Amidation of salicyluric acid and gentisuric acid: a possible role for peptidylglycine alpha-amidating monooxygenase in the metabolism of aspirin.

Bifunctional peptidylglycine alpha-amidating monooxygenase (PAM) catalyzes the copper-, ascorbate-, and O2-dependent cleavage of C-terminal glycine-extended peptides, N-acylglycines, and the bile acid glycine conjugates to the corresponding amides and glyoxylate. Two known metabolites of aspirin, salicyluric acid and gentisuric acid, are also substrates for PAM, leading to the formation of salicylamide and gentisamide. The time course for O2 consumption and glyoxylate production indicates that salicylurate amidation is a two-step reaction. Salicylurate is first converted to N-salicyl-alpha-hydroxyglycine, which is ultimately dealkylated to salicylamide and glyoxylate. The enzymatically generated salicylamide and N-salicyl-alpha-hydroxyglycine were characterized by mass spectrometry and two-dimensional 1H-13C heteronuclear multiple quantum coherence NMR.

The enzymatic formation of novel bile acid primary amides.

Bifunctional peptidylglycine alpha-amidating monooxygenase (PAM) catalyzes the copper-, ascorbate-, and O(2)-dependent cleavage of C-terminal glycine-extended peptides and N-acylglycines to the corresponding amides and glyoxylate. The alpha-amidated peptides and the long-chain acylamides are hormones in humans and other mammals. Bile acid glycine conjugates are also substrates for PAM leading to the formation of bile acid amides. The (V(MAX)/K(m))(app) values for the bile acid glycine conjugates are comparable to other known PAM substrates. The highest (V(MAX)/K(m))(app) value, 3.1 +/- 0.12 x 10(5) M(-1) s(-1) for 3-sulfolithocholylglycine, is 6.7-fold higher than that for d-Tyr-Val-Gly, a representative peptide substrate. The time course for O(2) consumption and glyoxylate production indicates that bile acid glycine conjugate amidation is a two-step reaction. The bile acid glycine conjugate is first converted to an N-bile acyl-alpha-hydroxyglycine intermediate which is ultimately dealkylated to the bile acid amide and glyoxylate. The enzymatically produced bile acid amides and the carbinolamide intermediates were characterized by mass spectrometry and two-dimensional (1)H-(13)C heteronuclear multiple quantum coherence NMR.

Induction of peptidylglycine alpha-amidating monooxygenase in N(18)TG(2) cells: a model for studying oleamide biosynthesis.

The fatty-acid primary amide, oleamide, is a novel signaling molecule whose mechanism of biosynthesis is unknown. Recently, the N(18)TG(2) cell line was shown to synthesize oleamide from oleic acid, thereby demonstrating that these cells contain the necessary catalytic activities for generating the fatty-acid primary amide. The ability of peptide alpha-amidating enzyme, peptidylglycine-alpha-amidating monooxygenase (PAM; EC 1.14.17.3), to catalyze the formation of oleamide from oleoylglycine in vitro suggests this as a function for the enzyme in vivo. This investigation shows that N(18)TG(2) cells, in fact, express PAM and that cellular differentiation dramatically increases this expression. PAM expression was confirmed by the detection of PAM mRNA, PAM protein, and enzymatic activity that exhibits the functional characteristics of PAM isolated from mammalian neuroendocrine tissues. The regulated expression of PAM in N(18)TG(2) cells is consistent with the proposed role of PAM in the biosynthesis of fatty-acid primary amides and further establishes this cell line as a model for studying the pathway.

N-acylglycine amidation: implications for the biosynthesis of fatty acid primary amides.

Bifunctional peptidylglycine alpha-amidating enzyme (alpha-AE) catalyzes the O2-dependent conversion of C-terminal glycine-extended prohormones to the active, C-terminal alpha-amidated peptide and glyoxylate. We show that alpha-AE will also catalyze the oxidative cleavage of N-acylglycines, from N-formylglycine to N-arachidonoylglycine. N-Formylglycine is the smallest amide substrate yet reported for alpha-AE. The (V/K)app for N-acylglycine amidation varies approximately 1000-fold, with the (V/K)app increasing as the acyl chain length increases. This effect is largely an effect on the KM,app; the KM,app for N-formylglycine is 23 +/- 0.88 mM, while the KM,app for N-lauroylglycine and longer chain N-acylglycines is in the range of 60-90 microM. For the amidation of N-acetylglycine, N-(tert-butoxycarbonyl)glycine, N-hexanoylglycine, and N-oleoylglycine, the rate of O2 consumption is faster than the rate of glyoxylate production. These results indicate that there must be the initial formation of an oxidized intermediate from the N-acylglycine before glyoxylate is produced. The intermediate is shown to be N-acyl-alpha-hydroxyglycine by two-dimensional 1H-13C heteronuclear multiple quantum coherence (HMQC) NMR.

