Characterization of Bone Marrow-Derived Immune Compartment in Murine Brain Metastases by Spectral Flow Cytometry

Abstract

Metastatic brain cancer is the most common intracranial cancer among adults, often leading to poor survival outcomes. The incidence of brain metastases is on the rise, likely associated with advanced systemic therapeutics which improve patient survival, providing sufficient time for micrometastases to develop into fully established metastases. Therefore, brain metastasis remains an unmet clinical need. The brain metastatic tumor microenvironment is a complex ecosystem composed of numerous brain, stromal, and immune cells with unique adaptations that represent a formidable challenge for treatment. Importantly, brain metastases are enriched in immune cells, especially myeloid resident and recruited cells considered highly plastic. Thus, it is critical to assess the immune landscape and functional phenotypes longitudinally and across treatment conditions in brain metastasis in a non-biased, comprehensive manner. Here, we outline a protocol to assess immune cell populations in murine brain metastasis samples, using a 23-color panel for spectral flow cytometry.

1. Introduction

A majority of cancer-related deaths are attributed to metastatic disease and not due to primary tumors [1, 2], emphasizing the importance of understanding how tumor cells colonize distal organs. Brain metastatic cancers are those that have spread to the brain from extracranial origins and are widely considered incurable. Brain metastases are the most common type of intracranial neoplastic lesions in adults [3], and it is estimated that 1–4 out of 10 patients with solid primary tumors will develop brain metastasis over time [4]. Unfortunately, the average median survival remains 8–16 months post-initial diagnosis [4]. Additionally, the incidence of brain metastases is alarmingly increasing, partially attributed to improved diagnostic tools allowing for earlier detection [3] and improved systemic therapeutic interventions leading to prolonged patient survival, which provides ample time for disseminated dormant tumor cells to “awaken” and develop into fully established metastases [5-7]. As systemic therapeutics for cancers continue to advance, it is imperative that we concurrently improve strategies against brain metastatic disease.

Historically, tumor cells were studied in silo; however, it is now widely appreciated that tumor microenvironment plays a critical role in cancer progression. This environment is composed of all parenchymal cells from the tissue the tumor cells have reached, resident and recruited immune and stromal cells, extracellular matrix, and a variety of soluble factors that surround the cancerous lesions [8]. There is constant communication between the microenvironment and the disseminated tumor cell, resulting in their coevolution throughout the neoplastic process [9]. The concept of immune privilege, once believed to isolate the central nervous system and provide an almost impenetrable physical barrier, has largely evolved, and we now know that immune cells do communicate with cells in the brain to elicit immunological responses, in healthy conditions and neuropathologies [10]. Technologies like single-cell transcriptomic, proteomics, cytometry by time of flight [11], flow cytometry, and functional assessments have shown that the immune compartment in brain metastases is abundant and vastly restructured [11-13]. Importantly, pro- and antitumor phenotypes among these immune cells have been observed concurrently, emphasizing the importance of probing this dynamic landscape in a comprehensive manner.

The study of brain metastasis has expanded significantly over the past decade largely due to the effort devoted to learning more technically challenging cell injection and microsurgery methods by more research teams. This has led to the development of more cell lines with enhanced brain tropism [14] and started to illuminate some of the important mechanisms that enable the outgrowth of disseminated tumor cells into the central nervous system [15-28]. Metastasis to the central nervous system can be established by injecting the brain tropic cells into the arterial circulation via intracardiac (left ventricle) or intracarotid injection, intracranial injection, or cisterna magna injection to establish leptomeningeal metastasis. Each type of injection delivers the cancer cells into a different body compartment, starting the metastatic process at different stages, which results in different selective pressures until a fully established brain metastatic tumor develops. Thus, it is critical to understand the advantages and disadvantages of each method to determine which one is the most appropriate for the biological question being explored. Our group is interested in understanding how the immune system contributes to the establishment and outgrowth of brain metastasis in a manner that better parallels human disease.

