Isolation of Pericytes from Mouse Cortical Tissue Using FACS for Single-cell Sequencing

Summary

To obtain mouse brain cortical pericytes meeting the demands of scRNA-seq, we present a critically optimized FACS protocol based on CD13⁺/CD31⁻ selection. Our key refinements significantly enhance cell viability and RNA integrity, enabling reliable transcriptomic profiling.

Abstract

Pericytes, as a core component of the neurovascular unit, play a vital role in maintaining the integrity of the blood-brain barrier and regulating cerebral hemodynamic homeostasis through modulation of microvascular tone. Dysfunction of pericytes has been implicated in a variety of neurological disorders. To investigate the heterogeneity of pericytes and elucidate their mechanisms in neurovascular diseases at single-cell resolution, we established an optimized and reproducible protocol for isolating mouse brain pericytes suitable for single-cell RNA sequencing (scRNA-seq). This method builds upon the established CD13⁺/CD31^- sorting strategy, but incorporates key enhancements tailored specifically for scRNA-seq applications. Briefly, brains were harvested from three adult C57BL/6 mice. The brain tissues were mechanically dissociated and subjected to enzymatic digestion to obtain single-cell suspensions. After sequential centrifugation, the final cell pellet was resuspended in 20% bovine serum albumin (BSA) solution, a pivotal step we introduced to minimize stress and aggregation, thereby maximizing viability and RNA quality for sequencing. This optimized FACS-based approach thus enables the robust isolation of viable mouse brain pericytes meeting the stringent requirements for scRNA-seq, offering a powerful tool for dissecting pericyte heterogeneity and their pathological roles in neurovascular diseases.

Introduction

Pericytes are an important type of cell located along microvessels, typically embedded within the capillary basement membrane and enveloping endothelial cells¹. In the brain, pericytes, together with endothelial cells, astrocytes, and other cell types, form the neurovascular unit and are integral to the structure and function of the blood-brain barrier (BBB). Pericytes play essential roles in the regulation of cerebral blood flow, vascular remodeling, and maintenance of vascular homeostasis. They maintain the integrity of the BBB through close physical interactions and paracrine signaling with neighboring cells²; dynamically regulate transendothelial transport via endocytosis and the expression of specific transporter proteins; and contribute to cerebral hemodynamic stability through their contractile and relaxant activities³. Dysfunction of pericytes, as well as alterations in their coverage or density, can lead to vascular dysregulation and neurological deficits, ultimately resulting in pathological changes associated with various neurological disorders⁴. However, the inherent heterogeneity of pericytes poses significant challenges for further mechanistic studies.

With the advancement of single-cell RNA sequencing (scRNA-seq) technology, an increasing number of studies have utilized scRNA-seq to explore cellular heterogeneity and identify key targets in the pathogenesis and progression of diseases. Commonly used pericyte markers include platelet-derived growth factor receptor-beta (PDGFR-β)⁵, neuron-glial antigen 2 (NG2), and alpha-smooth muscle actin (α-SMA). However, these markers have relatively poor specificity⁶, as they can simultaneously label other vascular cells, such as smooth muscle cells, and neural precursor cells like oligodendrocyte progenitor cells, making it difficult to distinguish pericytes from other brain cells. In search of a more robust alternative, we turned to CD13 (Aminopeptidase N, APN), a marker widely recognized for its strong and consistent expression in cerebrovascular pericytes in previous studies^7,8,9,10. In our study, we conducted immunofluorescence staining on brain tissue sections, and the results confirmed CD13 expression in pericytes, demonstrating its effectiveness in distinguishing pericytes from endothelial cells, as the CD13 signal was exclusively associated with perivascular cells and absent from CD31⁺ endothelial cells. (Figure 1A). However, it is important to note that CD13 expression is not entirely exclusive to pericytes; detectable levels are also observed in smooth muscle cells of arterioles and venules (Figure 1B,C).

