We describe the optical imaging of mice infected with Mycobacterium tuberculosis (M. tuberculosis) using reporter enzyme fluorescence (REF). This protocol facilitates the sensitive and specific detection of M. tuberculosis in pre-clinical animal models for pathogenesis, therapeutics and vaccine research.
Reporter enzyme fluorescence (REF) utilizes substrates that are specific for enzymes present in target organisms of interest for imaging or detection by fluorescence or bioluminescence. We utilize BlaC, an enzyme expressed constitutively by all M. tuberculosis strains. REF allows rapid quantification of bacteria in lungs of infected mice. The same group of mice can be imaged at many time points, greatly reducing costs, enumerating bacteria more quickly, allowing novel observations in host-pathogen interactions, and increasing statistical power, since more animals per group are readily maintained. REF is extremely sensitive due to the catalytic nature of the BlaC enzymatic reporter and specific due to the custom flourescence resonance energy transfer (FRET) or fluorogenic substrates used. REF does not require recombinant strains, ensuring normal host-pathogen interactions. We describe the imaging of M. tuberculosis infection using a FRET substrate with maximal emission at 800 nm. The wavelength of the substrate allows sensitive deep tissue imaging in mammals. We will outline aerosol infection of mice with M. tuberculosis, anesthesia of mice, administration of the REF substrate, and optical imaging. This method has been successfully applied to evaluating host-pathogen interactions and efficacy of antibiotics targeting M. tuberculosis.
The slow growth rate of M. tuberculosis is a major roadblock in rapid diagnosis of tuberculosis1,2,3. While culture based diagnosis takes weeks to produce results, acid-fast smear has diagnostic limitations4 in children5 and patients co-infected with human immunodeficiency virus6,7. Optical imaging technologies have recently been recognized as an alternative to traditional diagnostic methods for tuberculosis8,9. Fluorescence and bioluminescence can be used to optically image M. tuberculosis in live animals in real-time10,11,12,13,14,15,16,17,18,19. Optical imaging has the benefit of a rapid and specific assessment of an infection with M. tuberculosis20,21,22.
We outline details for optical imaging of M. tuberculosis in live mice using REF. This method is very specific and sensitive23,24 and, similar to other optical methods, is less expensive than other methods of tuberculosis (TB) imaging25, including computed tomography (CT)26, magnetic resonance imaging (MRI)27, and F-fluorodeoxyglucose positron emission tomography/CT (F-FDG PET/CT)28. REF utilizes custom fluorescent or bioluminescent substrates that upon cleavage by a bacterial enzyme, produce a fluorescent product8,29. Hence, it has the advantage of not requiring a recombinant mycobacterial reporter strain30,31. The FRET substrate described is comprised of a fluorochrome and a quencher connected by a β-lactam ring that is hydrolyzed by BlaC (β-lactamase), naturally constitutively expressed by tuberculosis-complex Mycobacterium8,32. The bacteria directly generate signal due to REF catalytic activity that allows amplification by many orders of magnitude and sensitive detection of M. tuberculosis.
The REF substrate used in this study has excellent tissue penetration in live animals and reduced background due to its long wavelength. With this long-wavelength substrate it is possible to achieve a threshold of detection for M. tuberculosis of nearly 100 colony forming units (CFU) in vitro and <1000 CFU in the lungs of mice in vivo (whole animal)8,33. REF can be used as a diagnostic tool for sputum, clinical materials and even directly in patients with micro endoscopic systems16,32,33,34 due its high sensitivity and specificity. REF can be applied to any tuberculosis clinical strain, because it uses a naturally produced bacterial enzyme, BlaC, for detection present in all strains. These characteristics make REF imaging a valuable tool in pre-clinical tuberculosis research in general to facilitate therapeutic and vaccine evaluation as well as analysis of pathogenesis, but may also ultimately be applied to diagnosis in tuberculosis patients.
Animal studies were performed in accordance with the guidelines and regulations set forth by the Institutional Animal Care and Use Committee of Texas A&M University. Review and approval from a biohazard safety officer or committee at your institution may be required.
