Effects of single and multiple OXA-48 mutants on enzyme activity, stability, and structure-activity relationship. Evolutionary consequences of antibiotic resistance development caused by OXA-48

Abstract

Antimicrobial resistance, primarily caused by the overuse of antimicrobials such as antibiotics, is becoming an increasing concern to public health. To that end, the global spread of the -lactamase OXA-48 is worrisome, as it readily catalyzes the hydrolysis of -lactam drugs, such as penicillins as well as our “last resort” carbapenems. On the contrary, OXA-48 exhibits only limited catalytic activity against 3rd generations cephalosporins like ceftazidime. However, naturally evolving variants and results from laboratory studies have shown that OXA-48 can expand its substrate profile, conferring increased ceftazidime resistance. Expansion of the substrate profile towards ceftazidime is seen to be accompanied by a trade-off towards carbapenems and penicillins, greatly reducing OXA-48 ability to catalyze the hydrolysis of penicillins and carbapenems. Here, X-ray crystallography, steady-state enzyme kinetics and differential scanning fluorimetry were used to characterize and analyze wild type (wt) OXA-48:wt and two variants, OXA-48:F72L and OXA-48:A33V/K51E/F72L/S212A/T213A (OXA-48:Q5), where the latter two were evolved towards increased ceftazidime resistance. Steady-state enzyme kinetics revealed that the two mutants had increased catalytic ability to hydrolyze ceftazidime. Such increases in kcat/Km hypothesized to arise from increased flexibility of the -loop, which was observed in the OXA-48:Q5 X-ray crystal structure in complex with piperacillin, is in line with previous studies. Further supporting the hypothesis, urea dependent kinetics and thermostability measurements show that these mutants likely exhibit increased dynamical behavior that would aid ceftazidime binding. OXA-48:F72L showed a bigger urea dependence on the enzyme activity with no activity at 4 M urea, whereas OXA-48:wt and OXA-48:Q5 needed 6 M urea to become inactive. This suggested that OXA-48:F72L is more flexible, and that OXA-48:Q5 regains some resistance to chemical denaturing by urea. The pH dependency showed higher piperacillin activity at pH 7.2 compared to 5.2 and 9.2 for all three variants. The increase in ceftazidime activity came along with a functional trade-off against the penicillin piperacillin as well as reduced thermostability of (OXA-48:F72L: -6.5C/OXA-48:Q5: -6.4C) compared to wt OXA-48, which may be caused by sub-optimal substrate positioning within the active site of OXA-48:Q5. This work provides experimental evidence, that during evolution of OXA-48 towards increased ceftazidime activity, structural changes can arise, likely affecting the chemical environment within the active site, causing increased enzyme flexibility, and ultimately shaping functional trade-offs.