Here, we describe a simple protocol for the isolation and staining of murine bone marrow cells to phenotype hemopoietic stem and progenitor cells along with the supporting niche endothelial and mesenchymal stem cells. A method to enrich cells located in endosteal and central bone marrow areas is also included.
The bone marrow (BM) is the soft tissue found within bones where hematopoiesis, the process by which new blood cells are generated, primarily occurs. As such, it contains hematopoietic stem and progenitor cells (HSPCs), as well as supporting stromal cells that contribute to the maintenance and regulation of HSPCs. Hematological and other BM disorders disrupt hematopoiesis by affecting hematopoietic cells directly and/or through the alteration of the BM niche. Here, we describe a method to study hematopoiesis in health and malignancy through the phenotypic analysis of murine BM HSPCs and stromal niche populations by flow cytometry. Our method details the required steps to enrich BM cells in endosteal and central BM fractions, as well as the appropriate gating strategies to identify the two key niche cell types involved in HSPC regulation, endothelial cells and mesenchymal stem cells. The phenotypic analysis proposed here may be combined with mouse mutants, disease models, and functional assays to characterize the HSPC compartment and its niche.
Flow cytometry is an invaluable method to characterize and prospectively isolate immune and hematopoietic cells. It is also increasingly being used to analyze stromal and epithelial populations of different tissues. The hematopoietic stem cell (HSC) has unique properties of self-renewal and multipotency. In adult mammals, HSCs primarily reside in the bone marrow (BM), where they receive quiescence and survival signals from the surrounding microenvironment or niche1. HSCs are formally defined according to functional assays2. Nevertheless, several landmark papers have shown the usefulness of flow cytometry to identify HSCs. Through the use of limited cell surface markers, it is possible to discriminate hematopoietic populations that are highly enriched in HSCs3. Flow cytometry is, therefore, a central method in the stem cell field. It has been extensively used to evaluate the impact of putative niche cell types and niche factors on HSCs. By combining flow cytometry with imaging and functional assays, it has been shown that HSCs are critically supported by perivascular mesenchymal stem cells (MSCs) and endothelial cells (ECs). BM MSCs are a heterogenous group and have different cytokine contributions4, but it is well established that leptin receptor (LepR)+ MSCs are key niche cells1. BM ECs are also highly heterogeneous and can be part of sinusoids, arterioles, and type H/transitional vessels5. Different studies have shown the nuanced contribution of these different ECs. For example, endosteal sinusoidal ECs are spatially closer to quiescent HSCs6, while non-migratory HSCs with lower levels of reactive oxygen species are located near arteriolar ECs7. The endosteal versus central location of niches is also very important. Endosteal type H vessels are associated with perivascular stromal cells that are lost with aging, leading to the loss of HSCs8. In acute myeloid leukemia, central ECs are expanded, while endosteal vessels and endosteal HSCs are lost9.
Most studies in the field have focused on hematopoiesis itself and on the cell's extrinsic regulation of HSCs. It has been, however, increasingly recognized that there is a need to better characterize the niches that regulate other progenitors, namely multipotent progenitors (MPPs), particularly considering that they are the main drivers of hematopoiesis in steady state10. In contrast with a fixed hierarchical structure, recent studies have shown that hematopoiesis is a continuum in which HSCs differentiate into biased MPPs at an early stage11. MPPs have been named after different classification schemes12, but a recent consensus paper by the international society for experimental hematology (ISEH) proposed MPPs to be discriminated as early MPPs and according to their lymphoid (MPP-Ly), megakaryocytic and erythroid (MPP-Mk/E), and myeloid (MPP-G/M) bias13. The use of flow cytometry will be critical in further studying the importance of BM niches in the regulation of these populations. Current flow cytometry methods use variable gating strategies to differentiate HSPCs and identify stromal cells, namely ECs, using inconsistent markers. The goal of the current method is to present a simple and reproducible workflow of BM staining to identify HSPC subpopulations, heterogeneous groups of ECs, and LepR+ MSCs. We believe this technique, although comparable with previously reported methods (see, for example, reference14), provides an updated and easy-to-implement protocol for the phenotypic analysis of hematopoietic cells in the two functional marrow areas, endosteal and central BM8,15, as well as BM stromal niche cells.
