Evidence for a metalloprotein structure of plasma membrane 5'-nucleotidase

Oxidative modification of rat liver 5'-nucleotidase: the mechanisms for protection and re-activation

The effect of oxidative stress catalysed by transition metals appears to have a critical relevance for the structure and function not only of membrane lipids but also of integral membrane proteins in a complex lipid-protein assembling, and membrane-dependent function. The integral membrane enzyme 5'-nucleotidase is susceptible to Fe((2+))-ion catalysed oxidative modification, and the extent of enzyme inhibition is in inverse relationship (r = -0.820) with lipid peroxidation (MDA) level. This work is also a comparative study about possible effectiveness of different Fe-ion chelators (deferoxamine, Na-citrate, Na-salicylate, ammonium oxalate and EDTA), antioxidants (GSH, GSH/GSH-Px system, Cu, Zn-SOD and mannitol) and metal cations (Mg(2+) and Mn(2+)) to protect or restore Fe(2+)-ion induced 5'-nucleotidase inhibition and to suppress Fe(2+)-ion enhanced lipid peroxidation. Among the examined chelators it was only deferoxamine and Na-citrate that exerted a fully protective and reactivating ability; among the antioxidants it was only GSH; among the metal cations it was only Mn(2+). The ability to protect or restore 5'-nucleotidase activity and to diminish chain-induced lipid peroxidation is explicable in terms of: metal-binding ability, capacity of taking iron away from a biological molecule, or ability of transferring the damage to itself. After a short incubation period, the iron associated with enzyme or lipid hydroperoxides could be in a labile coordinative linkage, still able to interact with possible ligands or metal cations.

Differences in the cellular zinc content and 5'-nucleotidase activity of normal and acrodermatitis enteropathica (AE) fibroblasts

The acrodermatitis enteropathica (AE) mutation affects zinc (Zn) metabolism in human fibroblasts. We hypothesize that the mutation affects the cell Zn content, which subsequently affects the activity of various zinc-dependent enzymes, such as 5'-nucleotidase. Therefore, normal and AE fibroblasts were grown in normal medium containing physiological levels of Zn (16 micromol/L) for approximately 24 h. The medium was replaced by normal medium (16 micromol/L Zn), Zn-depleted medium (1.5 micromol/L Zn), or Zn-supplemented medium (200 micromol/L Zn) for another 24 h. Regardless of the Zn concentration of the growth medium, the AE fibroblasts contained significantly less Zn than normal fibroblasts grown in comparable medium. Nevertheless, growth of the fibroblasts in 200 micromol/L Zn medium significantly increased the cell Zn content fourfold of both normal and AE fibroblasts. The activity of 5'-nucleotidase in the AE fibroblasts grown in 16 micromol/L Zn or 1.5 micromol/L Zn medium was also significantly lower than in normal fibroblasts. Changing the growth medium from 16 micromol/L Zn to 1.5 micromol/L Zn medium did not affect the activity of the enzyme in either genotype. Cells grown in 200 micromol/L Zn medium exhibited threefold greater 5'-nucleotidase activity in AE fibroblasts, but had no affect on enzyme activity in normal cells. In summary, altering the cell Zn content of normal fibroblasts did not result in a significant change in their 5'-nucleotidase activity. However, AE fibroblasts grown in 200 micromol/L Zn medium exhibited recovery of their 5'-nucleotidase activity to normal levels. These results support the hypothesis that the AE mutation affects the cellular Zn content. The lower cell Zn content subsequently affects the activity of 5'-nucleotidase.

Effect of phenylalanine and its metabolites on ATP diphosphohydrolase activity in synaptosomes from rat cerebral cortex

The in vitro effects of phenylalanine and some of its metabolites on ATP diphosphohydrolase (apyrase, EC 3.6.1.5) activity in synaptosomes from rat cerebral cortex were investigated. The enzyme activity in synaptosomes from rats subjected to experimental hyperphenylalaninemia (alpha-methylphenylalanine plus phenylalanine) was also studied. In the in vitro studies, a biphasic effect of phenylalanine on both enzyme substrates (ATP and ADP) was observed, with maximal inhibition at 2.0 mM and maximal activation at 5.0 mM. Inhibition of the enzyme activity was not due to calcium chelation. Moreover, phenylpyruvate, when compared with phenylalanine showed opposite effects on the enzyme activity, suggesting that phenylalanine and phenylpyruvate bind to two different sites on the enzyme. The other tested phenylalanine metabolites phenyllactate, phenylacetate and phenylethylamine) had no effect on ATP diphosphohydrolase activity. In addition, we found that ATP diphosphohydrolase activity in synaptosomes from cerebral cortex of rats with chemically induced hyperphenylalaninemia was significantly enhanced by acute or chronic treatment. Since it is conceivable that ATPase-ADPase activities play an important role in neurotransmitter (ATP) metabolism, it is tempting to speculate that our results on the deleterious effects of phenylalanine and phenylpyruvate on ATP diphosphohydrolase activity may be related to the neurological dysfunction characteristics of naturally and chemically induced hyperphenylalaninemia.

