Delivery of Antibodies into the Brain Using Focused Scanning Ultrasound

Summary

Presented here is a protocol to transiently open the blood-brain barrier (BBB) either focally or throughout a mouse brain to deliver fluorescently-labeled antibodies and activate microglia. Also presented is a method to detect the delivery of antibodies and microglia activation by histology.

Abstract

Only a small fraction of therapeutic antibodies targeting brain diseases are taken up by the brain. Focused ultrasound offers a possibility to increase uptake of antibodies and engagement through transient opening of the blood-brain barrier (BBB). In our laboratory, we are developing therapeutic approaches for neurodegenerative diseases in which an antibody in various formats is delivered across the BBB using microbubbles, concomitant with focused ultrasound application through the skull targeting multiple spots, an approach we refer to as scanning ultrasound (SUS). The mechanical effects of microbubbles and ultrasound on blood vessels increases paracellular transport across the BBB by transiently separating tight junctions and enhances vesicle- mediated transcytosis, allowing antibodies and therapeutic agents to effectively cross. Moreover, ultrasound also facilitates the uptake of antibodies from the interstitial brain into brain cells such as neurons where the antibody distributes throughout the cell body and even into neuritic processes. In our studies, fluorescently labeled antibodies are prepared, mixed with in-house prepared lipid-based microbubbles and injected into mice immediately before SUS is applied to the brain. The increased antibody concentration in the brain is then quantified. To account for alterations in normal brain homeostasis, microglial phagocytosis can be used as a cellular marker. The generated data suggest that ultrasound delivery of antibodies is an attractive approach to treat neurodegenerative diseases.

Introduction

Therapeutic ultrasound is an emerging technology aimed at treating brain diseases in a noninvasive manner, in part by facilitating access of therapeutic agents to the brain^1,2,3. As only a small fraction of therapeutic antibodies targeting brain diseases are taken up by and retained in the brain⁴, therapeutic ultrasound offers the possibility to increase their uptake and target engagement^5,6.

In our laboratory, we are developing therapeutic approaches for neurodegenerative diseases in which an antibody in various formats is delivered across the blood-brain barrier (BBB) using microbubbles. To achieve this, ultrasound is applied through the skull into the brain in multiple spots using a scanning mode we refer to as scanning ultrasound (SUS)⁷. The mechanical interaction between the ultrasound energy, the intravenously injected microbubbles and the brain vasculature transiently separates the tight junctions of the BBB in a given sonication volume, allowing antibodies and other cargoes including therapeutic agents to effectively cross this barrier^7,8,9. Moreover, ultrasound has been shown to facilitate the uptake of antibodies from the interstitial brain into brain cells, such as neurons, where the antibody distributes throughout the cell body and even into neuritic processes^5,10.

Alzheimer's disease is characterized by an amyloid-β and tau pathology¹¹, and a host of animal models is available to dissect pathogenic mechanisms and validate therapeutic strategies. A SUS approach, by which ultrasound is applied in a sequential pattern across the entire brain, when repeated over several treatment sessions, can reduce amyloid plaque pathology in the brains of amyloid-β-depositing amyloid precursor protein (APP) mutant mice and activate microglia which take up the amyloid, leading to improvement in cognitive function⁷. BBB opening with ultrasound and microbubbles also reduces tau pathology in pR5, K3 and rTg4510 tau transgenic mice^5,12,13. Importantly, whilst microglia remove extracellular protein deposits, one of the underlying clearance mechanisms for intraneuronal pathologies induced by SUS is the activation of neuronal autophagy¹².

Here, we outline an experimental process, by which fluorescently labeled antibodies are prepared, and then mixed with in-house lipid-based microbubbles, followed by retroorbital injection into anesthetized mice. Retroorbital injection is an alternative to tail vein injection which we have found to be equally efficacious and simpler to repeatedly perform. This is immediately followed by applying SUS to the brain. To determine the therapeutic antibody uptake, mice are sacrificed and the increased antibody concentration in the brain is then quantified. As a proxy of the change in brain homeostasis, microglial phagocytic activity is determined by histology and volumetric 3D reconstruction.

The generated data suggest that ultrasound delivery of antibodies is a potentially attractive approach to treat neurodegenerative diseases. The protocol can be similarly applied to other drug candidates, as well as model cargos such as fluorescently labeled dextrans of defined sizes¹⁴.

