This protocol provides an analysis of the macrophage subpopulations in the adult mouse central nervous system by flow cytometry and is helpful for the study of multiple markers expressed by these cells.
Numerous studies have demonstrated the role of immune cells, in particular macrophages, in central nervous system (CNS) pathologies. There are two main macrophage populations in the CNS: (i) the microglia, which are the resident macrophages of the CNS and are derived from yolk sac progenitors during embryogenesis, and (ii) the monocyte-derived macrophages (MDM), which can infiltrate the CNS during disease and are derived from bone marrow progenitors. The roles of each macrophage subpopulation differ depending on the pathology being studied. Furthermore, there is no consensus on the histological markers or the distinguishing criteria used for these macrophage subpopulations. However, the analysis of the expression profiles of the CD11b and CD45 markers by flow cytometry allows us to distinguish the microglia (CD11b+CD45med) from the MDM (CD11b+CD45high). In this protocol, we show that the density gradient centrifugation and the flow cytometry analysis can be used to characterize these CNS macrophage subpopulations, and to study several markers of interest expressed by these cells as we recently published. Thus, this technique can further our understanding of the role of macrophages in mouse models of neurological diseases and can also be used to evaluate drug effects on these cells.
The microglia are the parenchymal tissue-resident macrophages of the central nervous system (CNS). They play two key functional roles: immune defense and maintenance of the CNS homeostasis. In contrast to the MDM, which are renewed continually from the hematopoietic stem cells in the bone marrow, the microglial cells differentiate from primitive hematopoietic progenitor cells originating in the yolk sac (YS) that colonized the brain during embryonic development1,2,3. In rodents, the transcription factor Myb plays a crucial role in the development of all bone marrow derived monocytes and macrophages, but for YS derived microglia, this factor is dispensable and differentiation remains dependent on the transcription factor PU.14.
In the healthy CNS, microglia are dynamic cells that constantly sample their environment, scanning and surveying for invading pathogens or tissue damage5. The detection of such signals initiates a pathway to resolve the injury. The microglia rapidly switch from a ramified morphology to an amoeboid one, which is followed by phagocytosis and release of various mediators, such as pro- or anti-inflammatory cytokines. Thus, depending on their microenvironment, activated microglia can acquire a spectrum of distinct priming states6.
The microglia profoundly impact the development and progression of many neurological disorders. In the rodent models of Alzheimer's disease (AD)7, Amyotrophic Lateral Sclerosis (ALS)8, Multiple Sclerosis (MS)9 or Parkinson disease (PD)10, the microglia are shown to play a dual role, either inducing detrimental neurotoxicity or acting in a neuroprotective manner, which is dependent on the specific disease, the disease stage, and whether the disease was influenced by the systemic immune compartment7,8,9,10,11. Most of the CNS lesions observed in the diseases cited above contain a heterogeneous population of myeloid cells, including not only parenchymal microglia, but also perivascular and meningeal macrophages, as well as CNS-infiltrating MDM. These cell types may differentially contribute to the pathophysiologic mechanisms related to injury and repair7,12,13,14,15. The current challenge for investigators who are studying these disease models is to establish whether the peripheral monocytes and macrophages infiltrate the CNS and if so, to distinguish the resident microglia from these cells. Indeed, the microglial cells are very plastic; when they are activated, the microglia re-express markers that are usually expressed by peripheral monocytes and macrophages. The issue, therefore, relies on identifying markers that can distinguish the resident microglia from the infiltrating monocytes and macrophages.
The discrimination of these populations on brain slices by immunohistological applications is limited due to the lack of specific antibodies. However, flow cytometry analysis is an efficient technique to assess the expression of several markers and to distinguish cell populations (for example, lymphocytes, macrophages/MDM CD11b+CD45high, and microglia CD11b+CD45med), as well as cell subpopulations16,17,18. This protocol describes the procedures for isolating the mononuclear cells from the mouse CNS in neurologic disease models by using an optimized, enzymatic tissue dissociation and a density gradient centrifugation; as well as, a method for differentiating the microglia and MDM populations in the CNS by using flow cytometry.