Figure 1: Immunofluorescence staining shows CD13 expression in brain vascular pericytes and smooth muscle cells, with no detectable signal in endothelial cells. (A) Representative immunofluorescence imaging of the arteriole-capillary transition zone in the brain microvasculature. The merged image shows CD13⁺ pericyte (green) ensheathing CD31⁺ endothelial cells (red). α-SMA⁺ smooth muscle cells (cyan) are localized to the arteriolar region and are absent from capillaries. (B) Cross-section of an arteriole. The images demonstrate clear co-localization of CD13⁺ pericyte (green) and α-SMA⁺ smooth muscle cells (cyan) in the surrounding smooth muscle layer, confirming CD13 expression in arteriolar smooth muscle cells. (C) Cross-section of a venule. Similarly, CD13 (green) is detectable in the α-SMA smooth muscle cells (arrowheads) of the venular wall. Scale bar: 20 µm (A-C). Please click here to view a larger version of this figure.

To obtain mouse cortical vascular pericytes that meet the requirements for scRNA-seq, we performed fluorescence-activated cell sorting (FACS) using CD13⁺/CD31⁻ antibody labeling for pericyte isolation¹¹. Initially, CD45 and CD41 markers were used to exclude hematopoietic cells from brain tissue suspensions¹². Subsequently, pericytes were enriched by selecting for CD13⁺ cells, and endothelial cells were further excluded by gating for CD31^- populations, thereby isolating the target pericyte population¹³.

We acknowledge that the CD13-positive selection step concomitantly captures CD13-expressing smooth muscle cells from both arterioles and venules. To fulfill the critical requirements for scRNA-seq and maximize pericyte retention, we strategically selected this marker combination to prioritize pericyte yield at the sorting stage. Although this approach leads to the co-isolation of a minor population of smooth muscle cells, it effectively maximizes the recovery of all potential pericytes during the preliminary enrichment phase. This stepwise gating strategy significantly minimizes initial cell loss. In addition, to maximize pericyte viability and yield, we also employed a 20% BSA solution for post-dissociation cell resuspension and filtration, effectively removing debris while preserving sorted cell viability and purity¹⁴.

Using this method, we successfully distinguished brain vascular pericytes from other cell types, achieving high-purity and high-viability pericyte preparations that fulfill the stringent requirements for scRNA-seq. This technique provides a robust foundation for subsequent in-depth investigations into the roles of pericytes in various disease contexts.

Protocol

In this study, we used three adult C57BL/6J mice (3 months old, 20-25 g). All mice were housed under a 12 h light/dark cycle with strictly controlled temperature and humidity, with free access to food and water prior to experiments. All experimental procedures involving laboratory animals were approved by the Laboratory Animal Care and Ethics Committee of the Institute of Microcirculation, Chinese Academy of Medical Sciences, and Peking Union Medical College, and were conducted in strict accordance with relevant animal welfare guidelines.