In this chapter, we outline an approach to characterize the immune compartment in murine brain metastatic tissue samples using spectral flow cytometry. Spectral flow cytometry is different from traditional flow in that the full spectrum emission for each fluorophore across all lasers is collected [29]. Since each fluorophore has a unique spectrum emission signature, we can distinguish each fluorophore from one another through an unmixing process—allowing fluorophores with very similar, but different, peak emission to be discerned, thus allowing for a large number [12] of parameters to be probed in a single panel, which would not be possible with traditional flow. Given the plastic nature of certain immune cell populations, we emphasize the importance of including functional phenotypic markers, alongside identity markers to comprehensively assess the heterogenous brain metastasis landscape. As we continue to tease apart the complexities of the brain tumor microenvironment, it is critical that we correctly identify and characterize the functional phenotypes of bone marrow-derived immune cells for future more targeted experiments. As we previously described [30], mouse reporter strains may be used to perform fate-mapping of a particular chemotactic axis of interest (see Note 1). Last, although this chapter is specifically in the context of brain metastasis, it may also be useful for analysis of other neuro-pathological diseases.

2. Materials

2.1. Establishment of Brain Metastatic Tumor in Mice

1. Mice: Foxp3^DTR-GFP immunocompetent mice in the C57BL/6 J genetic background.

2. Cells: E0771-BrM tumor cells transduced with firefly luciferase vector, organotrophic selection.

3. Dulbecco’s phosphate-buffered saline (1× DBPS).

4. Trypsin/EDTA 0.25% 1× Solution.

5. RPMI-1640, supplemented with 10% fetal bovine serum (FBS), 1% penicillin–streptomycin, and 0.5% amphotericin.

6. Syringe: BD insulin syringe 1 mL, 26G, 0.5 in. detachable needle.

7. Isoflurane (Piramal Critical Care).

8. Anesthesia machine RC² Rodent Circuit Controller (VetEquip).

9. D-luciferin potassium (Syd Labs).

10. IVIS imaging system (Revvity).

2.2. Single-Cell Preparation of Brain Metastatic Tumors

1. Buffers: 1× DPBS, ACK Red Blood Cell Lysis (150 mM NH4Cl, 10 mM KHCO3, 0.1 mM EDTA).

2. Neural Tissue Dissociation Kit [31] (Miltenyi Biotec).

   1. Enzyme mix 1: 1900 μL of Buffer X + 50 μL of Enzyme P.
   2. Enzyme mix 2: 20 μL of Buffer Y + 10 μL of Enzyme A.

3. Debris Removal Solution (Miltenyi Biotec).

4. GentleMACS Octo Dissociator with heaters (Miltenyi Biotec).

5. GentleMACS C tubes (Miltenyi Biotec).

6. MACS SmartStrainer, 70 μm (Miltenyi Biotec).

2.3. Staining of Single-Cell Suspensions

1. Buffers: 1× PBS, MACS buffer (0.5% BSA, 2 mM EDTA in 1× PBS, pH 7).

2. Brilliant Stain Buffer (BD HorizonTM).

3. Zombie NIR^™ Fixable Viability Kit (Biolegend).

   1. Zombie NIR solution: 1 μL of DMSO-reconstituted dye +1000 μL of PBS (see Note 9).
4. Fc Block stock: Pure MS CD16/CD32 (2.4G2) (TONBO Biosciences).

   1. Fc solution: 45 μL MACS buffer +5 μL FC stock.

5. Antibodies (Table 1) (see Note 2).

Table 1.