Another approach is to eliminate the myelin and purify the cells by using magnetic beads conjugated to specific antibodies19,20,21. Myelin removal using anti-myelin magnetic beads is more expensive and affects the viability and the yield of isolated cells22. This step and the following immunomagnetic separation of the microglia, limit further studies of specific immune cell populations21,22.
These procedures provide an easy way to study the macrophage subpopulations in disease development, and to determine the drug effects or gene modifications on macrophage phenotypes and activation states.
All methods described here have been approved by the Institutional Animal Care and Use Committee at the ICM Institute and by the Darwin French Ethic Animal Committee, and are covered under the protocol 01407.02.
1. Preparation
2. Perfusion and Dissection
3. Cell Dissociation
4. Density Gradient
5. Cell Labeling for Flow Cytometry
After the density gradient centrifugation and antibody staining, the cells were acquired on a flow cytometer and analyzed using a morphological gating strategy as follows. A first gate was defined in the dot plot Forward-Scattered-Area (FSC-A) versus Forward-Scattered-Height (FSC-H) to discriminate single cells from doublets (Figure 3A). The single cells were then gated on FSC-A versusSide-Scattered-Area (SSC-A) dot plots to exclude cell debris and pyknotic cells, based on the relative cell size and cell granularity (Figure 3B).Then, the microglia and the MDM were identified based on their expression levels of the CD11b and CD45 markers (Figure 3C, 3D). Finally, other expression markers were analyzed for each macrophage subpopulation (Figure 3E-3G). The microglia were reported to express the CX3CR1 but not the CCR2 marker in contrast to MDM23. Recently, it was demonstrated that Tmem119 is a specific marker for microglia19. The following markers (Tmem119, CX3CR1, and CCR2) were assessed by flow cytometry using detector antibodies specific for each marker; all CD11b+CD45med cells were shown to be positive for the Tmem119 and CX3CR1 markers but not for the CCR2 marker, which validates this gating strategy (Figure 3E-3G).
In the absence of the density gradient isolation, the macrophage subpopulations could not be defined in the dot plot FSC-A versus SSC-A due to the presence of many cells and myelin debris (Figure 3H). Of note, the permeabilization treatment induces a cell shrinkage and thus the macrophage subpopulations appeared smaller in the FSC versus SSC dot plot (Figure 3I, J) as previously described24. The viability of the cells was determined using the fixable dead cell stain kit. After the density gradient, 85-95% of cells are alive (Figure 3K). The microglia and MDM populations can be defined by the CD45 and CD11b markers on a dot plot after isolation of the brain and spinal cord cells in a single sample (Figure 3L), and also from the brain or spinal cord alone (Figure 3M, 3N).
For flow cytometry labeling, it is important to know whether the protein and/or the epitope recognized by the antibody is intra- or extracellular. In the case of intracellular antigens, a permeabilization buffer is required. Tmem119 is a transmembrane protein but the epitope recognized by the antibody stains as intracellular19. In the absence of cell permeabilization, there was no labeling (Figure 3O). The absence of this permeabilization step could lead to false negative results. Since the Tmem119 antibody is not directly conjugated to a fluorophore, the cells were labeled with a secondary antibody; the negative control was secondary antibody alone (black line) (Figure 3E, 3O-3R). When directly conjugated antibodies are used, there are two main strategies to assess specific labeling: isotype antibodies as a negative control or Fluorescence Minus One (FMO) controls. FMO controls contain every labeling antibody in the mixture except the control antibody for that sample. Thus, an FMO control is necessary for each marker, while only one labeling mixture is needed for the isotype antibodies control, per sample. For the labeling of the brain immune cells, we chose the isotype antibodies control since a limited number of cells (400,000 cells) is obtained after the density gradient from one mouse brain.
Figure 1: Steps of the Cell Dissociation Procedure. (A) The tissue was placed in a tube with digestion cocktail enzymes. (B) The tissue was finely cut into small pieces with scissors. (C) The small pieces of tissue were incubated for 30 min (37 °C) for efficient tissue digestion. Please click here to view a larger version of this figure.