Protocol

All animal experiments were approved by the animal ethics committee of the University of Queensland.

1. In-house microbubble preparation

1. Weigh out a 9:1 molar ratio of 1,2-distearoyl-sn-glycero-3-phosphocholine and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[amino(polyethyleneglycol)-2000] (ammonium salt). 0.5 mg of lipid mixture is required per 1 mL of microbubble solution. Alternatively, lipids can be bought already in chloroform, if using pre-dissolved lipids proceed to step 1.3.
2. Dissolve the lipid in a small volume of chloroform in a glass beaker.
3. Evaporate the chloroform with an evaporator or a nitrogen stream.
4. Rehydrate the dried lipid film with 10 mL of PBS + 10% glycerol + 10% propylene glycol solution that has been filtered through 0.22 µm filter.
5. Place the rehydrated lipid solution in a water bath sonicator set to 55 ˚C (above the melting temperature of the lipids) and sonicate until fully dissolved.
6. Aliquot lipid solution into autoclaved 1.5 mL HPLC vials and screw on the septa caps.
7. Aspirate all of the air in the vial with a 5 mL syringe equipped with a 27 G needle and create a vacuum in the vial.
8. Add octofluoropropane to the vial with the included syringe, drawing up the gas from the canister. Fill the vial with 1-2 mL of octafluoropropane by reading the volume in the syringe.
9. Seal each vial with paraffin film and refrigerate.
10. On the day of the experiment, bring the vial to room temperature, add 0.5 mL of 0.9% NaCl solution to the vial, then place the vial in an amalgamator and agitate for 45 s (preset time) to produce the microbubbles.

2. Microbubble quality control using a coulter counter

1. Take the microbubble solution out of the amalgamator and vent the gas from the vial by piercing the septa with a 19 G needle.
2. Dilute the microbubble solution by performing two-step 1:5,000 serial dilutions by adding 100 µL of microbubble solution into 5 mL of filtered flow solution using a 1 ml syringe with 19 G needle, and then taking 100 µL of 1:50 diluted microbubble solution and pipetting into 10 mL filtered flow solution in a cuvette using a pipette.
3. Check that the electrolyte tank has sufficient flow solution and that the waste tank is empty.
4. Place the cuvette in the coulter counter platform and lock it into place. Use a 30 µm aperture for sample acquisition.
5. In the software, load the standard operating method (SOM), then select Edit SOM | concentration. Enter 5000x dilution.
6. In the software, choose a suitable file name. For example, microbubble_1_date.
7. Load and secure the cuvette into the platform.
8. In the software select Run| Preview and verify that the sample concentration is less than 10%. If this number is higher than 10%, perform a new dilution of the microbubbles with a higher dilution factor.
9. Select 'Start' to begin an acquisition of the sample to obtain the initial readout.
10. Rinse the aperture of the Coulter counter with filtered flow solution after each measurement. Repeat steps 2.2-2.9 to obtain 3 replicates.
11. Place a cuvette with diluted microbubble solution into a sonicator water bath and sonicate for 30 s.
12. Measure the solution with sonicated microbubbles and label as blank.
13. In the software, subtract the final readout from the initial readout. This subtracts any particles that are not microbubbles and do not contain gas.
14. Select Display Results in the software, to display the microbubble concentration, size distribution, average size, and volume concentration.

3. Fluorescent antibody labeling

1. Obtain a 1 mg/mL solution of mouse IgG in PBS without any additives.
2. Label 1 mg of mouse IgG with AlexaFluor 647 in 0.1 M sodium bicarbonate buffer by following the manufacturers' instructions located in the kit. This amount of fluorescently labeled IgG is sufficient to perform this procedure on 5-7 adult mice where a 5 mg/kg dose of antibody is administered.
3. Add the dye to the solution of IgG in 0.1 M sodium bicarbonate buffer and incubate for 15 min at room temperature.
4. Purify the fluorescently labeled antibody by pipetting the antibody solution into a spin column and centrifuging at 1,000 x g for 5 min. Free dye will remain in the column bed.
5. Use a spectrophotometer to measure the protein concentration. Measure the absorbance of the conjugate solution at 280 nm and 650 nm (A280 and A650). Calculate the concentration of protein in the sample using the equation:
   Protein concentration (M) = [A280 - (A650 x 0.03)] x dilution factor / 203,000.
6. Use a spectrophotometer to calculate the degree of labeling using the equation:
   moles dye per mole protein = A650 x dilution factor / 239 000 x protein concentration (M).
   NOTE: An acceptable degree of labeling is 3-7 moles dye per mole protein and typically obtained degree of labeling is around 6.

