Three stable dinuclear zinc(II) complexes, [Zn 2 L 1 (µ-NO 3 )(NO 3 ) 2 ] and [Zn 2 L 1 (µ-OMe)(NO 3 ) 2 ], whereaminomethyl}phenolate, were synthesized and characterized in the solid state and in aqueous solution. These complexes catalyze the hydrolysis of penicillin G and nitrocefin, serving as functional synthetic analogues of the metalloβ-lactamases, bacterial enzymes responsible for antibiotic resistance. The mechanism of the hydrolysis was studied in detail for the catalyst precursor [Zn 2 L 1 (µ-NO 3 )(NO 3 ) 2 ], which converts into [Zn 2 L 1 (µ-OH)(NO 3 ) n -(sol) 2-n ] (2-n)+ in the presence of water. The complex [Zn 2 L 1 (µ-OH)(NO 3 ) 2 ] (n ) 2) was characterized in the solid state. Initial coordination of the substrate carboxylate group is followed by the rate-limiting nucleophilic attack of the bridging hydroxide at the β-lactam carbonyl group in aqueous solution. The product is formed upon fast protonation of the intermediate. Mononuclear complexes Zn(cyclen)(NO 3 ) 2 and Zn(bpta)(NO 3 ) 2 are as reactive in the β-lactam hydrolysis as the dinuclear complexes. Consequently, the second zinc ion is not required for catalytic activity.

The stable dinuclear complex [Zn 2 (BPAN)(µ-OH)(µ-O 2 PPh 2 )](ClO 4 ) 2 , where BPAN ) 2,7-bis[2-(2-pyridylethyl)-aminomethyl]-1,8-naphthyridine, was chosen as a model to investigate the reactivity of (µ-hydroxo)dizinc(II) centers in metallohydrolases. Two reactions, the hydrolysis of phosphodiesters and the hydrolysis of -lactams, were studied. These two processes are catalyzed in vivo by zinc(II)-containing enzymes: P1 nucleases and -lactamases, respectively. The former catalyzes the hydrolysis of single-stranded DNA and RNA. -Lactamases, expressed in many types of pathogenic bacteria, are responsible for the hydrolytic degradation of -lactam antibiotic drugs. In the first step of phosphodiester hydrolysis promoted by the dinuclear model complex, the substrate replaces the bridging diphenylphosphinate. The bridging hydroxide serves as a general base to deprotonate water, which acts as a nucleophile in the ensuing hydrolysis. The dinuclear model complex is only 1.8 times more reactive in hydrolyzing phosphodiesters than a mononuclear analogue, Zn(bpta)(OTf) 2 , where bpta ) N,N-bis-(2-pyridylmethyl)-tert-butylamine. Hydrolysis of nitrocefin, a -lactam antibiotic analogue, catalyzed by [Zn 2 (BPAN)(µ-OH)(µ-O 2 PPh 2 )](ClO 4 ) 2 involves monodentate coordination of the substrate via its carboxylate group, followed by nucleophilic attack of the zinc(II)-bound terminal hydroxide at the -lactam carbonyl carbon atom. Collapse of the tetrahedral intermediate results in product formation. Mononuclear complexes Zn(cyclen)-(NO 3 ) 2 and Zn(bpta)(NO 3 ) 2 , where cyclen ) 1,4,7,10-tetraazacyclododecane, are as reactive in the -lactam hydrolysis as the dinuclear complex. Kinetic and mechanistic studies of the phosphodiester and -lactam hydrolyses indicate that the bridging hydroxide in [Zn 2 (BPAN)(µ-OH)(µ-O 2 PPh 2 )](ClO 4 ) 2 is not very reactive, despite its low pK a value. This low reactivity presumably arises from the two factors. First, the bridging hydroxide and coordinated substrate in [Zn 2 (BPAN)(µ-OH)(substrate)] 2+ are not aligned properly to favor nucleophilic attack. Second, the nucleophilicity of the bridging hydroxide is diminished because it is simultaneously bound to the two zinc(II) ions. IntroductionMetallohydrolases are hydrolytic enzymes that depend on metal ions, usually divalent zinc, cobalt, and manganese, for catalysis. 1 The mechanisms of enzymes containing a single metal ion, such as carboxypeptidase and carbonic anhydrase, have been extensively studied and are relatively well understood. 2 Dinuclear metallohydrolases have been studied to a lesser extent. 1 It is often proposed that a bridging hydroxide ion, which occurs in numerous dinuclear enzymes, serves as a nucleophile that attacks metal-bound substrates. 2,3 The low pK a value for the bridging water at dinuclear sites results in a higher local concentration of hydroxide ion compared to that afforded at mononuclear centers. However, simultaneous coordination of hydroxide to two metal ions also results in tighter binding and lowered nucle...

