Written By: Tim Bushnell, Ph.D.
All flow cytometer instruments have a certain 3 components you likely don’t know enough about. The way they are put together will dictate the performance of the system. As a user, you’ll be interacting heavily with these components, so you need to know both what they are and how they work.
The 3 components are the fluidics, optics, and electronics. Fluidics will be managed in order get interactions at the proper flow rate, ensuring that your data keep a tight CV. After cells have passed the laser intercept, you need a way of measuring the signal. This is where the second component will be used: optics. There are a few different types of optics you can use to measure your signal, like PMTs or APDs. Lastly, there are electronics, which process the photon into an electronic signal that is ultimately digitized and stored in a file known as the “FCS file.” This is key for interacting with your data once you have your results.
Let’s look at these 3 components in more detail…
Fluidics will take the cells from your tube and bring them to the interrogation point. That happens in what we call the “flow cell.” In Figure 1, you can see a flow cell represented by a stylized diagram. The red arrows indicate the laminar flow, which is a uniformed flow that runs in parallel sheaths. The fluid layers don’t mix, so when we inject our center sample into the sheath fluid, we constrain everything into what is known as the core stream. What this also allows, due to hydrodynamic focusing, is for the cells to align single-file in the axis. This way, when they pass the laser intercept, you’re seeing one cell at a time.
Figure 1: A stylized flow cell.
In Figure 1, you can see a picture of a flow cell. The sheath fluid is coming in at the top, and this is going to flow towards the bottom, where the flow cell is located. Just below where the cells enter the sheath fluid, we see the cone shape that positively accelerates the cells towards the the flow cell, where the cells pass through the laser intercept.The laser intercept is perpendicular to the flow of the sheath fluid. The detectors are going to be at 90 degrees relative to the laser intercept – except for the forward scatter detector, which would be in line with the laser. That’s where you’re going to get the light. Then you’ll measure the resulting light through your optics and electronics.
If you’re on an analyzer, then after the cells pass the laser intercept, they will go into either the waste or the sort chamber where the stream is charged, droplets are generated, and cells are sorted.
Since the sheath fluid velocity is fixed, the event rate is controlled by the core stream velocity. This is set by the differential pressure between the core stream and the sheath fluid. The low/medium/high choice that the researcher makes controls the differential pressure between the sheath fluid and the sample stream. The greater the difference, the wider the core stream as shown in Figure 2.
Figure 2: Increasing the differential pressure increases the core stream and the CV of the data.
This differential-pressure-induced increase in the CV is even more apparent when one looks at the consequences of increasing the flow rate during a cell-cycle experiment. The best practice is to run these experiments at low differential pressure to ensure the tightest CV for the G0/G1 population, as well as clear S and G2/M phases. This is shown in the left panel of Figure 3.
Figure 3: Effects of increasing differential pressure on a cell cycle experiment.
Due to the increase in differential pressure from low to medium, the G0/G1 peak has spread. More importantly, the G2/M peak has spread into the S region of the histogram. (Figure 3 – middle panel) Under high differential pressure, the whole histogram becomes almost impossible to analyze. As shown in the right panel of Figure 3, both peaks have spread dramatically.
The take-home message is to always concentrate your sample and run at the lowest possible differential pressure. That’s how you maintain a tight core stream and get some good, readable data. But if you’re going to operate at a higher differential pressure, run all your samples at the same pressure so that you don’t exclude samples due to fluidics.
Once the cells have passed the laser intercept, the system must guide the emitted photons to the detectors, which will convert the photons to a proportional number of electrons. This can be done using two major classes of detectors: photodiodes and photomultiplier tubes. There’s a special case called the Avalanche photodiode, or APD, which is becoming popular to use in certain instruments. APD has some advantages because it’s less expensive than a PMT, and it has improved sensitivity in the far red regions. Photodiodes, photomultiplier tubes, and APDs work in a similar fashion. The left panel of Figure 4 shows a photodiode, and as light hits this center, it causes a depletion zone. This in turn causes the negative and positive charges to move in opposite directions, generating a current that is turned into the photocurrent.
Figure 4: How photocurrent is generated by a photodiode and an APD.
The right panel of Figure 4 shows a schematic avalanche photodiode. Like a PD, the photon hits the center of the APD and generates an electron hole and the changes migrate to the n and p regions. Where the APD differs is that near the negative end of the depletion zone, there is an electric field. As the electrons enter this avalanche zone, there is an amplification of the negative charges moving to the n region. This increases the sensitivity of the APD. These detectors are generally less expensive than a photomultiplier tube, and they have improved sensitivity – especially in the red region of the spectrum.
Shown Figure 5 is a PMT. This is a vacuum tube. The light enters and hits the photocathode(1). As long as the work energy of the photon is above the work energy in the photocathode, it will inject an electron(2), which will be aimed at each of the dynodes. The dynodes have an increasing voltage, which causes secondary emission of electrons(3). The electrons travel around the dynodes until they reach the photoanode(4).