The animals used in this protocol were housed at the i3S animal facility under specific pathogen-free conditions in a 12 h light-dark cycle and temperature-controlled environment. Free access to standard rodent chow and water was provided. All the animals received humane care according to the criteria outlined by the Federation of European Laboratory Animal Science Associations for the care and handling of laboratory animals (EU Directive 2010/63/EU). The experimental procedure performed on the animals (euthanasia) was approved by the i3S Animal Ethics Committee (ref. DD_2019_15) and the Direção-Geral de Alimentação e Veterinária. Details of the materials used throughout this protocol can be found in the Table of Materials.
1. Preparation of solutions and staining cocktails
2. Sample extraction
3. Sample processing for analysis of hematopoietic cell populations in total bone marrow
4. Sample processing for the analysis of stromal cell populations in total bone marrow
5. Sample processing for the analysis of hematopoietic cell populations in crushed and flushed BM
6. Preparation of single-color controls (SCCs) for flow cytometry analysis
Representative plots of flow cytometry analysis of HSCs and MPPs in a healthy young adult C57Bl/6 mouse are shown in Figure 1. The gating strategy follows the latest harmonizing nomenclature proposed by the ISEH13. When analyzing the impact of a perturbation, such as infection or cancer, it is important to use a control mouse as a reference for normal gates. Fluorescence-minus-one (FMO) controls can be particularly useful to delineate the boundaries of the gates, but it should be noted that blindly setting up gate limits based on FMOs may omit target cells or include non-target cells. For example, the lower the intensity of the CD48 set, the higher the enrichment for quiescent long-term HSCs17. Furthermore, different properties of certain populations rely on the intensity of specific markers. For example, HSCs expressing higher CD150 levels are more myeloid-biased18.
The frequencies and absolute numbers obtained in the analysis of hematopoietic cell populations applied to a total of four animals are presented in Table 1.
Figure 2 shows representative plots of flow cytometry analysis of ECs and MSCs. The majority of hematopoietic cells are excluded following a Ter119/CD45 negative selection. LepR+ MSCs can then be readily identified, while true ECs require a subsequent selection of Sca-1+ cells. While Sca-1 is often used in immunofluorescence studies to specifically mark arterioles, this is not the case in flow cytometry, as this technique is highly sensitive, and ECs always express a certain (even if low) degree of Sca-1 at the cell surface. By not selecting only Sca-1+ cells, other non-endothelial cells would be included in the analysis, such as CD31-expressing myeloid cells, which might significantly impact the quantification of ECs in the BM. ECs can be analyzed as a whole population or based on specific markers that partially reveal their heterogeneity. Endomucin in combination with CD31 is very useful to identify the functionally distinct type H endothelium15 (Figure 2). Furthermore, it has been previously shown that ICAM-1 expression allows the discrimination between arteriolar (aBMECs) and sinusoidal ECs (sBMECs) (Figure 3)20.
The frequencies and absolute numbers obtained in the analysis of MSCs and endothelial cell populations applied to a total of four animals are presented in Table 2 and Table 3.
Although BM functional areas are not clearly delineated in evident histological regions, it is well established that bone-lining endosteal areas and central BM areas are enriched in certain cell types and events (e.g., the release of platelets/pro-platelets from megakaryocytes in the sinusoids of central BM areas). It is, therefore, useful to enrich for cell types in central and endosteal areas to separately analyze these two compartments. We applied a method of flushing the diaphysis to enrich for central cell types and crushing the metaphysis to enrich for endosteal cell types (Figure 4A). The quantification of aBMECs and sBMECs in flushed (central) and crushed (endosteal) BM tissue (Figure 4B) validates the applicability of this mechanical isolation method, as aBMECs are notoriously more abundant in endosteal regions, and sinusoids are more frequent in central BM areas.
