In Gram-negative bacteria, resistance against β-lactam antibiotics arises most often in the form of β-lactamase enzymes. β-Lactamases inactivate these antibiotics by hydrolysing their β-lactam pharmacophore. In this thesis, enzymatic drug modification by selected serine β-lactamases is studied using combined quantum mechanics/molecular mechanics (QM/MM) simulations.
Carbapenem breakdown by class A β-lactamases is inspected using so-called “computational assays”. The assays are simplified simulation protocols, which are still accurate enough to distinguish between active and inhibited enzymes. By limiting the simulation time and conformational sampling, a >99% reduction in required computational resources was achieved (compared to the original protocols), whilst still preserving the predictive power of the computational assay.
Further studies focus on class D β-lactamases, in particular the OXA-48 family. Most OXA-48 β-lactamases, including the wildtype OXA-48, are carbapenemases with specific preference for imipenem, but some variants have acquired activity against expanded-spectrum oxyimino cephalosporins. Ceftazidime breakdown for OXA-48-like enzymes is simulated to elicit the origins behind this enhanced cephalosporinase activity. Active site hydration was observed to correlate with the energy barriers for the rate-limiting reaction step. In addition to ceftazidime breakdown, carbapenem inactivation by OXA-48 is compared for imipenem and meropenem. QM/MM simulations are used to identify the preferred substrate orientation for deacylation, and to further illustrate that the difference in carbapenem hydrolysis comes from a subtle change in the active site hydrogen bonds.
Lastly, two suggested inhibition mechanisms of the non-β-lactam β-lactamase inhibitor avibactam against OXA-48 are compared to deduct the most plausible acylation pathway. The QM/MM potential energy profiles show that avibactam most likely utilises a similar mechanism to β-lactam substrates, where a carboxylated Lys73 acts as a proton acceptor in acylation. Based on these data, it is hypothesised that the avibactam inhibition most likely results from post-acylation decarboxylation of Lys73, which prevents any further reactivity.
|Date of Award||28 Sep 2021|
- The University of Bristol
|Supervisor||Marc W Van der Kamp (Supervisor) & Jim Spencer (Supervisor)|