Automatically Generated

Summary

Here we outline and demonstrate a protocol for primary mouse liver sinusoidal endothelial cell (LSEC) isolation. The protocol is based on liver collagenase perfusion, nonparenchymal cell purification by low-speed centrifugation, and CD146 magnetic bead selection. We also phenotype and characterize these isolated LSECs using flow cytometry and scanning electron microscopy.

Abstract

Liver sinusoidal endothelial cells (LSECs) are specialized endothelial cells located at the interface between the circulation and the liver parenchyma. LSECs have a distinct morphology characterized by the presence of fenestrae and the absence of basement membrane. LSECs play essential roles in many pathological disorders in the liver, including metabolic dysregulation, inflammation, fibrosis, angiogenesis, and carcinogenesis. However, little has been published about the isolation and characterization of the LSECs. Here, this protocol discusses the isolation of LSEC from both healthy and nonalcoholic fatty liver disease (NAFLD) mice. The protocol is based on collagenase perfusion of the mouse liver and magnetic beads positive selection of nonparenchymal cells to purify LSECs. This study characterizes LSECs using specific markers by flow cytometry and identifies the characteristic phenotypic features by scanning electron microscopy. LSECs isolated following this protocol can be used for functional studies, including adhesion and permeability assays, as well as downstream studies for a particular pathway of interest. In addition, these LSECs can be pooled or used individually, allowing multi-omics data generation including RNA-seq bulk or single cell, proteomic or phospho-proteomics, and Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq), among others. This protocol will be useful for investigators studying LSECs' communication with other liver cells in health and disease and allow an in-depth understanding of the role of LSECs in the pathogenic mechanisms of acute and chronic liver injury.

Introduction

Liver sinusoidal endothelial cells (LSECs) line the hepatic sinusoid walls and are the most abundant nonparenchymal cells in the liver¹. LSECs are distinguished from other capillary endothelial cells elsewhere in the body by the presence of fenestrae and the lack of a classical basement membrane or a diaphragm^2,3. Hence, the LSECs possess distinctive phenotypic and structural characteristics that enhance their permeability and endocytic capacity to eliminate a variety of circulating macromolecules, including lipids and lipoproteins. LSECs play a pivotal role in the crosstalk between parenchymal and nonparenchymal cells, such as stellate cells and immune cells. LSECs are key in maintaining liver homeostasis by keeping the stellate cells and Kupffer cells in a quiescent status⁴. LSECs modulate the composition of hepatic immune cells populations by mediating adhesion and trans-endothelial migration of circulating leukocytes^5,6. During acute and chronic liver injury⁷, including ischemia-reperfusion injury (IRI)⁸, nonalcoholic steatohepatitis (NASH)⁹, and hepatocellular carcinoma (HCC), LSECs undergo phenotypic changes known as capillarization characterized by defenestration and formation of basement membrane¹⁰. These phenotypic changes in LSECs are associated with LSECs dysfunction and the acquisition of prothrombotic, proinflammatory, and profibrogenic properties.

Several methods for the isolation of LSECs from mouse liver have been developed¹¹. Some techniques depend on separating nonparenchymal and parenchymal cells followed by density gradient centrifugation to purify the LSECs from nonparenchymal fractions. The limitation of this method is the presence of contaminating macrophages in the final steps of LSECs isolation, which could affect the purity of the isolated LSECs¹². This protocol is based on collagenase perfusion of the mouse liver and CD146⁺ magnetic beads positive selection of nonparenchymal cells to purify LSECs. LSECs isolated using this method show high purity and preserved morphology and viability. These LSECs are optimum for functional studies, including permeability and adhesion assay, as well as downstream studies for pathways of interest. Moreover, with the growing interest in generating big datasets in both clinical research and discovery science, these high-quality LSECs isolated from both healthy and diseased livers with nonalcoholic steatohepatitis (NASH) or other conditions can be pooled or used individually, allowing multi-omics data generation and comparison between health and disease^13,14. In addition, the isolated LSECs can be employed to develop two-dimensional as well as three-dimensional in vitro models like organoids to decipher the activated signaling pathway in LSECs and their intercellular communication with other liver cells under different noxious stimuli and in response to various therapeutic interventions.

Protocol

Animal protocols were conducted as approved by the Institutional Animal Care and Use Committee (IACUC) of Mayo Clinic. Eight-week-old C57BL/6J male mice were purchased from Jackson Laboratory. Mice were housed in a temperature-controlled 12:12-h light-dark cycle facility with free access to diet.