5'-Nucleotidase: molecular structure and functional aspects

Images

5'-nucleotidase from bull seminal plasma, chicken gizzard and snake venom is a zinc metalloprotein

Using flame atomic absorption spectrometry the tight association of zinc to three different purified 5'-nucleotidases at a molar ratio of 2 could be proven. These 5'-nucleotidases purified from bull seminal plasma (BSP), chicken gizzard (CG) and snake venom (SV) are thus zinc metalloproteins. Removal of zinc results in the loss of their AMPase activity, which could be fully restored after readdition of zinc at a molar ratio of 2, for BSP and CG, and 1.5, for SV 5'-nucleotidase. Reactivation of their AMPase activity after the removal of zinc could also be obtained by addition of cobalt and copper ions, which were found to also bind with a molar ratio of 2 to the three 5'-nucleotidases tested.

Purification of bovine liver cytosolic 5'-nucleotidase. Kinetic and structural studies as compared to the membrane isoenzyme

Cytosolic 5'-nucleotidase from bovine liver has been purified to homogeneity. Two affinity chromatographies on concanavalin A and 5'AMP-Sepharose columns result in a 12,000-fold purification. The sequential elution of glycoproteins from the concanavalin-A-Sepharose column with methyl alpha-D-glucoside and methyl alpha-D-mannoside greatly increases the degree of purification of the enzyme. Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate shows two subunits having apparent molecular masses of 65 kDa and 57 kDa respectively, while only one band at 70 kDa is observed in the case of the membrane-bound 5'-nucleotidase. Both the Stokes radii, measured by gel exclusion HPLC, and the sedimentation coefficient, determined by density gradient ultracentrifugation, indicate that the cytosolic enzyme is a heterodimer of about 130 kDa. This contrasts with the membrane-bound 5'-nucleotidase which is a homodimer of 140 kDa. Moreover, the antibodies raised against the membrane 5'-nucleotidase inhibited the cytosolic form indicating that a common antigenic determinant(s) exists between the two isoenzymes. However, structural differences are revealed by immunoblotting. In the same way, the effect of lectins suggests that differences in the structure of the carbohydrate chains exist between the two isoenzymes. The purified cytosolic enzyme has lower affinity for the nucleotides than does the membrane enzyme. In addition, while ADP, [alpha,beta-CH2]ADP and ATP were strong competitive inhibitors of the membrane enzyme, ADP and ATP activate the cytosolic form and [alpha,beta-CH2]ADP has no effect. Moreover, two pH optima at 7.5 and 9.5 are observed in the cytosolic enzyme while only one at 7.5 occurred in the membrane form. Finally the exogenous cations, MgCl2 and MnCl2, are necessary for the maximal activity of the cytosolic but not of the membrane 5'-nucleotidase. All these observations indicate that the two isoenzymes are different.

Topography, purification and characterization of thyroidal 5'-nucleotidase

1. Subcellular studies of bovine thyroid indicate that 5'-nucleotidase is predominantly associated with plasma membranes, although a considerable part of this ectoenzyme is also found internalized. 2. The enzyme displaying the features of a glycoprotein has been purified 1400 times by detergent solubilization and two subsequent affinity chromatographic steps. 3. Thyroidal 5'-nucleotidase can be classified as an unspecific metallo-dependent 5'-ribonucleotide phosphohydrolase. The native enzyme exists as a dimer (MW 150 kDalton), composed of two similar or identical subunits.

