β-lactamase expression is the most prevalent mechanism of bacterial resistance to
the β-lactam family of antibiotics, which includes the penicillins and cephalosporins. Their occurrence in many bacterial pathogens poses a
threat to public health and a challenge to medicinal chemists when developing new and more effective β-lactam antibiotics. Our work focuses on
the TEM family of class A β-lactamases, and on the class C β-lactamase, AmpC, an important resistance enzyme among Gram-negative nosocomial
pathogens.
What are the structural and energetic bases of ligand recognition at each step of the hydrolytic mechanism of Group I
β-lactamases? The hydrolysis of β-lactams by may be thought of as occurring in five steps: beginning with (A) a pre-covalent
encounter complex and proceeding through (B) a high energy intermediate to (C) an acylated ground state.  This ground state is attacked by a hydrolytic water proceeding through (D) a second and typically
rate-determining high-energy intermediate, which collapses to (E) a product complex. Although the overall steps are widely accepted, the details
are controversial and the energetic bases of recognition are poorly understood. We hope to address open questions such as the identity of the
catalytic base, the direction of water attack, the role of substrate assisted catalysis, and the distinction between β-lactam substrates and
β-lactam inhibitors of the enzyme.
To understand the mechanism of β-lactam hydrolysis by AmpC, we have determined the crystal structures of several key steps. Using a
mutant enzyme, AmpC S64G, we have captured the enzyme in complex with the substrate (step A) and product (step E) forms of cephalothin.
In addition, we have captured the wild-type enzyme in acyl complex with cephalothin (step C):
| By capturing the major milestones along the reaction pathway of cephalothin
hydrolysis by AmpC, it is clear that cephalothin undergoes a dramatic conformational change as it is hydrolyzed. In addition, these structures suggest
that the ligand itself is crucial for activating the enzyme at several steps in the hydrolytic pathway. (Beadle et al., Structure,
2002) |
 |
Do inhibitors bind differently than substrates? We have also determined the structures of acyl-adducts of both β-lactam
substrates (loracarbef, A below) and inhibitors (cloxacillin, B below) (Patera et al., Journal of the American Chemical Society,
2000), and of deacylation transition state analogs (boronic acid transition state analogs bearing the cloxacillin R1 side chain, C below, and
the cephalothin R1 side chain, D below) (Caselli et al., Chemistry & Biology, 2001).
Can we find non-β-lactam inhibitors of serine β-lactamases? We hypothesize that
non-β-lactams can be found to complement the structure of AmpC, that such inhibitors will inhibit the enzyme
potently, and that they will behave differently than β-lactamas as anti-resistance agents. Unlike viral
resistance, resistance to β-lactams, and indeed most antibiotics, occurs mostly through the transfer of
pre-evolved resistance mechanisms. Thus, when a new β-lactam is introduced, resistance quickly follows. There
are several reasons for this, one of the most important being that β-lactams are ancient and the resistance
mechanisms were invented millions of years ago. A good example of this is clavulanate, which is widely used to
inhibit β-lactamases in community acquired infections. This inhibitor is itself a β-lactam and is a natural
product of Streptomyces clavulageris. Clavulanate has its access blocked by porin channel mutations, is
hydrolyzed by mutant β-lactamases, and can up-regulate the expression of the very β-lactamases it is meant
to inhibit. Unlike β-lactam inhibitors such as clavulanate, non-β-lactam inhibitors would not be hydrolyzed
by mutant β-lactamases, might evade porin channel mutations, and should not up-regulate the expression of
β-lactamases.
 We are pursuing several strategies to discover novel inhibitors.
The simplest is the design of boronic acid transition state analogs that are decorated with substrate groups. In collaboration with Fabio Prati and
Paola Costi at the University of Modena, we have used structure to guide the synthesis of nanomolar inhibitors of AmpC (Caselli et al., Chemistry &
Biology, 2001). For example, a 20 nM inhibitor has been developed consisting of a boronic acid transition state analog with the ceftazidime R1 side
chain (figure at right). In a second strategy, we have used molecular docking to help us design novel inhibitors, and combined the structure-based
design with in parallel synthesis (Tondi et al., Chemistry & Biology, 2001). Several of the designed inhibitors reverse antibiotic resistance in
cell culture.
 The synergistic effect of the ceftazidime
side chain boronic acid (bottom disk) on ceftazidime (top disk) in bacterial plate assays. Left. E.
cloacae that does not produce AmpC -- no synergy. Right. A resistant strain of E. cloacae that
produces AmpC -- strong synergy.
Recent publications: - Morandi F, Caselli E, Morandi S, Focia PJ, Blazquez J, Shoichet BK, Prati F. Nanomolar inhibitors of AmpC beta-lactamase. J Am Chem Soc 125 (3), 685-95 (2003). [Pubmed | DOI | PDB 1MXO | PDB 1MY8 | News In Brief]
- Beadle BM, Shoichet BK. Structural basis for imipenem inhibition of class C beta-lactamases. Antimicrob Agents Chemother 46 (12), 3978-80 (2002). [Pubmed | DOI | PDB 1LL5 | Download PDF]
- Minasov G, Wang X, Shoichet BK. An ultrahigh resolution structure of TEM-1 ß-lactamase suggests a role for Glu166 as the general base in acylation. J Am Chem Soc 124 (19), 5333-40 (2002). [Pubmed | DOI | PDB 1M40 | Download PDF]
- Trehan I, Morandi F, Blaszczak LC, Shoichet BK. Using steric hindrance to design new inhibitors of class C ß-lactamases. Chem Biol 9 (9), 971-80 (2002). [Pubmed | DOI | PDB 1LL9 | PDB 1LLB | Download PDF]
- Wang X, Minasov G, Shoichet BK. The structural bases of antibiotic resistance in the clinically-derived mutant ß-lactamases TEM-30, TEM-32, and TEM-34. J Biol Chem 277 (35), 32149-56 (2002). [Pubmed | DOI | PDB 1LHY | PDB 1LI0 | PDB 1LI9 | Download PDF]
|