Caution: All the procedures require BSL3 containment. The personnel are required to wear personal protective equipment at all times. All manipulations are performed inside the Biosafety Cabinet (BSC) and all sharps are disposed of in sharps containers. Work surfaces and BSC are cleaned with buffered phenol and 70% ethanol before initiating work and after work. Procedures would normally be done at biosafety level 3 with Mycobacterium tuberculosis, but for illustration and filming purposes, the authors demonstrate these procedures at biosafety level 2 with less virulent bacteria.
1. Strains and culture conditions
Note: M. tuberculosis strain CDC1551 is used in this study, but any M. tuberculosis strain can be used in the same manner.
2. Aerosol infection of mice using a Madison Chamber
3. Animal anesthesia
4. Reporter enzyme fluorescence (REF) imaging
5. Analysis of REF imaging
6. Quantification of bacteria by CFU
REF imaging of mice infected with M. tuberculosis along with uninfected control mouse are shown in Figure 1A. The infected mice produced a significantly (P = 0.0057) higher fluorescence signal from the lungs at 6 weeks post-infection compared to control upon substrate administration. A typical time course to study the therapeutic efficacy against M. tuberculosis using REF could be imaging at week 1, 2, 4, 6, 15, 24 post-infection. Fluorescence intensity is quantified using epi-illumination for lung tissues (Figure 1B). The consistently increasing signal from week 2 to week 6 in lungs of infected mice suggests that REF is successfully able to detect M. tuberculosis in vivo. A drop in the background signal of control mice at a later time point (Figure 1B) could be attributed to the increase in body mass and volume of mice over a period of 6 weeks thereby reducing the excitation wavelength penetration. 3D FLIT reconstruction of fluorescence sources in mice infected with M. tuberculosis is represented in Figure 2A. The image sequences are acquired at multiple trans-illumination points in mouse using the same excitation and series of emission filters. These image sequences are then used for 3D FLIT reconstruction for fluorescent source distribution within the animal subject. Figure 2A-D demonstrates the different directions (coronal, sagittal and transaxial) of mouse tomography with lung organ registration. NTF efficiency maps for checking the reconstruction quality are represented in Figure 2E. NTF allows subtracting the background light leakage from trans-illumination images through an extra image captured with neutral density filter. The image taken with the specific excitation filter (Figure 2E-measured) is normalized to the transmission image measured with the same emission filter and an open excitation filter (Figure 2E- simulated) to give the signal produced by the substrate alone. The similar measured and simulated NTF efficiency profile in both the horizontal (Figure 2F) and vertical (Figure 2G) profiles with nearly 0% percentage error (Figure 2E-% error) provides evidence of a good quality 3D reconstruction with reduced artifacts and improved signal localization and sensitivity.
Figure 1. Imaging of M. tuberculosis with REF. (A) in vivo imaging of mice uninfected and infected with M. tuberculosis at 2, 4, and 6 weeks post-infection. The color bar represents the intensity of the fluorescent signal in photons per second per cm2 from low (yellow) to high (red). (B) Quantification of fluorescence intensity from mice infected with M. tuberculosis. Fluorescence values for both control (black) and infected (red) are represented along with the standard error for each time point. The significance of results was determined by Students t-test, p values of <0.05 were considered significant. Please click here to view a larger version of this figure.
Figure 2. 3D FLIT reconstruction of fluorescence sources in mice infected with M. tuberculosis. Mouse tomography represented in different directions; A) coronal, B) sagittal, and C) transaxial with D) lung organ registration. The 3D region of interest (ROI) is represented as a red cube in the lung for source measurements. (E) NTF efficiency maps of the measured and simulated for checking the reconstruction quality. The measured and simulated NTF efficiency profile was compared, providing good quality of 3D reconstruction (similar measured and simulated NTF efficiency). F) Horizontal and G) vertical signal profiles representing measured (blue) and simulated (red) NTF efficiency curve. The horizontal and vertical red bars indicate the source position. The color bar represents the intensity of the fluorescent signal in photons per second per cm2 from low (blue) to high (red). Please click here to view a larger version of this figure.
