Immunophenotypic Characterization of Normal Bone Marrow Stem Cells

The recent advances in flow cytometry technology and the emergence of new high-speed flow cytometers have given a valuable contribute to diminish this problem in two different (but complementary) aspects: 1) by reducing dramatically the acquisition time period, making it more reasonable to study minor cell populations; and 2) by increasing the number of parameters that can be analyzed per cell at the same time, which is critical to improve the immunophenotypic characterization of those not-well characterized cell populations that lack a specific known marker.

By definition, a stem cell is an undifferentiated cell with the potential ability of self-renewal and the capability of differentiation along different cell lineages (multipotency).MSC can be found on a great variety of adult tissues, where they play an important role in tissue regeneration, such as: bone marrow, adipose tissue, umbilical cord blood, umbilical cord matrix, menstrual blood, endometrium, placenta, dental pulp, skin and thymus, among others (Chamberlain et al., 2007;Ding et al. 2011;Kolf et al., 2007;Martins et al., 2009;Musina et al., 2005;Pittenger et al., 1999).
In addition to their presence in numerous adult tissues, MSC are relatively easy to isolate and have the capability to expand manyfold in culture without lose their stem cell properties.Moreover, when MSC are systemically transplanted, they are able to migrate to sites of injury and promote tissue repair, by producing growth factors or other soluble factors important to tissue regeneration, as well as by undergoing cellular differentiation (Chamberlain et al., 2007, Kolf et al., 2007;Mafi et al., 2011); such features explain the success of MSC transfusion therapy in genetic disorders affecting mesenchymal tissues (Horwitz et al., 2002;Undale et al., 2009).Furthermore, those cells have the ability of suppressing the immune response of a wide variety of immune cells, including T, B and NK lymphocytes, and antigen-presenting cells (Chamberlain et al., 2007;Stagg, 2007), and their importance in patients' clinical outcome has already been proven in severe acute graft-versus-host disease (Remberger et al., 2011;von Bahr et al., 2011).Moreover, the results achieved in animal models of autoimmune diseases are promising and encouraged the beginning of phase I clinical trials in multiple sclerosis (Constantin et al., 2009;Darlington et al., 2011;Siatskas et al., 2009).

Identification and quantification of bone marrow MSC
As referred previously, the study of minor cell populations with no known specific cell marker toke great advantage on the development of high-speed multi-parameter flow cytometers.The use of an 8-color FACSCanto II (Becton Dickinson Biosciences, BDB) flow cytometer allowed us to identify MSC in bone marrow, quantify them and further characterize their immunophenotypic profile.We employed a monoclonal antibody panel with a backbone of 3 common markers (CD13, CD45 and CD11b) for the identification of MSC (known to be CD13+CD45-CD11b-) in each tube that, at the same time, permitted the study of the expression of five more proteins on MSC per tube.
MSC are rare in bone marrow, being reported that they represent approximately 0,01% of all nucleated bone marrow cells (Chamberlain et al., 2007;Mafi et al., 2011), although is known that their number declines with aging (Caplan, 2007).Our data point to a percentage ranging between 0,01% and 0,03% of all nucleated bone marrow cells (Martins et al., 2009).

Flow cytometer quality control, compensation setup strategies and other technical issues
According to the manufacturer's recommendations, it is done a daily quality control using the Rainbow Beads (BDB).In what concerns to cytometer's compensation setup, it is made once per month by setting up the Rainbow Beads (BDB) values according to the EuroFlow consortium's guidelines and then by doing a general compensation for stable fluorochromes and a specific compensation for each monoclonal antibody conjugated with tandem fluorochromes.Although the compensation is automatic, it is always revised by experienced staff at the end of the process.
In order to detect cellular autofluorescence, a negative control was made for each sample, where the bone marrow sample was only stained for CD45 PO and CD34 PerCPcy5.5.
SSC and FSC light dispersion properties allow a good discrimination between viable and dead cells and the doublets were excluded based on FSC-Area versus FSC-Height characteristics.

