The Escherichia coli MurB gene encoding UDP-N-acetylenolpyruvylglucosamine reductase was expressed to a level of approximately 100 mg/L as a fusion construct with maltose binding protein. Rapid affinity purification, proteolysis, and anion exchange chromatography yielded homogeneous enzyme containing 1 mol/mol bound FAD. Enzyme was maximally activated by K+, NH4+, and Rb+ at cation concentrations between 10 and 50 mM. Steady-state enzyme kinetics at pH 8.0 and 37 degrees C revealed weak and strong substrate inhibition by NADPH and UDP-N-acetylenolpyruvylglucosamine, respectively, where the KiS were 910 microM and 73 microM. Substrate inhibition was pH dependent for both substrates. Initial velocity measurements as a function of both substrates produced patterns consistent with a ping pong bi bi double competitive substrate inhibition mechanism. Data at pH 8.0 yielded kinetic constants corresponding to Km,UNAGEP = 24 +/- 3 microM, Ki,UNAGEP = 73 +/- 19 microM, Km,NADPH = 17 +/- 3 microM, Ki,NADPH = 910 +/- 670 microM, and kcat = 62 +/- 3 s-1. A slow anaerobic exchange reaction with thio-NADP+ provided evidence for release of NADP+ in the absence of UNAGEP. Alternate reduced nicotinamide dinucleotides, including NHXDPH, 3'-NADPH, and alpha-NADPH, were substrates, whereas NADH was not. Several nucleotides, including ADP and UDP, were weak inhibitors of the enzyme with inhibition constants between 5 and 97 mM. Various analogs of NADP+, including 3'-NADP+, thio-NADP+, APADP+, NEthDP+, and NHXDP+, were inhibitors of the enzyme with respect to NADPH and yielded inhibition constants in the range of 110-1100 microM. Analogs without the 2'- or 3'-phosphate of NADPH or NADP+ were not substrates or inhibitors. Double inhibition experiments with varied APADP+ and UNAG produced inhibition patterns consistent with mutually exclusive inhibitor binding. The data suggest that NADPH and UNAGEP share a subsite that prevents both molecules from binding at once.

In order to understand the nature of ATP and t-glutamate binding to glutamine synthetase, and the involvement of Arg 339 and Arg 359 in catalysis, these amino acids were changed to cysteine via site-directed mutagenesis. Individual mutations (Arg --t Cys) at positions 339 and 359 led to a sharp drop in catalytic activity. Additionally, the K,,, values for the substrates ATP and glutamate were elevated substantially above the values for wild-type (WT) enzyme. Each cysteine was in turn chemically modified to an arginine "analog" to attempt to "rescue" catalytic activity by covalent modification; 2-chloroacetamidine (CA) (producing a thioether) and 2,2'-dithiobis (acetamidine) (DTBA) (producing a disulfide) were the reagents used to effect these chemical transformations. Upon reaction with CA, both R339C and R359C mutants showed a significant regain of catalytic activity (50% and 70% of WT, respectively) and a drop in K,,, value for ATP close to that for WT enzyme. With DTBA, chemically modified R339C had a greater kc,, than WT glutamine synthetase, but chemically modified R359C only regained a small amount of activity. Modification with DTBA was quantitative for each mutant and each modified enzyme had similar K,,, values for both ATP and glutamate. The high catalytic activity of DTBA-modified R339C could be reversed to that of unmodified R339C by treatment with dithiothreitol, as expected for a modified enzyme containing a disulfide bond. Modification of each cysteine-containing mutant to a lysine "analog" was accomplished using 3-bromopropylamine (BPA). The R339C mutant, upon modification with BPA, had a greater kc,, than WT enzyme; however, the R359C mutant did not show significant regeneration of activity with this reagent. The data are consistent with X-ray crystallographic studies showing Arg 339 and Arg 359 at the active site of glutamine synthetase (Liaw SH, Eisenberg D, 1994, Biochemistry 33:675-681) interacting with ATP, glutamate, and intermediates along the catalytic pathway. Because enzyme activity could be restored for the R339C mutant by monofunctional (amine) and bifunctional (amidine) reagents, Arg 339 most likely interacts with substrates in a monodentate fashion. Conversely, Arg 359 seems to interact bifunctionally with substrates because covalent modification of R359C with BPA did not lead to a significant regain of catalytic activity.

No abstract

Uridine 5'-diphospho-N-acetylenolpyruvylglucosamine reductase (MurB), the second enzyme in the peptidoglycan synthetic pathway of Escherichia coli, has been crystallized in two previously unreported forms, one orthorhombic and the other monoclinic. MurB (molecular mass 38kDa) crystallizes in a range of conditions that utilize polyethylene glycol fractions as precipitants, and crystals can be grown with or without the enzyme's substrate, uridine 5'-diphospho-N-acetylenolpyruvylglucosamine. X-ray diffraction from crystals of the orthorhombic form extends to 2 ,~ resolution and shows the symmetry and systematic absences of space group P2j2~2j. These crystals show significant variations in cell dimensions at room temperature and at 100 K. A crystal used to collect a 2.0,A resolution data set at a synchrotron source showed cell dimensions at ca 100K of a = 51.0, b---79.3 and c ---87.1 ,~, indicating one molecule peroasymmetric unit. The monoclinic crystals scatter X-rays to 3.0 A resolution consistent with space group P2~, unit-cell dimensions (ca 100K) a-----50.7, b=92.4, c--85.5A, and /5= 104 ° , and two molecules per asymmetric unit. Mercury derivatives have been prepared with both orthorhombic and monoclinic forms, and efforts are underway to exploit these derivatives to determine the structure of this protein.