Antibody clones, catalog numbers, and manufacturers listed

Antibody	Clone	CAT #	Manufacturer
BUV395-conjugated CD45	30-F11	564279	BD Horizon
PE/Fire640-conjugated CD11b	M1/70	101280	Biolegend
AlexaFluor700-conjugated Ly6C	AL-21	561237	BD Pharmingen
BV605-conjugated Ly6G	IA8	127639	Biolegend
BUV496-conjugated F4/80	T452342	750644	BD OptiBuild
BV711-conjugated CD11c	N418	117349	Biolegend
PerCP-eFlour 710-conjugated CD206	MR6F3	46-2061-82	Thermo Fisher eBiosciences
eFluor 450-conjugated MHC-II	M5/114.15.2	48-5321-82	Thermo Fisher eBiosciences
BV480-conjugated CD115	CSF-1R	746456	BD OptiBuild
PE-Dazzle 594-conjugated CD86	GL-1	105042	Biolegend
BUV615-conjugated SiglecF	1RNM44N	366-1702-82	Thermo Fisher eBiosciences
PE-conjugated TCRβ	H57-597	109208	Biolegend
BUV737-conjugated CD4	GK1.5	612761	BD Horizon
APC/Fire 750-conjugated CD8	53–6.7	100766	Biolegend
PE/Cy7-conjugated CD335 (NKp46)	29A1.4	25-3351-82	Thermo Fisher eBiosciences
BV421-conjugated CD25	PC61	562606	BD Horizon
BV510-conjugated CD127	SB/199	563353	BD Horizon
PE/Cy5-conjugated CD62L	Mel-14	104410	Biolegend
BV570-conjugated CD44	IM7	103037	Biolegend
BV650-conjugated CD19	6D5	115541	Biolegend
BV786-conjugated TCRγδ	GL3	744117	BD OptiBuild

2.4. Flow Cytometry Characterization of the Brain Metastasis Immune Landscape

1. Flow cytometer: Aurora (Cytek Biosciences).

2. Software: FlowJo v10.9.0 (Tree Star).

3. Methods

3.1. Establishment of Brain Metastatic Tumors in Mice

The protocol described in this section is optimized for brain tumors arising from intracardiac inoculation of tumor cells, where single cells reach the brain capillaries after actively going through the processes of surviving in the circulation, traversing the brain blood barrier at the single-cell stage, and surviving in the brain tissue as solitary cells before reinitiating growth:

1. Trypsinize 50–100,000 brain metastatic cells transduced with a vector encoding the firefly luciferase enzyme, and resuspend in a volume of 100 μL 1× DPBS. Keep on ice.

2. Anesthetize mice with isoflurane, administer subcutaneous analgesic, and inject cells into the left ventricle of the heart (intracardiac injection) of 8- to 10-week-old mice (see Note 3). To perform this injection, the needle is inserted vertically (90° from body) in the third intercostal space (between the third and fourth ribs), slightly to the right of the sternum (see Note 4). After injection, retract the needle, and apply light pressure to the injection site for 1 min. Place the mouse on a heating pad, and observe it until full recovery from anesthesia.

3. Monitor in vivo tumor growth kinetics by bioluminescence imaging using an IVIS (In Vivo Imaging System) imaging system 2–3 times a week post intracardiac injection by retroorbital administration of 150 mg/kg D-luciferin into anesthetized mice. Images are performed after 2 min of D-luciferin injection.

4. At the experimental end point, and immediately after imaging, perfuse mice via cardiac puncture with 10 mL DPBS (see Note 5). Dissect brains, and reimage to identify metastatic brain regions. Once metastatic brain regions are localized, dissect them with a sterile, cold scalpel, place in cold DPBS, and keep on ice until transfer to digestion cocktail.

3.2. Single-Cell Preparation of Brain Metastatic Tumors

1. Prepare the Neural Tissue Dissociation Kit (P) enzyme mixes 1 and 2, as described in Subheading 2.2, item 2, to digest one whole mouse brain (see Note 6).

2. Remove the metastatic brain fragments from cold DPBS, and place into a clean petri dish using sterile forceps. Mince the metastatic brain fragments using a sterile, cold scalpel until mushy consistency.

3. Once brain fragments are fully minced, add sample into C tube with enzyme mix 1, and pipette to mix well. Add enzyme mix 2 into C tube, and mix well by pipetting. Close the C tube tightly, until hearing a click. Invert the C tube upside down, ensuring that most of the brain fragments are located on the C tube cap (see Note 7).

