This protocol describes a technique for visualizing macrophage behavior and death in embryonic zebrafish during Mycobacterium marinum infection. Steps for the preparation of bacteria, infection of the embryos, and intravital microscopy are included. This technique may be applied to the observation of cellular behavior and death in similar scenarios involving infection or sterile inflammation.
Zebrafish is an excellent model organism for studying innate immune cell behavior due to its transparent nature and reliance solely on its innate immune system during early development. The Zebrafish Mycobacterium marinum (M. marinum) infection model has been well-established in studying host immune response against mycobacterial infection. It has been suggested that different macrophage cell death types will lead to the diverse outcomes of mycobacterial infection. Here we describe a protocol using intravital microscopy to observe macrophage cell death in zebrafish embryos following M. marinum infection. Zebrafish transgenic lines that specifically label macrophages and neutrophils are infected via intramuscular microinjection of fluorescently labeled M. marinum in either the midbrain or the trunk. Infected zebrafish embryos are subsequently mounted on low melting agarose and observed by confocal microscopy in X-Y-Z-T dimensions. Because long-term live imaging requires using low laser power to avoid photobleaching and phototoxicity, a strongly expressing transgenic is highly recommended. This protocol facilitates the visualization of the dynamic processes in vivo, including immune cell migration, host pathogen interaction, and cell death.
Mycobacterial infection has been demonstrated to cause host immune cell death1. For example, an attenuated strain will trigger apoptosis in macrophages and contain the infection. However, a virulent strain will trigger lytic cell death, causing bacterial dissemination1,2. Considering the impact these different types of cell death have on host anti-mycobacterial response, a detailed observation of macrophage cell death during mycobacterial infection in vivo is needed.
The conventional methods for measuring cell death are to use dead cell stains, such as Annnexin V, TUNEL, or acridine orange/propidium iodide staining3,4,5. However, these methods are unable to shed light on the dynamic process of cell death in vivo. The observation of cell death in vitro has already been facilitated by live imaging6. However, whether the results accurately mimic physiological conditions remains unclear.
Zebrafish have been an excellent model for studying host anti-mycobacterium responses. It has a highly conserved immune system similar to that of humans, an easily manipulated genome, and the early embryos are transparent, which allows for live imaging7,8,9. After infection with M. marinum, adult zebrafish form typical mature granuloma structures, and embryonic zebrafish form early granuloma like structures9,10. The dynamic process of innate immune cell-bacteria interaction has been explored previously in the zebrafish M. marinum infection model11,12. However, due to high spatial-temporal resolution requirement, the details surrounding the death of the innate immune cells remain largely undefined.
Here we describe how to visualize the process of macrophage lytic cell death triggered by mycobacterial infection in vivo. This protocol may also be applied to visualizing cellular behavior in vivo during development and inflammation.
Zebrafish were raised under standard conditions in compliance with laboratory animal guidelines for ethical review of animal welfare (GB/T 35823-2018). All zebrafish experiments in this study were approved (2019-A016-01) and conducted at Shanghai Public Health Clinical Center, Fudan University.
1. M. marinum Single Cell Inoculum Preparation (Figure 1)
2. Zebrafish Embryo Preparation
3. Infection via Bacterial Microinjection
4. Live Imaging of the Infection
5. Single Cell UV Irradiation to Induce Apoptosis and Live Imaging
6. Image Processing
Mycobacterium infection can trigger different host responses based on the routes of infection. In this protocol, zebrafish embryos are infected by intramuscular microinjection of fluorescently labeled bacteria into the midbrain or trunk (Figure 3) and observed by confocal live imaging. Infection via these two routes will locally restrict the infection causing innate immune cell recruitment and subsequent cell death.
Visualizing the details of innate immune cell death is challenging. Lytic cell death occurs over a very short time window and requires high-resolution microscopy to observe. Also, the high motility of innate immune cells allows them to migrate out of the observation area. In this protocol, we solve this issue by observing multiple embryos in parallel. An array of zebrafish embryos can be mounted on a single glass microscope slide for infection, and up to 10 embryos can be mounted on the same 35 mm glass bottom dish for live imaging (Figure 4). By taking advantage of a live data model of confocal microscopy, more than one embryo can be observed simultaneously. This enhances the efficiency of the live imaging and greatly increases the probability of capturing the entire lytic cell death process.
