia coli, we reconstructed complete inhibition of pro-sigmaK RIP by BofA and SpoIVFA, and found that SpoIVB serine protease activity could partially restore RIP, apparently by targeting SpoIVFA. Pulse-chase experiments demonstrated that SpoIVFA synthesized early during B. subtilis sporulation is lost in a SpoIVB-dependent fashion, coincident with the onset of pro-sigmaK RIP, supporting the idea that SpoIVB targets SpoIVFA to trigger RIP of pro-sigmaK. Loss of BofA depended not only on SpoIVB, but also on CtpB, a serine protease secreted from the mother cell. CtpB appeared to cleave BofA near its C-terminus upon coexpression in E. coli, and purified CtpB degraded BofA. We propose that RIP of pro-sigmaK involves a three-step proteolytic cascade in which SpoIVB first cleaves SpoIVFA, CtpB then cleaves BofA and finally SpoIVFB cleaves pro-sigmaK.     

Inactivation of bacterial glutamine synthetase by ADP-ribosylation

Glutamine synthetase from Escherichia coli was inactivated by chemical modification with arginine-specific reagents (Colanduoni, J. A., and Villafranca, J. J. (1985) Biochem. Biophys. Res. Commun. 126, 412-418). E. coli glutamine synthetase was also a substrate for an erythrocyte NAD:arginine ADP-ribosyltransferase. Transfer of one ADP-ribosyl group/subunit of glutamine synthetase caused loss of both biosynthetic and gamma-glutamyltransferase activity. The ADP-ribose moiety was enzymatically removed by an erythrocyte ADP-ribosylarginine hydrolase, resulting in return of function. The site of ADP-ribosylation was arginine 172, determined by isolation of the ADP-ribosylated tryptic peptide. Arginine 172 lies in a central loop that extends into the core formed by the 12 subunits of the native enzyme. The central loop is important in anchoring subunits together to yield the spatial orientation required for catalytic activity. ADP-ribosylation may thus inactivate glutamine synthetase by disrupting the normal subunit alignment. Enzyme-catalyzed ADP-ribosylation may provide a simple, specific technique to probe the role of arginine residues in the structure and function of proteins.     

The inactivation of glucosamine synthetase from bacteria by anticapsin, the C-terminal epoxyamino acid of the antibiotic tetaine

Incubation of anticapsin with the purified glucosamine synthetase (2-amino-2-deoxy-D-glucose-6-phosphate ketol-isomerase, amino transferring, EC 5.3.1.19) from Escherichia coli, Pseudomonas aeruginosa, Arthrobacter aurescens and Bacillus thuringiensis led to the formation of an inactive enzyme irreversibly modified. The inactivation reaction followed pseudo-first-order kinetics. The rate of the inactivation reaction at various concentrations of anticapsin exhibited saturation kinetics, implying that anticapsin binds reversibly to the enzyme prior to inactivation. The determined Kinact is in the range of 10(-5) M (B. thuringiensis) and 10(-6) M (E. coli, P. aeruginosa, A. aurescens ). The addition of glutamine protected the amidotransferase from inactivation by anticapsin . The anticapsin was demonstrated to be a mixed type or competitive inhibitor with respect to glutamine with a Ki value of 10(-6) to 10(-7) M. Reaction of anticapsin with the enzyme exhibits the characteristics of affinity labelling of the glutamine binding site. Chemical modification of the enzyme thiol group with various reagents, 5,5'-dithiobis-(2-nitrobenzoic) acid, 6,6'- dithiodinicotinic acid, 1,1'- dithiodiformamidine , N-ethylmaleimide and iodoacetamide, resulted in an inactive enzyme.     

Glutamine synthetase of pseudomonads: some biochemical and physicochemical properties.

The glutamine synthetases from several Pseudomonas species were purified to homogeneity, and their properties were compared with those reported for the enzymes from Escherichia coli and other gram-negative bacteria. The glutamine synthetase from Pseudomonas fluorescens was unique because it was nearly precipitated quantitatively as a homogeneous protein during dialysis of partially purified preparations against buffer containing 10 mM imidazole (pH 7.0) and 10 mM MnCl2. The glutamine synthetases from Pseudomonas putida and Pseudomonas aeruginosa were purified by affinity chromatography on Affi-blue gel. Dodecamerous forms of the E. coli and P. fluorescens glutamine synthetases had identical mobilities during polyacrylamide gel electrophoresis. Their dissociated subunits, however, migrated differently and were readily separated by electrophoresis on polyacrylamide gels containing 0.1% sodium dodecyl sulfate. This difference in subunit mobilities is not related to the state of adenylylation. Regulation of the Pseudomonas glutamine synthetase activity is mediated by an adenylylation-deadenylylation cyclic cascade system. A sensitive procedure was developed for measuring the average number of adenylylated subunits per enzyme molecule for the glutamine synthetase from P. fluorescens. This method takes advantage of the large differences in transferase activity of the adenylylated and unadenylylated subunits at pH 6.0 and of the fact that the activities of both kinds of subunits are the same at pH 8.45.     

Purification and characterization of protease Re, a cytoplasmic endoprotease in Escherichia coli.

Protease Re, a new cytoplasmic endoprotease in Escherichia coli, was purified to homogeneity by conventional procedures, using [3H]casein as the substrate. The enzyme consists of a single polypeptide of 82,000 molecular weight. It is maximally active between pH 7 and 8.5 and is independent of ATP. It has a pI of 6.8 and a Km of 10.8 microM for casein. Since diisopropyl fluorophosphate and phenylmethylsulfonyl fluoride inhibited this enzyme, it appears to be a serine protease. Protease Re was sensitive to inhibition by L-1-tosylamido-2-phenylethylchloromethylketone but not to that by 1-chloro-3-tosylamido-7-aminoheptanone, thiol-blocking reagents, chelating agents, or various peptide aldehydes. Re also degraded [125I]globin, [125I]glucagon, and 125I-labeled denatured bovine serum albumin to acid-soluble products (generally oligopeptides of greater than 1,500 daltons), but it showed no activity against serum albumin, growth hormone, insulin, or a variety of fluorometric peptide substrates. It also hydrolyzed oxidatively inactivated glutamine synthetase (generated by ascorbate, oxygen, and iron) four- to fivefold more rapidly than the native protein. Protease Re appears to be identical to the proteolytic enzyme isolated by Roseman and Levine (J. Biol. Chem. 262:2101-2110, 1987) by its ability to degrade selectively oxidatively damaged glutamine synthetase in vivo. Its role in intracellular protein breakdown is uncertain.