1. Reagent preparation

1. 30% (wt/vol) Glucose Solution: Dissolve 15 g glucose in 50 mL of UltraPure DNase/RNase-free dH₂O. To ensure complete dissolution, first add glucose to 20 mL of UltraPure DNase/RNase-free dH₂O and mix thoroughly. Then add UltraPure DNase/RNase-free dH₂O to a final volume of 50 mL. Filter the solution and store at 4 °C for up to 6 months.
2. DNase I (10 mg/mL ): Dissolve 50 mg DNase I in 5 mL of sterile ddH₂O. The stock solution can be stored at -20 °C for up to 2 months.
   NOTE: Repeated freeze-thaw cycles should be strictly avoided to prevent enzymatic activity loss. To minimize freeze-thaw effects, aliquot 150 µL of the solution into sterile microcentrifuge tubes prior to initial freezing, and thaw individual aliquots as needed for experiments.
3. 2% (vol/vol) Fetal Bovine Serum (FBS) in Phosphate-Buffered Saline (PBS): Prepare 1× PBS by mixing 5 mL of 10× PBS with 45 mL of UltraPure DNase/RNase-free dH₂O. Add 1 mL of FBS to 49 mL of 1× PBS and mix thoroughly.
   NOTE: Prepare this solution freshly during Step 2.1.2. Allocate 15 mL per tissue sample.
4. Collagenase and dispase (CD) enzyme stock solution: Dissolve 100 mg of Dispase and 100 mg of Collagenase separately in 1 mL of UltraPure DNase/RNase-free dH₂O each. Combine solutions to obtain a 2 mL final volume (100 mg/mL enzyme concentration).
   NOTE: Aliquot 500 µL into sterile microcentrifuge tubes and store at -20 °C. Vortex before use to avoid activity loss from freeze-thaw cycles. It is important to note that the powder form can cause severe skin/eye irritation and respiratory distress. Always handle it in a chemical fume hood while wearing nitrile gloves and an N95 mask. In case of exposure, immediately irrigate the affected area with copious water and seek medical attention.
5. CD enzyme working solution: Mix 1.5 mL of CD enzyme stock solution with 43.5 mL of 2% (vol/vol) FBS in PBS (from Step 1.3) by gentle inversion.
   NOTE: Prepare immediately before tissue dissociation (Step 2.3.4). This volume is optimized for three mice; adjust proportionally as needed.
6. 22% (vol/vol) Percoll solution: Combine 11 mL of Percoll, 5 mL of 10× PBS, and 34 mL of UltraPure DNase/RNase-free dH₂O to prepare a 50 mL solution.
   NOTE: Filter-sterilize through a 0.22 µm PES membrane. It can be stored at 4 °C for one month without significant loss of stability.
7. Mechanical dissociation buffer: Add 1 mL of DNase I (10 mg/mL) to 10 mL of 2% (vol/vol) FBS in PBS (from Step 1.3). Mix by pipetting.
8. BSA solutions: Prepare 1%, 20%, and 35% (wt/vol) BSA solutions separately as follows:
   1. 1% (wt/vol) BSA: Dissolve 0.15 g of BSA in 15 mL of Dulbecco's Phosphate Buffered Saline (DPBS).
   2. 20% (wt/vol) BSA: Dissolve 3 g of BSA in 15 mL of DPBS.
   3. 35% (wt/vol) BSA: Dissolve 0.7 g of BSA in 2 mL of DPBS.
      NOTE: Filter all solutions through a 0.22 µm PES membrane before use.
9. Hanks' Balanced Salt Solution (HBSS)/BSA/Glucose Buffer: Add 167 µL of 30% (wt/vol) glucose (from Step 1.1) and 1.4 mL of 35% (wt/vol) BSA (from Step 1.8) to 50 mL of HBSS, mix gently.
   NOTE: Prepare this solution freshly under sterile conditions during Step 2.1.2. Store at 4 °C until use.
10. Tissue Collection Buffer: Supplement 12 mL of HBSS/BSA/Glucose Buffer (from Step 1.9) with 800 µL of DNase I (10 mg/mL). Use for temporary tissue storage and pipette rinsing.
    NOTE: Prepare this solution freshly under sterile conditions during Step 2.1.2.
11. 1.25% (w/v) Tribromoethanol Solution: Dissolve 1.25 g of 2,2,2-Tribromoethanol in 2.5 ml of 2-Methyl-2-butanol , then add this mixture to 97.5 ml of sterile ddH₂O or physiological saline to obtain a clear 1.25% Tribromoethanol Solution.
    NOTE: Prepare the solution fresh before each use. Store it protected from light at room temperature and use it promptly.

