Category: PBMC Basics

One thing that every immunologist agrees on is that immune cells are heterogeneous, and human peripheral blood mononuclear cells (PBMC) contain many different immune cell types.  CD4⁺ T cells alone exist in dozens of uniquely functioning subsets.  Receptor mediated signal transduction events in response to signals such as cytokines, chemokines, various receptor ligands, and engagement of the T cell or B cell receptors represent an important aspect of immune cell communication and the many types of immune cells present in human PBMC are differentiated by these responses.  Trying to derive mechanisms from averages of complex PBMC populations obscures the understanding of the relationships that occur between expression of proteins and signaling states in unique cell populations and on the single cell level.

Dr. Gary Nolan and colleagues pioneered a method using flow cytometry to look at cell signaling in single cells using antibodies targeting phosphorylated protein sites, termed phospho-flow cytometry or “phosflow”.  In phosflow, PBMCs or other cell populations are stimulated with signaling receptor ligands and/or antagonists for a period of time.  Then the cells are fixed using paraformaldehyde-based buffers to lock the cells in their induced states of phosphorylation.  The cells are then permeabilized and stained with fluorescently-labeled antibodies against the phosphorylated proteins, antigens to identify cell populations, and other proteins of interest, and analyzed on a flow cytometer.

Using this method, our lab has assessed the response to many cytokines by analyzing phosphorylation of the STAT family of transcription factors, as well as phosphorylation events that occur downstream from T cell receptor (TCR) and B cell receptor (BCR) engagement, and Toll-like receptor ligation, such as NF-kB (p65), ERK, p38, and ZAP70.  BD Biosciences is a key resource for many of the reagents needed to perform these assays, providing several different buffers and antibodies against many important signaling targets.

There are several considerations for optimizing this protocol.  The first is selecting the proper permeabilization method.  Nolan and colleagues published a protocol in which 100% methanol is used for permeabilization following fixation, and this is the method I have used most frequently.  One key advantage with this method is that following methanol permeabilization, the cells can be stored for an extended period of time at -20ºC or -80ºC prior to staining with the antibodies and running the flow cytometry.  Thus, many samples can be stimulated on different days, but the flow cytometry analysis can be performed later in a large batch, in order to limit the variability between staining intensities that can otherwise occur between experiments.  This is very important because when analyzing phosflow data, the fluorescence intensities which translate to magnitude of the phosphorylation of the protein target is often the key factor being analyzed, although the percentage of cells able to respond to a given signal can also be an important parameter for analysis.  However, there are a number of commercially available buffer sets for these assays which should be selected based on the phospho-proteins and other antigens of interest.

Another important factor to consider when setting up these assays is selection of the target antigens for identifying cell populations of interest.  Some antigens do not survive permeabilization with some of these reagents, including methanol.  For instance, I have had poor luck trying to identify B cells using CD19 or monocytes using CD14. CD20 and CD33 have been used respectively as alternatives that work well with methanol.

Cytobank, an online resource dedicated to phosflow created by Gary Nolan’s lab, is great for assisting in selecting the proper antigens and antibody clones for phosflow.  This website lists antibodies, including the phospho-target antibodies, that have been successfully and unsuccessfully tested with BD Biosciences’ various permeabilization buffers.

In summary, phosflow is an excellent method for analyzing signal transduction events in complex cell populations.  Using single cell assays for studies on protein expression and signaling will drive our understanding of the immune system’s complexity to the next level.

 

Nolan lab Phosflow protocol:

Single-cell phospho-protein analysis by flow cytometry. Schulz KR, Danna EA, Krutzik PO, Nolan GP.Curr Protoc Immunol. 2012 Feb;Chapter 8:Unit 8.17.1-20.

 

BD Biosciences Phosflow Homepage:

http://www.bdbiosciences.com/research/phosflow/index.jsp?WT.ac=FP_Phosflow

 

Cytobank’s list of antibodies tested with various permeabilization buffers:

http://www.cytobank.org/facselect/

 

Further Reading:

Flow cytometric analysis of cell signaling proteins. Suni MA, Maino VC.  Methods Mol Biol. 2011;717:155-69.

Single-cell, phosphoepitope-specific analysis demonstrates cell type- and pathway-specific dysregulation of Jak/STAT and MAPK signaling associated with in vivo human immunodeficiency virus type 1 infection.  Lee AW, Sharp ER, O’Mahony A, Rosenberg MG, Israelski DM, Nolan GP, Nixon DF. J Virol. 2008 Apr;82(7):3702-12.