1. Preparation of collagen-coated culture dish or plate

1. To make 50 mL of 0.02 mol/L acetic acid, add 0.6 mL of glacial acetic acid to 49.4 mL of H₂O.
2. Make 50 µg/mL collagen type I in 0.02 mol/L acetic acid. Dilution depends on the concentration of the lot.
3. Coat 10 cm culture dishes with 3 mL of collagen solution. Incubate at room temperature (RT) for 1 h.
   NOTE: When using a different dish or plate for culture, the volume of coating solution needs to be adjusted based on culture area, generally use 6-10 µg/cm².
4. Remove excess fluid from the coated surface, and wash with Phosphate-buffered saline (PBS) 3 times. Allow the dishes to air dry.
5. If the collagen solution is not sterile, sterilize the collagen-coated dishes by exposure to ultraviolet (UV) light for 10 min in a sterile tissue culture hood.

2. Equipment setup

1. Set up the heated and humidified recirculating perfusion apparatus as shown in Figure 1B.
2. Rinse the perfusion system using 10% bleach for 5 min followed by sterile water for another 10 min.
3. Drain the rinsing liquid as much as possible before perfusing the collagenase solution.
4. Infuse collagenase solution through the perfusion system (as shown in Figure 1B) to prewarm it at 37 °C. Set the pump rate at speed 1 as shown on the speed control dial (equal to 40 mL/min).
5. Keep this speed the same throughout the whole procedure. Reuse the collagenase solution run in a closed circuit during the day of the LSEC isolation.
   ​NOTE: The pump speed is fixed at 1 to avoid mechanical pressure that could perturb LSECs biological condition.

3. Surgical procedure

1. Weigh the mouse.
2. Inject 90 mg/kg body weight of ketamine and 10 mg/kg bodyweight of xylazine intraperitoneally (IP).
3. Once the mouse is nonreactive to painful stimuli, secure the mouse to the surgical surface, as shown in Figure 1B.
4. Spray the mouse abdomen with 70% ethanol. Using surgical scissors, make an incision of ~5 cm long starting from the lower part of the abdomen up to the xiphoid process.
5. Next, make two lateral cuts using small iris scissors on each side of the abdomen to fully expose the abdominal organs.
6. Place the sheath of the intravenous (IV) catheter under the animal's back to lift and level the abdomen.
7. Using a regular curved dressing forceps, gently pull the intestines and the stomach off to the left of the animal.
8. Place a 5-0 surgical suture under the inferior vena cava (IVC) just below the exposed left kidney. Tie a loose hitch in the suture.
9. Place another 5-0 surgical suture around the hepatic portal vein (PV) just above the splenic vein branching point off the hepatic PV. Tie a loose hitch in the suture.
10. Using the PV suture as tension, insert the 20 G IV catheter in the hepatic PV 1 cm below where it branches to the right and left hepatic PV.
11. Slide the catheter up the vein but keep it below the branching area. Allow the blood to travel down the catheter until it begins to drip out.
12. With some of the Buffer A (Table 1) in an intravenous (IV) bottle positioned above the surgical area, use an IV line, and attach it to the catheter. Flush the liver with this solution while avoiding air entry into the system.
13. Tie off the suture around the inferior vena cava below the kidney. This will allow the liver to retro perfuse.
14. Cut the IVC below the suture to allow the animal to bleed out.
    NOTE: This step must be performed quickly to avoid congestion of the liver.
15. Secure the IV catheter with the PV suture.
16. Once perfusing, cut away the stomach, intestines, spleen, and other entrails attached to the liver.
17. Cut away the diaphragm and the major vessels from the thoracic cavity. Remove the liver from the animal and place it on the perfusion tray.
18. Carefully remove the IV line and hook up the collagenase solution in the recirculating chamber.
    NOTE: Steps 3.16-3.18 need to be done within 5 min, so the liver will not be perfused too long with Buffer A. Be very careful not to allow any air bubbles into the liver.
19. Allow the liver to perfuse until the capsule becomes mottled and appears mushy (10-15 min or more, period varies depending on the different lots of collagenase).
20. While liver is in perfusion, make sure the animal is deceased before disposing it in a necropsy bag.
21. Once digested, remove the liver from the chamber and place it in a 10 cm Petri dish with about 20 mL of serum-free Dulbecco's Modified Eagle Medium (DMEM).
22. Gently pick apart the liver with a couple of pipette tips and discard the biliary tree.
23. Filter the liver suspension through a 70 µm cell strainer into a 50 mL conical tube.
24. Centrifuge the cell suspension at 50 x g for 2 min at RT. Collect the supernatant containing nonparenchymal hepatic cells.
    ​NOTE: Hepatocytes are precipitated in the pellet, which can be further purified using gradient centrifugation as previously described¹⁵.