2. Experimental Procedure (using three mice as an example)

1. Pre-experiment preparation (Timing: 30 min)
   1. Autoclave required consumables and disinfect surgical instruments with 75% (vol/vol) ethanol.
   2. Prepare working solutions: 1× PBS (from Step 1.3), 2% (vol/vol) FBS in PBS (from Step 1.3), HBSS/BSA/glucose solution (from Step 1.9), and tissue collection buffer (from Step 1.10).
   3. Pour 10 mL of 1× PBS into a 10 cm cell cluture dish and keep it on ice.
   4. Prepare three ampoules, add 3 mL of tissue collection buffer to each, and keep them on ice.
   5. Prepare three 15 mL centrifuge tubes, add 10 mL of 2% (vol/vol) FBS in PBS to each, and keep them on ice.
2. Mouse brain harvesting (Timing: 30 min)
   1. Mice were deeply anesthetized by intraperitoneal injection of tribromoethanol (from Step 1.11) at a dose of 0.2 mL per 10 g body weight. After the loss of pedal reflexes, euthanasia was performed by cervical dislocation.
      NOTE: If the mouse did not reach a surgical plane of anesthesia (assessed by the absence of a pedal withdrawal reflex) within 5-10 minutes after injection, a supplemental intraperitoneal dose of tribromoethanol, equal to one-third of the initial dose, was administered.
   2. Spray the head/neck region with 75% ethanol. Make a midline scalp incision from occipital to frontal regions to expose the skull. Perform a transverse incision anterior to the bilateral orbits, followed by lateral incisions along the inferior margins of the cerebellum to detach the skull base. Finally, incise the calvarium longitudinally along the sagittal suture from posterior to anterior. Lift the cranial flap to expose the dura-enclosed brain. Gently elevate the brain while severing the optic nerves and the medulla oblongata-spinal cord connection with surgical scissors¹⁵. Place the intact brain in a 10 cm dish containing 10 mL of ice-cold 1× PBS.
      NOTE: Handle tissue gently to avoid damage.
   3. Transfer the brain to aluminum foil. Using forceps, lift the dura mater from the margins and excise it along the midline sulci. After complete dura removal, expose the subarachnoid space. Rinse the subarachnoid space with HBSS solution, then dissect vessels using a blunt needle and carefully remove the pia mater¹⁶. Section the brain coronally into 2-3 mm thick slices using a razor blade. Discard the olfactory bulb and caudal-most sections. Isolate the cerebral cortex along the corpus callosum and transfer cortical tissue to ampoules containing 3 mL of ice-cold tissue collection buffer¹¹.
   4. Pericyte isolation in this protocol specifically targets the cerebral cortex. For other regions of interest: Place coronal sections in a two-well slide with one drop of 2% (vol/vol) FBS in PBS. Identify target regions under dissection microscopy, isolate manually, and proceed with identical subsequent processing steps.
3. Tissue dissociation (Timing: 3-3.5 h)
   1. Mince tissues in ampoules using surgical scalpels into fragments of approximately 1 mm × 1 mm.
      NOTE: Complete mincing within 5 min to preserve tissue freshness.
   2. Pre-rinse pipette tips with tissue collection buffer. Slowly aspirate minced fragments toward the tip apex while maintaining vertical orientation (tip downward). Gently expel fragments into pre-chilled 15 mL centrifuge tubes (prepared in Step 2.1.5).
      NOTE: Avoid aspirating fragments into the upper barrel to prevent adherence to the pipette wall.
   3. Wash fragment-containing ampoules (from Step 2.3.1) with 2% (vol/vol) FBS in PBS and transfer the wash to centrifuge tubes. Centrifuge at 300 x g at 4 °C for 5 min.
   4. Discard supernatants post-centrifugation. Resuspend pellets in 3 mL of CD enzyme working solution (from Step 1.5) per tube. Aliquot each suspension equally into three tubes (total 9 tubes, 1 mL/tube), then supplement with 4 mL of CD enzyme working solution (final volume: 5 mL/tube).