When using imaging techniques, such as REF, there are key strategies that allow generation of robust and consistent data. Optical imaging produces scattered light in tissues which can impact the depth of penetration, since it is difficult to capture light emitted in all directions. Use of a near infra-red fluorophore (NIR) substrate for REF imaging having an excitation and emission wavelength in the Querange of 700 – 900 nm facilitates minimal absorption of the fluorescent signal by mammalian tissues. The custom designed substrate was constructed by linking a NIR fluorophore, IRDye 800Cw to a quencher, IRDye QC-1, by a lactam ring allowing fluorescence resonance energy transfer-based quenching. IRDye has excellent tissue penetration and light scattering characteristics and does not have any evident harmful effect on mammals36, being cleared from the blood and organs by 24 h. Fluorescence signal significantly increases beginning 4 h after substrate administration, reaching maximal levels 6 h after administration.
The bacterial infectious dose, mode of administration of infectious dose and the substrate as well as time-points of imaging post infection by performing pilot studies should be standardized before embarking on large, complex, experiments. Pilot studies can greatly reduce time and cost when imaging a large number of animals because a standard procedure can be optimized prior to doing the key experiment. Bacterial loads should be determined in the organs/tissues of interest following whole body imaging using trans-illumination and ex vivo lung imaging using epi-illumination to validate the source of the signal and determine the quality of correlation with bacterial numbers present8. Pilot studies will provide insight into the threshold of detection, dynamic range of the technique as well as determination of the optimal experimental conditions for imaging.
The primary advantages of using REF imaging as compared to other fluorescent and bioluminescent strategies are its high sensitivity and ability to image natural M. tuberculosis strains. REF imaging utilizes the catalytically robust enzyme BlaC that is conserved in all M. tuberculosis clinical isolates and tuberculosis-complex strains. The great sensitivity of REF imaging is due to the rapid catalytic rate of BlaC in combination with retention of the cleaved fluorescent product by the host cell. Signal continuously increases as long as the substrate is available resulting in nearly limitless build-up of the signal within the infected cells and tissues. This increased sensitivity of REF imaging as compared to alternative approaches of imaging allows specific detection of M. tuberculosis both in vitro and in vivo8,23,24,29.
REF imaging can be used for bacterial detection without genetic modifications enabling its direct application to any infection model, either laboratory animals8,37 or human clinical materials29,38. REF can be used to detect and image a wide array of pathogens39,40, since fluorogenic substrates can be developed for numerous enzymatic targets other than BlaC such as proteases, kinases, ureases, and β-galactosidases. However, careful thought should be given to the target in order to ensure it displays optimal characteristics for imaging. BlaC represents a good model enzyme for characteristics that will ensure successful application of this strategy. REF imaging provides an immediate read-out on bacterial load present in lungs during infections, which greatly speeds progress in study of tuberculosis pathogenesis, since determination of bacterial numbers normally requires three to six weeks, but even in more rapid-growing organisms this approach will save a great deal of time. REF could also be used to discriminate carcinomas from tuberculosis, a key problem in diagnosis of nodular lesions in patients41,42. REF serves as a novel tool to accelerate translational tuberculosis imaging and can even be applied to humans, potentially allowing rapid prediction of therapeutic outcomes.
Isoflurane | VETONE | 501027 | |
CNIR800 | Custom synthesized | ||
Fatal Plus solution | Vortech Pharmaceutical Ls, Ltd | ||
7H9 Middlebrook broth | BD | 271310 | |
OADC Middlebrook enrichment | BD | 212351 | |
Sporcidin | RE-1284F | ||
7H11 Middlebrook Agar | BD | 212203 | |
Madison Chamber | |||
IVIS Spectrum | Perkin Elmer | 124262 | |
XGI-8-gas Anesthesia System | Perkin Elmer | ||
Living Imaging software | Perkin Elmer | ||
Transparent nose cones | Perkin Elmer | ||
M. tuberculosis strain CDC1551 | ATCC | ||
Female BALB/C mice, 5-7 weeks | Jackson Laboratory |