Azide-Masked Fluorescence Turn-On Probe for Imaging Mycobacteria

Abstract

A fluorescence turn-on probe, an azide-masked and trehalose-derivatized carbazole (Tre-Cz), was developed to image mycobacteria. The fluorescence turn-on is achieved by photoactivation of the azide, which generates a fluorescent product through an efficient intramolecular C-H insertion reaction. The probe is highly specific for mycobacteria and could image mycobacteria in the presence of other Gram-positive and Gram-negative bacteria. Both the photoactivation and detection can be accomplished using a handheld UV lamp, giving a limit of detection of 10³ CFU/mL, which can be visualized by the naked eye. The probe was also able to image mycobacteria spiked in sputum samples, although the detection sensitivity was lower. Studies using heat-killed, stationary-phase, and isoniazid-treated mycobacteria showed that metabolically active bacteria are required for the uptake of Tre-Cz. The uptake decreased in the presence of trehalose in a concentration-dependent manner, indicating that Tre-Cz hijacked the trehalose uptake pathway. Mechanistic studies demonstrated that the trehalose transporter LpqY-SugABC was the primary pathway for the uptake of Tre-Cz. The uptake decreased in the LpqY-SugABC deletion mutants ΔlpqY, ΔsugA, ΔsugB, and ΔsugC and fully recovered in the complemented strain of ΔsugC. For the mycolyl transferase antigen 85 complex (Ag85), however, only a slight reduction of uptake was observed in the Ag85 deletion mutant ΔAg85C, and no incorporation of Tre-Cz into the outer membrane was observed. The unique intracellular incorporation mechanism of Tre-Cz through the LpqY-SugABC transporter, which differs from other trehalose-based fluorescence probes, unlocks potential opportunities to bring molecular cargoes to mycobacteria for both fundamental studies and theranostic applications.

Scheme 1. Transport and Utilization of Trehalose in Mycobacteria

The scheme was created with BioRender.com

Scheme 2. Synthesis of Tre-Cz and Mal-Cz and the Photochemical Conversion of Tre-Cz to the Fluorescent Product P

See the Supporting Information for the syntheses of IV, V, and VI. DMF: N,N-dimethylformamide.

Figure 1

(A) Absorption spectra of Tre-Cz before and after irradiating a solution of Tre-Cz in methanol (50 μM) with a handheld UV lamp (intensity: 0.5 mW/cm² at 365 nm) every 10 s for a total of 120 s. Direction of arrows indicates increasing irradiation time. Inset is the expanded region showing decreases of absorbance at 329 nm with increasing irradiation time. (B) Change of absorbance at 329 and 378 nm vs irradiation time. The red and black curves are the first-order exponential decay fits of the experimental data. (C) Fluorescence spectra of Tre-Cz in methanol (50 μM) before (red) and after irradiation with a handheld UV lamp for 180 s (black). Excitation: 378 nm.

Figure 2

Uptake of Tre-Cz is specific for mycobacteria. (A) Fluorescence intensity of M. tuberculosis H37Ra (10⁸ CFU/mL) after treating with different concentrations of Tre-Cz (0–500 μM). (B) Fluorescence intensity of M. smegmatis mc²155, S. epidermidis ATCC 35984, P. aeruginosa PAO1, and E. coli ATCC 25922 after treating with Tre-Cz. The concentration of all bacteria strains was 10⁸ CFU/mL. (C) Fluorescence intensity of mixed bacteria of M. smegmatis, S. epidermidis, P. aeruginosa, and E. coli after incubating with Tre-Cz at different concentrations of M. smegmatis (10²–10⁸ CFU/mL) and fixed concentrations of the other three strains (each at 10⁸ CFU/mL). (D) Fluorescence intensity of different concentrations of M. smegmatis mc²155 incubated with Tre-Cz. (E) Fluorescence intensity of mixed bacteria incubated with Tre-Cz at different concentrations of E. coli, S. epidermidis, and P. aeruginosa (each at 10²–10⁸ CFU/mL) and a fixed concentration of M. smegmatis (10⁸ CFU/mL). (F) Uptake of Tre-Cz or Mal-Cz (100 μM) by M. smegmatis mc²155 (10⁸ CFU/mL). The fluorescence intensity of the bacteria only was subtracted from each data point. Results are presented as means ± standard error of the mean (SEM) of three independent experiments. Data were analyzed by one-way analysis of variance (ANOVA) (*P < 0.1, **P < 0.01, ***P < 0.001, ****P < 0.0001, n.s.: not significant, and P > 0.05). Groups were compared to (A) [Tre-Cz] = 0, (B) M. smegmatis, and (F) Tre-Cz.

Figure 3

Fluorescence turn-on imaging of mycobacteria treated with Tre-Cz. Confocal fluorescence, differential interference contrast (DIC), and overlay images of M. smegmatis mc²155 (10⁸ CFU/mL) incubated with Tre-Cz (100 μM) (A) before and (B) after irradiation. Images were taken at the same locations in both (A,B). (C) M. smegmatis mc²155 incubated with the photoproduct P (100 μM). (D) M. smegmatis-spiked sputum processed and incubated with Tre-Cz (Subject 1). Images of samples from subjects 2–5 are shown in Figure S7. Fluorescence intensities of the cell pellets and cell washes of all the five samples are shown in Figure S8. The scale bars are 20 μm.

Figure 4

Uptake of Tre-Cz requires metabolically active bacteria: (A) heat-killed vs viable bacteria, (B) log-phase vs stationary-phase bacteria, and (C) untreated vs bacteria treated with varying concentrations of INH for 3 h. The viability of INH-treated bacteria after incubating with Tre-Cz was determined by plating on 7H10 agar and counting the colonies (CFU/mL). The data are shown on top of each column. In all experiments, ∼10⁸ CFU/mL M. smegmatis mc²155 was used. The fluorescence intensity of bacteria only was subtracted from each data point. The data are presented as means ± SEM from three independent experiments. Data were analyzed by one-way ANOVA (*P < 0.1, **P < 0.01, ***P < 0.001, ****P < 0.0001, n.s.: not significant, and P > 0.05). Groups were compared to [INH] = 0 in (C).

Figure 5

Fluorescence intensity of M. smegmatis mc²155 (10⁸ CFU/mL) treated with Tre-Cz in the presence of varying concentrations of trehalose. Data were analyzed by one-way ANOVA (****P < 0.0001). Groups were compared to [Trehalose] = 0.

Figure 6

Confocal fluorescence microscopy images of (A) ΔAg85C, (B) ΔlpqY, (C) ΔsugC, and (D) ΔsugC::sugC after treating with Tre-Cz. The scale bars are 20 μm. (E) Fluorescence intensity of M. smegmatis mc²155 wild type, ΔlpqY, ΔsugA, and ΔsugB after treating with Tre-Cz. (F) Fluorescence intensity of M. smegmatis mc²155 wild type, ΔsugC, and ΔsugC::sugC after treating with Tre-Cz. (G) Fluorescence intensity of M. smegmatis mc²155 wild type, ΔAg85C, and ΔlpqY after treating with different concentrations of Tre-Cz. In (E–G), the fluorescence intensity of the bacteria only was subtracted from each data point. The results are means ± SEM from three independent experiments. Data were analyzed by one-way ANOVA (*P < 0.1, **P < 0.01, ***P < 0.001, ****P < 0.0001, n.s.: not significant, and P > 0.05).