The mechanism of ceftazidime and cefiderocol hydrolysis by D179Y variants of KPC carbapenemases is similar and involves the formation of a long-lived covalent intermediate

Abstract

Klebsiella pneumoniae carbapenemase (KPC) variants have been described that confer resistance to both ceftazidime-avibactam and cefiderocol. Of these, KPC-33 and KPC-31 are D179Y-containing variants derived from KPC-2 and KPC-3, respectively. To better understand this atypical phenotype, the catalytic mechanism of ceftazidime and cefiderocol hydrolysis by KPC-33 and KPC-31 as well as the ancestral KPC-2 and KPC-3 enzymes was studied. Steady-state kinetics showed that the D179Y substitution in either KPC-2 or KPC-3 is associated with a large decrease in both k_cat and K_M such that k_cat/K_M values were largely unchanged for both ceftazidime and cefiderocol substrates. A decrease in both k_cat and K_M is consistent with a decreased and rate-limiting deacylation step. We explored this hypothesis by performing pre-steady-state kinetics and showed that the acylation step is rate-limiting for KPC-2 and KPC-3 for both ceftazidime and cefiderocol hydrolysis. In contrast, we observed a burst of acyl-enzyme formation followed by a slow steady-state rate for the D179Y variants of KPC-2 and KPC-3 with either ceftazidime or cefiderocol, indicating that deacylation of the covalent intermediate is the rate-limiting step for catalysis. Finally, we show that the low K_M value for ceftazidime or cefiderocol hydrolysis of the D179Y variants is not an indication of tight binding affinity for the substrates but rather is a reflection of the deacylation reaction becoming rate-limiting. Thus, the hydrolysis mechanism of ceftazidime and cefiderocol by the D179Y variants is very similar and involves the formation of a long-lived covalent intermediate that is associated with resistance to the drugs.

Keywords: KPC-31 β-lactamase; KPC-33 β-lactamase; antibiotic resistance; cross-resistance; enzyme variants; transient kinetics; β-lactamase.

Fig 1

Structure of β-lactams used in this study: (A) ceftazidime and (B) cefiderocol. (C) Kinetic model for serine-active site β-lactamases. Minimal scheme for the reaction with E as a free enzyme, S as substrate, ES as enzyme-substrate complex, EAc as acyl-enzyme intermediate, and P as a product. (D) Equations for k_cat, K_M, and k_cat/K_M based on the kinetic scheme shown in C. A simplified model is also shown based on the assumption that k₋₁ >> k₂.

Fig 2

Rates of ceftazidime and cefiderocol hydrolysis plotted versus substrate concentration for KPC-2 (A and B, respectively), KPC-33 (KPC-2 D179Y) (C and D), KPC-3 (E and F), and KPC-31 (KPC-3 D179Y) (G and H). For KPC-2 and KPC-3, it was not possible to reach saturating substrate concentrations. The second-order rate constant at steady-state, k_cat/K_M, was determined by fitting the progress curves to the equation v = k_cat/K_M [E][S], where [S] << K_M. For KPC-33 and KPC-31, the steady-state parameters were obtained by a non-linear least squares fit of the data to the Michaelis-Menten equation v = k_cat[S]/(K_M + [S]).

Fig 3

Pre-steady-state experiments. Enzymes were mixed with 50 µM ceftazidime (A and B) or cefiderocol (C and D) and absorbance at 260 or 259 nm, respectively, was monitored for 15 s. The sign of ΔConcentration (µM) was changed from negative to positive. (A) Progress curves of KPC-2 (blue) and KPC-33 (red) with ceftazidime. Black lines correspond, respectively, to fit burst equation and a single exponential fit for KPC-33 and KPC-2. (B) Progress curves of KPC-3 (blue) and KPC-31 (red) with ceftazidime fit to a single exponential and burst equation, respectively (black lines). (C) Progress curves of KPC-2 (blue) and KPC-33 (red) with cefiderocol fit to a single exponential and burst equation, respectively (black lines). (D) Progress curves of KPC-3 (blue) and KPC-31 (red) with cefiderocol fit to a single exponential and burst equation, respectively (black lines).

Fig 4

Example of results obtained during pre-steady-state experiments. The change in ceftazidime concentration was monitored at 260 nM and plotted versus time (red points) for KPC-33 (KPC-2 D179Y) with ceftazidime concentrations of 50 µM (A), 100 µM (B), 150 µM (C), and 200 µM (D). The sign of ΔConcentration (µM) was changed from negative to positive. The data were fit to the burst equation (black line; see Materials and Methods). The individual k_obs values at each substrate concentration were then fit to a hyperbola to estimate K_D (dissociation constant) and k₂ (E) (see Materials and Methods).