An Efficient Computational Assay for β-Lactam Antibiotic Breakdown by Class A β-Lactamases

Breakdown of β-lactam antibiotics by β-lactamases is one of the most common resistance mechanisms against these drugs. Here, we present a computationally efficient combined quantum mechanics/molecular mechanics simulation protocol for hydrolysis of the β-lactamase acylenzyme with meropenem, a carbapenem antibiotic. Starting from the previously published computational assay, we have successfully reduced both the conformational space and the sampling time, which leads to a more than 99% reduction on computer resources needed for this assay (less than 24 CPU hours per reaction). This simplified in silico protocol distinguishes accurately between class A β-lactamases that are able or unable to break down carbapenems. We further demonstrate that these protocols can be used to elucidate mechanistic details of β-lactam breakdown. The new efficient computational assay

may be used to predict carbapenem hydrolytic activity of new β-lactamase variants as they arise, and in identifying routes to development of new, more resilient β-lactam antibiotics or new generations of mechanism-based β-lactamase inhibitors.