flow cytometry fallacies

The 3 Most Common Flow Cytometry Fallacies

Flow cytometry is fast evolving from a method only revered by immunologists, to one used by nearly every biological specialty. It’s pretty much my favorite tool. Unfortunately, as with most lab techniques, much of flow cytometry is taught on the job without a lot of standards. And too often bad habits are passed along like gospel, when in fact they are fallacies.

But you can avoid getting on that train! Below are the 3 most common rookie mistakes with flow cytometry.

3 Common Flow Cytometry Fallacies

1. Believing Isotype Controls Are “Real” Controls

First, let’s all get on the same page. Flow cytometry is dependent on the detection of fluorophores conjugated to antibodies directed against a specific antigen. Antibodies (or immunoglobulins) are formed from two immunoglobulin heavy chains and two light chains. In immunology, an antibody isotype refers to variant immunoglobulin heavy and light chains. In humans there are five heavy chain isotypes and two light chain isotypes.

So…what’s the deal with the immunology primer? Well, there’s always the possibility that antibodies can bind non-specifically, either to a low affinity, non-specific target or to intracellular targets as a cell begins to die due to compromised membrane integrity. Thus, to ensure that our data is reliable, we need to account for this non-specific binding. One way to do this is to use isotype controls. These control antibodies have the same isotype as the antibody used in the assay, but are not directed against a specific protein. For instance, if we choose to use an anti-CD3 FITC antibody with an isotype of IgG1, kappa in an assay, we would choose a FITC antibody with a IgG1, kappa isotype as an isotype control.

So obviously, isotype controls have their place and purpose. The problem is when people use isotype controls as actual assay controls, or use it to discern between positive and negative populations. All that isotype controls really do is determine non-specific binding and if your cells have not been properly blocked before staining. So please, PLEASE don’t use them in place of a real control in your experiment (i.e. unstained cells, unstimulated cells, etc).

2. Not Running Controls/FMOs

As any good scientist knows, setting up the proper controls for your experiment is key for getting reliable data. Therefore, you absolutely with no uncertain terms MUST run control samples with your experimental flow cytometry samples. They can be as simple as unstained or unstimulated cells if you happen to be looking at activatio.  But it is imperative that you have actual negative controls to compare your data to – and before you ask, no you can’t compare to your negative compensation control.

Just as important in running control samples is another kind of control that is sadly overlooked way too often, especially with newbies to flow cytometry: FMO.

FMO, or Fluorescence Minus One, is a basic concept with huge implications. As a quick reminder: the use of fluorescent molecules in flow cytometry means that we need to concern ourselves with minimizing spectral overlap and data spread, two issues that are unavoidable but fixable. The first is addressed through the process of compensation, and the later by running FMOs.

As someone who has run too many FMO panels to count, they are a pain. Seriously, they’re the worst. But, when balanced against how important of a control they are to a flow cytometry experiment, it is easy to be motivated to take the time to perform them, and perform them well.

3.  Performing Manual Compensation

As mentioned above, compensation is the process of correcting for fluorescent spillover. Lucky for us, most software used to capture raw flow cytometry data (i.e. FACS Diva) also calculate and make a compensation matrix without the user ever having to lift a finger. Still, many researchers feel the need to manually compensate, usually based on nothing more than flow plot aesthetics. This is a very bad idea.

You see, there are specific algorithms in place that calculate compensation based on the information you provide the instrument (i.e. compensation controls and experimental set up)1. So if you decide to play with these numbers, you run the risk of getting incorrect data. But, if you take the time to set up your experiment as perfectly as possible, you can avoid the knee-jerk reaction to manually compensate your data. You’ll save yourself a big headache in the end.

As a final note, if you’re having any difficulties with the above points or pretty much anything else with flow cytometry, a great source that’s likely under your nose is your institution’s Flow Cytometry Core. Those people are legit.

What are your favorite flow cytometry fallacies? Comment below!

References

  1.  Bagwell and Adam (1993).  Fluorescence spectral overlap compensation for any number of flow cytometry parameters.  Ann NY Acad Sci 20:167-184.

Image Credit:  http://klarititemplateshop.com/

Leave a Comment