Material and methods
The immunophenotypic characterization of bone marrow MSC were performed in fresh EDTA-collected bone marrow samples from healthy individuals.After collection the samples were stored at 4 ºC and processed within 24 hours.
Whole bone marrow samples were stained for surface cell markers using a stain-lyse-andthen-wash direct immunofluorescence technique.200 μl of whole bone marrow were aliquoted in different tubes and stained with the following combinations of monoclonal antibodies in an 8-color staining protocol, detailed in table 1.Data acquisition was performed in a FACSCanto II flow cytometer (BDB), using FACSDiva acquisition software (BDB).The total bone marrow cellularity of the whole sample was acquired (5 x 10 6 events, minimum) for each tube.Bone marrow MSC were identified as CD13 + /CD45 -/CD11b -, as shown in Figure 1.
Several studies on adhesion molecules and chemokine receptors expression have been made in order to shed light on MSC migratory and homing ability.CD29 (integrin β 1 -subunit) and CD106 (vascular cell adhesion molecule 1, VCAM-1) seem to be important in the adhesion of MSC to endothelial cells (Chamberlain et al., 2007;Kolf et al., 2007;Stagg, 2007) and CD29, which when dimerized with CD49e (integrin α 5 -subunit) forms a receptor that binds to fibronectin and invasin, is likely to promote MSC-extracellular matrix interaction (Gu et al., 2009).CD146 (Muc18) plays an important role in cell-cell and cell-extracellular matrix adhesion and an increased expression of these marker on tumor cells is associated with an increased cell motility and invasiveness/ metastasis capability (Bardin et al., 2001;Zeng et al., 2011).The glycoprotein CD90 (Thy-1) regulates as well cell-cell and cell-extracellular matrix interactions, being involved in adhesion to endothelial cells, migration, metastasis and tissue regeneration (Jurisic et al., 2010;Rege & Hagood, 2006).
The enzyme CD73 is an ecto-5'-nucleotidase that produces extracellular adenosine.In animal tumor models, CD73-generated adenosine inhibits both homing and expansion of T cells via adenosine-receptor signaling.In fact, recent research shows that adenosine suppresses T cell immune response both in activation and effector phases, as well as NK cell immune activity (Wang et al., 2011;Zhang et al., 2010).In what concerns to growth factor receptors, NGFR (nerve growth factor receptor, CD271) is expressed in a wide variety of tissues and, depending on the cell type, signaling through this receptor regulates NF-kB activation, apoptosis, tissue regeneration, immune cell activation, proliferation and cell differentiation (Micera et al., 2007;Rogers et al., 2010).Finally, CD105 (endoglin) is one of the receptors for TGF-β, a growth factor involved in the regulation of development, maintenance and proliferation of MSC (Stagg, 2007), and also known to play an important role in tissue repair.
Some discrepancies described in the expression of adhesion molecules, chemokine receptors and other proteins, may be the reflex of the microenvironmental differences present in different studies.Although there are a great similitude in the phenotypic profile of MSC isolated from different tissues, differences do exist (Chamberlain et al., 2007;Kolf et al., 2007;Martins et al., 2009).As well as different cultures conditions can also change the MSC phenotype (Chamberlain et al., 2007;Halfon et al., 2011;Stagg, 2007;Tormin et al., 2011).This could be a clue of MSC highly sensitiveness to microenvironment alterations, and their potential to change their protein expression profile could be of great importance in giving an appropriate response to physiological or pathological challenges: by changing their migratory pattern, by initiating an immunomodulatory or immunosuppressive response, by modifying the production and release of soluble factors, or by undergoing cell differentiation.
As a minor bone marrow cell population easy to expand in vitro, it is attractive to characterize the MSC immunophenotype after culture cell expansion.Nevertheless, characterizing these cells directly (without previous culture) enables an analysis closest to their physiological conditions, excluding the phenotypic alterations induced by factors present in the culture medium.Moreover, this direct approach allows an accurate quantification of MSC in bone marrow.Also, this same strategy can be applied to MSC from other tissues.

