CYTOMETRY/CELL ANALYSIS: Optical cell volume measurement

ETHAN SCHONBRUN

Cell volume is a critical parameter in both biology and medicine: Understanding cell growth and death, quantifying intracellular concentrations of ions and proteins, and diagnosing most hematological disorders all require accurate measurement of cell volume.

Optical technology offers many benefits for cytometry, but obtaining an accurate optical measure of cell volume is challenging.¹ Often, cells have complex, three-dimensional shapes and are composed of heterogeneous materials that have various optical properties. In addition, because there is considerable cell-to-cell variability in a population, making assumptions based on average cell properties is of limited value.

A clever solution for finding the volume of an object with a complex shape and composed of an unknown material was proposed by Archimedes more than 2000 years ago. By submerging the object in a fluid-filled container, the displaced volume can be easily measured and the measurement corresponds absolutely to the object volume (see Fig. 1a). As long as the object is not porous, the displaced volume is nearly independent of the physical properties of the object. The volume displacement method works effectively even on objects with extremely complex three-dimensional shapes that would be incredibly difficult to quantify with any other method. These same advantages make volume displacement appealing for measuring cell volume.²

Not coincidently, an electrical analogue to the volume displacement principle, called the Coulter counter, has become the most widely used method for quantifying cell volume. Cells are immersed in a conducting electrolyte and the suspension is passed through a small aperture (see Fig. 1b). A constant current is applied across the aperture and voltage pulses are obtained as cells traverse the opening. If the cells are non-permeable to the electrolyte and assumed to be insulators, the voltage pulse height is proportional to the cell volume. In order to retrieve an accurate absolute volume, however, a shape factor needs to be used: While the method is nearly independent of the cell's electrical properties, it is not (unlike Archimedes' displacement) completely independent of a cell's geometry.

An optical Coulter counter

Scientists at Harvard University's Rowland Institute have developed a volume displacement method based not on displaced fluid volume or free electrons, but on light absorption.

Cells are immersed in a buffer that contains an absorbing dye, and, in direct analogy to the electrical Coulter counter, are passed through a microfluidic channel with a finite height. The presence of a cell in the field of view displaces a number of dye molecules that is proportional to the cell volume and the dye concentration. The concentration of the dye can be measured before beginning the experiment and is the only parameter that needs to be known a priori—unlike the sample's shape factor in the Coulter counter.

For an object that does not scatter light, a single-intensity measurement would be enough to reconstruct an accurate volume estimate from the measured intensity. Cells do scatter light, however, and the measured intensity has scattering contributions in addition to dye exclusion. We have minimized the contribution from scattering in two ways. First, we added bovine serum albumin (BSA) to raise the refractive index of the absorbing buffer to approximately match that of the cell. In addition, we image each cell with a second color that is not absorbed by the dye and consequently can be used to subtract out the remaining scattering component.

In this setup, a color camera measures at a throughput of approximately 1000 cells per minute. Figure 2 shows results for optical displacement imaging of leukemia cells and a comparison of volume distributions for two different strains. Intracellular scattering is nearly eliminated (see Fig. 4d), and the height map does not suffer from speckle, haloing, shade-off, or any other artifacts commonly found in quantitative phase microscopes.

FIGURE 2. In optical displacement imaging, color images using blue and green illumination are collected by a color camera (a). The green channel is used to compensate for intracellular scattering (b), while the blue channel contains contributions from both volume displacement and scattering (c). After processing, a thickness map can be retrieved where contributions from scattering have been removed (d). Measurements enable display of the thickness map and volume distribution for 648 HL60 leukemia cells (e) and 552 K562 leukemia cells (f) [3].

No need for sphering

The most common application of cell volume measurements is in hematology where complete blood cell counts quantify red blood cell volume and distribution width. Using these measurements, hematologists can diagnose whether the patient has macrocytic (in which cells are too large) or microcytic (cells are too small) anemia.