4. Ultrasound set-up

1. Using the focused ultrasound system, add the 5 mm spacer to the water bolus to position the ultrasound focus 9 mm below the bottom of the water bolus.
2. Fill the water bolus with approximately 300 mL of deionized water that has been degassed with an inline degasser for 20 min (oxygen content should be below 3 ppm). Place the annular array into the filled water bolus and use a dental mirror to check that there are no air bubbles on the surface. If present on the surface, remove the annular array and replace it in the water bolus.
3. Launch the application software. In the waveform menu, select Set waveform duty cycle. Settings are PRF (Hz) 10, duty cycle 10%, focus 80 mm, center frequency 1 MHz, amplitude (MPa peak negative pressure) 0.65 MPa, mechanical index = 0.65. Press Set to define the waveform and store it in memory.
   NOTE: The focused ultrasound system is pre-calibrated by the manufacturer from measurements taken by a calibrated hydrophone.
4. In the focused ultrasound system software, define a treatment plan. This requires defining a treatment region consisting of multiple individual treatments sites, and defining actions to be taken at each of those treatment sites. In this case, the treatment zone is one hemisphere of the mouse brain.
5. In the motion controller window, go to the Scan tab and Enter start, stop and increment value for motion in the x dimension, and start, stop and increment value for motion in the y direction. Enter values for X: start -4, stop 3.50 and Y: -3.00, stop 3, Increment 1.5, # loops: 1.
6. Define the actions for treatment sites. In the motion controller window, select the Event button. In the Script Editing Window, select a list of actions that will be executed in the order selected at each treatment site. Set Movement Type to raster grid at the top of the script window. In the events tab select Add Actions to move them to the script panel, and add Move Synchronously, Start trigger arb, wait, Stop trigger arb. Click on the wait action and select a wait time of 6, 000 ms.
   NOTE: These settings will make the treat a 6 x 5 grid of treatment spots spaced 1.5 mm apart, with each spot having a treatment duration of 6 s. The total duration to sonicate a mouse brain is approximately 3 min. This size treatment grid is suitable for adult C57/Bl6 mice weighing approximately 30 g. The size of the grid of treatment spots can be adjusted up or down depending on the size of the mouse.

5. Animal preparation

1. Weigh the mouse with a balance accurate to 0.1g.
2. Anesthetize mouse with 90 mg/kg ketamine and 6 mg/kg xylazine intraperitoneally. Test for absence of reflexes with a toe pinch. Alternatively, mice can be anesthetized with isoflurane using an appropriate inhalational anesthetic apparatus with an appropriate face mask. If using isoflurane, the mouse should be placed on a heat pad during the ultrasound to prevent hypothermia.
3. Use an electric razor to shave the hair from the head of the animal, then apply hair removal cream with a cotton bud, leave on for 2-3 min or until the hair is wiped away clean with a damp piece of gauze. Take care that the hair removal cream does not get in the eyes of the mouse.
4. Mark the center of the mouse’s head with a permanent marker. The transducer has a hole in its center and the transducer focus and focal spot can be aligned visually.
5. Fill a small weigh boat that has previously had the bottom cut off and replaced with plastic wrap glued to the bottom of the weigh boat with ultrasound gel. This serves as an 8 mm spacer and provides good coupling to the head of the mouse and allows visual inspection of the focus of the transducer aligned with the head of the mouse.

6. Microbubble preparation

1. Warm a microbubble vial to room temperature. To activate, add 0.5 mL of 0.9% NaCl solution to the vial and place in an amalgamator to agitate for 45 s to produce the microbubbles.
2. Vent the vial by piercing the septa with a 27 G needle.