4. Separation of nonparenchymal hepatic cells and LSECs purification

NOTE: Purify the CD146⁺ LSECs using the immunomagnetic beads, following the manufacture's instructions.

1. Centrifuge the supernatant at 300 x g for 5 min at 4 °C.
2. Collect the cell pellet and resuspend in 1 mL of isolation buffer (Table 1), determine the cell number using an automatic cell counter following the manufacturer's instructions.
3. Centrifuge the cell suspension at 300 x g for 10 min at 4 °C, aspirate the supernatant completely.
4. Resuspend the pellet with 90 µL of isolation buffer per 10⁷ total cells.
5. Add 10 µL of CD146 magnetic beads per 10⁷ total cells. Mix well and incubate for 15 min at 4 °C.
6. To wash the cells, add 1-2 mL of isolation buffer per 10⁷ cells, centrifuge at 300 x g for 10 min, and then aspirate the supernatant completely.
7. Resuspend up to 10⁹ cells in 500 µL of isolation buffer (Table 1).
   NOTE: For higher cell numbers, scale up buffer volume accordingly.
8. Prepare the separation column, rinse it with 3 mL of isolation buffer.
9. Apply the cell suspension onto the column stacked with 70 µm pre-separation filters.
10. Wash the column with the 3 mL of isolation buffer 3 times.
    NOTE: When washing the column, the isolation buffer should be added as soon as the column reservoir is empty.
11. Remove the column from the separator and place it on a 15 mL centrifuge tube. Pipette 5 mL of isolation buffer onto the column.
12. To recover the magnetic beads labeled cells, firmly push the plunger into the column to flush out the cells.
13. Centrifuge at 300 x g for 5 min at 4 °C. LSECs are ready for microscopic examination and downstream analysis.

5. LSECs immunophenotyping and purity assessment by flow cytometry

1. Determine the cell numbers of isolated LSECs using an automatic cell counter following the manufacturer's instructions.
2. Centrifuge the cells at 300 x g for 5 min, aspirate the supernatant completely.
3. Take 1 x 10⁶ cells/tube and resuspend with 90 µL of staining buffer (Table 1).
4. Add 10 µL/tube of mouse FcR block and 1 µL of viability dye, incubate at 4 °C for 10 min.
5. Stain the cells with a combination of CD45, CD146, and stabilin-2 antibodies diluted at 1:50 with staining buffer. Incubate at 4 °C for 20 min.
6. Wash the cells with 5 mL of staining buffer and centrifuge at 300 x g for 10 min, then aspirate the supernatant completely.
7. Resuspend the cells with 300 µL of staining buffer and run through the flow cytometer.

6. LSECs culture and examination

1. Seed 1 x 10⁶ cells/well in a 6-well plate and culture it with endothelial cells growth medium consisting of 5% fetal bovine serum (FBS), 1% endothelial cells growth supplement, and 1% primocin solution.
2. Examine the isolated LSECs by light microscopy. Acquire bright-field images with a 10x magnification.

7. LSECs morphology and fenestrae examination by scanning electron microscopy

1. Pre-coat the cell culture inserts (3 µm pore size) with collagen solution.
2. Culture isolated LSECs (120,000 cells) on the insert with endothelial cells growth medium in a 24-well plate at 37 °C warm and humidified atmosphere. Allow cells to settle down and adhere for 2 h.
3. For cell fixation, add an equivalent volume of prewarmed at 37 °C Trump's fixative to the cell culture media.
4. After incubation for 10 min, replace 50% diluted Trump's fixative with undiluted Trump's fixative.
5. Fix the cells in Trump fixative for 2 h, then incubate for 1 h in 1% osmium tetroxides.
6. Proceed with samples dehydration, drying in a critical point drying device, mounting, sputter coating, and examination using a scanning electron microscope.