Kinetic mechanism and intrinsic isotope effects for the peptidylglycine alpha-amidating enzyme reaction.

The bifunctional peptidylglycine alpha-amidating enzyme catalyzes the C-terminal amidation of glycine-extended peptides. The first enzyme activity, peptidylglycine alpha-hydroxylating monooxygenase, catalyzes the oxygen-, ascorbate-, and copper-dependent formation of alpha-hydroxyglycine derivatives. These are substrates for the second enzyme activity, peptidylamidoglycolate lyase, which catalyzes their breakdown to the corresponding C-terminal amidated peptide and glyoxylate as final products. Kinetic and isotope effect studies were carried out with N-benzoylglycine as a substrate at pH 6.0 using monofunctional and bifunctional monooxygenase activities. Kinetic data indicate an equilibrium ordered mechanism, with hippuric acid binding first followed by oxygen. A potentially important difference between the two monooxygenase activities is that product release occurs more slowly from the bifunctional enzyme, indicating an influence of the lyase domain on release of alpha-hydroxyglycine product to solution. Intrinsic isotope effects for the C-H bond cleavage were measured for the monofunctional form of the enzyme using a double-label tracer method, yielding 10.6 +/- 0.8 and 1.20 +/- 0.03 for the primary and alpha-secondary deuterium intrinsic isotope effects, respectively. These values are identical to previous measurements for the analogous enzyme system, dopamine beta-monooxygenase [Miller, S. M., and Klinman, J. P. (1985) Biochemistry 24, 2114-2127]. The identity of intrinsic isotope effects for peptidylglycine alpha-hydroxylating monooxygenase and dopamine beta-monooxygenase with substrates of comparable reactivity (N-benzoylglycine and dopamine, respectively) extends similarities between the two enzymes significantly beyond sequence homology and cofactor requirements.

Structural and functional investigations on the role of zinc in bifunctional rat peptidylglycine alpha-amidating enzyme.

Bifunctional peptidylglycine alpha-amidating enzyme (alpha-AE) catalyzes the two-step conversion of C-terminal glycine-extended peptides to C-terminal alpha-amidated peptides and glyoxylate. The first step is the ascorbate-, O2-, and copper-dependent hydroxylation of the alpha-carbon of the glycyl residue, producing an alpha-hydroxyglycine-extended peptide. The second step is the ascorbate-, O2-, and copper-independent dealkylation of the carbinolamide intermediate. We show that alpha-AE requires 1.1 +/- 0. 2 mol of zinc/mol of enzyme for maximal (S)-N-dansyl-Tyr-Val-alpha-hydroxyglycine dealkylation activity. Treatment of the enzyme with EDTA abolishes both the peptide hydroxylation and the carbinolamide dealkylation activities. Addition of Zn(II), Co(II), Cd(II), and Mn(II) partially restores carbinolamide dealkylation activity to the EDTA-treated enzyme. Addition of Co(II) produces the greatest restoration of dealkylation activity, 32% relative to a control not treated with EDTA, while Mn(II) addition results in the smallest restoration of dealkylation activity, only 3% relative to an untreated control. The structure and coordination of the zinc center has been investigated by X-ray absorption spectroscopy. EXAFS data are best interpreted by an average coordination of 2-3 histidine ligands and 1-2 non-histidine O/N ligands. Since catalytic zinc centers in other zinc metalloenzymes generally exhibit only O/N ligands to the zinc atom, a zinc-bound water or hydroxide may serve as a general base for the abstraction of the hydroxyl proton from the carbinolamide intermediate. Alternatively, the zinc may function in a structural role.

