PBMCs can be roughly differentiated into 3 main populations on FSC/SSC (left panel: A- Lymphocytes, B- Monocytes and C- Granulocytes). If RBCs are not eliminated from the sample they overwhelm the PBMC minority and make analysis of these cells impossible (as shown by the identical sample pre-RBC lysis, right panel).
RBCs can be eliminated by several mechanisms including ACK (ammonium) lysis, water lysis (both lyse the RBCs preferentially via distinct mechanisms) and CD45 fluorescent thresholding (whereby CD45 labels all differentiated hematopoetic cells except RBCs and platelets, allowing for them to be “ignored” instead of lysed). A summary of these protocols and pros and cons of each approach can be found here.
Subsequent multicolour staining will be highly dependent on the target cell population of interest (T-cell subset, progenitor cell, etc.). For many cell types there is an OMIP (online multicolour immunophenotyping panel) which have recommended staining parameters (antibody clones, concentrations, fluorophores). Often these OMIPs will discriminate between subsets beyond what you need (i.e. specific NK subsets and activation state when all you want is total NK cells). In that case you can be selective as to what subset markers you use or seek the advice of the core facility staff.
With all motile cells, capping can occur if the antibody crosslinks surface receptors. This leads to antigen shedding and cell death but this can be blocked via the inclusion of 0.01% Sodium Azide. Note that you cannot look at cell stimulation or energy dependent processes if you include azide.
In conjunction with the above recommendations, immunophenotyping then becomes multicolour flow cytometry specific to the experiment in question. As such there is no general protocol that can be provided, but general rules are as follows:
Use the brightest fluorophores for your least abundant antigens (specifically keep in mind that some abundant markers are down regulated following activation: e.g. CD3).
Stay away from short wavelength fluorophores if you have primary cells, fibroblasts and other cells with high background autofluorescence (i.e. use “redder” fluorophores and longer wavelength lasers; see below figure).
If you are using limited colours use all the lasers to limit your compensation requirements (i.e. for 3 colours use VioBlue, PE, and APC instead of FITC, PE and PE-Cy7).
If you are using 3 or more colours use fluorescence minus one (FMO) controls to determine gating boundaries (see below figure). It is important to do this in stimulated and unstimulated samples if applicable.
Fc blocking is critical if you are using FcReceptor-positive cells (PBMCs or phagocyte containing samples).
If your cells are hard to come by, use compensation beads rather than single stained cells. You MUST use the same fluorophore and fluorescence-protein ratio (ideally the same antibody). See the links section for supplier information on compensation bead providers at time of writing.
Always perform surface marker staining, then fix, permeabilize and stain for intracellular markers if you are staining for both extracellular and intracellular molecules. This is so you can optimally label surface epitopes prior to fixation.
Always stay away from PE and PE conjugate dyes when doing intracellular staining if possible. PE is a HUGE molecule and can be difficult to get into the cell and wash unbound antibody out of the cell.
Short wavelengths (higher energy) are able to excite electrons on biological molecules within cells more easily than long wavelengths (lower energy). As a result, channels off of the 405nm laser and 488nm laser exhibit worse autofluorescence. This is worth keeping in mind when selecting a staining panel as this higher background can limit the resolution of dim/low abundance antigens.
FMO Controls (Fluorescence Minus One) are done by excluding only the fluor of interest (i.e. a FITC FMO is stained for all markers except the FITC conjugated antibody). This allows for accurate setting of gating boundaries for the FITC channel. The FMO controls would be done for all fluors in the panel (see table to right) in an experiment using >3 colours. As you can see from the figure below, FMOs are crucial in setting these gating boundaries. Note that the “Stained (Unstimulated)” population using the unstained control boundaries (top row, 3rd panel) is falsely considered positive for the APC channel. With the APC FMO gating boundaries (bottom row, 3rd panel) this unstimulated population is correctly considered negative for the APC channel (in this case a marker of NK cell activation).