Deformability of Tumor Cells versus Blood Cells

The potential for circulating tumor cells (CTCs) to elucidate the process of cancer metastasis and inform clinical decision-making has made their isolation of great importance. However, CTCs are rare in the blood, and universal properties with which to identify them remain elusive. As technological advancements have made single-cell deformability measurements increasingly routine, the assessment of physical distinctions between tumor cells and blood cells may provide insight into the feasibility of deformability-based methods for identifying CTCs in patient blood. To this end, we present an initial study assessing deformability differences between tumor cells and blood cells, indicated by the length of time required for them to pass through a microfluidic constriction. Here, we demonstrate that deformability changes in tumor cells that have undergone phenotypic shifts are small compared to differences between tumor cell lines and blood cells. Additionally, in a syngeneic mouse tumor model, cells that are able to exit a tumor and enter circulation are not required to be more deformable than the cells that were first injected into the mouse. However, a limited study of metastatic prostate cancer patients provides evidence that some CTCs may be more mechanically similar to blood cells than to typical tumor cell lines.


Fig. S4. Passage time versus buoyant mass characteristics of murine tumor cell lines
The mouse tumor cell lines were measured with an applied pressure of 1.5 psi, which is the same condition as the measurements of mouse CTCs in the main text (Fig. 4). Plots are on a log-log scale, with an X-axis ranging from 3 pg to 600 pg, and a Y-axis ranging from 0.008 s to 300 s. Dotted grid lines on the X-axis are at 10 pg and 100pg, while dotted grid lines on the Y-axis are at 0.001 s, 0.01 s, 0.1 s, 1 s, 10 s, and 100 s.   The density distribution of PC3 cells measured by the SMR was used to convert CTC diameters to buoyant masses.

Fig. S8
. Entry and transit velocity ratios A) Entry velocity versus volume for EpCAM hi YFP lo and EpCAM lo YFP hi cells. B) Transit velocity versus volume for EpCAM hi YFP lo and EpCAM lo YFP hi cells. C) Entry and transit velocity ratios for each replicate of experiment are shown, comparing EpCAM hi YFP lo to EpCAM lo YFP hi as well as comparing EP5 cells treated with platelets to those treated with buffer. The velocity ratios are calculated in the same manner as the passage time ratios described in the Supplementary Materials and Methods. The entry and transit velocities change similarly, with neither change dominating the other. D) For each measurement, the entry velocity ratio was divided by the transit velocity ratio to more clearly compare the contribution of each in both cell lines having undergone an EMT or co-incubation with platelets. There is no significant difference between the ratios for the two cases (p = 0.4, two-sided Wilcoxon rank-sum test). Also, in both cases the ratios (entry velocity ratio : transit velocity ratio) are not significantly less than 1 (p = 0.5 for EpCAM hi YFP lo :EpCAM lo YFP hi and p = 0.875 for EP5-Platelet:EP5-Buffer, using a one-sided sign test), indicating that frictional roles do not have a dominant contribution to the changes seen in passage time measurements. Table   Table S1. Methods for measuring single-cell deformability

Technology Local or Global Measurement Description References
Atomic Force Microscopy Local  Applies small forces (pN to µN) to locally deform cell's surface  Measures Young's modulus [1][2][3] Microrheology Local  Uses small particles to passively or actively probe viscoelastic properties of cell regions  Magnetic twisting can probe the cell surface [4][5][6][7][8] Micropipette aspiration Local or Entire cell  Suction cell partially or fully into micropipette  Cell geometry provides insight into elastic and viscoelastic properties 9,10 Optical Stretcher Entire cell  Two laser beams apply stretching force to cell  Cell images provide size and deformation information as measure of compliance 11,12 Hydrodynamic deformation Entire cell  Hydrodynamic deformation applied to cells in microfluidic channel  High throughput measurement of aspect ratio or circularity of cell [13][14][15] Microfluidic constrictions Entire cell  Cells are pushed through a constricted channel  Deformability is indicated by amount of time taken to pass through the constriction [16][17][18][19][20]

