What is Fluorescence Spillover?
Assessing Relative Contribution
Compensation refers to correcting a phenomenon called fluorescence spillover in flow cytometric analysis. This is the removal of the signal of any given fluorochrome from all detectors except the one dedicated to measuring a specific fluorochrome.
Compensation in flow cytometry is poorly understood; nevertheless, compensation is critical for several aspects of flow cytometry, particularly in determining antigen density. Due to poor understanding, compensation is not set appropriately.
Every fluorescent molecule can emit light with a spectrum characteristic of that molecule. However, the emission Spectra of two molecules can overlap. For example, fluorescein (FITC) and phycoerythrin (PE) emission Spectra overlap. To discriminate between FITC fluorescence and PE fluorescence, the emission light can be split according to wavelength and distributed to each respective detector. Each of these detectors can be fitted with a different filter that eliminates light outside a defined narrow-spectrum region.
Therefore, FITC fluorescence will dominate a detector with a 530 nm filter, and the PE fluorescence will dominate a detecter with a 530nm filter. However, some of the FITC fluorescence can appear in the detective designated for PE; the signal left is recorded as a result is termed fluorescence spillover.
As the name suggests, some of them spill over into the PE detector. Another way to describe this is spectral overlap between fluorophores. Compensation must be applied in both directions to represent the data correctly.
It is impossible to design filters specific to one fluorescent dye. Therefore when specific combinations of fluorophores are used, fluorescence spillover will always occur – as is the case with FITC and PE. Calculating how much of the 575nm signal is from PE and how much of it is from FITC is termed compensation. This is a form of correction, i.e., correcting the PE detected signal to account for the amount of FITC fluorescence part of the PE band.
The gold standard approach for correcting spectral overlap is automatic compensation. This is mathematically encoded in a flow cytometer equipped with an automatic compensation package carry. Alternatively, data can be collected uncompensated and fed through a third-party software package to perform automated compensation.
To conduct compensation, there must be a single stained control for every parameter used in the experiment. In addition, there are rules for good compensation controls:
- Controls must be at least as bright, or brighter than, any sample the compensation will be applied to – this is because compensation cannot be extrapolated, i.e., by making sure that the compensation controls are the equivalent to the brightest part of the experiment; this ensures that all signals in experimental samples are properly compensated
- The background fluorescence should be the same for the positive and negative controls
- The compensation controls must match the exact experimental fluorochrome, ensuring accuracy, i.e., GFP compensates for GFP. An exception to this circumstance is the tandem dye. i.e., PE-Cy5. In this example, the PE will be activated by the 561nm it will emit, subsequently activating the Cy5 component. The fluorescence emitted by the component is ultimately the wavelength detected by the channels. It is subsequently good practice to use the same sample vial of antibody for the construction of controls as used for the dye in the experiment; this will remove any errors that will be caused by lot-to-lot variation
- Collect enough events: accurate compensation relies on calculating the median fluorescence intensity of both negative and positive populations for every control. In the absence of an adequate number of events, this calculation will not be accurate, and the compensation will suffer.
Image Credit: OpturaDesign/Shutterstock.com
After acquiring compensation control in the software, third-party software is used to calculate compensation. Compensation is a mathematical calculation; the software is, therefore, more accurate than a manual calculation. ‘Cowboy compensation’ is the process of manually adjusting compensation values by eye to adjust for populations. This will result in significant errors in compensation as this will not account for the effects of the adjustments on all of the other channels. Therefore, manual adjustments are considered poor practices and should be avoided.
It is good practice to inspect the compensation matrix after it has been made in the software. This can be achieved by looking at the flow cytometric N×N plots.e., plotting every fluorochrome against every other in the channel. By doing so, examples of over-or under-compensation can be identified.
Overcompensation results in the ‘frowning’ of fluorescent signals leaning towards the X-axis. Therefore, the values have been over-subtracted in those two channels. A consequence of under-compensation is observing a ‘smile,’ i.e., curve into the center of the plot. In under-compensation, not enough correction has been applied between the two channels.
The compensation controls will need to be repeated to troubleshoot in these cases.
Compensation is always necessary for flow cytometry to be able to differentiate between distinct populations of cells - some of the compensation that accounts for the significant background noise that occurs when the added brightness of one fluorochrome creates a significant background noise for another.
In summary, this is achieved by measuring the overlap in the spectrum between the different fluorochromes used. These measured values can subsequently be used to create a matrix. By inverting the matrix, the true compensation values are determined, and these values are used to correct the overlap in each detector per fluorochrome.