Figure 2: Schema of the Density Gradient After Centrifugation Step. After centrifugation, myelin is present at the top of the density gradient and immune cells containing macrophage populations are found at the interphase between the 37% and 70% density gradient layers. Please click here to view a larger version of this figure.
Figure 3: Representative Flow Cytometry Analysis of Macrophages. A logical gating strategy was applied: (A) The doublet discrimination refers to dot plot FSC-A versus FSC-H. (B) The morphological gating strategy refers to dot plot FSC-A versus SSC-A. (C) The macrophage subpopulations refers to dot plot CD45 versus CD11b. (D) The dot plot of cells stained with isotype antibodies as a control for CD45 and CD11b staining. The analysis of the markers expressed by these two macrophage subpopulations: Tmem119 (E), CX3CR1 (F), and CCR2 (G) (10,000 events in gate P2 were acquired). Dot plots showing the cell size (FSC-A) and granularity (SSC-A) of the brain and spinal cord cells before the density gradient (H) or after the density gradient isolation and with permeabilizing treatment (I) or without (J). (K) Flow cytometry histogram illustrates the percentage of dead cells after the density gradient. Gates shown in the dot plots illustrate the CD11b+CD45med microglia population and the CD11b+CD45high MDM population in the brain and spinal cord (L) or in the brain (M) or the spinal cord (N) separately. Analysis of Tmem119 expressed by the macrophage subpopulations in the brain and spinal cord cells permeabilized (O) or non-permeabilized (P), and in the brain (Q) or the spinal cord (R) permeabilized cells. Please click here to view a larger version of this figure.
Volume for 1 tube (mL) | Density gradient medium 1x | PBS 10x | PBS 1x |
Density gradient medium 30% | 1.5 | 0.15 | 3.35 |
Density gradient medium 37% | 1.85 | 0.185 | 2.965 |
Density gradient medium 70% | 3.5 | 0.35 | 1.15 |
Table 1: Composition of the Different Percentages of the Density Gradient Medium.
It has been demonstrated that the microglia and MDM have different functions and phenotypes in the CNS, and thus the identification and the analysis of these macrophage subpopulations are essential in order to better understand neurological diseases9,18,25. Flow cytometry analysis using two markers (CD11b and CD45) allows for the distinction between each subpopulation (Figure 3C). This strategy was previously validated by using other specific markers, such as CCR2 for the MDM and CX3CR1 for the microglia, and a recently developed approach with an antibody to Tmem119 (Figure 3E–3G). We performed experiments with CCR2 and CX3CR1 antibodies (Figure 3F, 3G) as we previously published18. The CD11b+CD45med population highly expressed Tmem119 and CX3CR1 but not CCR2 in contrast to the CD11b+CD45high population. The CD11b+CD45high population is composed of various subpopulations with or without CCR2 expression. The identification and characterization of these macrophage subpopulations (perivascular macrophages, infiltrating macrophages, etc.) are an extensive area of research. Furthermore, the markers for these populations can change depending on the pathological processes. Indeed, CCR2 is down-regulated when the monocytes cross the blood brain barrier26.
The CD45 and CD11b antibodies recognized the cell surface antigens, so the permeabilization treatment is unnecessary. Therefore, this technique could also be used to purify each macrophage subpopulation by cell sorting for mRNA analysis or other functional assays. It is of note that the use of the enzyme and density gradient can induce, to a certain extent, microglia activation. We have previously observed that after isolation, the microglia and macrophages express the inducible nitric oxide synthase (iNOS). Nevertheless, increased iNOS expression level observed in the pathologic versus the normal condition can be highlighted using this method18. Furthermore, this technique achieves a well-separated cell population (singlets) with a high proportion of live cells.
For the successful replication of this protocol, there are several critical parameters. The setup of the gradient is most critical. Our experience indicates that preparing the gradients in 15 mL tubes versus 50 mL ones achieves better separations. If no cells are seen in the 70-37% interface, the amount of CNS tissue used for cell isolation should be considered. We usually collect ~400,000 cells from one mouse brain. Although this yield fluctuates from various experimental conditions, one for a better visualization of the 70-37% interface is to make the 37% density gradient layer with PBS and phenol red.
