A whole-cell hypersensitive biosensor for beta-lactams based on the AmpR-AmpC regulatory circuit from the Antarctic Pseudomonas sp. IB20

Abstract

Detecting antibiotic residues is vital to minimize their impact. Yet, existing methods are complex and costly. Biosensors offer an alternative. While many biosensors detect various antibiotics, specific ones for beta-lactams are lacking. To address this gap, a biosensor based on the AmpC beta-lactamase regulation system (ampR-ampC) from Pseudomonas sp. IB20, an Antarctic isolate, was developed in this study. The AmpR-AmpC system is well-conserved in the genus Pseudomonas and has been extensively studied for its involvement in peptidoglycan recycling and beta-lactam resistance. To create the biosensor, the ampC coding sequence was replaced with the mCherry fluorescent protein as a reporter, resulting in a transcriptional fusion. This construct was then inserted into Escherichia coli SN0301, a beta-lactam hypersensitive strain, generating a whole-cell biosensor. The biosensor demonstrated dose-dependent detection of penicillins, cephalosporins and carbapenems. However, the most interesting aspect of this work is the high sensitivity presented by the biosensor in the detection of carbapenems, as it was able to detect 8 pg/mL of meropenem and 40 pg/mL of imipenem and reach levels of 1-10 ng/mL for penicillins and cephalosporins. This makes the biosensor a powerful tool for the detection of beta-lactam antibiotics, specifically carbapenems, in different matrices.

FIGURE 1

Relative expression of the ampC gene in Pseudomonas sp. IB20 after exposure to beta‐lactam antibiotics. The expression levels of the ampC gene from IB20 were determined using RT–qPCR after exposure to subinhibitory concentrations (1/4 of MIC values) of several beta‐lactams. For comparison, the expression levels of ampC in P. aeruginosa PAO1‐V were also evaluated. The data obtained indicated that the ampC gene is highly induced in IB20, with the highest values observed in the presence of imipenem, ampicillin, and carbenicillin. In comparative terms, the expression levels of the IB20 ampC gene were 2.4‐fold higher than those observed in PAO1 in the presence of imipenem, 2‐fold higher in the presence of ceftazidime, and 270‐fold higher in the presence of carbenicillin. Three biological and technical replicates were used for this assay, and rpsL was used as the normalizing gene (MICs PAO1‐V: Imipenem 1 μg/mL; ceftazidime 1 μg/mL; carbenicillin 16 μg/mL; ampicillin 1024 μg/mL. MICs Pseudomonas sp IB20: Imipenem 1 μg/mL; ceftazidime 2 μg/mL; carbenicillin 4096 μg/mL; ampicillin 16,384 μg/mL).

FIGURE 2

Scheme illustrating the construction of transcriptional fusions for biosensor generation. (A) Original ampRC gene‐locus from Pseudomonas sp. IB20; (B) Biosensor 1; (C) Biosensor 2; (D) mCherry biosensor control. Key components: ampC betalactamase, Coding sequence for the AmpC betalactamase; ampR regulator, Coding sequence for the AmpR protein; mCherry reporter, coding sequence for the mCherry fluorescent protein; P3, Strong promoter for constitutive expression. Buffer sequence; pampC, promotor sequence from ampC‐coding sequence; PampR, promotor sequence from ampR‐coding sequence; RBS, improved ribosomal binding site.

FIGURE 3

Microscopic visualization of biosensors exposed to sub‐inhibitory concentrations of ampicillin and imipenem. Raw fluorescence production was evaluated by microscopic visualization of biosensors 1 and 2 in cultures grown on black 96‐well plates with a transparent bottom. The cultures were grown from an initial OD₆₀₀ of 0.05 using LB medium supplemented with ampicillin (1000 ng/mL) or imipenem (125 ng/mL) corresponding to 1/4 of their respective MIC values. For comparative purposes, strain SN0301 carrying the empty vector pSEVA541 was also included in the testing. Microscopic visualization was performed after 24 h of culture using a green light excitation filter (540 nm) and a 10× magnification on a fluorescence microscope.

FIGURE 4

Determination of the optimal response time of the constructed biosensors. Cultures of each biosensor were supplemented with subinhibitory concentrations of imipenem to evaluate its effect on fluorescence production over time. The raw fluorescence values were normalized to OD₆₀₀, and the biosensor induction capacity was expressed as the fold change of the relative fluorescence compared to the absence of the drug. (A) Biosensor 1; (B) Biosensor 2; (C) mCherry control biosensor. The red bars represent concentrations of 2 ng/mL (1/256 of the MIC), 16 ng/mL (1/32 of the MIC) and 125 ng/mL (1/4 of the MIC), from the lightest to the darkest.

FIGURE 5

Effect of initial cell density on the fluorescent response of biosensors exposed to subinhibitory concentrations of imipenem. Fluorescence production of biosensors was evaluated in microplates at two different initial cell densities: OD₆₀₀ 0.2 (exponential phase) and 0.4 (stationary phase). (A and D) Biosensor 1; (B and E) Biosensor 2; (C and F) mCherry control biosensor. The red bars represent the relative inductions of fluorescence production at 0.04, 0.08, 0.16, 0.3, 0.6, 1.25, 2.5, 5.0, 10 and 20 ng/mL, from the lightest to the darkest bar.

FIGURE 6

Detection of different beta‐lactam antibiotics by the constructed biosensors. (A) Low concentrations of imipenem; (B) High concentrations of imipenem; (C) Meropenem; (D) Amoxicillin‐clavulanate; (E) Cefuroxime; (F) Ceftazidime; (G) Ampicillin; (H) Cabernicillin. Raw fluorescence values were normalized to OD₆₀₀ and relative to the basal fluorescence produced in the absence of antibiotics (y‐axes). The data was obtained after 10 h of incubation at 37°C. The antibiotic ranges vary for each compound according to their respective minimum inhibitory concentration (MIC). The black bars correspond to biosensor 1, the grey bars to biosensor 2, and the red bars to the mCherry biosensor control.

FIGURE 7

Dose–response profiles of biosensors against drugs belonging to the same beta‐lactam subclass. (*penicillin + beta‐lactamase inhibitor)