Green-Light-Activatable Penicillin for Light-Dependent Spatial Control of Bacterial Growth, Biofilm Formation, and In Vivo Infection Treatment

Abstract

Our ability to prevent, treat, and cure bacterial infections is nowadays seriously threatened by the rise of (multidrug) antimicrobial resistance (AMR), and novel molecular approaches in the antibacterial arsenal are urgently needed. To fight the development of AMR, the field of photopharmacology aims to develop photoresponsive antimicrobials allowing for noninvasive activation of the drug only at the site needed, with spatiotemporal precision, reducing the bacterial exposure to the active antimicrobial in the environment. This study reports the development and application for the first time of a green-light-activatable variant of penicillin (Penicillin-PPG), designed through the incorporation of a photocleavable protecting group. Here, we demonstrate that Penicillin-PPG shows no antimicrobial activity in the dark, while it can be precisely activated through irradiation with green light. Furthermore, we show Penicillin-PPG's utility to spatially control bacterial growth, achieve light-dependent inhibition of biofilm formation, and showcase the unprecedented usage of a photoactivatable antimicrobial in vivo in a small animal infection model. Furthermore, we apply Penicillin-PPG in combination with a λ-orthogonally photocaged bioactive compound to achieve photocontrol over antimicrobial activity dependent on two distinct colors of light.

1

Design of a photocleavable penicillin analogue. Left: A schematic representation of the similarity between the C-terminal substrate peptide of transpeptidase and penicillin, highlighting the common carboxylic acid moiety (red). Right: The proposed design of a photocleavable penicillin analogue through blockage of the carboxylic acid moiety with a photocleavable protecting group.

2

Synthesis and photochemical characterization of Penicillin-PPG. (a) Synthetic scheme for the formation of Penicillin-PPG. (b) UV–vis absorption spectra of samples of Penicillin-PPG (20 μM, LB with 1% DMSO, 37 °C), before and after irradiation (λ = 530 nm, black and green lines, respectively, total irradiation time 25 min). (c) ¹H NMR spectra of samples of Penicillin-PPG (2 mM, DMSO-d ₆/D₂O 1:1) in the dark and after irradiation with green light (λ = 520 nm). The release of penicillin was confirmed through spiking with a penicillin-potassium salt (top spectrum).

3

Growth of E. coli in LB with or without green light-irradiated Penicillin-PPG. Growth curves of E. coli DH5α were determined in triplicate, and shown are the averages and standard deviations. 80 μM Penicillin-PPG (a) or OAc-PPG (b) was used. The final concentration of DMSO was 3%. Cells were grown overnight at 37 °C.

4

Spatial control of bacterial growth with Penicillin-PPG. (a) Schematic representation of the experimental setup, in which part of an agar plate inoculated with E. coli DH5α was covered with an aluminum sticker (‘dark’) while the other part was irradiated (‘light’, λ = 530 nm, 1 h). (b) Experimental results of agar plates containing Penicillin-PPG, (c) DMSO control, or (d) penicillin (90 μM Penicillin-PPG or penicillin, 3% DMSO in LB-agar). Of note, the color of the plate containing Penicillin-PPG appears slightly red, which is caused by the dissolved Penicillin-PPG.

5

Prevention of S. epidermidis biofilm formation by green light-activated Penicillin-PPG. S. epidermidis ATCC 35984 was used to inoculate growth medium in microtiter plates supplemented with Penicillin-PPG, penicillin or without supplementation (DMSO control). After irradiation for 2 h with green light or incubation in the dark, biofilms were grown on coverslips for 24 h at 37 °C. Subsequently, the coverslips were stained with Crystal Violet to detect biofilm formation. (a, b) Biofilm growth in the absence of antibiotic, regardless of irradiation with green light. (c, d) Absence of biofilm formation in the presence of penicillin (52 μM), regardless of irradiation with green light. (e) Unimpaired biofilm formation in the presence of Penicillin-PPG (52 μM) upon incubation in the dark. (f) Impaired biofilm formation in the presence of Penicillin-PPG after irradiation with green light.

6

Kaplan–Meier survival curves of G. mellonella after S. aureus infection and treatment with Penicillin-PPG. N = 30 larvae per group at t = 0. (a) There are no significant differences in the survival of uninfected larvae that received (1) no injection, (2) Penicillin-PPG injection and subsequent 2 h irradiation with green light, (3) Penicillin-PPG injection and subsequent incubation in the dark, and (4) PBS injection. (b) Compared to S. aureus-infected larvae that received no treatment (blue line), S. aureus-infected larvae that were injected with Penicillin-PPG and received green light irradiation for 2 h (green line) showed a significantly increased survival. This was not the case for S. aureus-infected larvae treated with Penicillin-PPG that were kept in the dark (black line). *P < 0.05; ns, not significant.

7

(a) Synthetic scheme for the preparation of Tazobactam-PPG. (b) Partial ¹H NMR spectra of samples of Tazobactam-PPG (2 mM, DMSO-d ₆/D₂O 1:1) in the dark and after irradiation with violet light (λ = 400 nm) for the times indicated. The release of tazobactam was confirmed through a spike with tazobactam (top spectrum).

8

(a) Overlay of UV–vis absorption spectra of Tazobactam-PPG (blue) and Penicillin-PPG (red), both at 20 μM in water, 0.5% DMSO, 37 °C. (b) UV–vis absorption spectra of a sample of Tazobactam-PPG (20 μM, water, 0.5% DMSO, 37 °C) before and during irradiation (λ = 395 nm, black and blue lines, respectively, total irradiation time 5 min). (c) Time-dependent, normalized absorption values at 410 nm of the absorption spectra shown in (b) (blue curve) as well as relative UPLC peak area (RPA) of the peak corresponding to Tazobactam-PPG in UPLC-MS chromatograms upon increasing irradiation time (pink curve). (d) Absorption spectra of a mixture of Tazobactam-PPG and Penicillin-PPG (both 20 μM, water, 1% DMSO, 37 °C) in the dark (black line) and during irradiation with violet light (λ = 395 nm, 50 s, blue line). (e) Absorption spectra of the mixture of Tazobactam-PPG and Penicillin-PPG shown in (d) after irradiation with violet light for 50 s (black line) followed by irradiation with green light (λ = 525 nm, 2 min, red lines). (f) Time-dependent absorption values at 410 and 505 nm of the absorption spectra shown in parts (d) and (e), showing the effect of violet light irradiation (first 50 s, violet box) and green light irradiation (after 50 s, green box). (g) Relative peak areas based on integration of the UPLC-MS peaks corresponding to Tazobactam-PPG and Penicillin-PPG after irradiation of a mixture of the two compounds (both 20 μM, water, 1% DMSO), with violet or green light for the indicated irradiation times. (h) Structures of Tazobactam-PPG and Penicillin-PPG.

9

(a) The effect of irradiation on the growth of β-lactam resistant E. coli DH5α. Irradiation with light of both colors resulted in bacterial growth inhibition. (b) Irradiation with solely one of the two colors allowed for bacterial growth. (c) Spatial control of bacterial growth in a dual-color system. (i) Schematic representation of the layout of the plate experiment; two orthogonal slits allowed for an overlap of the two irradiation colors solely in the middle of the plate. Note that green light illumination took place from the top of the Petri dish (as in Figure ), whereas violet light illumination was performed from the bottom of the dish. (ii) Results of the plate experiment with Penicillin-PPG and Tazobactam-PPG (80 and 180 μM, respectively). (iii) Control for light cytotoxicity (containing 80 μM penicillin).