Predictive Computational Model of Microfluidic Deformability Cytometry Utilizing a Cross-Slot Configuration

S. Hymel, H. Lan, D. Khismatullin
Tulane University,
United States

Keywords: deformability cytometry, microfluidics, deformation index, elongation index

Summary:

Various techniques have been developed to measure deformability of living cells, which can be classified into 1) single-cell methods such as micropipette aspiration and AFM (1-5) and 2) microfluidics-based deformability cytometry (6-11). Microfluidic methods provide several advantages for disease diagnosis over traditional single-cell techniques: 1) high throughput and easy operation, 2) physiological flow conditions, 3) reduced risk of cell activation, and 4) ability to detect the stage of the cell cycle during deformability measurement. Microfluidic deformability cytometry (mDC) (6-9) in which the cells are stretched by extensional flow at the stagnation point of a cross-slot microchannel is one of the most promising assays for high-throughput screening of the mechanical properties (“mechanotype”) of diseased cells. In this method, the cells with different mechanotype are distinguished by comparing the “eyes” of deformation index (DI) vs. cell size density plots. However, mDC does not account for changes in DI due to a shift of the cells from the flow centerline (offset) in the inlet channels or due to pressure fluctuations between two inlet channels. This leads to significant measurement errors and, in particular, to the artificially wide distribution of the DI and cell size. Using our custom computational algorithm for deformable cell migration (VECAM) (12), we simulated cell motion and deformation in a cross-slot microchannel (Fig. 1). In VECAM, the cell and its external environment are a multiphase continuum with moving interfaces (e.g., cell’s cortical layer) tracked by the volume-of-fluid (VOF) method. The viscoelasticity of the cell cytoplasm is described by the Oldroyd-B model. To mimic mDC experiment, we randomized the initial cell placement, cell size, and cell shear elasticity via pseudo-random normal sampling. We also induced pressure fluctuations at the inlets. Our numerical simulation indicates that 1) bigger cells experience larger deformation in the cross-slot area even in the absence of changes in shear elasticity (Fig. 2a); 2) DI rapidly decreases with the cell offset in the lateral direction (y-axis), especially if the cell has a large diameter (Fig. 2b); and 3) mDC loses sensitivity to cell elasticity changes when the Y-offset distance exceeds 5 μm (Fig. 2c). Cell offset in both lateral and transverse (z-axis) directions further decreases the DI (Fig. 2d). If the inlet pressure fluctuates at just 10%, DI decreases by 38% from its value at undisturbed, symmetric flow; and 1% fluctuation leads to a 10% change in DI (Fig. 2e,f). Our data showed excellent quantitative agreement for DI measurement of HL60 cells (9, Fig. 3a). With pseudo-random normal sampling, we get the DI vs. cell size plots (Fig. 3b,c) similar to mDC data (Fig. 3d). From the numerical data in Fig. 2, we identified functional relationships between the offset distance and DI. By applying these relationships to the data in Fig. 3b, we significantly reduced the DI data variability and increased its sensitivity to shear elasticity changes (Fig. 3e). When integrated with predictive computational models of deformable cell migration such as VECAM, deformability cytometry techniques provide a sensitive measurement of the mechanical properties of living cells.