Purification, characterization and cellular localization of 5'-nucleotidase from Torpedo electric organ

5'-Nucleotidase was isolated from the electric organ of the electric ray Torpedo marmorata after solubilization in Triton X-100 and deoxycholate by affinity chromatography on concanavalin A-Sepharose and AMP-Sepharose. The purified enzyme has a Km for AMP of 38 microM, with a maximal velocity of 31 units/mg of protein. Of the purine and pyrimidine mononucleotides, AMP is hydrolysed most effectively. beta-Glycerophosphate, phosphoenolpyruvate and p-nitrophenyl phosphate are not substrates for the enzyme. Adenosine 5'-[alpha, beta-methylene]diphosphate, ADP and ATP are competitive inhibitors in this order of potency. Concanavalin A inhibits enzyme activity in a non-competitive manner. Whereas Mg2+, Ca2+ and Sr2+ activate enzyme activity in the millimolar range, Hg2+, and in particular Pb2+ and Zn2+, inhibit enzyme activity. On SDS/polyacrylamide-gel electrophoresis the enzyme has an apparent Mr of 62000, whereas that of the native deoxycholate-enzyme complex is 131000. An antiserum raised against the native enzyme inhibits enzyme activity. Inhibition studies suggest the presence of tissue-specific variants of the enzyme. By immunohistochemical analysis the enzyme can be localized to the ramifications of nerve terminals in the electric organ.

Chemical modification of essential carboxyl group and histidine residue in the plasma-membrane 5'-nucleotidase

An investigation, using specific chemical reagents, of the amino acids involved in the catalytic activity of the purified 5'-nucleotidase (5'-ribonucleotide phosphohydrolase, EC 3.1.3.5) from bovine liver plasma membranes, was carried out. The enzyme was irreversibly inactivated by N-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinoline (EEDQ). The inhibition kinetics were of the first-order type and decreased partially in the presence of nucleotides and divalent cations. These results indicate for the first time that a carboxyl group is essential for the catalytic process of 5'-nucleotidase. Moreover, chemical modification by diethylpyrocarbonate also produced inactivation of the enzyme and showed a differential spectrum with a peak at 240 nm characteristic of N-carbethoxyhistidine residues. This inactivation was efficiently released upon decarbethoxylation by hydroxylamine only when the extent of inactivation, due to low concentration of diethylpyrocarbonate, was limited. The time-dependent inactivation followed first-order kinetics and nucleotides afforded significant protection against diethylpyrocarbonate modification. The results indicate the involvement of the histidine residue in catalysis.

Inhibition of smooth muscle 5'-nucleotidase by imidazole and its reversal by magnesium

Imidazole, commonly used as an effective pH-buffering reagent in aqueous media maintained at pH 7-8, was found to depress the 5'-nucleotidase (5'-ribonucleotide phosphohydrolase, EC 3.1.3.5) activity of microsomal membrane fraction isolated from rat vas deferens smooth muscle in a dose-dependent manner in the absence of added Mg2+. Such an inhibitory effect of imidazole on the smooth muscle 5'-nucleotidase was not dependent upon the purity or integrity of the membrane fractions used and could be fully reversed by the inclusion of 5-10 mM Mg2+ in the assay medium. Of the five different pH-buffering reagents tested, imidazole was specific in exerting inhibitory effect on the 5'-nucleotidase in the absence of Mg2+ and this inhibition could not be accounted for by the impurities present in the imidazole. Differential effects of chelating reagents and other divalent metal ions on the 5'-nucleotidase activity were also observed in imidazole and Tris buffer solutions. The 5'-nucleotidase activity was not affected if the membranes were preincubated and washed with a large volume of 50 mM imidazole and subsequently assayed in 50 mM Tris in the absence of Mg2+. Similar findings were obtained with EDTA treated membrane. These results suggest that imidazole does not act by removal of the activating metal ion but rather interacts directly with 5'-nucleotidase and alters the metal-enzyme interactions.

Isolation and kinetic properties of 5'-nucleotidase from guinea-pig skeletal muscle

5'-Nucleotidase (EC 3.1.3.5) has been solubilized and purified 1200-fold from guinea-pig skeletal muscle, to a specific activity of 40 U/mg protein. The purified enzyme yields a single protein band on polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. Guinea-pig skeletal muscle 5'-nucleotidase is extremely sensitive to inhibition by nucleoside di- and triphosphates. The inhibition is of the competitive type, and can be reversed only by strong excess of Mg2+. Nucleoside diphosphates are more powerful inhibitors than nucleoside triphosphates. The Ki values for ADP and ATP are 0.036 and 0.28 microM, respectively. The purified enzyme does not require exogenous cations for maximal activity and is inhibited by EDTA. This inhibition is reversed by divalent cations. This indicates that the enzyme contains a tightly bound metal cation.