Fatty acid amide biosynthesis: a possible new role for peptidylglycine alpha-amidating enzyme and acyl-coenzyme A: glycine N-acyltransferase.

Fatty acid primary amides have recently been recognized as mammalian hormones [Cravatt et al. (1995) Science 268, 1506-1509]. The route to their biosynthesis is unknown. Many mammalian peptide hormones also possess a C-terminal alpha-amide moiety that arises from the posttranslational oxidative cleavage of a C-terminal glycine-extended precursor. The enzyme that catalyzes this reaction is peptidylglycine alpha-amidating enzyme, which is known to preferentially amidate peptide substrates containing a penultimate, hydrophobic amino acid [Tamburini et al. (1990) Int. J. Pept. Protein Res. 35, 153-156]. We show that N-myristoylglycine is a substrate for peptidylglycine alpha-amidating enzyme with a (V/K)app that is 55 +/- 4% of the value measured for D-Tyr-Val-Gly. N-Fatty acylglycines are enzymatically produced in mammals from fatty acyl-coenzyme A (CoAs) and glycine by acyl-CoA:glycine N-acyltransferase. The sequential actions of acyl-CoA:glycine N-acyltransferase and peptidyl-glycine alpha-amidating enzyme would lead to the biosynthesis of fatty acid amides.

The irreversible inactivation of two copper-dependent monooxygenases by sulfite: peptidylglycine alpha-amidating enzyme and dopamine beta-monooxygenase.

Peptidylglycine alpha-amidating enzyme (alpha-AE) and dopamine beta-monooxygenase (D beta M), two copper-dependent monooxygenases that have catalytic and structural similarities, are irreversibly inactivated by sodium sulfite in a time- and concentration-dependent manner. Studies with alpha-AE show that the sulfite-mediated inactivation is dependent on the presence of redox active transition metals free in solution, with Cu(II) being the most effective in supporting the inactivation reaction. Sulfite inactivation of alpha-AE is specific for the monooxygenase reaction of this bifunctional enzyme and amidated peptides provide protection against the inactivation. Consequently, the sulfite-mediated inactivation of alpha-AE and D beta M most likely results from the transition metal-catalyzed oxidation of sulfite to the sulfite radical, SO3-.

The inactivation of bifunctional peptidylglycine alpha-amidating enzyme by benzylhydrazine: evidence that the two enzyme-bound copper atoms are nonequivalent.

Peptidylglycine alpha-amidating enzyme catalyzes the two-step conversion of C-terminal glycine-extended peptides to C-terminal alpha-amidated peptides and glyoxylate in a reaction that requires O2, ascorbate and 2 mol of copper per mole of enzyme [Kulathila et al. (1994) Arch. Biochem. Biophys. 311, 191-195]. Peptides with a C-terminal alpha-hydroxyglycine residue are intermediates in the amidation reaction. Benzylhydrazine inactivates the enzymatic conversion of dansyl-Tyr-Val-Gly to dansyl-Tyr-Val-NH2 in a time- and concentration-dependent manner. In contrast, the enzymatic conversion of dansyl-Tyr-Val-alpha-hydroxyglycine to dansyl-Tyr-Val-NH2 is unaffected by benzylhydrazine. The plot of 1/(inactivation rate) vs 1/[benzylhydrazine] is parabolic, indicating that the inactivation results from the interaction of 2 mol of benzylhydrazine per mole of enzyme. EPR spectra obtained from benzylhydrazine inactivation reactions carried out in the presence of a radical trap, alpha-(4-pyridyl-1-oxide)-N-tert-butylnitrone, show the formation of a carbon-centered benzyl radical. The benzyl radical most likely results from redox chemistry between benzylhydrazine and the enzyme-bound Cu(II) ions because EPR studies show that enzyme-bound Cu(II) is reduced to Cu(I) in the presence of benzylhydrazine. The kinetic constants for benzylhydrazine as a reductant in the amidation reaction were determined at benzylhydrazine concentrations too low to cause significant enzyme inactivation. Mimosine exhibits mixed inhibition vs benzylhydrazine; however, previous results have shown that benzylhydrazine is competitive vs ascorbate [Miller et al. (1992) Arch. Biochem. Biophys. 298, 380-388]. This change in kinetic mechanism coupled with the nonlinear inactivation kinetics have lead to a proposal that the two enzyme-bound Cu(II) atoms are nonequivalent with respect to their reduction by benzylhydrazine.