Figure 1: Analysis of hematopoietic cell populations. Representative plots showing the gating strategy for the flow cytometry analysis of hematopoietic cell populations in total BM using the software provided. Abbreviations: FSC-H = forward scatter – height, FSC-A = forward scatter – area, Lin = lineage, LKS = Lin– Sca-1+ c-Kit+ hematopoietic progenitor cells, MPPsLy = multipotent progenitors – lymphoid, MPPsG/M = multipotent progenitors – granulocyte/macrophage, MPPsMk/E = multipotent progenitors – megakaryocyte/erythrocyte, MPPs = multipotent progenitors, HSCs = hematopoietic stem cells. Please click here to view a larger version of this figure.
Figure 2: Analysis of stromal cell populations. Representative plots showing the gating strategy for the flow cytometry analysis of stromal cells in total BM using the software provided. Abbreviations: FSC-H = forward scatter – height, FSC-A = forward scatter – area, MSCs = mesenchymal stem cells, LepR = leptin receptor, ECs = endothelial cells. Please click here to view a larger version of this figure.
Figure 3: Analysis of endothelial cells. Representative plots showing the gating strategy for the flow cytometry analysis of ECs in total BM using the software provided. Abbreviations: FSC-H = forward scatter – height, FSC-A = forward scatter – area, ECs = endothelial cells, aBMECs = arterial bone marrow endothelial cells, sBMECs = sinusoidal bone marrow endothelial cells. Please click here to view a larger version of this figure.
Figure 4: Analysis of EC populations in central and endosteal areas. (A) Representation of the different parts of a murine long bone. (B) Plots showing the statistically significant differences in the frequencies of CD31+ endothelial cells in aBMECs (enriched in the crushed sample) and sBMECs (enriched in the flushed sample). Each dot represents a mouse (n = 9), and samples are paired. The statistical test used was the paired T-test. **** denotes two-tailed P-value < 0.0001. Please click here to view a larger version of this figure.
Cell population | Mean of freq. of BM live cells ± SD (n=4) in total BM | Mean of cell count ± SD (n=4) per femur in total BM | Mean of freq. of BM live cells ± SD (n=4) in endosteal BM | Mean of freq. of BM live cells ± SD (n=4) in central BM |
Lin– | 3.05 ± 0.13 | 337610 ± 59414 | 3.56 ± 0.21 | 3.53 ± 0.19 |
LKS | 0.096 ± 0.026 | 10206 ± 1794 | 0.102 ± 0.025 | 0.133 ± 0.022 |
MPPsLy | 0.058 ± 0.011 | 6141 ± 716 | 0.058 ± 0.011 | 0.085 ± 0.007 |
MPPsG/M | 0.011 ± 0.004 | 1233 ± 326 | 0.013 ± 0.003 | 0.014 ± 0.003 |
MPPsMk/E | 0.0010 ± 0.0002 | 113 ± 21 | 0.0014 ± 0.0004 | 0.0012 ± 0.0005 |
MPPs | 0.0092 ± 0.0045 | 940 ± 374 | 0.0105 ± 0.0048 | 0.0118 ± 0.0061 |
HSCs | 0.0054 ± 0.0023 | 572 ± 191 | 0.0055 ± 0.0023 | 0.0059 ± 0.0016 |
Table 1: Quantitative data for hematopoietic cell populations. Mean of frequency in BM live cells of the different hematopoietic cell populations analyzed in total, endosteal, and central BM and cell counts per femur in total BM. Data shown as mean ± standard deviation (SD), n = 4.
Cell population | Mean of freq. of BM live cells ± SD (n=4) | Mean of cell count ± SD (n=4) per tibia |
ECs | 0.080 ± 0.011 | 4523 ± 1521 |
Type H ECs | 0.041 ± 0.006 | 2299 ± 752 |
Type L ECs | 0.038 ± 0.005 | 2103 ± 736 |
MSCs | 0.180 ± 0.048 | 10270 ± 4282 |
Table 2: Quantitative data for stromal cell populations. Mean of frequency in BM live cells and cell counts per tibia for the different stromal cell populations analyzed in total BM. Data shown as mean ± standard deviation (SD), n = 4.
