A Fluorescent Probe for Detecting Mycobacterium tuberculosis and Identifying Genes Critical for Cell Entry

Abstract

The conventional method for quantitating Mycobacterium tuberculosis (Mtb) in vitro and in vivo relies on bacterial colony forming unit (CFU) enumeration on agar plates. Due to the slow growth rate of Mtb, it takes 3-6 weeks to observe visible colonies on agar plates. Imaging technologies that are capable of quickly quantitating both active and dormant tubercle bacilli in vitro and in vivo would accelerate research toward the development of anti-TB chemotherapies and vaccines. We have developed a fluorescent probe that can directly label the Mtb cell wall components. The fluorescent probe, designated as DLF-1, has a strong affinity to the D-Ala-D-Ala unit of the late peptidoglycan intermediates in the bacterial cell wall. We demonstrate that DLF-1 is capable of detecting Mtb in both the actively replicating and dormant states in vitro at 100 nM without inhibiting bacterial growth. The DLF-1 fluorescence signal correlated well with CFU of the labeled bacteria (R² = 1 and 0.99 for actively replicating and dormant Mtb, respectively). DLF-1 can also quantitate labeled Mtb inside of cells. The utility of DLF-1 probe to quantitate Mtb was successfully applied to identify genes critical for cell invasion. In conclusion, this novel near infrared imaging probe provides a powerful new tool for enumerating Mtb with potential future use in bacterial virulence study.

Keywords: Mycobacterium tuberculosis; dormant M. tuberculosis; fluorescence; imaging.

Figure 1

Structure of DLF-1. DLF-1 is synthesized by conjugating glycyl-vancomycin with a near infrared dye Cy5.5.

Figure 2

Thermodynamic evaluation of Ac-L-Lys-D-Ala-D-Ala-OH binding to DLF-1. Top: Raw curve of power vs. time. Bottom: Fit of integrated peak areas (heat released per injection) used to determine the thermodynamic parameters presented in Supplementary Table 1. (A) Representative ITC data measuring the thermodynamics of binding of peptide to Vancomycin. Peptide (600 μM) was titrated into a solution containing vancomycin (32 μM) in 1X PBS at 25°C. (B) Representative ITC data measuring the thermodynamics of binding of peptide to DLF-1. One percent of DMSO was added to control and test titrations, as it increased solubility of DLF-1.

Figure 3

Mycobacterial growth curves in the presence or absence of DLF-1. (A–C) represent the growth curve of each mycobacterium in media containing 0, 100 nM, 1 μM of DLF-1, respectively. (A) Mycobacterium smegmatis; (B) Mycobacterium bovis BCG; (C) Mycobacterium tuberculosis CDC1551. Error bars represent standard errors of means calculated based on results from three independent experiments.

Figure 4

Direct labeling of different mycobacteria with DLF-1. Fluorescent intensities of DLF-1 labeled bacteria were plotted against the numbers of bacteria. (A–C) represent plots for M. Smegmatis, M. bovis BCG, and Mtb CDC1551, respectively. Linear trend lines with R-squares are also displayed for each plot. Linear range: M. smegmatis 4.3 × 10⁵−4.3 × 10⁷; M. bovis BCG 3.5 × 10⁴−3.5 × 10⁷; and Mtb 5.3 × 10⁴−5.3 × 10⁷. (D) Minimum number of Mtb within the detectable linear range. ^**P < 0.01; ^*P < 0.05. Error bars represent standard errors of means calculated based on results from three independent experiments.

Figure 5

Imaging infection of human epithelial cell line (A549) and human macrophage cell line (THP-1) with DLF-1 (100 nM) labeled Mtb. (A) Fluorescence detected from cells infected with DLF-1 labeled Mtb at different MOIs in vitro. Fluorescent intensities were measured for the labeled DLF-1 (Cy5.5) and for the endogenous fluorescent tdTomato protein, respectively. Error bars represent standard errors of means calculated based on results from three independent experiments. One-way ANOVA test was conducted for assessing overall differences among groups, and Turkey's multiple comparison tests were applied to assess differences between two groups. ^**P < 0.01; and ^***P < 0.001. (B) Images of DLF-1 labeled Mtb inside of cells. THP-1 and A549 were infected with a tdTomato-expressing Mtb strain which has been labeled with DLF-1. Red, pseudo-colored DLF-1; Green, pseudo-colored tdTomato; and Blue, DAPI. White scale bars represent 10 μm.

Figure 6

DLF-1 labeling of dormant Mtb cultured in Wayne model. (A) Correlation between fluorescence measured with DLF-1 wavelengths and bacterial numbers of dormant Mtb. Mtb was labeled with DLF-1 at 100 nM. (B) Threshold detection of DLF-1 labeling dormant Mtb in culture. ^*P < 0.05. (C) DLF-1 (100 nM) labeling dormant Mtb smeared on slides, and imaged with a fluorescent microscope. (D) Microscopy images of the DLF-1 (100 nM) labeled dormant Mtb strain carrying a genome-integrated tdTomato gene inside of THP-1 macrophages. Red, Cy5.5; Green, tdTomato; Blue, DAPI stained nuclei.