Bone marrow hematopoietic stem cells
The multipotent hematopoietic stem cell is mainly located in the bone marrow of adult animals and has the ability to differentiate along all hematopoietic cell lineages.A number of studies based on in vitro cell culture, xeno-transplantation of hematopoietic human cells in immunodeficient mice and in pre-immune animal fetuses, were carried out to identify the human hematopoietic stem cell and unveil the hematopoietic precursors hierarchy (Nimer, 2008;Yin et al., 2007), becoming clear that CD34-positive cells were able to differentiate and give rise to all blood cells.There are evidences that, within this heterogeneous population, the more immature CD34+ HSC expresses CD133 and are CD38-negative/dim.It is also known that the CD34+CD133+ subpopulation can arise from the CD133+CD34-CD38subset (Goussetis et al., 2006;Nimer, 2008;Yin et al., 1997).

Identification and quantification of the different bone marrow CD34+ HSC cell compartments
As already referred, CD34-positive cells are an heterogeneous bone marrow cell population, consisting in various cell compartments differing in immunophenotype, size and lineage commitment.The immunophenotypic pattern of each compartment is well described and, with a relatively low number of markers, the majority of those subsets can be accurately and easily identified.
Attending only to the immunophenotypic features, is possible to identify the following bone marrow CD34+ cell subsets by flow cytometry: uncommitted (more immature) precursors, neutrophil precursors, B cell precursors, monocytic precursors, plasmacytoid dendritic cells precursors, erythroid precursors, basophil precursors and mast cell precursors.

A single-tube protocol to identify the different bone marrow CD34+ HSC compartments
Recently, we developed an 8-color single-tube protocol to identify the different bone marrow CD34+ HSC subsets by flow cytometry.
The single-tube protocol we propose here was constructed to allow an accurate, quick and easy identification and quantification of those cellular compartments.Attending to the monoclonal antibodies and fluorochrome-conjugation available on the market and to compensation issues, and based on our experience and knowledge on the hematopoietic maturation dynamics, we elected the best markers to identify with precision the cell populations of interest.

Material and methods
The immunophenotypic characterization of bone marrow CD34+ precursors were performed in fresh EDTA-collected bone marrow samples from healthy individuals.After collection, the samples were stored at 4 ºC and processed within 24 hours.The quality control and compensation strategies are described in detail in section 2.2.1.
A stain-lyse-and-then-wash direct immunofluorescence protocol was used, and the monoclonal antibodies were combined as presented on table 3. Table 3. Panel of monoclonal antibodies used for the identification and quantification of the different subpopulations found in bone marrow CD34+ HSC Data acquisition was performed on a FACSCanto II flow cytometer (BDB), using FACSDiva acquisition software (BDB).In a first step of acquisition, the whole bone marrow cellularity was stored (100.000events).In a second step, only events within the CD34+ electronic gate were acquired (5.000 to 10.000 CD34+ events).

How to identify the different CD34+ HSC compartments with the single-tube protocol?
1.The most immature (uncommitted) compartment of bone marrow C34+ HSC The most immature compartment can be easily identified based on their positivity to CD133 marker (CD133 hi ).To differentiate this subset from CD34+ neutrophil precursors and CD34+ plasmacytoid dendritic cell precursors, also expressing CD133 (CD133 int ), other important phenotypic characteristics have to be taken into account: CD35 -/CD34 hi /HLA-DR hi /CD117 hi /FSC int /SSC int /CD123 -. Figure 3 presents a detailed immunophenotype of this compartment considering all the markers used in this protocol.3. Bone marrow CD34+ neutrophil precursors Neutrophil precursors show high reactivity to CD44 antigen, as the plasmacytoid dendritic cell precursors (CD44 hi ), but in the absence of CD123 marker.Other important immunophenotypic features of this CD34+ compartment are: CD133 int /CD35 -/HLA-DR hi /CD117 hi /CD45 int/dim /FSC hi /SSC hi (Figure 6).

