Imaging Neutrophil Migration in the Mouse Skin to Investigate Subcellular Membrane Remodeling Under Physiological Conditions

All animal experiments and procedures were approved by the National Cancer Institute (National Institutes of Health, Bethesda, MD, USA) Animal Care and Use Committee (protocols LCMB-031 and LCMB-035) and were compliant with all relevant ethical regulations. Both male and female mice aged between 2 and 6 months were used for the experiments. mT/mG and wild-type (WT) host mice are in an FVB/NJ background while LyzM-Cre x mTmG mice are in a C57BL/6 background.

1. Materials and preparation of reagents and tools

1. Coat tubes, dishes, and needles/syringes with PBS + 1% BSA (without Calcium and Magnesium) overnight under agitation.
2. Carefully clean surfaces and tools/instruments to be used for animal work using either 70% ethanol or by autoclaving.
   1. For neutrophils purification, prepare the following: 3 x 15 mL tubes, 1 x 60 mm dish, 1 x 1.5 mL tube; HBSS containing 10 mM HEPES (pH 7.3); red blood cell lysis solution (ACK); Histopaque 1077; and Histopaque 1119.
3. For neutrophil injections, prepare the following: 1 x low volume syringe (10 μL), 1 x 33 G beveled needle, isoflurane, and anesthesia solution (Ketamine, Xylazine, and Acepromazine at 80 mg·kg^-1, 2 mg·kg^-1 and 4 mg·kg^-1, respectively, in saline solution).

2. Purification of neutrophils from a donor mouse

1. Euthanize the donor mouse according to the local institutional regulations.
2. Collect long bones from the animal (femur, tibia, and humerus), paying attention to preserving bone integrity and avoiding cutting the bones' heads during collection²⁴.
3. In a BSA-coated 60 mm dish, remove muscles and fat tissue. Cut the bones' heads to get access to the marrow, and then flush and homogenize the bone marrow using a syringe (27 G needle) filled with HBSS.
4. Filter the solution into a 15 mL BSA-coated tube using a 40 µm strainer and centrifuge it at 400 x g for 5 min at room temperature (RT).
5. Aspirate the supernatant, resuspend the pellet in 1 mL of ACK for 30 s to lyse the red blood cells, and then add 9 mL of HBSS to stop the lysis. Centrifuge the solution at 400 x g for 5 min at RT.
6. Prepare a density step gradient in a 15 mL BSA-coated tube by gently placing 4 mL of 1119 Histopaque at the bottom of the tube, and then gently layering 4 mL of 1077 Histopaque on the top. Take extra care to avoid mixing the two layers.
7. Aspirate the supernatant and resuspend the cell pellet in 2 mL of HBSS. Gently layer the solution on top of the step gradient prepared above. Spin at 1,000 x g for 30 min at RT with the lowest possible acceleration and deceleration speed ("no brake" is preferred).
8. From the top of the tube, gently aspirate half of the 1077 Histopaque layer. Collect the remaining half of the 1077 Histopaque layer and the top half of the 1119 Histopaque layer into a fresh BSA-coated tube. The cloudy cell suspension between the two density layers contains predominantly neutrophils.
9. Add HBSS to reach a final volume of 15 mL and gently mix by inverting the tube. Spin at 400 x g for 5 min at RT.
10. After centrifugation, resuspend and wash the cell pellet with 10 mL of HBSS and spin at 400 x g for 5 min at room temperature. Repeat the procedure, and then resuspend the pellet in 15 mL of HBSS to determine the number of cells using a cell counter. Finally, spin and resuspend cells in 1 mL of HBSS and transfer to a 1.5 mL BSA-coated tube.
11. After purification, keep the neutrophil suspension in the BSA-coated tube under gentle agitation for 30 min on a tube rotator at RT until labeling and/or injection.
    NOTE: A successful purification yields 10-20 x 10⁶ neutrophils per mouse with a 95% purity. Purity is assessed via flow cytometry, using previously described neutrophil markers (Ly6G⁺, Ly6C^–, Gr1⁺, and CD11b⁺)^25,26.
    ​NOTE: Protocols on neutrophil purification from human peripheral blood and mouse bone marrow are available previously^24,27.

