Multicolor Flow Cytometry Panel Development
Flow cytometry allows researchers to test sophisticated biological hypotheses using flow cytometry. With multicolor flow cytometry single cells can be probed with multiple markers and protein expression levels can be correlated using multiple antibodies. Such techniques allow researchers to determine not only the percentage of each cell that expresses each marker but also the overlap. However, multiparameter flow cytometry panel development is a complex endeavor, with many pitfalls that can impact data quality.
Here are some critical considerations for successful panel design.
Markers and Fluorophores
First, develop a strategy for discriminating between cells based on the expression of the selected phenotyping markers. For each antigen, select a compatible fluorophore. Each antigen/fluorophore combination needs to be distinguishable from the rest. To avoid overlap spread out the emission spectra of the fluorophores and use the minimum number of colors required for the project.
Also, avoid saturating your panel with high-abundance antigen which can result in loss of the signal from low-abundance antigens. For improved detection, markers for rare cell types and those with low abundance should be paired with bright fluorophores. Live/dead control dyes should be included to screen irrelevant cells and to reduce background
While gating is performed during data analysis, your gating strategy should be planned for during the panel development. Your experimental setup should incorporate gates to exclude dead or clumped cell and exclusion of irrelevant cells based on size, granularity, or cell-type marker.
Compensation and Controls
When a fluorescent marker is excited, some of its emission may fall outside the desired range creating spectral overlap with emission from other fluorophores. When only staining for a few markers, select fluorophores with minimal overlap. When designing larger multiparameter flow cytometry experiment, each new color increases the chance of non-specific spill over into almost all other channels. Compensation will be required to differentiate spillover from the desired signals. Controls using either cells or commercial beads bound to the same fluorophore can be used for compensation can be used to increase the resolution. Flow cytometry software can then be used to subtract the signal from the irrelevant fluorophores, leaving you with just the signal you want.
Prepare a single-color compensation control for each fluorophore, plus an unstained control for autofluorescence. Be sure to collect enough events, both positive and negative, from each compensation control to allow the software to calculate the required compensation. Make sure they are brighter than any of your samples, since brightness is essential for proper compensation. Fluorescence minus one (FMO) controls — cells or beads stained with all but one fluorophore– can also be used to determine spillover for each spectral emission. FMOs are important for measuring rare cell types and low-expression markers, and in cases where it is difficult to distinguish between negative and positive cells.
To ensure statistical validity, collect enough events from each sample to detect sufficient positive and negative cells, given the rarity of your cell type.
Take time to develop an optimized staining protocol. A poorly optimized protocol can lead to increased background due to cell damage, nonspecific staining, and/or precipitation. Avoid buffers that increase autofluorescence and include a “blocking” buffer to prevent aggregation.
Extracellular stains can be applied without permeabilizing cells, and when cells need to remain viable for sorting and collection. Intracellular markers can be stained after fixing and permeabilizing cells. It can be difficult for antibody-conjugated dyes to reach intracellular targets, so brighter and smaller fluorophores are preferable.
After completing all these steps, you’ll need to test the panel to ensure the controls are working, the results make sense, and you have the resolution you need. If not, it’s time to go back and troubleshoot.