4. Attach the upside-down C tube onto the sleeve of the gentleMACS Octo Dissociator with Heaters. Run the gentleMACS Program 37C_NTDK_1 program.

5. When the program is finished, detach the C tubes from the instrument. Perform a quick spin to collect entire sample to the bottom of the upright C tube.

6. Add 10 mL of cold DPBS to sample, and mix by pipetting. Transfer the sample through a 70 μm MACS SmartStrainer to a 50 mL conical tube. Pipette an additional 10 mL of cold DPBS through the strainer, to thoroughly wash strainer. Discard strainer.

7. Centrifuge sample at 300 G for 10 min, at 4 °C. Aspirate supernatant by pipetting.

8. Resuspend pellet with 3100 μL of cold DPBS, and transfer to a clean 15 mL conical tube. Do not vortex.

9. Add 900 μL of cold Debris Removal Solution and mix well by pipetting gently (see Note 8). Do not vortex.

10. Carefully, and slowly, overlay 4 mL of cold DPBS on top of the sample (see Note 9).

11. Centrifuge at 3000 G for 10 min, at 4 °C with full acceleration and soft deacceleration. Three phases are formed during this step: Aspirate the top two phases by pipetting carefully and discard them.

12. Gently tap the tube to resuspend the remaining pellet, and fill 15 mL tube with cold DPBS to wash. Gently invert the tube three times. Do not vortex.

13. Centrifuge at 1000 G for 10 min, at 4 °C with full acceleration and full deacceleration. Aspirate supernatant by pipetting.

14. Resuspend cell pellet in 1 mL of cold ACK Lysis Buffer. Do not vortex.

15. Incubate for 10 min on ice. Add 10 mL of cold DPBS to wash, and mix by pipetting.

16. Centrifuge at 300 G for 10 min, at 4 °C. Aspirate supernatant by pipetting.

17. Resuspend cell pellet in 100 μL of cold PBS for flow cytometry staining. Keep on ice until ready to stain.

3.3. Staining of Single-Cell Suspensions

1. Transfer samples into V-well plate, and keep on ice.

2. Add 100 μL diluted Zombie NIR^™ dye (prepared as indicated in Subheading 2.3, item 3a) to each sample well. Incubate on ice for 15 min, covered with foil.

3. When incubation is complete, top off each well with 100 μL MACS buffer, and mix thoroughly by pipetting to wash (see Note 10).

4. Centrifuge at 245 G for 7 min, at 4 °C. Dump the supernatant.

5. Resuspend pellet in 50 μL of Fc Block solution (prepared as indicated in Subheading 2.3, item 4a), per well. Incubate on ice for 10 min, covered with foil. When incubation is complete, top off each well with 200 μL MACS buffer, and mix thoroughly to wash.

6. Centrifuge at 245 G for 7 min, at 4 °C. Dump the supernatant.

7. Resuspend pellet in 100 μL of antibody cocktail mix (Table 2) for full stain samples, fluorescence-minus-one (FMO) antibody cocktail mix for FMO controls, single-color antibodies for compensation controls, or MACS buffer for unstained controls (see Note 11). Incubate on ice for 20 min, covered with foil.

8. When incubation is complete, top off each well with 200 μL MACS buffer, and mix thoroughly to wash.

9. Centrifuge at 245 G for 7 min, at 4 °C. Dump the supernatant.

10. Resuspend each well in 100 μL of PBS, cover with foil, and take to flow cytometer on ice. Samples are run on a Cytek Aurora instrument (Cytek Biosciences, USA), using the “HI” setting and plate loader. Acquisition settings are set to stopping criteria of 80 μL collection volume or 5,000,000 cells (see Note 12).

Table 2.