3. Neutrophil labeling

NOTE: To visualize the neutrophils, they were purified from LyzM-Cre mT/mG mice, in which myeloid cells express a membrane-targeted peptide fused with the GFP. In addition, purified neutrophils from wild-type (WT) mice were labeled in vitro. The following steps describe the procedure used to label purified neutrophils with a commercial cell-permeable green fluorescent dye and can be adapted to any other MPM compatible fluorescent probe of choice.

1. Add the green fluorescent dye (manufacturer's recommended concentration; here, 1 µM) to the neutrophil suspension (purified from WT mouse) in a 1.5 mL BSA-coated tube. Incubate for 30 min at RT under gentle agitation in a tube rotator.
2. Wash three times using HBSS, spin at 400 x g for 5 min at RT, and resuspend the pellet in 1 mL of HBSS.
3. Keep the neutrophil suspension in the BSA-coated tube under gentle agitation in a tube rotator at RT until the injection procedure.
   NOTE: Injection of labeled neutrophils performed right after labeling is highly preferred. However, if necessary, labeled neutrophils can be kept under agitation for an additional 10-15 min post-labeling without compromising the integrity of neutrophil response in vivo. Long-term agitation (more than 60 mins post labeling) will lead to clumping, neutrophil death, and misleading observations in vivo.
4. Just before the injection, resuspend the cell pellet in saline solution to reach a density of 2-5 x 10⁶ cells per 20 µL.

4. Neutrophil injection

NOTE: Anesthesia must be performed following the local institutional Animal Care and Use Committee guidelines.

1. Place the animal in a closed chamber and administer 3%-5% isoflurane (flow rate of 2 L/min) using a calibrated vaporizer connected to an oxygen concentrator.
2. Assess whether the animal is unconscious by testing paw withdrawal reflex upon pinching the hind footpad. Then, transfer the animal to a warming pad (37° C) to maintain its body temperature. Continue to administer the isoflurane (1%-2%; flow rate of 0.5 L/min) through a nose cone.
   SAFETY NOTE: To limit personnel exposure to toxic isoflurane, the use of a dispenser system equipped with a filtering unit to recapture circulating isoflurane is recommended.
3. To optimize imaging quality, remove the hairs from the ear using a fine trimmer. This procedure could also be performed a day or two before the experiment.
   NOTE: Use of hair-removal cream is discouraged since it is known to exacerbate inflammatory response in the mouse skin and introduce imaging artifacts.
4. Place the animal on its side. Using surgical tape, hold the edge of the ear, and flatten it along the warming pad. Gently apply the tape to prevent any damage to the ear.
5. Fill a syringe equipped with a 33 G needle with the neutrophil suspension (2-5 x 10⁶ cells per 20 µL) and mount it on a micromanipulator.
   NOTE: The injection angle between the needle and the ear on a warming pad must be between 10° and 30° to minimize tissue damage and increase the success of cell injection.
6. Gently move the needle (with the bevel up) toward the ear, and slowly pierce the skin. Once the needle is inside the skin, slowly push the piston and deliver 2-3 µL of cell suspension.
   NOTE: Areas with visible blood vessels should be avoided as damaging a blood vessel results in an undesired and complex immune response to injection and issues with the imaging.
7. Repeat the injection in 2-3 different areas of the ear, preferably 2-3 mm apart. Do not "overfill" the ear with the cell suspension.
   NOTE: The injection must be performed in the center of the ear. The periphery of the ear is thin and must be handled with care as it can be easily damaged, whereas the center part is thicker containing larger blood vessels and fewer hair follicles, and can withstand larger injection volumes.
8. Carefully remove the surgical tape and let the animal regain consciousness under supervision. Allow the animal to rest for 1 h before imaging to let the tissue and the cells recover from the injection procedure.