   5. Incubate tubes in a 37 °C hybridization oven for 100 min with orbital shaking at 100 x g¹⁴.
   6. Prepare tissue dissociation solution during incubation by adding 1 mL of DNase I solution to 10 mL of 2% (vol/vol) FBS in PBS (from Step 1.10).
   7. Terminate digestion by centrifugation at 300 x g at 4 °C for 5 min.
      NOTE: Assess digestion efficiency microscopically using a 10-20 µL aliquot. The ideal endpoint manifests as bead-like vascular fragments; elongated vascular segments indicate under-digestion, whereas predominant single cells signify over-digestion.
   8. Resuspend pellets in 1 mL of tissue dissociation solution per tube. Dissociate mechanically by pipetting 100 times¹¹.
      NOTE: Avoid full pipette volume aspiration to minimize bubble formation and preserve cell viability.
   9. Pool all suspensions (from a total of 9 digestion tubes representing the brain tissues of 3 mice) into a 50 mL centrifuge tube. Rinse each digestion tube with 2 mL of 2% (vol/vol) FBS in PBS and combine washes. Centrifuge at 300 x g at 4 °C for 5 min.
   10. After centrifugation, distinct layers are visible. Carefully discard the uppermost layer (containing impurities and small cellular debris), while retaining the intermediate layer (consisting of cells or cell clusters with lower density) and the bottom layer (intact cells and partially undigested tissue fragments). Add 12.5 mL of 20% BSA solution and gently invert the tube to resuspend the pellets. Centrifuge again at 1000 × g at 4 °C for 20 min.
   11. Collect the bottom pellet (enriched in intact cells with higher purity) and transfer the supernatant (primarily containing myelin debris, lipids, and a small number of cells) into a new 50 mL centrifuge tube. Resuspend by gentle inversion and repeat centrifuge at 1,000 x g at 4 °C for 20 min.
   12. Resuspend each pellet obtained from the two centrifugation steps (step 2.3.10 and Step 2.3.11) in 2 mL of 2% (vol/vol) FBS in PBS each (final combined volume: 4 mL).
       NOTE: Avoid touching the tube walls during resuspension to minimize contamination by debris.
   13. Combine the two resuspended suspensions from the previous step into a 15 mL centrifuge tube. Centrifuge at 300 × g, 4 °C for 5 min. Discard the supernatant (waste buffer containing BSA, cell debris, and potentially trace amounts of residual myelin lipids). Resuspend the pellet (the final purified and washed, high-purity intact target cells) in 2 mL of PBS solution containing 2% FBS.
       NOTE: This step is to transfer the cells into a clean, suitable suspension buffer for downstream applications.
   14. Prepare four 15 mL centrifuge tubes, each containing 4 mL of 22% Percoll solution. Gently layer 0.5 mL aliquots of the resuspension onto the Percoll gradient (four aliquots total) to maximize cell yield. Centrifuge at 560 x g at 4 °C for 10 min.
       NOTE: Set the centrifuge to "no brake" to avoid vortex-induced disruption of the density gradient, ensuring myelin debris remains stabilized in the supernatant.
   15. Carefully aspirate supernatants. Resuspend pellets in 1 mL of HBSS/BSA/glucose buffer per tube. Pool suspensions into a new 15 mL tube. Rinse Percoll tubes with 2 mL of HBSS/BSA/glucose buffer and combine washes. Centrifuge at 300 x g at 4 °C for 5 min).
4. Flow cytometry sorting preparation and acquisition (Timing: 50 min)
   1. Carefully aspirate the supernatant. Resuspend the cell pellet in 1 mL of HBSS/BSA/glucose buffer.
   2. Add 5 µL of Fc receptor (FcR) blocking reagent to the cell suspension and incubate at 4 °C for 5 min.
   3. Add the following fluorescently conjugated antibodies to the cell suspension at the indicated volumes (sufficient for tissue derived from 3 mice): 5 µL of CD45-PE; 5 µL of CD41-PE; 20 µL of CD31-APC; 15 µL of CD13-FITC.
   4. Incubate the cells on ice for 20 min in the dark.