Recipe to prepare full stain antibody cocktail mix for one sample. For multiple samples, multiply accordingly

Antibody cocktail mix for one sample
Antibody	Volume (μL)
BUV395-conjugated CD45	0.5
PE/Fire640-conjugated CD11b	0.5
AlexaFluor700-conjugated Ly6C	0.8
BV605-conjugated Ly6G	0.8
BUV496-conjugated F4/80	0.5
BV711-conjugated CD11c	0.5
PerCP-eFlour 710-conjugated CD206	0.5
eFluor 450-conjugated MHC-II	0.5
BV480-conjugated CD115	0.5
PE-Dazzle 594-conjugated CD86	0.5
BUV615-conjugated SiglecF	0.5
PE-conjugated TCRβ	0.5
BUV737-conjugated CD4	0.5
APC/Fire 750-conjugated CD8	0.5
PE/Cy7-conjugated CD335 (NKp46)	0.5
BV421-conjugated CD25	0.5
BV510-conjugated CD127	0.5
PE/Cy5-conjugated CD62L	0.5
BV570-conjugated CD44	0.5
BV650-conjugated CD19	0.5
BV786-conjugated TCRγδ	0.5
Reagent	Volume (μL)
MACS buffer	78.9
Brilliant stain buffer	10

3.4. Flow Cytometry Characterization of the Brain Metastasis Immune Landscape

1. Gate samples to exclude doublets and dead cells using Zombie NIR (Fig. 1).

2. Distinguish between microglia and recruited immune cell populations by gating on CD45 versus CD11b [30, 32] (see Note 13). Resting resident microglia are identified as CD45^loCD11b+, while bone marrow-derived myeloid cells are identified as CD45^hiCD11b+ (Fig. 1). Lymphocytes can be identified as CD45+CD11b− (see Note 14).

3. In the myeloid lineage (CD45^hiCD11b+), distinguish monocytes (Ly6C^hiLy6G−) from granulocytes (Ly6G+) (see Note 15). From the granulocyte population, assess the level of SiglecF to identify eosinophils (see Note 16) and CD115 to identify neutrophils (CD115−) (see Note 17) [33].

4. Next, identify two macrophage populations (F4/80+Ly6C^lo and F4/80+Ly6C^hi). F4/80+Ly6C^lo macrophages are typically associated with tissue healing phenotypes, while F4/80+Ly6C^hi macrophages are associated with inflammatory phenotypes (see Note 18). From both macrophage populations, assess the levels of MHC-II, CD86, CD206, and CD115 (see Note 19).

5. Inside the lymphoid lineage (CD45+CD11b−), identify two main populations:

   1. Conventional T cells (TCRβ+ TCRγδ−), which are distinguished from γδ T cells (TCRβ− TCRγδ+). Within the conventional T cells, identify helper T cells (CD4+CD8−) and cytotoxic T cells (CD4−CD8+). From the helper T cell population, GFP reporter is used to identify regulatory T cells (GFP+CD25±), using our Foxp3^DTR-GFP mice (see Note 20). For both cytotoxic T cells and helper T cells, identify effector T cells (CD44+CD62L−), central memory T cells (CD44+CD62L+), and naïve T cells (CD44−CD62L+) (see Note 21).
   2. From the non-T cell population (TCRβ− TCRγδ−), we identify B cells (CD19+MCHII±) and dendritic cells (CD11c+CD11b±) (see Note 22).
   3. Within the lymphocyte gate, identify natural killer cells (CD335+TCRβ−) and natural killer T cells (CD335+-TCRβ+).

Fig. 1.