5. Anesthesia for IVM

NOTE: A deeper anesthesia significantly improves the imaging quality by decreasing the motion artifacts due to heartbeat and respiration. Neutrophil-injected mice are subjected to chemical anesthesia, which provides deeper sedation than gas-induced anesthesia. Anesthesia must be performed following the local institutional Animal Care and Use Committee guidelines.

1. One hour after recovery from the neutrophil injection, anesthetize the animals by a subcutaneous injection containing a mixture of Xylazine/Ketamine/Acepromazine (80 mg·kg^-1, 2 mg·kg^-1, and 4 mg·kg^-1, respectively) in saline solution.
   NOTE: To ensure reproducibility and reliability of the results, animals must be kept warm from the time of anesthesia to the end of the experiment using a heating pad or water circulating pad connected to a warming pump.
2. To maintain anesthesia beyond 1 h, inject a mixture of Ketamine/Acepromazine (40 mg·kg^-1, and 2 mg·kg^-1 respectively) subcutaneously every 40 min. Monitor anesthetized mice periodically for signs of consciousness by testing paw withdrawal reflex.
3. Additionally, for imaging procedures lasting more than 1 h, apply a sterile non-medication ophthalmic ointment to the eyes to prevent corneal drying during anesthesia and maintain body hydration by subcutaneous injections of saline solution (200 µL every hour).
   ​NOTE: Alternately, a digitally controlled perfusion-based system can be used to ensure a periodic and smooth injection of the anesthesia solution and/or provide fluids. An automated syringe driver can be set up to deliver the desired volume of anesthetic and provide hydration for longer periods. To this end, a winged infusion needle can be inserted subcutaneously into the dorsal skin of the animal and secured with surgical tape.

6. Imaging

NOTE: The animal setup and the imaging parameters described here are optimized for an inverted multiphoton microscope equipped with a single laser line to excite and simultaneously collect emissions from the different fluorophores.

1. Before placing the anesthetized animal on the microscope stage, ensure that the microscope is turned on and that both stage and lenses are pre-heated to 37 °C. Cover the hole on the stage with a glass coverslip (thickness #1 or 1.5) that is aligned with the heated lens.
2. Move the animal onto the stage carefully. If imaging is planned for a long duration (>1 h), apply ophthalmic ointment and secure the winged infusion-based perfusion system as described above.
3. Place a drop of saline on the center of the coverslip and place the neutrophils-injected ear on top of it. Gently flatten the ear using a sterile cotton swab across the center of the coverslip to remove air pockets.
   NOTE: Application of saline (or PBS) between the ear and coverslip reduces light refraction due to air pockets, thereby ensuring even refractive index and superior imaging quality.
4. Secure the ear by gently pressing a wooden stick (removed from the cotton swab) to the side of the ear closer to the animal's head and locking it using tape (Figure 1A).
   NOTE: A secured wooden stick minimizes motion artifacts due to heartbeat and respiration and dramatically increases the imaging quality. Importantly, the wooden stick must be fastened just enough to stabilize the ear without impairing the blood flow.
5. Using the microscope eyepiece, find an area of interest, and then switch to multiphoton acquisition mode.
6. Set the laser excitation wavelength to 900-930 nm to allow simultaneous acquisition of GFP/green fluorescent dye, mTomato as well as collagen-I (via SHG).
7. Set appropriate set of mirrors and filter combination on the detectors to collect the emitted light (bandpass filters: Blue = 410-460 nm, Green = 495-540 nm, Red = 580-640 nm).
   NOTE: Injected neutrophils, when labeled with a green-fluorescent dye, exhibit a strong green fluorescence making them visible under the microscope eyepiece even if injected into deeper portions of the ear.
8. Determine the setting most appropriate to the hardware configuration. Even if migrating cells can be tracked with a 30 s to 1 min interval, image subcellular events with a higher speed (at least <10 s interval). In the example below, two different imaging setups are presented to show the difference between classical IVM and ISMic.
   1. In the classical IVM approach to track cell migration and investigate the host mouse tissue (Figure 1), use a 30x lens and a galvo scanner with an image size of 512 x 512 pixels (pixel size of 0.83 µm), and set z-axis displacement using the motorized stage with a step size of 2 µm, allowing imaging of a 30 µm deep volume every 30 s.
   2. In the ISMic approach to image highly dynamic membrane remodeling at a higher magnification and resolution (Figure 2), use a 40x lens and a resonant scanner with 3x averaging and image size of 512 x 512 pixels (pixel size of 0.25 µm), and set z-axis displacement using a piezo with a step size of 1 µm, allowing imaging of 20 µm deep volume every 4-5 s.
9. Trigger neutrophil migration by inducing a sterile laser injury. Focus the excitation laser at a high power sufficient to reach at least 80 mW²², on a narrow area of the tissue (20 µm x 20 µm) for 10 s.
10. Identify the laser-induced injury by its strong auto-fluorescent signals appearing in all the channels and by the resulting alteration of collagen arrangement (Figure 1D).
    NOTE: To obtain reliable results, the injury must always be induced as far as possible from the blood vessels; if ruptured, blood cells released into the tissue will impact imaging quality as well as result in a strong overall immune reaction.
11. Save the data at the end of the experimentation for further analyses.
12. Euthanize the animal according to the local institutional guidelines. If needed, tissue can be collected for fixation and further processing.