Figure 5: Schematic of a photomultiplier tube.
Many PMTs can amplify a signal over 10,000 to 100,000 times. There are some PMT systems that can detect a single photon of light. An important point to remember is that the photons coming in are proportional to the emerging photocurrent.
Finding the best voltage can be a very difficult process, especially for the new users. With digital cytometers, it is possible to determine a minimum starting voltage. One way to do this is the “Peak 2 Method,” published by Maecker and Trotter in 2006. Using this method, you run a dim particle over a voltage range and plot the CV on the Y-axis versus the voltage, which results in a curve similar to that shown in Figure 6.
Figure 6: Typical results from the peak 2 method for PMT optimization
At low voltage, the CVs are relatively high. With increasing voltage, there is a decrease in the CVs – until you get to the inflection point, where the CVs start to level out. This “inflection point” then becomes the minimum useable voltage. (You could improve that voltage using voltage optimization, but that’s an entirely separate topic!) One important thing about PMTs and PDs: They generally react to any form of light – any photon that hits them. So you have to control the signal that reaches the detector, and for that you’ll use filters like the the two optical elements shown in Figure 7, a long pass and a short pass filter.
Figure 7: Long and Short pass filters.
As the name suggests, a long pass filter, such as an LP500, will let light in above 500 nanometers. A short pass filter, like an SP500, lets in light below 500 nanometers. You can use those filters to move light around, especially if you’re able to couple them with some form of mirror, like a dichroic mirror.
Figure 8: Bandpass filters.
What you’ll often put right in front of the PMT is a bandpass filter, and that filter should be named according to the following convention: a number followed by a slash and a second number, such as “520/20.” Some people can get confused about this terminology. In Figure 8, you can see the spectral scan of a fluorochrome. Shown in green is a 520/20 filter – this means means it’s centered at 520 nanometers with a 20-nanometer window, plus or minus 10 nanometers. That’s the range of light that will be allowed through to the PMT – in this case, the range from 510 to 530 nm.
Electronics are the third flow cytometer component. The electronics take the signal from the detector and turn it into a digital value that is stored as a Listmode file for downstream analysis by your preferred software. Figure 9’s left panel shows a typical electronic pulse being generated by the detector.
Figure 9: A signal processing diagram.
As the pulse comes off the detector, two things happen, which are depicted in the middle panel of Figure 9. The first process is that this pulse is sampled. This happens at a very fast rate, between 10 to 100 MHz. With each sampling, the pulse is digitized by the analog to digital converter(ADC). The power of this ADC provides the dynamic range for the signal to be binned. For example, a 10 bit ADC has 210 bins or 1024. At the upper end, there are cytometers using 24 bit ADCs, resulting in 224 = 16,777,216 bins. The boxes in shown in the middle panel, under the curve, illustrate what each of the heights would be during the sampling. The more sampling you do, the more accurate the representation of your pulse.
The digitized values are then put into a listmode file (Figure 9 – right panel), and depending on how the system is setup, you might just measure the height of that pulse. That might even be the only thing you measure, but you could also look at the width of the pulse or the time of flight. If you take the integral of the height and the width, you can get the area of the pulse. A lot of people recommend using area measurements on the newer digital instruments because it’s more accurate when looking at the whole distribution of the fluorescence, and not just a single peak. If you had one bright peak, like just one little spot on the cell, that would give you a higher height value. Of course, a relatively low signal would then be integrated over the whole cell.
Users will be interacting frequently with all 3 of the flow cytometer components covered in today’s blog. The fluidics are managed to get interactions at the right flow rate, which will ensure that your data keep a tight CV. Then you can run the same flow rate for all your samples, and you won’t have different CVs for every sample. There are also different optics you can use, like PMTs, APDs, and PDs. It’s important to remember the bandpass filters because they let you know the detector on which your signal will be measured. And with the newer generation of instruments, you can actually change out bandpass filters and design the flow cytometer to your specifications – just make sure you cite the specific bandpass filter that you’re using. Finally, there are electronics, which process the photon into an electronic signal that is ultimately digitized and stored in a file called the “FCS file.” Analysis can be performed on this file at a later time.
To learn more about the 3 Components Of Every Flow Cytometer You Don’t Know Enough About, and to get access to all of our advanced materials including 20 training videos, presentations, workbooks, and private group membership, get on the Flow Cytometry Mastery Class wait list.
My other passions include grilling, wine tasting, and real food. To be honest, my biggest passion is flow cytometry, which is something that Carol and I share. My personal mission is to make flow cytometry education accessible, relevant, and fun. I’ve had a long history in the field starting all the way back in graduate school.
Latest posts by Tim Bushnell (see all)
- Understanding Reproducibility in Flow Cytometry – It’s the Antibodies! - October 24, 2019
- 3 Components Of Every Flow Cytometer You Don’t Know Enough About - October 10, 2019
- 4 Ways To Analyze Tissues By Flow Cytometry - September 26, 2019