Gating strategy to analyze immune cell landscape in brain metastasis samples using spectral flow cytometry. Shown is a representative murine brain metastasis sample, from a Foxp3^DTR-GFP mouse using E0771-BrM cells via intracardiac injection. Cell populations were defined as follows: resting resident microglia (CD45^loCD11b+), bone marrow-derived myeloid cells (CD45^hiCD11b+), lymphocytes (CD45+CD11b−), inflammatory monocytes (CD45^hiCD11b+ Ly6C^hiLy6G−), granulocytes (CD45^hiCD11b+Ly6C^loLy6G+), macrophage population #1 (CD45^hiCD11b+F4/80+Ly6C^lo), macrophage population #2 (CD45^hiCD11b+F4/80+Ly6C^hi), conventional T cells (CD45+CD11b− TCRβ+ TCRγδ−), γδ T cells (CD45+CD11b− TCRβ− TCRγδ+), helper T cells (CD45+CD11b− TCRβ+ TCRγδ− CD4+CD8−), cytotoxic T cells (CD45+CD11b− TCRβ+ TCRγδ− CD4-CD8+), regulatory T cells (CD45+CD11b− TCRβ+ TCRγδ− CD4+CD8-Foxp3GFP +CD25±), B cells (CD45+CD11b− TCRβ− TCRγδ− CD19+MCHII±), dendritic cells (CD45+CD11b− TCRβ− TCRγδ− CD11c+CD11b±), natural killer cells (CD45+CD11b− CD335+TCRβ−), and natural killer T cells (CD45+CD11b− CD335+TCRβ+)

4. Notes

1. When using mouse reporter strains to track a particular chemotactic axis, ensure that the fluorophore reporter does not interfere with other fluorophores in the panel.

2. Investigator must select antibody fluorophores based on the laser availability and instrument filter configurations on the flow cytometry instrument available to the investigator. Marker to fluorophore matching will depend on antigen density, co-expression profiles on cells of interest, classification of marker expression (primary, secondary, or tertiary antigens), level of fluorophore brightness (brightness index), cross-laser excitation, and spectral spillover. If not familiar with these parameters, arrange consultation with your flow cytometry core facility.

3. It is critical to leave some air in between the syringe plunger and the cell suspension, to allow for visualization of the pulsating blood into the syringe needle hub. Visually inspect the cell suspension in the syringe to ensure there are no air bubbles; if so, gently tap them away.

4. To ensure needle is in the right position for injection, pulsating arterial blood must be visualized in the syringe needle hub. If there is no blood pulsation into syringe needle hub, reposition the needle, and retry the injection.

5. Complete perfusion (until clear) is necessary for experiments in which tissue-resident immune cell populations are analyzed. Inadequate perfusion could lead to left-over circulating immune cell infiltrates, which would artificially sway experimental results.

6. Accurate volumes are calculated by weighting the brain tumor dissected fragment and calculate the volume of enzyme mixes 1 and 2 from the kit as per the manufacturer’s instructions. Volume listed in the protocol corresponds to a tissue of ~400 mg of weight or less. If more tissue volume is used, adjust enzyme mixes accordingly.

7. Since the GentleMACS Octo Dissociator rotates the C tube via stator and rotator on the tube cap, it is essential to ensure brain fragments are completely on the bottom of the upside-down tube (i.e., on the cap of the C tube). If brain fragments remain on the sides of the tube, they will not come into contact with the rotator on the tube cap and will not be digested.

8. Myelin is removed from cell suspension to obtain a cleaner flow run and can be removed in different ways. A popular alternative is a myelin removal kit, but we have used myelin removal and debris removal protocols and have consistently obtained better results using the debris removal solution when using adult brains.

9. To ensure proper overlay, we first gently tilt the 15 mL conical tube at a 20° angle to achieve a diagonal, slanted meniscus. Next, we slowly dispense 4 mL of cold DPBS on the side of the tube, allowing the liquid to slowly slide into the sample with a slow, steady stream (over a period of ~1 min). Then slowly tilt 15 mL conical tube upright, and keep tube upright until it is in the centrifuge.

10. For Zombie NIR^™ Fixable Viability dye, the manufacturer recommends diluting at 1:100–1:1000 with 1× PBS, depending on different cell types. It is recommended to titrate the amount of dye to determine dilution with optimal performance for each investigator’s samples. Additionally, the manufacturer indicates cells must be resuspended in PBS buffer (without protein, no Tris–HCl buffer) for the viability stain. After the viability stain incubation period is complete, sample can be transferred back into MACS buffer.