Bone marrow CD34+ monocyte precursors
Using this single-tube approach, the monocyte precursors are primarily identified by exclusion of all the other myeloid CD34+ precursors.Is noteworthy that a large percentage of monocyte-committed CD34+ precursors express CD35, being discriminated from CD34+ erythroid precursors by their CD117 dim/-/HLA-DR + /CD45 hi phenotype.Although classically the identification of this CD34+ subset was made focusing on the expression of CD64, this marker seems to be also present on CD34+ plasmacytoid and myeloid dendritic cell precursors.In line with this, CD35 might be a good option to the identification of CD34+ monocyte precursors.The immunophenotype of this population is depicted in Figure 7.

Bone marrow CD34+ B cell precursors
Even in the absence of an B-cell lineage specific marker, as CD19 or CD79a, CD34+ B cell precursors are clearly identified by the low expression of CD44 and CD45, along with low light scatter properties (Figure 8).Our protocol allows the identification of basophil precursors using the classical markers and attending to the immunophenotype HLA-DR -/dim /CD123 int/hi .Of note, this CD34+ subset presents the lowest expression of CD44 among all bone marrow myeloid CD34+ cells, being easy to differentiate this precursors from all the other myeloid precursors by using CD44 marker (Figure 9).

Bone marrow CD34+ plasmacytoid dendritic cell precursors
The plasmacytoid dendritic cell precursors are identified using the classical markers, as being HLA-DR hi /CD123 hi/int .The most immature forms of this precursor express CD133 (CD133 int ).The immunophenotypic characteristics of this population are represented on Figure 10.

The maturation dynamic of bone marrow CD34+ hematopoietic stem cell
The possibility of a multiparameter analysis in a single cell basis conduct to a broader knowledge on the immunophenotypic characteristics of bone marrow CD34+ compartments and how it varies along the differentiation through different hematological cell lineages.

Conclusion
The emergence of high-speed multi-parameter flow cytometers have given an important contribute to unveil the phenotypic characteristics of minor cell populations and/or populations without a known specific cell marker.
Using flow cytometry to characterize bone marrow MSC directly (without in vitro cell culture) represents a great advantage by enabling an analysis closest to the physiologic conditions of the cells, excluding all the phenotypic alterations induced by factors present in the culture medium.Moreover, this direct analysis allows an accurate quantification of these cells in bone marrow.In addition, the strategy used for bone marrow can also be applied in MSC from other tissues, allowing their direct quantification and characterization.
A broader knowledge about the immunophenotypic characteristics of the different compartments of bone marrow HSC could improve their identification, allow a more accurate quantification of those compartments, as well as shed light on the protein expression patterns in the earliest stages of maturation of each hematological cell lineage.Furthermore, a better knowledge of those protein expression patterns might contribute to the development of new strategies to identify aberrant phenotypes in hematological diseases affecting the more immature bone marrow cells compartments, which can be helpful in the classification of acute leukemias, diagnosis of myelodysplastic syndromes and detection of minimal residual disease.A more extensive understanding of the phenotype of CD34+ hematopoietic stem cells in the different maturational stages could also be useful to monitoring and investigate if different mobilization regimens have the capability of mobilizing distinct CD34+ hematopoietic stem cells subpopulations.
Here, we presented a simple, quick and economic approach to identify and quantify the different bone marrow CD34+ HSC compartments.

Fig. 12 .
Fig. 12. Maturational dynamic of bone marrow CD34+ HSC.Uncommitted CD34+ cells are presented in red, lineage committed CD34+ cells are presented in blue and the lineage committed CD34-cells correspond to grey events , presented on table 2.

Table 2 .
Distribution of the different cell compartments of bone marrow CD34+ HSC.The results are expressed as mean ± standard deviation (range).Adapted fromMatarraz et al.Leukemia 2008