C-terminal amidated peptides: production by the in vitro enzymatic amidation of glycine-extended peptides and the importance of the amide to bioactivity.

Peptidylglycine alpha-amidating enzyme (alpha-AE) can be used in an in vitro reaction to convert C-terminal glycine-extended peptides to peptide hormones with a C-terminal amino acid amide. Structure-activity data for 45 bioactive peptides show that the C-terminal amide is required for the full biological activity of most amidated peptide hormones. These data emphasize the role alpha-AE can have in amidated peptide production.

Bifunctional peptidylglcine alpha-amidating enzyme requires two copper atoms for maximum activity.

The conversion of C-terminal glycine-extended peptides to C-terminal alpha-amidated peptides occurs in two distinct reactions, both of which are catalyzed by bifunctional peptidylglycine alpha-amidating enzyme. The first step is the alpha-hydroxylation of the C-terminal glycine residue and the second step is the dealkylation of the alpha-hydroxyglycine-extended peptide to the alpha-amidated peptide and glyoxylate. We show that the bifunctional enzyme requires 1.9 +/- 0.2 mol of copper/mol of enzyme for maximal dansyl-Tyr-Lys-Gly amidation activity under the conditions of high enzyme concentration (approximately 80 microM) required to measure initial rates for this poor substrate. The enzyme, as purified, contains a substoichiometric amount of copper and has only trace levels of amidation activity. Addition of exogenous Cu(II) ions stimulates amidation activity approximately 3000-fold at the optimum copper stoichiometry and the enzyme is then inhibited by excess Cu(II). No stimulation of amidation activity is observed upon the addition of the following divalent metal ions: Mn(II), Fe(II), Ni(II), Cd(II), and the oxovanadium cation, VO(II). The enzyme-catalyzed dealkylation of alpha-hydroxyhippuric acid to benzamide shows no dependence on copper, indicating that the copper dependence of the amidation reaction must be attributed to a copper dependence in peptide alpha-hydroxylation.

Transition-state analysis of AMP deaminase.

The transition state of the allosteric AMP deaminase from Saccharomyces cerevisiae has been characterized by 14C and 15N Vmax/Km heavy-atom kinetic isotope effects. The primary 6-14C isotope effect was measured with [6-14C]AMP, and the 6-15N primary isotope effect was measured by isotope ratio mass spectrometry using the natural abundance of 15N in AMP and by using 15N release from ATP as a slow substrate. Isotope effects for AMP as the substrate were measured in the presence and absence of ATP as an allosteric activator and GTP as an allosteric inhibitor. Kinetic isotope effects with [6-14C]AMP were 1.030 +/- 0.003, 1.038 +/- 0.004, and 1.042 +/- 0.003 in the absence of effectors and in the presence of ATP and GTP, respectively. Isotope effects for [6-15N]AMP averaged 1.010 +/- 0.002. Allosteric activation increased the 15N isotope effect to 1.016 +/- 0.003. A primary 15N kinetic isotope effect with ATP, which has a Vmax/Km 10(-6) that for AMP, was 1.013 +/- 0.001. The presence of D2O as solvent caused a marginally significant decrease in the [6-15N]AMP kinetic isotope effect from 1.011 +/- 0.001 to 1.007 +/- 0.002. Previous studies have established that the solvent D2O effect is inverse (0.34) for slow substrates with two or more protons transferred prior to transition state formation and remains inverse (0.79) with AMP as substrate [Merkler, D. J., & Schramm, V. L. (1993) Biochemistry 32, 5792-5799]. Bond vibrational analysis was used to identify transition states for AMP deaminase that are consistent with all kinetic isotope effects.(ABSTRACT TRUNCATED AT 250 WORDS)

Catalytic mechanism of yeast adenosine 5'-monophosphate deaminase. Zinc content, substrate specificity, pH studies, and solvent isotope effects.