 

T cells present in human peripheral blood mononuclear cells (PBMC) are complex populations of different subtypes with unique functions.  CD4⁺ and CD8⁺ T cells are most simply classified as naïve, central memory, effector memory, and terminally differentiated effector subtypes, and T cells are currently thought to differentiate along this progression path following antigen experience.  In a previous post, I discussed using CD3, CD4, CD8, CD45RA or CD45RO, and CCR7 or CD62L as markers for differentiating T cells into these subtypes by multiparametric flow cytometry.

In addition, CD4⁺ T cells can be even further characterized as to their specific functions.  The cytokine milieu present during T cell activation through the T cell receptor (TCR) guides T cells to differentiate into memory and effector T helper subtypes (T_H) including T_H1, T_H2, and T_H17 subsets, all of which are present in PBMC.

Each of these T_H subtypes functions by expressing a unique set of cytokines and chemokine receptors in order to elicit specific responses in different tissue settings.  It is important to note that memory T cells maintain flexibility in their functionality and can be re-polarized into different subtypes if re-stimulated under an alternate T_H inducing condition.

Using multiparametric flow cytometry, each of these T_H subsets can be identified by expression of specific cytokines, transcription factors, and surface markers including chemokine receptors, although the various markers used in the literature to classify the same subset do not overlap entirely. Identification of T_H subsets by surface markers simply requires surface staining and is the easiest assay to perform.  Transcription factor expression can be assessed, which requires intracellular/nuclear permeabilization protocols such as BD Biosciences’ FoxP3 Buffer Set.  Finally, intracellular cytokine production can be detected following TCR stimulation with anti-CD3 and anti-CD28 antibodies or the combination of Phorbol 12-Myristate 13-Acetate (PMA) and the Ca2+ ionophore, ionomycin.  This generally requires 4-6 hours of stimulation in the presence of protein transport inhibitors: either brefeldin A or monensin, which are optimal for different cytokines.

T_H1 cells promote cellular immunity against viruses and intracellular bacteria.  T_H1 cells have been characterized in many ways.  Production of IFNg following TCR stimulation is probably the most common and direct functional representation to assess T_H1 in PBMC.   Alternatively, the expression of the chemokine receptors CXCR3 and CCR5 have been associated with T_H1 but not T_H2 or T_H17 cells.  The transcription factor T-bet (TBX21) is the major factor regulating T_H1 differentiation and is also a marker of T_H1 cells that can be assessed by flow cytometry, although T-bet expression is promiscuous and can be expressed in T_H17 cells.  Interestingly, expression of any of these markers may not overlap entirely with IFNg expression.  For instance, in central memory T cells, CXCR3+ cells are considered pre-T_H1 cells and do not express IFNg, while in the effector memory population, a fraction of CCR5⁺ and CXCR3⁺ cells express IFNg.  As always, care must be taken when making comparisons between percentiles and functions of populations identified using different markers.

T_H2 cells are the CD4 helper subtype that promotes immunity against extracellular pathogens and are involved in allergic inflammation.  T_H2 cells in PBMC can be classified by production of T_H2-specific cytokines including IL-4, IL-5, and IL-13 following TCR stimulation or PMA/ionomycin.  However, expression of these cytokines can be difficult to detect and expression of the surface receptors CRTH2, CCR4, or CCR3 are alternatives.  Again, expression of CRTH2, CCR4, or CCR3 does not mark all existing IL-4, IL-5, and IL-13 expressing T_H2 cells and vice versa, although CRTH2 has been demonstrated to be quite effective.

T_H17 cells are known as the subtype associated with inflammatory autoimmune diseases and protection from fungal infections.  T_H17 cells are classified by expression of IL-17 or surface marker expression, being CCR6⁺CCR4⁺.   Expression of the T_H17-specific transcription factor RORg/gt can also be assessed.  A fraction of IL-17 expressing cells can also express IFNg and T_H17 associated markers CCR6 and RORg/gt. Some have even suggested that IL-17⁺IFNg⁺ cells are a transitional state when more plastic T_H17 cells differentiate into effector T_H1 cells.

As cytokine expression may be the most discrete method of describing a T_H cell regarding its function, I have used cytokine expression as a measure of T_H1 and T_H17 cells in human PBMC.  However in the case of T_H2 cells, expression of IL-4 and IL-13 was hard to detect by flow cytometry, even with PMA and ionomycin stimulation and thus I have instead utilized CRTH2.  The protocol that I have optimized for identification of T_H1, T_H2, and T_H17 cells involves a 4 hour stimulation of PBMCs with anti-CD3 and anti-CD28 antibodies or with PMA + ionomycin in the presence brefeldin A.  Following this, I first stain for surface markers including CRTH2 to identify T_H2 cells.  Then the cells are fixed and permeabilized using BD Biosciences’ Cytofix Cytoperm buffer set and stained for IFNg (T_H1) and IL-17 (T_H17).  Using this method I have never seen co-expression for any of these factors with the exception of a small population of cells that co-stain with IFNg and IL-17, as commonly described in the literature.