The innate immune system is the first line of defense against mycobacterial infection, and two key components are the macrophage and the neutrophil. Here we utilize previously reported Tg(coro1a:eGFP;lyzDsRed2) and Tg(mpeg1:loxP-DsRedx-loxP-eGFP;lyz:eGFP) to distinguish the macrophages and the neutrophils in vivo16,17,18. A macrophage heavily engorged with bacteria became round and displayed reduced motility, with eventual cytoplasmic swelling, rupturing of the cell membrane, and quick dissemination of the cytoplasmic content. These events are typical morphological changes of lytic cell death as previously reported (Figure 5A)16. UV irradiation has been used to trigger cells to undergo apoptosis in zebrafish20,21. Consistent with this notion, UV irradiated macrophages showed typical apoptotic cell phenotypes, such as cell shrinkage, nuclear fragmentation, and chromatin condensation (Figure 5B)22,23. Combined with the use of Cerulean-fluorescent M. marinum19, three color live imaging of the interaction among macrophage, neutrophil, and M. marinum was achieved in vivo. We also observed that macrophages can actively phagocytose and disseminate M. marinum (Supplemental Figure 3A). However, neutrophils had limited phagocytic capability and quickly underwent lytic cell death without obvious bacterial engorgement (Supplemental Figure 3B). Neutrophils could be triggered by the phagocytosis of only a few dead M. marinum that do not express Cerulean-fluorescence, or simply by phagocytosis of limited dead cell debris.
Figure 1: Schematic diagram of single cell bacteria preparation. Single cell Cerulean-fluorescent M. marinum stocks were generated following the described process. Please click here to view a larger version of this figure.
Figure 2: Diagram of zebrafish embryo mounting for microinjection. (A) Schematic diagram of the mounting process. (B) Zebrafish embryos were mounted laterally for infection of the trunk region. (C) Zebrafish embryos were mounted with their heads directed upwards for infection of the midbrain. Please click here to view a larger version of this figure.
Figure 3: Positioning for microinjection. (A) The red arrow indicates the injection site for infection of the trunk region. (B) The red arrow indicates the injection site for midbrain infection. Please click here to view a larger version of this figure.
Figure 4: Mounting zebrafish embryos for live imaging. (A) For midbrain infection, zebrafish embryos were mounted with their heads directed downwards. (B) For the trunk region infection, zebrafish embryos were mounted laterally with the injection site close to the bottom of the glass bottom dish. Please click here to view a larger version of this figure.
Figure 5: Typical morphological changes in M. marinum infection induced macrophage lytic cell death and UV induced macrophage apoptosis. (A) Time lapse imaging of a macrophage (Mac) undergoing lytic cell death once it is heavily engorged with M. marinum. The midbrain of the 3 dpf Tg(coro1a:eGFP;lyzDsRed2) zebrafish embryo is infected by Cerulean-fluorescent M. marinum (~500 cfu) via microinjection. Images for both the overlay channel (upper panel) and DIC channel (lower panel) are provided. T 00:00 is 5 h 20 min post infection. White dashed lines = outline of the cell membrane; black dashed lines = swelling cytoplasm; black arrows = ruptured cell membrane; red dashed lines = quickly lost cytoplasmic content. (B) Time lapse imaging of UV irradiated macrophage. One GFP+ cell in the midbrain region of 3 dpf Tg(mfap4:eGFP) is irradiated by UV and followed by time lapse imaging. White dashed lines = outline of the cell membrane; black arrows = nuclear fragmentation and chromatin condensation. Scale bar = 15 μm. Please click here to view a larger version of this figure.