      NOTE: Protect from light. Avoid stirring.
   5. Add 6 mL of HBSS/BSA/glucose buffer to wash the cells. Centrifuge at 300 x g at 4 °C for 5 min.
   6. Aspirate the supernatant. Resuspend the cell pellet in 1 mL of HBSS/BSA/glucose buffer. Add 200 µL of 7-aminoactinomycin D (7-AAD) solution (to exclude dead cells) and incubate on ice for 10 min. Subsequently, adjust the volume to 2 mL with HBSS/BSA/glucose buffer.
      NOTE: Protect from light.
   7. Add 1% BSA solution to sorting collection tubes (15 mL conical tubes), gently swirl to rinse the inner surfaces. Aspirate excess solution, leaving exactly 1 mL of 1% BSA solution in each tube to receive sorted cells. Store temporarily on ice.
   8. Filter stained cell suspension through a 40 µm strainer immediately before sorting.
   9. Establish an initial gate based on forward scatter (FSC-A) and side scatter (SSC-A) parameters to exclude cellular debris exhibiting low FSC-A/SSC-A signals.
   10. Depletion of hematopoietic lineage cells was achieved through dual-negative gating for CD45 and CD41, effectively eliminating contamination from leukocytes, immune cells, and platelets.
   11. Capture putative pericytes by CD13-positive selection from vascular wall cells.
   12. Evaluate membrane integrity via 7-AAD fluorescent probing; exclude apoptotic cells by 7-AAD⁻ gating.
       NOTE: Collect high-purity CD31⁻/CD13⁺ pericytes through CD31-negative selection. Collect the sorted pericytes directly into tubes containing 1% BSA (from Step 2.4.7) and place them on ice. Immediately prepare single-cell suspensions for counting. Then load the samples directly for sequencing without any intermediate culture steps.
5. scRNA seq and Analysis (Tming: 3-4 days)
   1. Single-cell suspensions of sorted pericytes were prepared following the protocol from the 10x Genomics "GEM-X Universal 3' Gene Expression v4 4-plex Reagent Kits User Guide" (CG000731)^17,18. The cell suspension prepared for loading had a viability of > 90% and was diluted to a concentration within the recommended range of 300-2000 cells/µL.
   2. Combine single-cell suspension with barcoded gel beads and enzymes, then partition into Gel Beads-in-Emulsion (GEMs) via microfluidic droplet encapsulation¹⁹. Perform in situ cell lysis, reverse transcription, and barcode-cDNA linkage.
   3. Break GEMs to recover products. Amplify cDNA by PCR using the following protocol: initial denaturation at 9 °C for 4 s; followed by cycling (13 cycles for <500 target cells, 12 cycles for 501-6000 target cells) consisting of denaturation at 9 °C for 2 s, annealing at 9 °C for 3 s, and extension at 7 °C for 6 s per cycle; and a final extension at 72 °C for 1 min followed by a hold at 4 °C. Validate the amplified products for fragment size (200-300 bp) and yield.
      NOTE: The amplification products can be stored at 4 °C for up to 72 h or at -2 °C for one week, or they can be processed directly in the next step.
   4. For qualified cDNA: Fragment, perform end repair, ligate P7 adapter, and amplify with index primers. Size-select fragments to construct sequencing libraries.
   5. Sequence libraries on the Illumina NovaSeq platform following QC validation.
   6. Demultiplex sequencing data to extract cell barcodes, unique molecular identifier (UMIs), and RNA sequences. Align reads to the reference genome.
   7. Quantify UMI counts, estimate valid cell numbers, and generate a gene-barcode expression matrix²⁰.
   8. Remove doublets using the DoubletFinder package (v2.0.6) in R. Predict based on artificial nearest-neighbor identification to ensure data quality.
   9. Conduct principal component analysis (PCA) and Uniform Manifold Approximation and Projection (UMAP) dimensionality reduction using Seurat (v5.0). Identify cell clusters and visualize pericyte-specific marker genes to validate cellular identity.