11. Investigator must titrate each staining antibody to determine optimal stain for cells of interest in investigator’s samples. Investigators may refer to each antibody’s datasheet as a starting point for testing antibody titrations. Single-color controls are required to set up an experiment on the flow cytometer instrument. Critically, the final antibody dilution for single-color compensation controls should match dilution used in the full antibody cocktail mix; for example, for a BUV395-conjugated CD45 single-color control, we would prepare the mix as follows: 0.5 μL antibody +99.5 μL MACS buffer and 100 μL to compensation sample. FMO controls are recommended to identify upper boundary of background signals to help gate positive populations. FMOs are highly recommended for large panels (> 10 fluorophores), as well as secondary (antigen expression on a continuum) and tertiary antigens (low density antigen expression), regardless of panel size. The final antibody dilutions for FMO controls should match dilutions used in the full antibody cocktail mix. For example, for a BUV395-conjugated CD45 FMO control, we would prepare the mix as Noted in Table 1, with the modifications: no BUV395-conjugated CD45 antibody, 79.4 μL of MACS buffer, and 100 μL to FMO sample. Unstained controls are required to set up an experiment on the spectral flow cytometry instrument.

12. These acquisition setting stop criteria are to ensure most sample is collected from a 100 μL volume reconstituted sample. If final reconstitution volumes are adjusted, then collection volume stopping criteria may be adjusted. Given the low proportion of immune cells in these tissues, it is important to maximize the amount of sample collected.

13. It is critical to distinguish microglia, the brain tissue-resident immune cells, from recruited immune cell populations. This is particularly tricky, since there is considerable overlap in surface protein expression between these two cell populations [34]. Reporter mouse strains have been used to distinguish microglia from bone marrow-derived macrophages mostly under physiological conditions [35-37]. However, it is important to note that these cells are exceedingly plastic and may up- or down-regulate expression in the brain metastatic microenvironment [38].

14. We report percentages or median fluorescence intensity from our analysis; however if absolute counts are desired, then counting beads may be used.

15. The granulocyte population (Ly6C^loLy6G+) is observed under nontumor bearing conditions [39]. However, this population experiences dramatic expansion under tumor-bearing conditions in circulation [39]. Importantly, granulocyte populations exhibit heterogenous transcriptional profiles and phenotypes in cancer [40], which could be due to their high plasticity [41]. Additional studies must be performed to investigate granulocyte subpopulations and to assess whether these subpopulations are transcriptionally and/or phenotypically distinct.

16. SiglecF is the eosinophil lineage marker; however, it is important to note that in cancer, other granulocyte cells such as neutrophils may express SiglecF [42].

17. The distinction between neutrophils and G-MDSCs using identity markers is difficult, since G-MDSCs are believed to be a functional state of neutrophils and not a different cell type [31]. In mice, CD115 (CSF-1 receptor) is expressed in G-MDSCs, but not in neutrophils, making it an attractive marker to identify neutrophils (CD115−) [33]. Of note, suppressive functional assessments must be performed to accurately define “G-MDSCs”; thus we simply call the CD115+ population “CD45^hiCD11b+Ly6C^loLy6G+CD115+ cells.” Excitingly, the myeloid field is currently teasing apart the functional, phenotypic, and transcriptomic differences across such plastic cells.

18. The transition from inflammatory to tissue healing phenotypes is critical in the context of wound healing. In cancer, this phenotypic transition may be hijacked to promote progression and metastasis [43].

19. MHC-II binds antigen peptides and presents them to CD4+ T cells, for the purpose of eliciting an immunological response. CD86 serves as costimulatory molecules to naïve T cells. CD206 is a pattern recognition receptor for mannoglycoproteins present on microbes. CD115 is a marker for macrophage migration.

20. If investigators are using mice without the FoxP3 reporter, the same can be achieved by intracellular staining of the Foxp3 transcription factor.

21. CD127 (IL-7Ra) versus CD62L may also be used to identify central memory T cells (CD127^hiCD62L^hi) and peripheral effector memory T cells (CD127^hiCD62L^lo).

22. MHC-II and CD86 are used to assess level of antigen presentation and co-stimulation on B cells and dendritic cells. CD206 may be used to assess immature state in dendritic cells.