Current clinical hematology analyzers use optical scattering to quantify both cell volume and hemoglobin concentration. But because scattering couples geometry to optical properties, cells are required to be "sphered" (that is, processed to swell from disc- to sphere-shape) before measurement. Departure from a spherical shape produces artifacts in measurement that lead to inaccurate results.

Instead, we have applied the optical displacement method to measure cell volume while cells are in their natural, non-spherical state. In addition, we have combined volume displacement with a simultaneous hemoglobin absorption measurement to quantify single cell hemoglobin mass.⁴ Our results are similar to those obtained with a clinical hematology analyzer, Siemens Advia 2120, but we measure a smaller coefficient of variation of hemoglobin concentration (see Fig. 3). Studies have demonstrated that hemoglobin concentration is tightly regulated in the body⁵ and a more narrow distribution of hemoglobin concentration might more accurately represent the actual physiological state.

FIGURE 3. Complete blood counts analyze red blood cell volume and hemoglobin mass and are one of the most frequent clinical tests. To measure both these parameters, the system uses two-color absorption, where red light is used to retrieve cell volume and blue light retrieves hemoglobin mass (a). Color, thickness, and mass maps represent a discoid and parachute red blood cell (b). A comparison of the optics-based system (red) and a clinical hematology analyzer (blue) demonstrates that in addition to volume and mass, the optical system captures images of every cell, which enables study of the morphology of outliers (c) [4].

Multiplex-able

Another major strength of an optical technique, compared to an electrical one, is that it is straightforward to multiplex with other optical techniques. In our tests, we combined this method with optical absorption to create a method for measuring red blood cell volume and hemoglobin mass—a method that we believe is more accurate than the clinical standard. In addition, cell volume measurements can be matched with a fluorescence image or signal to correlate volume and organelle morphology.

Figure 4 shows thickness maps of neutrophils that are captured simultaneously with fluorescence images of their nuclei. Neutrophil volume distributions have been linked to early detection of infection.⁶ Nuclear morphology of neutrophil cells has also been studied in the context of infection due to the fact that nuclei become more segmented as the cell ages. We have begun a project to correlate these two behaviors in blood samples. A better measurement of cell age could have major implications in the early detection of infection when the body is trying hard to make more white blood cells.

FIGURE 4. Images of height maps of neutrophil cells collected by the optical Coulter counter are shown in red. The nuclear morphology is also simultaneously observed using a nucleic acid fluorescence stain, Syto16, shown in blue.

Simpler and full of potential

By reducing and correcting for scattering, we have enabled this optical method to be, like Archimedes' approach, nearly invariant to both the object's shape and physical properties. Unlike volume measurements based on optical scattering and interferometry, the optical Coulter counter does not rely on a priori knowledge or measurement of a cell's refractive index, which results in a significant simplification of the system.

We are also investigating correlations between cell volume and nuclear morphology in white blood cells in the hope of building a diagnostic system for the early detection of infection. The ability of optics to multiplex measurable quantities will further enable cell volume to be correlated with a host of other biochemical and morphological cell properties.

REFERENCES

1. A. Tsur, J. K. Moore, P. Jorgensen, H. M. Shapiro, and M. W. Kirschner, PLoS One, 6, e16053 (2011).

2. W. H. Grover et al., Proc. Nat. Acad. Sci., 108, 10992–10996 (2011).

3. E. Schonbrun, G. Di Caprio, and D. Schaak, Opt. Exp., 21, 8793–8798 (2013).

4. E. Schonbrun, R. Malka, G. Di Caprio, D. Schaak, and J. M. Higgins, J. Cytometry, 85, 332–338 (2014).

5. J. M. Higgins and L. Mahadevan, Proc. Nat. Acad. Sci., 107, 20587–20592 (2010).

6. F. Chaves, B. Tierno, and D. Xu, Am. J. Clin. Pathol., 124, 440–444 (2005).

Ethan Schonbrun, Ph.D., is a junior fellow and the principal investigator of the Optofluidics Cytometry Group at Harvard University's Rowland Institute; e-mail: [email protected]; http://www2.rowland.harvard.edu/book/ethan-schonbrun.