7. Ultrasound treatment

1. Invert the vial of microbubbles and gently draw up 1 µL/g bodyweight of the solution. To this add solution of fluorescently labeled antibody and mix gently in the syringe. The maximum volume injected is 150 µL.
   NOTE: In-house prepared microbubbles are approximately 60-fold less concentrated than the clinically used (e.g., Definity microbubbles). Adjust the volume or concentration such that the number of microbubbles injected are similar to those clinically used (ie. Definity 1.2 x 10⁸ microbubbles/kg body weight).
2. Inject the microbubble and antibody solution retroorbitally, taking care to inject gently and slowly. Then, apply ophthalmic ointment to the mouse's eyes.
3. Place the mouse in the head holder (see Table of Materials) and fix the nose of the mouse. Then place the ultrasound gel-filled small weigh boat on top of the head.
4. Lower the water bolus until it sits on top of the ultrasound gel in the weigh boat.
5. Use the joystick to move the transducer focus to the center of the head. Select Reset Origin in the motion tab.
6. Select complete scan. Steps 7.3-7.6 will take 2 min.
7. For consistency, set a timer to ensure a 2 min delay between injecting microbubbles and selecting complete scan.
8. After the treatment is complete, apply ophthalmic ointment to the eyes and place mouse in a warmed recovery chamber. If hypothermia is observed during the procedure, a warming pad can be placed under the mouse to provide supplemental heat during the procedure.

8. Tissue harvesting and processing

1. At the timepoint of interest after ultrasound delivery (at least 1 h to detect high levels of antibody in the brain) deeply anaesthetise the mouse with an overdose of Pentobarbitone solution (100 mg/kg) and transcardially perfuse the mouse with 30 mL PBS. A good perfusion is required to specifically detect fluorescently-labelled antibodies that have been delivered to the brain.
2. Collect the brain and fix by immersion in 4% PFA for 24 h at 4 °C, then wash with PBS.
3. Image the brain in an infrared scanner by placing the brain on the tray and by acquiring an image in the 700 nm channel.
   NOTE: After fixation, the brain is removed from the PFA, washed in PBS and sectioned. Alternatively, it can be placed in ethylene glycol cryoprotectant solution for 24 h at 4 °C or until submerged, and then moved to a new cryoprotectant containing vial and placed at -20 °C for long term storage.
4. Section the brain cold in PBS using a vibratome. Glue the brain to the platform and cut 30-40 µm sections and collect into PBS.
   NOTE: Lysosomal autofluorescence (prevalent in sections from animals older than about 12 months) should be bleached by overnight illumination of the sections in a light chamber box, at room temperature and in PBS containing azide (0.02%) to block bacterial growth.

9. Tissue staining and image acquisition

1. Transfer sections to blocking solution (5% BSA in 0.2% Triton/PBS) for 2 h at room temperature, then wash the sections 3x by replacing the solution with 0.2% Triton/PBS.
2. Incubate sections at 4 °C overnight with the primary antibodies against Iba1 (dilution 1:1,000) and CD68 (dilution 1:500), in 0.2% Triton/PBS, followed by 3x washes with 0.2% Triton/PBS.
3. Incubate sections for 2 h at room temperature with fluorescent secondary antibodies (dilution 1:500), followed by 3x washes with 0.2% Triton/PBS only.
   NOTE: The nuclei can also be stained using DAPI solution (0.5 µg/mL).
4. Transfer the sections to a slide and coverslip with hard-set mounting media. Allow sufficient time for the mounting medium to solidify before visualization and image acquisition (overnight).
5. With a confocal microscope obtain z-stack images (at least 10 µm depth and 0.3 µm step size, 10 images per animal) of the ultrasound targeted brain area by using a confocal microscope and an objective of at least 40x magnification.
   NOTE: Be careful to acquire images within the signal dynamic range, avoiding under-/over-exposure. Save the image files in the corresponding software format used by the microscope.

10. Image analysis

1. Convert the z-stack microscopy files using the image analysis software file importer.
2. Open the converted files into the software and adjust the intensity of the Iba1 channel to observe microglial cells.
3. Crop a single microglial cell by drawing a box around it, select ‘crop’ and save the new file.
4. Build the 3D surface rendering of the IBA1 signal by:
   1. Open the single cell Imaris file generated in the previous step.
   2. Add a surface to the file.
   3. Select the channel corresponding to the Iba1 staining.
   4. Apply a threshold that should overlap the Iba1 staining
   5. Select only the structure of interest that is forming the microglial cell of interest
5. Finalize the process
   1. Obtain and record the volume of the Iba1 staining
   2. Build the 3D surface rendering of the CD68 staining
   3. Inside the IBA1 surface subfile, mask the CD68 channel and remove the voxels outside of the IBA1 staining
   4. Add a new surface to the file
   5. Select the channel corresponding to the CD68 staining
   6. Apply a threshold that should overlap the CD68 staining and is as closely as possible matching the staining volumes
   7. Finalize the process, as only the intra-microglial CD68 structures will be present in this file and, thus, they do not require another threshold filtering step
   8. Obtain and record the number and average volume of the CD68 positive structures
   9. Calculate the relative volume of CD68 structures per its corresponding IBA1 structures as a measure of microglial activation in a phagocytic state.