Results

To achieve precise isolation of brain pericytes, cells dissociated from brain tissue were stained with PE Texas Red-conjugated anti-mouse CD41/CD45, FITC-conjugated anti-mouse CD13, and APC-conjugated anti-mouse CD31, establishing a multi-step sorting strategy integrating surface markers and viability detection (Figure 2). Initial gating based on forward scatter (FSC-A) and side scatter (SSC-A) parameters effectively excluded cellular debris (Figure 2, top-left ...

Discussion

This study describes a systematically optimized and validated FACS strategy based on CD13⁺/CD31⁻ multicolor labeling for the highly efficient isolation of mouse brain vascular pericytes²⁴. Flow cytometric and subsequent scRNA-seq analyses confirmed that the method precisely discriminates pericytes from endothelial cells and yields cells that consistently meet all critical quality control metrics for scRNA-seq, including viability, aggregation rate, live cell concentration, and nucleate...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Fetal Bovine Serum (FBS)	Gibicol	A5669701	Aliquot and store at -20°C; can be stored at 4°C for up to 1 week after thawing
Phosphate Buffered Saline(10×PBS, pH 7.4)	Life Technologies	70011-044	Dilute before use for cell washing
D-(+)-Glucose	Sigma-Aldrich	G7021-1KG	For preparing cell energy buffer
Bovine Serum Albumin (BSA)	Sigma-Aldrich	A1933-25g	Protects cells and prevents adhesion
Dulbecco's  Phosphate Buffered Saline (DPBS,1×)	Gibco	14190-144
Hanks' Balanced Salt Solution (HBSS,1×)	Life Technologies	14175-095	For cell washing and resuspension
UltraPure DNase/RNase-Free Distilled Water	Life Technologies	10977-015	For preparing nuclease-free reaction systems
Collagenase/Dispase	Sigma-Aldrich	11097113001-100mg	For tissue digestion and cell dissociation; aliquot and store at -20°C
DNaseI(RNase-Free)	TIANGEN	RT411-1500U	Prevents DNA aggregation; aliquot and store at -20°C
Percoll	Sigma-Aldrich	P1644-100ML	For cell separation; store at 4°C
FITC Rat Anti-Mouse CD13	BD Biosciences	558744	Fluorescent conjugate; store protected from light at 4°C for up to 1 month
APC Rat Anti-Mouse CD31	BD Biosciences	551262	Store protected from light at 4°C for several months
PE Rat Anti-Mouse CD41	BD Biosciences	558040	Store protected from light at 4°C for several months
PE Rat Anti-Mouse CD45	BD Biosciences	553081	Store protected from light at 4°C for several months
FITC Rat IgG1, κ Isotype Control	BD Biosciences	552916	Fluorescence control; store protected from light at 4°C
APC Rat IgG2a, κ Isotype Control	BD Biosciences	553932	Store protected from light at 4°C for several months
PE Rat IgG2b, κ Isotype Control	BD Biosciences	553989	Store protected from light at 4°C for several months
RNase AWAY Decontamination Reagent	Ambion	10328011	For eliminating RNase contamination
Purified Rat Anti-Mouse CD16/CD32 (Mouse BD Fc Block)	BD Pharmingen	553141	Prevent non-specific antibody binding
7-Aminoactinomycin D (7-AAD)	BD Pharmingen	559925	For dead cell exclusion.
2,2,2-Tribromoethanol	Sigma-Aldrich	T48402-25g	For the preparation of 1.25% (w/v) Tribromoethanol Solution
2-Methyl-2-butanol	Psaitong	M20008-500mL	For the preparation of 1.25% (w/v) Tribromoethanol Solution
C57BL/6 mouse	Jiangsu Huachuang Xinnuo Pharmaceutical Technology Co., Ltd.	SYXK (Su) 2020-0041
50 mL centrifuge tube	Corning	430291
15 mL centrifuge tube	Corning	430791
cell cluture dish 100mm×20mm	Corning	430167
Cell Ranger		3.1.0/4.0.0	Single-Cell Sequencing Data Analysis Software
Loupe Cell Browser		4.1	Cell Browser
STAR		2.5.1b	RNA Alignment
irlbpy		-	Principal Component Analysis
tsne		0.15	Barnes-Hut t-SNE (t-Distributed Stochastic Neighbor Embedding) Analysis
k-means		0.17	K-Means Clustering
graph-based clustering		0.17	Graph-Based Clustering
FastQC		v0.11.2	Data Quality Control (QC)