Representative Results

Experimental schematics and equipment set up:
In this protocol, mouse liver was digested using a closed perfusion circuit, then nonparenchymal cells and hepatocytes were separated by low-speed centrifugation at 50 x g for 2 min. Primary LSECs were isolated using CD146 magnetic beads selection from the nonparenchymal fraction. The experimental schematics are shown in Figure 1A. The cannula was placed through the PV while the inferior vena cava was tied up to ensure unidirectional perfusion of collagenase through the liver (Figure 1B). The in-house perfusion chamber is equipped with a heating and humidification system to ensure a 37-40 °C warm and humidified apparatus air. The perfused collagenase circulates through a closed system and can be reused for two or three mice during the isolation day to make the isolation more cost-effective.

Assessment of isolated LSECs purity and surface markers by flow cytometry:
The purity of the isolated LSECs was assessed by employing well-characterized specific LSEC surface markers CD146 and stabilin-2^16,17,18. Stained cells were analyzed on a flow cytometer; data were analyzed using FlowJo software.

LSEC population was gated (Figure 2A) and analyzed singlets for the following gating strategy; viable cells were defined using viability dye staining before labeling cells with specific markers (Figure 2B). Then CD45^– population was gated to exclude any immune cells contamination. The isolated LSECs reached 94.8% viability (Figure 2C). The percentage of CD45^– cells was 89.7% (Figure 2D), and 92.3% of the cells were CD146 and stabilin-2 double-positive (CD45^–CD146⁺stabilin-2⁺) LSECs (Figure 2E).

Morphology of the isolated LSECs:
The isolated LSECs were seeded in a 6-well plate at 1 x 10⁶ cells/well, cultured in a complete growth medium, and examined by light microscopy after 6 h of culture (Figure 3A). LSECs were plated on a 3 µm pore-sized culture insert to visualize the LSECs fenestrae by scanning electron microscope (SEM). After fixation and processing, the fenestrae were identified, as shown in Figure 3B. As previously reported, LSECs lose their fenestrae in culture overnight¹⁹; hence, it is recommended to process the cells as soon as possible after the isolation for downstream studies to avoid dedifferentiation in vitro.

Figure 1: Experimental schematics and equipment set up. (A) Experimental schematics were designed using Bio render. (B) Equipment setup and process. The liver was harvested and perfused in a closed collagenase perfusion apparatus. The liver was then transferred into a dish and dissociated. The liver suspension was collected. The nonparenchymal cell fraction was separated by centrifugation at 50 x g for 2 min. LSECs were purified using magnetic beads. Please click here to view a larger version of this figure.

Figure 2: Assessment of isolated LSECs purity and surface markers by flow cytometry. Data analysis and gating strategy as shown, (A) gated LSECs and (B) singlets based on forward scatter (FSC)/side scatter (SSC). Isolated LSECs were stained with a viability dye and a combination of CD45, CD146, and stabilin-2 antibodies. Singlets were gated and analyzed, (C) viable LSECs were gated on (D) CD45 population and analyzed for (E) CD146 and stabilin-2 expression. All gates were determined by fluorescence minus one (FMO). Please click here to view a larger version of this figure.

Figure 3: Morphology of isolated LSECs. (A) Bright field picture of cultured LSECs. Isolated LSECs were cultured in the complete growth medium for 6 h and examined by light microscopy (Scale bar 100 µm). (B) LSEC was examined by scanning electron microscope (SEM). White arrows indicate LSEC fenestrae (Scale bar, left panel 5 µm, right panel 1 µm). Please click here to view a larger version of this figure.

KRH stock solution (10x)
Reagents	Concentration
NaCl	1.15 M
HEPES (free acid)	0.2 M
KCl	0.05 M
KH2PO4	0.01 M
Buffer A
Reagents	Concentration
KRH solution, PH=7.4	1x
EGTA	0.5 mM
Adjust PH to 7.4, filter through a 0.2 μm filter before use.
Stored at room temperature for up to 6 months.
Buffer B
KRH solution, PH=7.4	1x
CaCl2	1 mM
Adjust PH to 7.4, filter through a 0.2 μm filter before use.
Stored at room temperature for up to 6 months.
Collagenase solution
Reagents	Concentration
Buffer B, PH=7.4	125 mL
Collagenase	35–40 mg, add BSA up to 100 mg
Percoll solution
Reagents	Concentration
Percoll	22.5 mL
PBS (10x), PH=7.4	2.5 mL
Isoaltion/Staining Buffer
Reagents	Concentration
MACS rinsing buffer	1250 mL
BSA stock	125 mL

Table 1: Buffers recipe. The table comprises the composition of the buffers and solutions used in this study.