Representative Results

Using this protocol fluorescently-labeled antibodies are delivered to the brain and can be detected, along with microglia activation. The conclusion that can be drawn is the use of focused ultrasound and microbubbles markedly enhances brain uptake of antibodies and can deliver antibodies to the whole brain or hemisphere of a mouse when used in a scanning mode. Figure 1 shows the TIPS ultrasound application device (different components labeled) that is used to open the BBB. Figure 2 shows the representative results from Coulter counter measurements of size and concentration which should be obtained when the microbubbles are produced correctly. To easily visualize the delivery, the antibodies were labeled with a far-red fluorescent dye. Antibody uptake by the brain can be easily visualized in whole brain or sections using an infrared scanner or using fluorescent microscopy on brain sections. Brain sections show the location of the fluorescently labeled antibody at a microscopic level. Representative results for scanning ultrasound delivery to the hippocampus of fluorescently labeled anti-tau antibody RN2N is shown in Figure 3. To observe any alteration of normal brain homeostasis as a consequence of SUS and antibody delivery, one read-out was the microglial lysosomal content in relation to phagocytosis. Figure 4 shows representative staining for microglia using Iba1 and the microglial lysosome specific marker CD68 to determine whether microglia become more phagocytic following the delivery of the antibody.

Figure 1: The focused ultrasound system used for delivering ultrasound with critical components being labeled. (A) The home-made gel holder serving as an 8 mm spacer. (B) View through the transducer showing how the target is aligned with the focus visually (C). Please click here to view a larger version of this figure.

Figure 2: Quality control measurements of in-house prepared microbubbles. (A) Coulter counter equipment used to obtain summary statistics (B) and size distribution of microbubbles in number (C) and volume distribution (D). Please click here to view a larger version of this figure.

Figure 3: Delivery of fluorescently-labeled antibody into the brain using SUS. (A) A fluorescently labeled antibody fragment specific for a tau isoform delivered on its own, SUS on its own, and a combination treatment revealed increased uptake of the antibody by a sonicated hemisphere when combined with SUS using near infrared fluorescent imaging of one hemisphere of the brain. A look-up-table (LUT) was applied, with higher fluorescence intensity in arbitrary units displayed in warmer colors. (B) Quantification of fluorescence was done without subtracting the SUS-only control levels of background fluorescence. Mean ± SEM shown. (C) Increased uptake of the fluorescently labeled antibody by hippocampal neurons shown in low and high magnification images of brain sections. In the combination treatment, the antibody distributes into cell bodies and even dendrites as shown for hippocampal neurons. Blue=DAPI, magenta=antibody fragment. Scale bar: 50 µm. Please click here to view a larger version of this figure.

Figure 4: Representative immunofluorescence labeling for microglia, showing the expected morphology of microglia and rendered in 3D with the image analysis software. Microglia morphology is observed in green and levels and distribution of CD68 is observed in red. Scale bar 10 µm. Green=Iba1, red=CD68. Please click here to view a larger version of this figure.

Discussion

Fluorescently-labeled antibodies can be delivered to the brain using focused ultrasound together with microbubbles applied in a scanning mode. Antibody delivery, microglial morphology and lysosomal enlargement can be detected by fluorescence microscopy following scanning ultrasound. Microglia can take up into their lysosomes antibodies and antigens that the antibodies have bound to in an Fc-receptor-mediated process⁴.

