Characterizing cellular mechanical phenotypes with mechano-node-pore sensing

The mechanical properties of cells change with their differentiation, chronological age, and malignant progression. Consequently, these properties may be useful label-free biomarkers of various functional or clinically relevant cell states. Here, we demonstrate mechano-node-pore sensing (mechano-NPS), a multi-parametric single-cell-analysis method that utilizes a four-terminal measurement of the current across a microfluidic channel to quantify simultaneously cell diameter, resistance to compressive deformation, transverse deformation under constant strain, and recovery time after deformation. We define a new parameter, the whole-cell deformability index (wCDI), which provides a quantitative mechanical metric of the resistance to compressive deformation that can be used to discriminate among different cell types. The wCDI and the transverse deformation under constant strain show malignant MCF-7 and A549 cell lines are mechanically distinct from non-malignant, MCF-10A and BEAS-2B cell lines, and distinguishes between cells treated or untreated with cytoskeleton-perturbing small molecules. We categorize cell recovery time, ΔTr, as instantaneous (ΔTr~0 ms), transient (ΔTr⩽40 ms), or prolonged (ΔTr>40 ms), and show that the composition of recovery types, which is a consequence of changes in cytoskeletal organization, correlates with cellular transformation. Through the wCDI and cell-recovery time, mechano-NPS discriminates between sub-lineages of normal primary human mammary epithelial cells with accuracy comparable to flow cytometry, but without antibody labeling. Mechano-NPS identifies mechanical phenotypes that distinguishes lineage, chronological age, and stage of malignant progression in human epithelial cells. Supplementary information The online version of this article (doi:10.1038/micronano.2017.91) contains supplementary material, which is available to authorized users.


Figure S2
Signal processing by customized MATLAB code. The acquired signal (a) is first low-pass filtered (b) to remove noise. The base-line is then normalized (c) to remove any drift. (d) A derivative cut-off detection is subsequently employed as an index to determine the start and end point of each pulse. (e) Finally, the current pulse magnitude and duration are measured based on this index. Within each blue box, the central line is the median and the edges of the box correspond to 25% and 75% of the wCDI distribution.

Figure S4
Relationship between mechanical properties and wCDI. (a) Comparison of wCDI with cortical tension as determined by micropipette aspiration of Jurkat, MCF7, and MCF10A cells. The wCDI is inversely related to cortical tension. Error bar indicates standard deviation for wCDI and standard error for cortical tension. (b and c) Comparison of wCDI with the elastic modulus, as measured by AFM, of breast cell lines (b) and lung cell lines (c). Within each blue box, the central line is the median and the edges of the box correspond to 25% and 75% of the wCDI distribution. The orange symbols are the reported elastic modulus of each cell line [3][4][5][6][7][8][9][10] . The trend of wCDI over various cell lines is inversely proportional to the elastic modulus.

Figure S5
Computational modeling of the electric field when a cell transits each section of the mechano-NPS microfluidic channel. The fine lines correspond to the calculated electric-field lines in each section of the microfluidic channel, and the white circle corresponds to a cell. As determined, the electric-field density, J, in the contraction channel is greater than that in the node. Computational simulation was performed using Comsol Multiphysics 5.0.

Figure S6
Schematic and representative mechano-sensing current pulses produced by an HMEC to illustrate the defined cellular recovery types after compressive deformation. (a) Instant recovery: The current drop (red dashed line) with respect to the baseline (blue dashed line) at the node-pore before and after the contraction channel are defined as ΔI np and ΔI r , respectively. We define "instant recovery" when a cell recovers to its original size and shape immediately after exiting the contraction channel and ΔI r = ΔI np . In this case, ΔT r~0 . (b) Transient recovery corresponds to the case when the cell recovers to its original size and shape, again defined as ΔI r = ΔI np , within the span of the nodepore sequence immediately following the contraction channel. Here, ΔT r ≤ 40 ms. (c) Prolonged recovery corresponds to the case when the cell does not recover to its original size and shape. In this specific case, ΔI r ≠ΔI np over the time scale recorded by mechano-NPS (ΔT r 440 ms). All schematic drawings (a-c, top) show the idealized mechano-NPS current pulse. The representative current pulses (a-c, bottom) show that the current at the "node" does not reach to the baseline current and has a more peak-like shape. This is due to the fast flow rate of the cells and the short length of the "node" segment.

STATISTICAL ANALYSIS TO COMPARE wCDI WITH FACS ANALYSIS FOR PRIMARY HMEC STRAINS
We employed a χ 2 test to determine whether there were any statistically significant differences between the obtained wCDI and FACS results: The observed values, O i , and expected values E i , were the number of MEP and LEP cells as measured by mechano-NPS and FACS, respectively. Supplementary Table 3 shows the χ 2 values for the different HMEC strains. For a P-value = 0.05, χ 2 = 3.841. Thus, there are no statistically significant differences between mechano-NPS and FACS. Polystyrene microspheres (Polysciences, #64155) suspended in PBS were measured with our NPS platform to determine the node-pore channel's effective diameter, D eff.np (n = 30) and the effective diameter of the contraction channel D eff,cont (n = 12). The d avg , σ d , ΔI/I, L, and σ eff correspond to the average diameter of the microspheres, the diameter standard deviation, the ratio of the current drop to baseline current, the channel length, and the effective diameter standard deviation, respectively.   To ensure adequate power to detect differences within experimental groups, we measured the power of each group using 2-sample and 1-sided power analysis with 95% confidence interval. The analyzed sample size, N a , provided the adequate power value (≥0.80) throughout the all experimental cases. In this table, N.A indicates power analysis is not applicable due to the high P-value (P ≥ 0.05).
Screening cell mechanical phenotypes electronically J Kim et al