Blood Cell Preparation
Whole human blood was purchased from Research Blood Components, LLC (Brighton, MA). Mononuclear cells were separated using Histopaque-1077 (Sigma-Aldrich 10771). Polymorphonuclear lymphocytes were enriched by using Lympholyte-poly (Cedarlane CL5070). Red blood cells were measured by diluting whole blood in PBS with 1% (w/v) Kolliphor P188 (Sigma-Aldrich), since the concentration of red blood cells is orders of magnitude higher than that of other cell types. Healthy BALB/c mouse blood was obtained via cardiac puncture. Red blood cells were removed by lysis (150 mM NH4Cl (8.02 g/L), 10 mM KHCO3 (1.00 g/L), 0.1 mM Na2EDTA, pH 7.23) or Histopaque-1077. To isolate live leukocytes and ensure removal of clots and platelets, the remaining cell solution was stained for CD45 (Biolegend 103121) and CD41 (Biolegend 101319) for 15 min at room temperature after pre-incubation with Fc block (Biolegend 101319). DAPI (Life Technologies D1306) was also used to stain dead cells. After resuspending the cells in PBS with 0.5%FBS and 2mM EDTA, cells were sorted using fluorescence-activated cell sorting (FACS) and the CD45 hi /CD41 lo /DAPI lo cells were collected for analysis in the SMR.

Buoyant mass to volume conversion
For measurements where the difference between cell types may be subtle, buoyant mass is converted to volume as a metric for size, since the volume of the cell is more consistently related to passage time than is buoyant mass. The single-cell densities of platelet-treated EP5 cells, buffer-treated EP5 cells, EpCAM hi YFP lo MMTV-PyMT cells, and EpCAM lo YFP hi MMTV-PyMT cells were measured on each day of experiment using the SMR as previously described 25 . The density of the culture medium in which the cells were measured was determined by the resonant frequency of the SMR. The volume of the cell was determined by the ratio of the buoyant mass to the difference in the average cell density (typical interquartile range: 0.003-0.007 g/mL) and the fluid density.

Calculating passage time ratio, removing secondary population of cells
The passage time ratios, as shown in Fig. 2C and F, were calculated as previously described 17 . In brief, the passage time versus volume data was plotted on a log-log scale. A line was fit to each data set, having the same slope, but variable intercepts. The difference between the two intercepts corresponds to the log10 of the ratio of the passage times, which is then converted to the actual passage time ratio by exponentiation.
To obtain accurate linear fits, the data was considered within its linear region (on a log-log scale) containing the majority of the cells for analysis, eliminating small particles or debris. Thus, the lines were fit for EP5 cells having volumes between 800 and 3500 μm 3 , and for MMTV-PyMT cells greater than 1000 μm 3 . As noted in the figure caption (Fig. 2), the EP5 buffer-treated and platelet-treated cells have a second population of cells presumed to be doublets due to their being twice the volume of the majority of the cells in the population. To obtain accurate linear fits, the second population of cells was removed by excluding cells below a given line roughly parallel to the major axis of the second population of cells. The same line was used to remove the second population on both the platelet-treated and buffer-treated cells for all three replicates of the experiment. Fig. S2 shows an example of the removal of the larger population of cells.

Converting imaged cell diameter to buoyant mass
The PC3 prostate cancer cell line was used to estimate the single-cell density of prostate cancer CTCs. PC3 cells were stained for EpCAM and resuspended in PBS with 1% Kolliphor for SMR measurement as would typically be done with patient samples. The single-cell densities of one aliquot of cells were measured in the SMR, while another aliquot was placed in a 24-well plate where they were fixed in 4% paraformaldehyde (PFA) and imaged. From image processing, we found that the maximum Feret's diameter determined by ImageJ software corresponded well to the volume measured by the SMR. The maximum Feret's diameter found for each CTC is shown in Fig. S6. Since the number of CTCs was low and the density distribution of the PC3 cells in 1% Kolliphor buffer was found to be quite spread (interquartile range: 0.017 g/mL, Fig. S7) compared to typical density measurements, instead of using one set density value to convert all of the CTC diameters to buoyant mass, a simulation was used to randomly assign a density value from the measured distribution to each CTC. Buoyant mass was then calculated by the following equation: where R corresponds to the radius of the imaged cell, cell is the density of the cell, and fluid is the density of the buffer used for measurement. The result is plotted in Fig. 5B and 5E. To determine the number of cells that fell in the buoyant mass range of 50 pg to 100 pg, or greater than 100 pg, the simulation was repeated 1000 times. The average number of cells in each buoyant mass range of interest was calculated and reported in the main text.