Urease catalyzes the hydrolysis of urea into ammonia and carbamic acid, with the latter compound decomposing into a second molecule of ammonia and carbonic acid. The enzyme from Klebsiella aerogenes is composed of three subunits (UreA, UreB, and UreG) and assembles into a trimer of heterotrimers (UreABC)3. The in vivo formation of its di-nickel, carbamylated lysine-bridged active site requires four accessory proteins: UreD, UreE, UreF, and UreG. The sequential binding of UreD, UreF, and UreG to urease or the binding of an accessory protein complex consisting of UreD:UreF:UreG primes the enzyme for activation by nickel that is delivered by UreE. Two of my projects focused on determining the role of UreD in this activation process and characterizing the properties of a soluble UreD:UreF:UreG accessory protein complex. I characterized a soluble, urease accessory protein complex containing a translational fusion of the maltose binding protein with UreD (MBP-UreD), UreF, and UreG (termed MBP-UreDFG) that was formed in vivo. This complex bound nickel weakly and existed as a dimer of heterotrimers in solution, with two UreF protomers located at the interfacial site. (MBP-UreDFG)2 dissociated to the heterotrimer as it bound to urease apoprotein or holoenzyme. The interaction with the apoprotein was disrupted by the presence of nickel and (UreG)2 dissociated from (MBP-UreDFG)2 in buffer containing GTP, magnesium, and nickel. I used mutagenesis approaches to examine the interaction of UreD with both urease and UreF and to explore the function of this protein in the transfer of nickel into the active site of urease. On the basis of a multiple sequence alighnment and a UreD homology model I generated from the Helicobacter pylori UreH:UreF:UreG crystal structure (UreH is homologous to UreD), 26 residues were selected for substitution. None of the variants were affected in their urease:UreD and UreD:UreF interactions in vivo. In vivo activation studies using UreD variants produced in the context of the complete urease gene cluster identified the D63A, D63Q, S85K, D142A, E176A, and E176Q UreD variants as being deficient in urease activation. The substituted residues mapped to a buried water tunnel identified in silico which originates at the putative UreF:UreD interface and exits at the opposite face of UreD. Purified urease activated in vivo by these variants contained substoichiometric amounts of nickel and varied amounts of zinc and iron per UreABC. My final project focused on characterization of the iron-containing urease from Helicobacter mustelae, UreA2B2, by resonance Raman spectroscopy. Previous studies showed the diferrous active site of this enzyme is rendered inactive in the presence of O2, yielding a μ-oxo bridged, diferric metallocenter. In contrast to earlier results, I observed downshifts of the νs and νas(Fe-O-Fe) following anaerobic reduction in H218O and subsequent chemical oxidation relative to analysis of enzyme in H216O, consistent with exchange of the μ-oxo atom. The νs(Fe-O-Fe) was downshifted by 10 cm-1 when the enzyme was incubated with urea or the slow-binding substrate phenyl phosphorodiamidate, but not following incubation with the inhibitor acetohydroxamic acid or the chaotropic agent guanidinium chloride. Rapid bulk-solvent exchange studies identified a urea- and solvent-sensitive mode at ~ 530 cm-1 which downshifted to ~511 cm-1 in D2O or H218O and upshifted by 10 cm-1 in the presence of urea. This result was consistent with a terminal Fe-OH stretch where the hydroxyl group is not displaced on substrate binding. I also identified a slow substrate-binding form of UreA2B2.
Read