Discussion

In the current manuscript, we describe a protocol for LSEC isolation from mouse liver consisting of two-step collagenase perfusion and subsequent magnetic-activated cell sorting (MACS). This protocol consists of the following three steps: (1) Perfusion through the PV with a calcium-free buffer followed by a collagenase-containing buffer to achieve liver cell dispersion; (2) Exclusion of hepatocytes with low-speed centrifugation; and (3) MACS-based positive selection of LSECs from nonparenchymal cells (NPCs) using anti-CD146 magnetic beads. The whole procedure could be completed within 3 h. Furthermore, the cost for this procedure, including all the consumable supplies, is around 150 USD per mouse, suggesting that this method of LSEC isolation is overall efficient and cost-effective. We use steps (1) and (2) for isolation of primary mouse hepatocytes as well, which is beyond the scope of the current manuscript.

There are a few critical steps in portal cannulation and liver perfusion: (i) Catheter positioning: the catheter tip should be placed no less than 3 mm distal to the PV bifurcation of an adult mouse, which is usually identified in the hilum of the liver. Deep placement of the catheter tip in the PV will compromise the perfusion of some lobes of the liver. Poor perfusion of a lobe manifests as an absence of a color change of the lobe and can be rectified if noticed immediately by a slight repositioning of the catheter into the distal PV. This maneuver could possibly optimize lobe perfusion. (ii) Air bubbles: the infusion route between the pump and the PV should be kept completely free from air bubbles. Minute air entry into the PV can cause an air embolism, which results in incomplete liver perfusion. To eliminate air in the catheter after removing the inner needle, Buffer A is injected with a syringe proximal to the bubble to expel any air before connecting the infusion line to the catheter. (iii) Collagenase strength: To maintain collagenase activity during the perfusion, it is essential to assure that Buffer B infused into the PV has the precise concentration of collagenase and pH and is kept at approximately 37 °C. As shown in Figure 1, a customized chamber is used to keep the whole perfusion system in a 37 °C warm and humidified atmosphere; in addition, the oxygen supply is maintained during collagenase perfusion to achieve complete digestion. An alternative option includes a water bath to pre-warm the collagenase solution²⁰, and an open collagenase perfusion system in situ where the collagenase solution drains out of the liver through the cut IVC²¹.

Although the protocol uses relatively bold catheters (20 G) to secure an adequate perfusion flow, a thinner catheter will also work well based on previous reports^22,23,24. However, it is sometimes difficult to successfully cannulate the PV especially when mice are used for specific disease models or if they have PV anomalies. If PV cannulation fails, mechanical and enzymatic liver digestion using an enzyme kit and gentle tissue dissociator is an alternative method for steps (1) and (2) to obtain nonparenchymal cell (NPC) suspension. We have confirmed that the cell yield and viability in this alternative method of dissociation without liver perfusion was comparable to the original method (data not shown). Portal vein perfusion for liver digestion has been employed by others and us^13,16,20. Furthermore, this method avoids the theoretical risk of LSECs phenotypic or functional alterations induced by mechanical force during liver digestion. Therefore, the use of the PV perfusion-based method depicted in this protocol is so far highly recommended.

Here, the optimum LSEC yield and purity when employing the current protocol are shown. We obtained approximately 2-5 x 10⁶ cells per mouse from both healthy mice and mice with diet-induced NASH using the current protocol. As for purity, the positivity of the same surface marker used for immunomagnetic separation (CD146) supports the accuracy of our isolation technique. In addition, there are several well recognized LSEC surface markers, such as lymphatic vessel endothelial hyaluronic acid receptor 1 (LYVE-1), CD32b, and stabilin-2. Controversies still exist around these markers; for example, i) LYVE-1 is present also in lymphatic endothelial cells, and ii) LSECs in a periportal area lack CD32b expression¹⁸. Thus, to examine the purity of these isolated LSECs with flow cytometry, stabilin-2 was employed as an established specific LSEC marker in addition to CD146. Using this methodology, we confirmed that the isolated LSECs had >94% viability and >90% purity (Figure 2). The LSEC-specific phenotype of the isolated cells was further confirmed using SEM, where most of the cells have fenestrae (Figure 3B), the characteristic morphological feature of LSECs.