Inactivation of beta-lactam antibiotics by metallo-beta-lactamase enzymes is a well-recognized pathway of antibiotic resistance in bacteria. As part of extensive mechanistic studies, the hydrolysis of a beta-lactam substrate nitrocefin (1) catalyzed by dinuclear zinc(II) model complexes was investigated in nonaqueous solutions. The initial step involves monodentate coordination of the nitrocefin carboxylate group to the dizinc center. The coordinated substrate is then attacked intramolecularly by the bridging hydroxide to give a novel intermediate (2') characterized by its prominent absorbance maximum at 640 nm, which affords a blue color. The NMR and IR spectroscopic data of 2' are consistent with it being zinc(II)-bound N-deprotonated hydrolyzed nitrocefin that forms from the tetrahedral intermediate upon C-N bond cleavage. Protonation of the leaving group is the rate-limiting step in DMSO solution and occurs after the C-N bond-breaking step. Addition of strong acids results in rapid conversion of 2' into hydrolyzed nitrocefin (3). The latter can be converted back to the blue species (2') upon addition of base. The low pK(a) value for the amino group in hydrolyzed nitrocefin is explained by its involvement in extended conjugation and by coordination to zinc(II). The blue intermediate (2') in the model system resembles well that in the enzymatic system, judging by its optical properties. The greater stability of the intermediate in the model, however, allowed its characterization by (13)C NMR and infrared, as well as electronic, spectroscopy.

Four palladium(II) aqua complexes catalyze hydrolytic decomposition of urea into carbon dioxide and ammonia. The initial rates of carbon dioxide formation at 313 K and pH 3.3 fall in the range 6.7 x 10(-)(5) to 1.6 x 10(-)(4) M min(-)(1), depending on the catalyst. The pseudo-first-order rate constant for the formation of carbon dioxide is 1.7 x 10(-)(3) min(-)(1) in the presence of 0.30 M cis-[Pd(en)(H(2)O)(2)](2+) as the catalyst at 313 K and pH 3.3. This reaction is ca. 1 x 10(5) times faster than the uncatalyzed decomposition of urea. The reaction catalyzed by cis-[Pd(en)(H(2)O)(2)](2+) is monitored by (13)C and (15)N NMR spectroscopic methods. The following steps in the mechanism of this reaction are studied quantitatively: binding of urea to the catalyst, formation of carbamic acid (H(2)NCOOH) coordinated to palladium(II) via the nitrogen atom, and conversion of this intermediate into carbon dioxide and ammonia. These products are formed also by another pathway that does not involve carbamic acid. Kinetic effects of added acid and inhibition of the reaction by addition of thiourea and of bases are interpreted quantitatively. Ammonia inhibits the decomposition. When, however, this product is sequestered by metal cations, the reaction becomes relatively fast and catalytic turnover is achieved. The most effective of these sequestering agents is the silver(I) cation. Although the simple palladium(II) complexes are very different from the enzyme urease, which contains nickel(II) ions, the decomposition of urea catalyzed by both kinds of agents involves carbamic acid as the intermediate. Kinetic and mechanistic studies with metal complexes contribute to the understanding of the enzymatic mechanism.

No abstract