Taken together, selection of markers for assessing these populations must be done carefully, bearing in mind that cytokines characteristic of a T_H subtype are not always co-expressed in all cells identified as that T_H subtype by the surface markers and transcription factors discussed here.  Flow cytometry panels containing markers for all of these T_H subtypes can allow measurement of more specific T_H populations from PBMC by gating based on specific expression as well as exclusion of the other T_H subtype markers.

 

Additional Reading

Heterogeneity of CD4+ memory T cells: functional modules for tailored immunity.  Sallusto F, Lanzavecchia A. Eur J Immunol. 2009 Aug;39(8):2076-82.

Chemokine receptor expression identifies Pre-T helper (Th)1, Pre-Th2, and nonpolarized cells among human CD4+ central memory T cells.  Rivino L, Messi M, Jarrossay D, Lanzavecchia A, Sallusto F, Geginat J. J Exp Med. 2004 Sep 20;200(6):725-35.

Human T cells that are able to produce IL-17 express the chemokine receptor CCR6.  Singh SP, Zhang HH, Foley JF, Hedrick MN, Farber JM. J Immunol. 2008 Jan 1;180(1):214-21.

Selective expression of a novel surface molecule by human Th2 cells in vivo.  Nagata K, Tanaka K, Ogawa K, Kemmotsu K, Imai T, Yoshie O, Abe H, Tada K, Nakamura M, Sugamura K, Takano S. J Immunol. 1999 Feb 1;162(3):1278-86.

Flow cytometry allows assessment of protein expression on a single cell level.  Because of the diversity of populations comprising peripheral blood mononuclear cells (PBMC), flow cytometry represents the best method for studying the functional and phenotypic properties of these cell subsets.  There have been a number of publications that present multiparametric flow cytometry staining panels for assessing PBMC populations, and as an easy first step into these assays, these panels can be referenced (see a few examples below).

However, use of these established panels may not suit further exploratory questions, and thus an understanding of the best methods for designing flow cytometry panels is needed.  Because of the many identified populations and the upper limits of markers that can be simultaneously used in flow cytometry, a lot of thought must go into planning these assays.

Flow cytometry utilizes fluorophore-conjugated antibodies targeting antigens of interest.  Antigenic markers uniquely present on PBMC population subsets have been actively sought after by the immunology community. The upper limit of the number of markers that can be simultaneously assessed depends on several parameters.  Most obviously, this depends on the specs of the flow cytometer instrument available.  I presently use a modified version of the BD LSR Fortessa, with five lasers, and allowing detection of up to 17 parameters plus forward and side scatter.

Unfortunately, because of overlap in the excitation and emission spectrums of available fluorophores, this doesn’t mean that I am able to successfully co-analyze 17 PBMC markers.  The fluorescence emitted from a given fluorophore may be detected or “spill over” into the range of detectors optimized for other fluorophores.  Although flow cytometry software is capable of calculating the percent spillover from this spectral overlap and subtracting this estimated fluorescence from the affected channels by a process known as compensation, this is not a perfect science. Due to spectral overlap, there are a few points to consider when designing an antibody “staining panel” for your immunological assessments.

First, select the fluorophores you hope to use in your panel.  To do this, determine the laser and detector configurations of your cytometer.  This will tell you for which fluorophores the excitation and emission spectrums are optimal on your instrument.

Next, choose detection channels to try first based on minimal excitation and emissions overlap.  Alternatively, consult previously published panels to see which fluorophore combinations have been recommended.  As an example, for my instrument’s 561nm laser, there are 5 detectors, so if I had to choose 3 of these, it would simply be every-other one based on the detector’s spectral range.  However, fluorophores can overlap in not only their emission but also their excitation spectrums.  Thus, it is important to look across the detectors for different lasers.  For instance on my instrument, there is a 780/60BP detector on the 561nm laser for detection of PE-Cy7, and the same detector on the 640nm laser for detection of APC-Cy7.   Thus the use of these channels together may be complicated if the fluorophores have a wide excitation range.  BD Biosciences has an awesome fluorescence spectrum viewer for envisioning the spill-over between parameters based on your cytometer configurations.  Other companies such as Life Technologies, eBioscience, and Biolegend also have similar tools to reference for fluorophores not sold by BD Biosciences.