Among the LSECs isolation methodologies, immunomagnetic beads selection is the technique used the most to isolate the LSECs from the NPC fraction^24,25,26,27. In contrast, other methods include centrifugal elutriation and selective adhesion-based separation^22,28,29. The non-immunomagnetic methods have been shown to deliver high yields of LSECs (approximately 9 x 10⁶ cells per mouse)¹¹. Hence, a relatively lower yield of isolated LSECs could be a limitation of the current protocol. On the other hand, the non-immunomagnetic selection-based methods require high technical expertise and sometimes show inconsistent results on cell yields and purities¹¹, while the immunomagnetic selection-based method holds the advantage of consistent yield and purity over other existing methods. Moreover, for immunomagnetic beads selection-based cell isolation, the specificity of cell surface markers used is always a matter of concern. For example, CD31, which has been used previously as a marker for immunomagnetic LSEC separation, is dominantly expressed on the PV endothelium and capillarized LSECs³. Although CD146 has been reported to be expressed also on natural killer cells and hepatic stellate cells^30,31,the high purity of LSECs isolated using this protocol mitigates concerns regarding the specificity of the endothelial marker CD146. Comparison of the pros and cons of various methodologies for LSEC isolation is discussed elsewhere^11,18,32.Comparison of cell yields and purities between other methods and this method is beyond the scope of the current manuscript and warrants future investigations.

In conclusion, we present a primary mouse LSEC isolation protocol based on one of the most widely used and accepted approaches. Isolated LSECs display high purity and preserved function. Moreover, this protocol is efficient and applicable to the healthy mouse liver as well as the liver from different disease mouse models. High-quality LSECs from these mice can be employed for multiomics data set generation^13,14. Thus, this approach facilitates the identification of molecular mechanisms regulating the function of LSEC and improves our understanding of their roles in intercellular communication in the liver in health and disease.

Materials

2.0-inch 20 G Intra Venous (IV) catheter	Terumo, SOmerset, NJ, USA	SR-OX2051CA
2–3-inch perfuion tray with a hole in the center			customized; made in house
405/520 viability dye	Miltenyi, Bergisch Gladbach, Germany	130-110-205
4-inch regular curved dressing forceps	Fisher Brand	FS16-100-110
5-0 Perma-Hand silk suture	Ethicon, Raritan, NJ, USA	A182H
Anti-stabilin-2 (Mouse) mAb-Alexa Fluorà 488	MBL International, Woburn, MA, USA	D317-A48
BSA stock	Miltenyi, Bergisch Gladbach, Germany	130-091-376
Anti-CD146 (LSEC)-PE, anti-mouse	Miltenyi, Bergisch Gladbach, Germany	130-118-407
CD146 (LSEC) MicroBeads, mouse	Miltenyi, Bergisch Gladbach, Germany	130-092-007
Anti-CD45-Viogreen, anti-mouse	Miltenyi, Bergisch Gladbach, Germany	130-110-803
Collagen type I	Corning, Corning, NY, USA	354236
Collagenase II	Gibco, Waltham, MA, USA	17101-015
Endothelial cells growth medium	ScienCell Research Laboratories, Carlsbad, CA, USA	211-500
FcR blocking reagent, mouse	Miltenyi, Bergisch Gladbach, Germany	130-092-575
FlowJo software, version 10.6	Becton, Dickinson and Company
Hardened Fine scissors	F.S.T, Foster city, CA, USA	14091-11
Heated (37 °C) and humidified recirculating perfusion apparatus equipped with Oxygen injection at a rate of 10psi.			customized; made in house
Hitachi S 4700 scanning electron microscope	Hitachi Inc, Pleasanton, CA, USA	SEM096
LS columns	Miltenyi, Bergisch Gladbach, Germany	130-042-401
MACS pre-separation filters (70 μm)	Miltenyi, Bergisch Gladbach, Germany	130-095-823
MACS rinsing buffer	Miltenyi, Bergisch Gladbach, Germany	130-091-222
MACS Smart Strainer (70 μm)	Miltenyi, Bergisch Gladbach, Germany	130-098-462
MACSQunt flow cytometer	Miltenyi, Bergisch Gladbach, Germany
Millicell Cell Culture Insert	Millipore Sigma, Burlington, MA, USA	PITP01250
Nexcelom cell counter	Nexcelom bioscience, Lawrence, MA, USA	Cellometer Auto T4 Plus
Percoll	GE Healthcare, Chicago, IL, USA	17-0891-01
Surgical scissors	F.S.T, Foster city, CA, USA	14001-12
Very small curved dressing forceps	F.S.T, Foster city, CA, USA	11063-07