Adenosine 5'-monophosphate (AMP) deaminase from baker's yeast is an allosteric enzyme containing a single AMP binding site and two ATP regulatory sites per polypeptide [Merkler, D. J., & Schramm, V. L. (1990) J. Biol Chem. 265, 4420-4426]. The enzyme contains 0.98 +/- 0.17 zinc atom per subunit. The X-ray crystal structure for mouse adenosine deaminase shows zinc in contact with the attacking water nucleophile using purine riboside as a transition-state inhibitor [Wilson, D. K., Rudolph, F. B., & Quiocho, F. A. (1991) Science 252, 1278-1284]. Alignment of the amino acid sequence for yeast AMP deaminase with that for mouse adenosine deaminase demonstrates conservation of the amino acids known from the X-ray crystal structure to bind to the zinc and to a transition-state analogue. On the basis of these similarities, yeast AMP deaminase is also proposed to use a Zn(2+)-activated water molecule to attack C6 of AMP with the displacement of NH3. The pKm and pKi profiles for AMP and a competitive inhibitor overlap in a bell-shaped curve with pKa values of 7.0 and 7.4. This pattern is characteristic of a rapid equilibrium between AMP and the enzyme, thus confirming the rapid equilibrium random kinetic patterns [Merkler, D. J., Wali, A. S., Taylor, J., Schramm, V. L. (1989) J. Biol. Chem. 264, 21422-21430]. The Vmax of the reaction requires one unprotonated and one protonated group with pKa values of 6.4 +/- 0.2 and 7.7 +/- 0.3, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Yeast AMP deaminase. Catalytic activity in Schizosaccharomyces pombe and chromosomal location in Saccharomyces cerevisiae.

The AMP deaminase gene was mapped to chromosome XIII of Saccharomyces cerevisiae strain JM1901. The AMP deaminase gene is located near SUP5, GAL80, SUF7, and SUF22. The presence of AMP deaminase in the fission yeast Schizosaccharomyces pombe was examined by comparing DNA hybridization, protein immunoreactivity, and catalytic activity from S. cerevisiae, known to contain the protein, to S. pombe. DNA hybridization experiments using the cloned S. cerevisiae AMP deaminase gene failed to hybridize to the genomic DNA from S. pombe strain 972h-s. Protein extracts from S. pombe and S. cerevisiae were analyzed in parallel and exhibited comparable AMP deaminase activities. Analysis of reaction intermediates in cell extracts of S. pombe established that IMP is formed directly from AMP without intervening steps. The AMP deaminase of S. pombe was purified 1,100-fold to a specific catalytic activity of 67 mumol/min/mg of protein. Purified protein interacted weakly with polyclonal antibodies prepared against S. cerevisiae AMP deaminase. AMP deaminases from both S. cerevisiae and S. pombe were activated by ATP with micromolar activation constants, are inhibited by coformycin, and are specific for AMP when compared to other purine nucleosides and nucleotides. The results establish that S. pombe contains an AMP deaminase with catalytic properties similar to that from S. cerevisiae, even though the DNA sequences of the genes and the immunoreactivity of the protein from S. pombe differs considerably from the AMP deaminase of S. cerevisiae. Genetic analysis of the pathways of purine metabolism in S. pombe (Pourquié, J., and Heslot, H. (1971) Genet. Res. 18, 33-44) had indicated the absence of AMP deaminase. The presence of a regulated AMP deaminase in S. pombe supports the hypothesis that eukaryotes regulate adenine nucleotide pools by the activity of AMP deaminase.

Production of recombinant salmon calcitonin by in vitro amidation of an Escherichia coli produced precursor peptide.

Salmon calcitonin (sCT) is a 32 amino acid peptide hormone that requires C-terminal amidation for full biological activity. We have produced salmon calcitonin by in vitro amidation of an E. coli produced precursor peptide. Glycine-extended sCT, the substrate for amidation, was produced in recombinant E. coli as part of a fusion with glutathione-S-transferase. The microbially produced soluble fusion protein was purified to near homogeneity by affinity chromatography. Following S-sulfonation of the fusion protein, the glycine-extended peptide was cleaved from the fusion by cyanogen bromide. The S-sulfonated peptide was recovered and enzymatically converted to the amidated peptide in a reaction with recombinant peptidylglycine alpha-amidating enzyme (alpha-AE) secreted from Chinese hamster ovary (CHO) cells. After reformation of the intramolecular disulfide bond, the sCT was purified with a step yield of 60%. The ease and speed of this recombinant process, as well as its potential for scale-up, make it adaptable to production demands for calcitonin, a proven useful agent for the treatment of post-menopausal osteoporosis. Moreover, the relaxed specificity of the recombinant alpha-AE for the penultimate amino acid which is amidated allows the basic process to be applied to the production of other amidated peptides.