With the help of different lasers and filters in the flow cytometer, this protocol can quantify several markers expressed by each of the macrophage subpopulations18. This technique is also useful for assessing the involvement of other immune cells such as T cells and B cellsin the same sample. Indeed, techniques using immunomagnetic enrichment only allow for the study of the populations previously selected21.
Overall, this technique constitutes a useful tool to characterize the different macrophage subpopulations in mouse models of neurological diseases. This unique alternative can assess the effects of treatment on inflammatory processes in pathological conditions.
The authors have nothing to disclose.
This work was supported by grants from Agence Nationale pour la Recherche (ANR-12-MALZ-0003-02-P2X7RAD), Association France Alzheimer and Bpifrance. Our laboratory is also supported by Inserm, CNRS, Université Pierre et Marie-Curie and the program "Investissements d'avenir" ANR-10-IAIHU-06 (IHU-A-ICM). We would like to thank the assistance of the CELIS cell culture core facility.
5-month-old Mice | Janvier | C57BL/6J | |
Liberase TL Research Grade | Sigma-Aldrich | 5401020001 | Digestion enzyme |
Deoxyribonuclease I from bovine pancreas | Sigma-Aldrich | DN25 | |
Percoll | GE Healthcare Life Sciences | 17-0891-01 | Density gradient medium |
Cell Strainer size 70 µm Nylon | Corning | 731751 | |
Venofix 25G | BRAUN | 4056370 | |
Piston syringe 10 mL | Terumo | SS+10ES1 | |
Pasteur pipette 230 mm | Dustcher | 20420 | |
1,5 mL tube | Eppendorf | 0030 123.328 | |
15 mL tube | TPP | 91015 | |
50 mL tube | TPP | 91050 | |
5 mL polystyrene round bottom tube | BD Falcon | 352054 | |
D-PBS (1X) without Ca2+/Mg2+ | Thermo Fisher Scientific | 14190-094 | |
D-PBS (10X) without Ca2+/Mg2+ | Thermo Fisher Scientific | 14200-067 | |
Fetal bovine serum | Thermo Fisher Scientific | 10270-106 | |
EDTA | Sigma-Aldrich | E4884 | |
Bovine Serum Albumin solution 30% | Sigma-Aldrich | A7284 | |
Paraformaldehyde 32% Solution | Electron Microscopy Sciences | 15714-S | Caution -Toxic |
Saponin | Sigma-Aldrich | S2002 | |
Sodium Azide | Sigma-Aldrich | 47036 | |
PerCPCy5.5 Rat anti-mouse CD11b (clone M1/70) | eBioscience | 45-0112 | |
Rat IgG2b K Isotype Control PerCP-Cyanine5.5 | eBioscience | 45-4031 | |
BV421 Rat anti-mouse CD45 (clone 30-F11) | BD Biosciences | 563890 | |
BV421 Rat IgG2b, κ Isotype Control RUO | BD Biosciences | 562603 | |
Rabbit anti-mouse TMEM119 (clone28-3) | Abcam | ab209064 | |
AlexaFluor 647 Donkey anti-rabbit IgG | Life Technologies | A31573 | |
Anti-Mouse CD16/CD32 Purified | eBioscience | 14-0161 | Mouse Fc Block |
Fixable Dead Cell Stain Kits | Invitrogen | L34969 | |
Mouse CCR2 APC-conjugated Antibody | R&D | FAB5538A | |
Rat IgG2B APC-conjugated Isotype Control | R&D | IC013A | |
Mouse CX3CR1 PE-conjugated Antibody | R&D | FAB5825P | |
Goat IgG PE-conjugated Antibody | R&D | IC108P | |
Centrifuge | Eppendorf | 5804R | |
Cell analyzer | BD Biosciences | BD FACSVERSE | |
Data Analysis Software | FlowJo LLC | FlowJo | |
Fine scissors | F.S.T | 14090-11 | |
Standard Pattern Forceps | F.S.T | 11000-13 | |
Mayo Scissors | F.S.T | 14010-15 | |
Dumont #5 Forceps | F.S.T | 11251-20 |