Cell population | Mean of freq. of BM live cells ± SD (n=4) | Mean of cell count ± SD (n=4) per tibia |
ECs | 0.064 ± 0.006 | 2255 ± 150 |
aBMECs | 0.041 ± 0.010 | 1419 ± 286 |
sBMECs | 0.023 ± 0.006 | 819 ± 250 |
Table 3: Quantitative data for endothelial cell populations. Mean of frequency in BM live cells and cell counts per tibia for the different endothelial cell populations analyzed in total BM. Data shown as mean ± standard deviation (SD), n = 4.
Supplementary Table 1: Details of the animals used in the study. Strain, sex, age, and weight of the animals used to produce the data shown in Figure 1, Figure 2, and Figure 3 and Table 1, Table 2, and Table 3. Abbreviation: g = grams. Please click here to download this File.
While the protocol described is simple and easy to perform, special attention should be brought to specific steps. For example, when obtaining flushed BM (step 5.2), the volume or number of times indicated to pass PBS 2% FBS through the inside of the central part of the bone should not be exceeded, as this might result in significant contamination of the flushed sample by endosteal cell populations.
Alterations to the protocol can be made to facilitate its execution by the investigator. In sample extraction (section 2), legs and/or bones can be stored in PBS 2% FBS at 4 °C for up to 24 h before proceeding to sample processing and analysis. However, this time should be minimized when possible. During incubation with HSCs primary staining cocktail (step 3.10) and secondary staining cocktail (step 3.11), while a 15 min incubation at RT is indicated, this can be exchanged for a 30 min incubation at 4 °C.
The same applies to incubation with stromal primary staining cocktail (step 4.7) , stromal secondary staining solution (step 4.8), ECs staining cocktail (step 4.12) and incubation with antibodies for SCCs (step 6.5); while a 15 min incubation at RT is indicated, this can be exchanged for a 30 min incubation at 4 °C.
Another protocol for BM isolation by centrifugation of murine long bones was recently described21. While this protocol presents the advantage of faster isolation of the BM cells, the protocol described here allows for the analysis of cell populations residing in different areas of the bones.
A particular limitation of the current method is that it is solely based on phenotypic analysis by flow cytometry. This is particularly relevant in the case of HSCs, which would require further functional validation. This method can, however, be combined with the sorting of enriched populations and the study of these cells in long-term reconstitution assays and in vitro colony assays.
Regarding stromal cell analysis, in particular MSCs and ECs, it has been previously shown that, despite the optimization of BM processing for flow cytometry analysis, a significant number of cells are not isolated from the tissue and there is a suboptimal quantification of these cells when compared with other methods such as whole-mount imaging19. Nevertheless, flow cytometry is extremely useful to perform the quantitative and qualitative analysis of BM ECs and MSCs and to prospectively isolate them for functional and expression assays.
Another limitation of the current method is the use of a mechanical technique to separate the endosteal and central fractions, which are to some extent inevitably contaminated by cells from the other compartment. Nevertheless, the quantification of aBMECs and sBMECs as explained in the representative results section shows that the mechanical isolation is a robust method to enrich for cell populations in these areas.