Characterization of a bifunctional peptidylglycine alpha-amidating enzyme expressed in Chinese hamster ovary cells.

Peptidylglycine alpha-amidating enzyme (alpha-AE) catalyzes the conversion of glycine-extended prohormones to their biologically active alpha-amidated forms. We have derived a clonal Chinese hamster ovary cell line that secretes significant quantities of active alpha-AE. Enzyme production was increased by selection for methotrexate-resistant cells expressing a dicistronic message. Amplification of the alpha-AE gene was monitored by Southern blot analysis, enzyme activity, and immunoreactive protein throughout the selection process. The soluble enzyme is bifunctional as determined by the ability to convert either the glycine-extended substrate, dansyl-Tyr--Val--Gly, or the intermediate, dansyl-Tyr--Val--alpha-hydroxyglycine, to the dansyl-Tyr--Val--NH2 product. The recombinant alpha-AE was purified by a simple two-step chromatographic process. The purified enzyme is partially glycosylated and the glycosylated and nonglycosylated forms of the enzyme were separated on a Con A-Sepharose column. The kinetic constants for dansyl-Tyr--Val--Gly, dansyl-Tyr--Val--alpha-hydroxyglycine, ascorbate, and catechol were the same for both forms of alpha-AE. In addition, mimosine is competitive vs ascorbate with K(is) = 3.5 microM for the nonglycosylated alpha-AE and K(is) = 4.2 microM for the glycosylated alpha-AE. Therefore, the presence or absence of asparagine-linked oligosaccharide does not affect the catalytic efficiency of the enzyme. Overexpression of the recombinant enzyme in CHO cells greatly enhances expression of the endogenous gene, implicating a feedback mechanism on the alpha-AE gene.

Rapid fluorimetric assay for the detection of the peptidyl alpha-amidating enzyme intermediate using high-performance liquid chromatography.

Peptidylglycine alpha-amidating enzyme catalyzes the conversion of glycine-extended peptides to their corresponding amidated peptides via a stable alpha-hydroxyglycine intermediate. Using a new rapid fluorimetric reversed-phase high-performance liquid chromatographic assay, we have demonstrated that the substrate and product of the amidation reaction, as well as both stereoisomers of the alpha-hydroxyglycine intermediate, can be separated and detected in quantities as low as 1 pmol. The method is highly reproducible and requires less than 11 min for separation and quantification.

18O isotopic 13C NMR shift as proof that bifunctional peptidylglycine alpha-amidating enzyme is a monooxygenase.

The biosynthesis of C-terminal alpha-amidated peptides from their corresponding C-terminal glycine-extended precursors is catalyzed by peptidylglycine alpha-amidating enzyme (alpha-AE) in a reaction that requires copper, ascorbate, and molecular oxygen. Using bifunctional type A rat alpha-AE, we have shown that O2 is the source of the alpha-carbonyl oxygen of pyruvate produced during the amidation of dansyl-Tyr-Val-[alpha-13C]-D-Ala, as demonstrated by the 18O isotopic shift in the 13C NMR spectrum of [alpha-13C]lactate generated from [alpha-13C]pyruvate in the presence of lactate dehydrogenase and NADH. In addition, one-to-one stoichiometries have been determined for glyoxylate formed/dansyl-Tyr-Val-Gly consumed, pyruvate formed/dansyl-Tyr-Val-D-Ala consumed, dansyl-Tyr-Val-NH2 formed/ascorbate oxidized, and dansyl-Tyr-Val-NH2 formed/O2 consumed. Quantitative coupling of NADH oxidation to dansyl-Tyr-Val-NH2 production using Neurospora crassa semidehydroascorbate reductase showed that two one-electron reductions by ascorbate occurred per alpha-AE turnover. The stoichiometry of approximately 1.0 dansyl-Tyr-Val-NH2 produced/ascorbate oxidized observed in the absence of a semidehydroascorbate trap resulted from the disproportionation of two semidehydroascorbate molecules to ascorbate and dehydroascorbate.