It is important to mention that fluorophores are improving all the time.  Newer chemistries have led to development of product lines such as the Brilliant Violet dye series and Q-Dots with highly improved excitation and emissions spectrums for minimal spill-over.  Thus, for each channel, there could be several fluorophores that fit the optimal excitation and emission properties but select the one with optimal qualities of not only brightness but also their excitation and emission spectrums.

Once you have selected a likely handful of combinable fluorophores, it is time to figure out which cellular markers to target with which fluorophores.  To do this, first determine which markers of interest in your staining panel are co-expressed.  If there is no co-expression of your selected markers, it becomes easier to utilize antibodies tagged with spectrally overlapping fluorophores.  For instance, Pacific Blue and V500 have an approximately 10-20% spillover into the other’s channel on my LSR Fortessa’s configurations.  However, when not co-expressed, for instance CD4 and CD8, the fluorescent spill-over and hence compensation issues between these fluorophores are minimized, i.e. if there is no expression of Pacific Blue, then there is nothing to subtract from the V500 channel and vice versa.  However, if markers are co-expressed, such as CD4 and CD45RA, then issues with compensation between these parameters may lead to over or under-subtraction of fluorescence values in CD4⁺CD45RA⁺ populations compared with single positive populations.  But if your objective is only to gate these populations for analysis of other markers or to determine population percentages, then markers such as these may be successfully utilized together if these populations are visually discernible on cytometry data plots.  Another key thing to mention is for the lower expressing antigens, choose the brightest fluorophores.

But we’re not finished yet.  Next, determine for which markers your objective is to gain information about fluorescence intensity.  Now take the example from above but imagine that your question involved the magnitude of CD45RA expression on the CD4 versus CD8 population of PBMC T cells.  I have not done this example directly, but I can imagine that having high levels of compensation between one but not another parameter (in this case between CD4 and CD45RA but not between CD8 and CD45RA), may lead to inaccuracies in the relative fluorescence intensities.   Thus, in this example, if CD45RA is the marker for which expression intensity is the question, then the overlapping spectral fluorophores Pacific Blue and V500 would be best suited for identifying CD4 and CD8, while a fluorophore with minimal compensation issues with these channels, such as PE, would be best for CD45RA.  Thus, for markers in which determining fluorescence intensity is your objective, choose fluorophores with the least spectral overlap with all of the others in your panel.

Finally here’s another trick:  you can use two exclusively expressed markers in the same channel, as long as there is a third marker to differentiate these populations.  For example, I have very successfully used CD4-PerCP-Cy5.5 (CD4⁺ T cells) together with CD20-PerCP-Cy5.5 (B cells) along with CD3-V450 to differentiate the T cells from B cells in human PBMC.  In this case, CD4⁺ T cells will be double positive in V450 and PerCP-Cy5.5 channels, while CD20⁺ B cells will be PerCP-Cy5.5⁺ but V450^–, and CD8⁺ T cells will be PerCP-Cy5.5^– and V450⁺.

In summary, multiparametric flow cytometry analysis of PBMCs can be tricky, but with these and other tricks up your sleeve it becomes much easier to successfully design panels optimal for the questions you are asking and to maximize the number of parameters in your flow cytometry staining panels.

 

 

Examples of established multiparametric flow cytometry panels:

Standardizing immunophenotyping for the Human Immunology Project.  Maecker HT, McCoy JP, Nussenblatt R.  Nat Rev Immunol. 2012 Feb 17;12(3):191-200.

Multiparameter flow cytometry monitoring of T cell responses.  Maecker HT.  Methods Mol Biol. 2009;485:375-91.

11-color, 13-parameter flow cytometry: identification of human naive T cells by phenotype, function, and T-cell receptor diversity.  De Rosa SC, Herzenberg LA, Herzenberg LA, Roederer M. Nat Med. 2001 Feb;7(2):245-8.

Nine-color flow cytometry for accurate measurement of T cell subsets and cytokine responses. Part I: Panel design by an empiric approach. McLaughlin BE, Baumgarth N, Bigos M, Roederer M, De Rosa SC, Altman JD, Nixon DF, Ottinger J, Oxford C, Evans TG, Asmuth DM. Cytometry A. 2008 May;73(5):400-10.