Supplemental Figure 1: Environmental chamber set up for live imaging. (A) Set the digital controller to keep the temperature at 28.5 °C. (B) Set wet wipes inside the chamber to provide humidity and prevent the evaporation of egg water. (C) Close the cover of the chamber and wait for at least 30 min for temperature stabilization before beginning live imaging. Please click here to view a larger version of this figure.
Supplemental Figure 2: Confocal panel setting for live imaging. (A) Representation of acquisition panel setting. (B) Representation of laser power and spectrum settings. (C) Representation of multiple jobs and loop setting in live data mode. Please click here to view a larger version of this figure.
Supplemental Figure 3: Macrophages disseminate infection and neutrophils undergo lytic cell death after M. marinum infection. (A) Macrophage disseminating M. marinum in the trunk of a 2 dpf Tg(coro1a:eGFP;lyz:DsRed2) zebrafish embryo infected by Cerulean-fluorescent M. marinum (~100 cfu). (B) Neutrophil (Neu) undergoing lytic cell death without obvious M. marinum laden in the trunk region of 3 dpf Tg(mpeg1:LRLG;lyz:eGFP) zebrafish embryo infected by Cerulean-fluorescent M. marinum (~100 cfu) via microinjection. Green color is assigned to LRLG and red color is assigned to eGFP for better visualization of the lytic cell death process. Arrows in cyan indicate target cells. Arrows in red point to the cells that are about to release cytoplasm contents in the next frame. Arrows in green point to the dead cells that have just lost their cytoplasm content. Scale bar = 25 μm. Please click here to view a larger version of this figure.
Video 1: A macrophage heavily laden with M. marinum undergoes lytic cell death, related to Figure 5A. Time-lapse imaging (63x objective) for 9 min and 18 s at 3 frames per second (fps) of the midbrain region of a 3 dpf Tg(coro1a:eGFP;lyzDsRed2) zebrafish embryo infected with Cerulean-fluorescent M. marinum. Please click here to view this video (Right click to download).
Video 2: A macrophage undergoes apoptosis after UV irradiation, related to Figure 5B. Time-lapse imaging (63x objective) of 74 min at 6 fps of the midbrain region of a 3 dpf Tg(mfap4:eGFP) zebrafish embryo. One GFP+ cell in the midbrain region of the embryos is irradiated by UV and followed by time lapse imaging. Please click here to view this video (Right click to download).
Video 3: A macrophage disseminates M. marinum, related to Supplemental Figure 3A. Time-lapse imaging (63x objective) of 24 min at 3 fps of the trunk region of 2 dpf Tg(coro1a:eGFP;lyz:DsRed2) zebrafish embryo infected with Cerulean-fluorescent M. marinum. Please click here to view this video (Right click to download).
Video 4: A neutrophil undergoes lytic cell death without obvious M. marinum engorgement, related to Supplemental Figure 3B. Time-lapse imaging (63x objective) of 7 min 30 s at 3 fps of the trunk region of 3 dpf Tg(mpeg1:LRLG;lyz:eGFP) zebrafish embryo infected with Cerulean-fluorescent M. marinum. Please click here to view this video (Right click to download).
This protocol describes the visualization of macrophage death during mycobacterial infection. Based on factors such as the integrity of the cell membrane, infection driven cell death can be divided into apoptosis and lytic cell death24,25. Lytic cell death is more stressful for the organism than apoptosis, because it triggers a strong inflammatory response 24,25. Observation of lytic cell death in vivo is difficult, due to the requirement of high spatial-temporal resolution, proper confocal microscopy settings, and strong transgenic expression.
Proper microinjection requires several critical steps. The bacterial stock must be thoroughly sonicated to remove all clumps before injection. We improved the zebrafish mounting for microinjection by embedding them on a glass slide between two layers of low melting agarose. After applying the second layer of agarose, the slide is transferred to an ice box or cold surface to accelerate solidification and prevent dehydration of the agarose. If the embryos need to be mounted on different slides, make sure to keep the top layer of agarose moist by adding extra egg water.