The described method allows for the study of hematopoiesis in different settings and stromal niches of HSPCs. The phenotypic analysis here presented may be useful to study HSPC niches when combined with specific mouse mutants, namely EC and MSC Cre lines that enable the selective manipulation of gene expression in these populations. For example, the Cdh5(PAC)-CreERT222 and the LepR-Cre23 lines can be used to induce the expression of certain genes selectively in ECs and MSCs, respectively. The method is useful to study their impact on the stromal compartment as well as on HSPCs. Available Cre lines to study the vascular BM niche have been recently reviewed in Mosteo et al.5. The prospective study of LT-HSCs without transplantation has been limited by the lack of robust reporter lines. The recently described Mds1GFP/+Flt3Cre 6 has been, however, shown to achieve high enrichment of LT-HSCs and good discrimination from downstream progenitors expressing Flt324. The combination of this mouse line and other hematopoietic strains, such as Vav-iCre25, with flow cytometry will allow the study of HSPCs and their relationship with the niche. Hematological malignancies are frequently associated with the disruption of hematopoiesis from the earliest stages, which is reflected in the alteration of the frequencies of hematopoietic stem and progenitor cells and of stromal populations9 that we have here presented for healthy C57Bl/6. Future studies should explore the application of similar methods to the one here described using new equipment based on spectral flow cytometry that enables a more extensive phenotypic characterization through the analysis of more markers (above 30 simultaneously).
The authors have nothing to disclose.
LM was supported by a grant from the Lady Tata Memorial Trust. JR was supported by a PhD fellowship from Fundação para a Ciência e Tecnologia (FCT; FCT fellowship UI/BD/150833/2021). ML was supported by a PhD fellowship from FCT (FCT fellowship 2021.04773.BD). DD was supported by grants from the American Society of Hematology, the Pablove Foundation, FCT (EXPL/MED-ONC/0522/2021), and the Portuguese Society of Hematology. We thank the support from Dr. Catarina Meireles and Emilia Cardoso of TRACY facility at i3s.
Alexa Fluor 647 anti-mouse CD54/ICAM-1 antibody | BioLegend | 116114 | |
APC Streptavidin | BioLegend | 405207 | |
APC/Cyanine7 anti-mouse CD117 (c-kit) antibody | BioLegend | 105826 | |
APC/Cyanine7 anti-mouse CD45 antibody | BioLegend | 103116 | |
APC/Cyanine7 anti-mouse TER-119/erythroid cells antibody | BioLegend | 116223 | |
Biotin anti-mouse CD3ε antibody | BioLegend | 100304 | |
Biotin anti-mouse CD4 antibody | BioLegend | 100404 | |
Biotin anti-mouse CD8a antibody | BioLegend | 100704 | |
Biotin anti-mouse Ly-6G/Ly-6C (Gr-1) antibody | BioLegend | 108404 | |
Biotin anti-mouse TER-119/erythroid cells antibody | BioLegend | 116204 | |
Biotin anti-mouse/human CD11b antibody | BioLegend | 101204 | |
Biotin anti-mouse/human CD45R/B220 antibody | BioLegend | 103204 | |
Brilliant Violet 510 anti-mouse CD150 (SLAM) antibody | BioLegend | 115929 | |
Calibrite 2 Color Beads | BD Biosciences | 349502 | |
Collagenase IV | Merck Life Science | C1889 | |
Dispase II | Merck Life Science | D4693 | |
Fetal Bovine Serum, qualified, heat inactivated, E.U.-approved, South America Origin | ThermoFisher Scientific | 10500064 | |
Hanks' Balanced Salt Solution (HBSS) | ThermoFisher Scientific | 14175095 | |
Mouse Leptin R Biotinylated Antibody | R&D systems | BAF497 | |
NucBlue Fixed Cell Reagent (DAPI) | ThermoFisher Scientific | R37606 | DAPI reagent |
PE anti-mouse endomucin antibody | ThermoFisher Scientific | 12-5851-82 | |
PE anti-mouse Flk2 (CD135) | ThermoFisher Scientific | 12-1351-82 | |
PE/Cyanine7 anti-mouse CD31 antibody | BioLegend | 102524 | |
PE/Cyanine7 anti-mouse CD48 antibody | BioLegend | 103424 | |
PerCP anti-mouse Ly-6A/E (Sca-1) antibody | BioLegend | 108122 | |
Phosphate-buffered saline (PBS) tablets | Merck Life Science | P4417 | |
Purified anti-mouse CD16/32 antibody | BioLegend | 101302 | |
RBC lysis buffer 10x | BioLegend | 420302 | |
Zombie Violet Fixable Viability Dye | BioLegend | 423114 | fluorescent dye |