7. Representative data analysis

NOTE: Data can be visualized and analyzed with either the microscope software, third-party software, or customized programs. The procedure and the tools utilized depend on the specific needs of the investigators. Shown here is one example of a workflow to quantify membrane curvature and local area changes at the leading edge and the rear of the migrating neutrophils.

1. Open the imaging files with Imaris or Fiji to visualize the migration of the neutrophils in 4D and determine the quality of the experiment.
   NOTE: Imaris and Fiji can read the imaging format of several microscope brands.
2. Process the data with the following steps by running customized MATLAB code scripts. Read the selected imaging files with the Bio-Formats²⁸ package for MATLAB. Segment each frame of the 3D volume with the Otsu algorithm²⁹ to identify the individual cells.
3. Track the identified objects based on the minimum displacement criterion³⁰.
4. Determine object's active contours and boundaries.
   1. Generate a max-projection for each object identified in a time frame.
   2. Apply the bwboundaries function in MATLAB to the max-projection of each identified object from step 7.2 to generate 100 boundry points.
   3. Optimize the boundary points identified with the dynamic contour algorithm to achieve subpixel precision³¹ (Figure 3A) using the functions in the MATLAB package downloaded via http://www.iacl.ece.jhu.edu/static/gvf/.
   4. Overlay the boundaries for the object and the object trajectory index determined by the tracking in step 7.3 with the max-projection of the original image to generate an output of a 2D overlayed movie for visualization and manual check.
5. In the overlayed movie, manually select all the objects identified as migrating cells by excluding non-cell objects, touching objects, and objects with incomplete boundaries. Record the tracking indices, start, and end frames of each cell trajectories in a spread sheet to be used as the input for the following steps implemented by MATLAB code scripts.
6. Track the cell boundary points from each frame to the next one according to the minimum displacement criterion. Use the Polyarea function in MATLAB to calculate the area underlined by adjacent boundary points in two consecutive time frames (Figure 3C). If the local cell boundary moves outward from the cell, assign a positive sign to the area change. If the local cell boundary moves inward to the cell, assign a negative sign to the area change.
7. Measure the local membrane curvature at each boundary point for each frame by fitting its neighbor boundary points to a circle (5 per side; total 11)³² (Figure 3B).
8. Generate kymographs using the imagesc function in MATLAB (Figures 3D,E) to show the changes of the subcellular features over time and relate to the position of the cell.