For live imaging, a high-resolution objective lens is required to observe the details of cell death. This requirement is always accompanied by short working distance, and thus requires positioning the infection site close to the cover slide. A long working distance water lens is ideal for imaging the deeper tissue and will allow for more room for proper embryo mounting. Extended live imaging using a laser with high intensity will cause tissue damage or the death of the embryo. Thus, it is very important to keep the intensity of the laser as low as possible to avoid photobleaching and toxicity. A strongly expressing transgenic can facilitate the observation using a laser with low intensity. Because GFP expression is stronger in Tg(coro1a:eGFP) than Tg(mpeg1:eGFP), we used Tg(coro1a:eGFP;lyz:DsRed2) instead of Tg(mpeg1:eGFP;lyz:DsRed2) in this study. Setting up a mobile workstation for microinjection close to the confocal machine is best for observing quick responses. Chilling the low melting agarose on ice to accelerate solidification time can also help reduce time between injection and live imaging.
In this protocol, we focus on observing macrophage behavior. However, the detailed study of neutrophil behavior during mycobacterial infection can also be informative. For example, how neutrophil extracellular traps (NETs) are involved in killing extracellular mycobacterium remains largely undefined. Combining the imaging technique described in this protocol with a histone protein labeling transgenic will greatly facilitate the visualization of NETs in vivo.
Currently, zebrafish are recognized as a very robust system for studying innate immune cell behavior. Statistical data of phagocytosis and cell death could be achieved using this protocol. Combined with the powerful gene editing tools available today, this protocol can provide an effective platform for further understanding the effect of a variety of factors on host-pathogen interaction in vivo.
The authors have nothing to disclose.
We thank Dr. Zilong Wen for sharing zebrafish strains, Dr. Stefan Oehlers and Dr. David Tobin for sharing M. marinum related resources, Yuepeng He for assistance in figure preparation. This work was supported by the National Natural Science Foundation of China (81801977) (B.Y.), the Outstanding Youth Training Program of Shanghai Municipal Health Commission (2018YQ54) (B.Y.), Shanghai Sailing Program (18YF1420400) (B.Y.), and Open Fund of Shanghai Key Laboratory of Tuberculosis (2018KF02) (B.Y.).
0.05% Tween-80 | Sigma | P1379 | |
10 mL syringe | Solarbio | YA0552 | |
10% OADC | BD | 211886 | |
3-aminobenzoic acid | Sigma | E10521 | |
5 μm filter | Mille X | SLSV025LS | |
50 μl/ml hygromycin | Sangon Biotech | A600230 | |
7H10 | BD | 262710 | |
7H9 | BD | 262310 | |
A glass bottom 35 mm dish | In Vitro Scientific | D35-10-0-N | |
Agarose | Sangon Biotech | A60015 | |
Confocal microscope | Leica | TCS SP5 II | |
Enviromental Chamber | Pecon | temp control 37-2 digital | |
Eppendorf microloader | Eppendorf | No.5242956003 | |
Glass microscope slide | Bioland Scientific LLC | 7105P | |
Glycerol | Sangon Biotech | A100854 | |
Incubator | Keelrein | PH-140(A) | |
M.marinum | ATCC BAA-535 | ||
Microinjection needle | World Precision Instruments | IB100F-4 | |
Microinjector | Eppendorf | Femtojet | |
Micromanipulator | NARISHIGE | MN-151 | |
msp12:cerulean | Ref.: PMID 25470057; 27760340 | ||
Phenol red | Sigma | P3532 | |
PTU | Sigma | P7629 | |
Single concavity glass microscope slide | Sail Brand | 7103 | |
Sonicator | SCICNTZ | JY92-IIDN | |
Spectrophotometer (OD600) | Eppendorf AG | 22331 Hamburg | |
Stereo Microscope | OLYMPUS | SZX10 | |
Tg(mfap4:eGFP) | Ref.: PMID 30742890 | ||
Tg(coro1a:eGFP;lyzDsRed2) | Ref.: PMID 31278008 | ||
Tg(mpeg1:LRLG;lyz:eGFP) | Ref.: PMID 27424497; 17477879 |