There are a number of critical steps to achieve repeatable BBB opening and antibody delivery using this method. It is critical to ensure good coupling between the ultrasound transducer and the head of the mouse. Remove all of the hair on the head of the mouse and ensure that there are no air bubbles in the coupling. The characteristics of the in-house prepared microbubbles are critical to success. Microbubbles must have a sufficient concentration of 10⁸ microbubbles/mL and a size distribution such that over 90% of the microbubbles are below 10 µm in diameter. This is because larger microbubbles are known to be filtered out of the circulation by the lungs. An acceptable median size is between 1 -3 µm. Microbubbles must be handled and injected gently to avoid destroying them in the syringe. It is important that the ultrasound be applied no later than 2 min following the injection of the microbubbles which can be achieved with practice. Targeting of a whole brain or entire hemisphere is easily achieved with the SUS approach and accuracy of targeting is unlikely to be a problem when targeting large regions. A smaller brain region such as the hippocampus or striatum can also be successfully targeted but in this case it is important that the focus overlaps the targeted region. The height of a mouse brain is similar to the axial length of the focused ultrasound beam at 1 MHz using a typical ultrasound transducer, so that the transducer needs only be moved in the x and y dimension and not the z dimension. This can be determined by the knowledge of stereotaxic coordinates for a particular brain structure and by viewing of the lambdoid and sagittal sutures through the depilated skin of the mouse.

Here we demonstrate a technique that uses retroorbital injections to deliver microbubbles and antibody. An alternative to retroorbital injections is tail vein injections which is also an effective technique to deliver microbubbles and antibodies. The advantages of retroorbital injection is it is technically less challenging than tail vein injections, and can be repeated multiple times (alternating eye of injection) with very minimal risk of tissue damage.

If no fluorescently labeled antibody is detected in the brain it is probable that the BBB did not open. Troubleshooting should focus on obtaining a concentrated microbubble solution and injecting it so as not to destroy the microbubbles and delivering the ultrasound within two minutes of the injection time. If no BBB opening occurs, the peak negative pressure setting can be increased, with the caveat that higher peak negative pressures increase the chance of causing microhemorrhages which we do not detect at a peak negative pressure setting of 0.65 MPa using the settings described. Depending on the antigen specificity of the antibody the staining pattern will be different. The staining pattern obtained when injecting an anti-tau antibody is shown in Figure 2.

This technique can be applied to a range of antibodies and as long as consistent BBB opening is obtained, binding of the antibody to a target in the brain can be assessed. The scanning ultrasound approach achieves opening of the BBB across an entire mouse brain in a reproducible manner.

A limitation of this technique is that the occurrence and extent of the BBB opening is not observed while the mouse is alive. This limitation could be overcome by including MRI imaging with gadolinium contrast agent to the procedure, but this significantly increases the time and cost of the procedure.

Described here is a single sonication and single administration of antibody protocol which can be used to determine how much increased uptake of antibodies can be achieved, as well as where they are located in the brain after delivery. The protocol can also be used in a longitudinal study to assess therapeutic effects of antibody delivery. In a treatment study the protocol can be repeated with an inter-treatment interval of one week or longer in order to evaluate the therapeutic potential of an antibody delivered to the brain. The therapeutic potential of the antibody delivered by ultrasound can be assessed in transgenic mouse models of neurodegenerative diseases. Read-outs of therapeutic effect could include behavioral tests, and histology and biochemistry for the levels of pathological proteins for example tau, amyloid-β or synuclein.

In conclusion, we have outlined a method to open the blood brain barrier in mice to deliver fluorescently-labeled antibodies. This method will be of interest to researchers evaluating therapeutic approaches for neurodegenerative diseases.

Materials

Name	Company	Catalog Number	Comments
1,2-distearoyl-sn-glycero-3-phosphocholine	Avanti	850365C
1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[amino(polyethyleneglycol)-2000]	Avanti	880128C
AlexaFluor 647 antibody labeling kit	Thermo Fisher	A20186
CD68 antibody	AbD Serotec	MCA1957GA	Use 1:1000 dilution
Chloroform	Sigma-Aldrich	372978
Coulter Counter (Multisizer 4e)
Glycerol	Sigma-Aldrich	G5516
Goat anti-rabbit IgG, Alexa Fluor 488	Thermo FIsher	A-11008	Use 1:500 dilution
Goat anti-rabbit IgG, Alexa Fluor 488	Thermo Fisher	A-11077	Use 1:500 dilution
head holder (model SG-4N, Narishige Japan)
Iba1 antibody	Wako	019-19741	Use 1:1000 dilution
Image analysis software	Beckman Coulter	#8547008
Isoflow flow solution	Beckman Coulter	B43905
Near infrared imaging system Odyssey Fc	Licor	2800-03
Octafluoropropane	Arcadophta	0229NC
Propylene Glycol	Sigma-Aldrich	P4347
TIPS (Therapy Imaging Probe System)	Philips Research	TIPS_007
	Bitplane