New device design brings unmatched confidence to cell measurements
Measuring the number and properties of cells moving in a flow – a process called flow cytometry – is of critical importance to diagnostic medicine, pharmaceutical research and biomedical science. Now, researchers at the National Institute of Standards and Technology (NIST) have developed a way to bring unprecedented improvements to the technique.
Flow cytometry typically involves labeling cells with a fluorescent material, shining a laser light on them as they pass through a particular point in a fluid channel the size of a human hair – narrow enough for the cells to generally move in a single file – and record the light emitted by the markers on the cell. Emission analysis reveals various characteristics such as cell type, size, DNA content and stage of the cell division cycle.
But the conventional single-measurement method lacks the means to quantify variations in its readings. For example, a cytometer that measures cancer biomarkers can use a specific measurement value to decide between healthy or diseased cells. A cell may return a slightly lower value, identifying it as healthy. But the range of values the instrument might return for repeated measurements of this cell is unknown.
This is problematic because it is not known how often the value can skew the diagnosis. For example, high confidence is imperative when counting circulating tumor cells, which lead to cancer metastasis; there may be one tumor cell for every million others in a blood sample. Improving the ability to resolve these rare events could lead to earlier detection and better treatment options.
Similarly, inaccurate counts of different cell populations, such as immune cells, in a sample can lead to incorrect diagnosis of disease or misinterpretation of the success of drug treatment. It can be difficult to tell whether different measurements come from distinct cell types or simply from experimental variability.
To address this situation, NIST researchers created and validated a new system that directly records the amount of variation in flow cytometry measurements to quantify uncertainty. Their instrument takes a series of multiple readings of the same particle whose position and velocity are precisely controlled as it moves through the channel, allowing them to observe measurement-dependent variations as low as 1%.
“This is the first time we can directly measure per-object uncertainty in a flow cytometer,” said Matthew DiSalvo, lead author of the NIST team who published their findings July 20, 2022 in the journal Lab. we have Chip. “When you measure something once, and only once, you can’t verify accuracy,” he said. “To do that, you need repeatability.”
The new chip-based serial flow cytometer solves this problem by taking four measurements of each particle: two readings at each of two different points separated by 16 millimeters; (about 0.6 inches) in the channel with flow rates as high as 100 or more particles per second.
Several critical challenges had to be overcome to create the new cytometer, which was initially tested using uniform plastic beads about 15 micrometers (µm, millionths of a meter) in diameter, about one-tenth the width of a human hair and the same size as average white blood cells.
One of the challenges was to fabricate a channel (about 40 µm by 80 µm) incorporating two identical areas over which the laser light was applied to the passing particle and to ensure that the greatest amount of emitted fluorescent light was captured. by a pair of detectors which measured upstream and downstream. To do this, DiSalvo and project leader Gregory Cooksey developed advanced waveguides that minimize losses and prevent emission light from spreading. Cooksey has extensive experience in pioneering microfluidic designs, and the cytometer team built on this body of work.
Another challenge, particularly demanding, was to find a way to ensure that the emissions recorded in each of the two zones of the channel came from the same particle. This could only be done by monitoring the position and velocity of the particle so exactly that the transit time from node to node was known precisely, allowing the system to match the measurements at the first node with those of the second.
To do this, the researchers developed a “hybrid” technique allowing the particle to be focused at a precise point in the channel. In conventional cytometers, this is usually done by using two different fluid pumps: one to inject the particles and one to create a separate fluid sheath around the inside perimeter of the channel, effectively forming a tube of liquid within the solid channel. This acts to confine the sample to the central core of the channel.
Or, at least, that’s the presumption. But actually, Cooksey said, in small chains “it’s the worst place you could put it.”
The reason is that the inertia of the fluid in the small channels adds two forces on the particle. A force acts to move the particle away from the wall. But another force pulls the particle away from the center of the channel due to the difference in velocity of the fluid acting on either side of the particle. As a result, the particle naturally tends to decenter towards an equilibrium position in which the forces are balanced. For the specific design used in this study, the centerline displacement varies with fluid velocity, from about 11 µm at 0.75 meters per second to 14 µm at 1.35 m/s.
To control this position very precisely, DiSalvo and Cooksey used four pumps to drive the sheath, allowing them to adjust the pressures from side to side as well as up and down. “We use hydrodynamics to put the particle in the position where we know it wants to go at the very beginning of its transit,” DiSalvo said. “So when a particle accelerates in the channel, it is already at its equilibrium position. It’s very tightly held along the entire length of the cytometer.
Therefore, the device is able to perform high precision measurements with the particles moving at a high speed of 1 m/s. (1 m/s, or about 2.3 mph, may seem slow. But at that speed, a 10 µm particle is moving 100,000 times its body length per second. An automobile doing this would be moving at about 1 million of mph.)
Since the experiments were reported in the new journal article, the team has been measuring white blood cells in the device. “The first thing we looked at is where the cell is in its division cycle,” Cooksey said. “We observe the amount of DNA in the cell. The fluorescent signal allows us to quantify how many cells are actively dividing, in the process of DNA synthesis.
Researchers also use different laser wavelengths to reveal different properties of the sample.
“We’ve significantly modified the waveguides to increase sensitivity by incorporating microlenses and other enhancements,” DiSalvo said. “We believe that we will soon be able to provide measurements of uncertainty while matching the sensitivity of commercial single-measurement systems that cost hundreds of thousands of dollars.”
The cytometer project includes collaborators with expertise in mathematics and modeling, leading to additional new analytical methods that will improve quantification and classification. NIST researchers Paul Patrone and Anthony Kearsley used signal analysis to separate measurement variations due to variations in flow rate and particle size. Their methods also allow for an unparalleled ability to identify and separate objects that would previously have been too close together to be distinguished to any conventional extent. Failure to distinguish separate objects prevents counting accuracy and may result in missing unique sample components, such as circulating tumor cells or interacting immune cells.