 

BD Bioscience’s Fluorescence Spectrum Viewer:

http://www.bdbiosciences.com/research/multicolor/spectrum_viewer/index.jsp

Human peripheral blood mononuclear cells (PBMC) are comprised of complex populations of T cells, B cells, NK cells, monocytes, and dendritic cells.  In addition, there are more complicated cell types such as NKT cells that are thought of as T cells that share many properties of NK cells.  Within all of these basic populations are many functionally unique subsets.  Because of the diversity of populations in PBMC, flow cytometry represents the best method for studying functional and phenotypic properties of these cell subsets.  In the next few blog posts, I will discuss selection of markers for studying PBMC populations using flow cytometry and the best way to arrange these markers in flow cytometry staining panels.

The PBMC cell type I have the most experience with characterizing is T cells.  CD3, a T cell specific marker, is necessary to differentiate T cells from other populations, simply because CD4 and CD8 can be expressed by other cell types.  CD8 can be expressed on NK cells, while CD4 can be expressed on populations of monocytes and dendritic cells.  CD4 and CD8 are also necessary markers for identification of these two major T cell populations.

CD4 and CD8 T cells are most simply classified as naïve or antigen experienced populations including central memory, effector memory and effectors.  Central memory and effector memory populations are known to differ in their effector functions and ability to home to different anatomical sites.  Two markers are necessary to differentiate naive, central memory, effector memory and effector T cell populations present in PBMC.

The first is CD45, a protein tyrosine phosphatase regulating src-family kinases, is expressed on all hematopoietic cells.  CD45 can be expressed as one of several isoforms by alternative splicing of exons that comprise the extracellular domain. CD45RA is expressed on naïve T cells, as well as the effector cells in both CD4 and CD8.  After antigen experience, central and effector memory T cells gain expression of CD45RO and lose expression of CD45RA.  Thus either CD45RA or CD45RO is used to generally differentiate the naïve from memory populations.

However, differentiation between central and effector memory populations and between naïve and effector populations can be achieved by adding a second marker.   There are several markers that have been used for this purpose and these tend to mark these populations at slightly different stages of the differentiation pathway that is thought to occur in T cells as they change from central to effector memory cells.  The chemokine receptor CCR7 is considered the gold standard for this discrimination, and the lymph node homing receptor CD62L is a close second choice.  Naïve and central memory cells express these receptors in order to migrate to secondary lymphoid organs, while the absence of these receptors allows for effector memory and effector cells to accumulate in peripheral tissues.  CCR7 has been classically thought to be difficult to stain for due to low expression levels, and as a result I have never used it.  However, a recent review article pointed out that a new brighter staining antibody to CCR7 has been developed (clone 150503), and is something I hope to test in the near future.  CD62L has generally been my marker of choice, however CD62L expression is lost following density gradient centrifugation, cryopreservation, TCR ligation or activation with PMA/ionomycin.  Thus CD62L as a marker is best utilized following overnight culturing subsequent to thawing cryopreserved cells.  Other potential markers are CD27 and CD28 which are also more highly expressed by the central memory and naive populations.  However, it is important to note that each of these markers, CCR7, CD62L, CD27, and CD28, marks slightly different populations of cells and care must be taken when making comparisons between populations defined by different markers for assessing frequencies or functions.

In summary, naïve T cells are CD45RA⁺CD45RO^–CCR7⁺CD62L⁺, central memory T cells are CD45RA^–CD45RO⁺CCR7⁺CD62L⁺, effector memory T cells are CD45RA^–CD45RO⁺CCR7^–CD62L^–, and effector cells are CD45RA⁺CD45RO^–CCR7^–CD62L^–.  Thus, CD3, CD4, CD8, CD45RA or CD45RO, and CCR7 or CD62L or CD27, are a great starting point for designing flow cytometry panels for the assessment of T cells present in human PBMC.  T cells populations, however, are much more complex, and may be further classified by helper subtypes and activation status, which will be discussed in a later blog.

The use of CD45RA and CD62L to differentiate naive, central memory, effector memory and effector T cell populations present in PBMC.  Naïve CD4 and CD8 T cells are CD45RA⁺CD45RO^–CCR7⁺CD62L⁺ (B, F), central memory T cells are CD45RA- CD45RA^–CD45RO⁺CCR7⁺CD62L⁺ (A, E), effector memory T cells are CD45RA^–CD45RO⁺CCR7^–CD62L^– (C, G), and effector cells are CD45RA⁺CD45RO^–CCR7^–CD62L^– (D, H).

 

Further Reading:

Central memory and effector memory T cell subsets: function, generation, and maintenance.   Sallusto F, Geginat J, Lanzavecchia A. Annu Rev Immunol. 2004;22:745-63.