UCLA

To carry out physiological functions in the human body, cells deform during biological processes including migration, invasion, and extravasation. The ability of a cell to change shape in response to applied stress, specifically cell deformability, is a characteristic that encompasses multifaceted intracellular changes including alterations occurring in the cytoskeleton and nucleus. Recent technological advances in the mechanobiology field have allowed for measuring features of cell deformability, which are valuable for deepening the understanding of disease states and investigating underlying molecular mechanisms that facilitate physical changes to an individual cell. Measurements of cell deformability are especially pivotal in assessing cancer malignancy, where mutations occurring on a single-cell level contribute to metastasis. While deformability is a valuable indicator of molecular mechanisms and abnormalities occurring internal to the cell, cells are also sensitive and can respond to external forces from the microenvironment. Recent advances exploit mechanical deformations to induce transient poration of the plasma membrane and altered epigenetics, which can be leveraged for cell engineering applications through the development of enabling tools and technologies.

Fluidic technologies can be built by tunable device designs and harnessed to control fluid flow to investigate cells at the micrometer level and simulate physiological geometries. In the introductory chapter, I present an overview of fluidic methods to assess cell deformability including filtration approaches and single-cell microfluidics. With filtration assays, the major component is a membrane containing micrometer-scale pores in which a suspension of cells are driven to deform through the pores, and the resultant volume indicates relative deformability. While filtration assays are simpler to implement, single-cell microfluidic devices are powerful in extracting multiple features of an individual cell. Single-cell microfluidic devices — coupled with high-speed imaging, microfabrication techniques, and automated computational analysis — involve the rapid deformation of cells from fluid shear stresses, extensional flow, or geometric constrictions.

In the second chapter I present a microfluidic technique called Quantitative Cyclical Deformability Cytometry (qc-DC) where single cells deform through multiple constrictions. To implement physically phenotyping, the timescale for cell transit is measured as features in which shorter transit times indicate increased cell deformability. Using qc-DC to characterize drug resistance in leukemia, I find that leukemia cells that survived chemotherapy treatment have altered physical phenotypes and increased cell deformability. To further assess the predictive power of physical phenotyping, single cells are computationally randomized into mixed populations containing varying proportions of drug-treated, surviving cells. Using supervised machine learning, I demonstrate that the transit time features obtained from qc-DC can be used to accurately classify mixed populations. These findings show the potential for physical phenotyping as a prognostic approach to predict relapse and improve patient outcomes.

Additionally, this dissertation presents a high-throughput platform to exert a controlled number of rounds of mechanical deformation on cells to stimulate altered cellular expression. Custom built using 3D printing technologies and soft lithography, the platform can subject cells to 0, 20, 40, or 80 rounds of deformations. With the ability to control the magnitude and duration of deformations that cells experience, the platform provides potential to expose cells to mechanical stimuli in a dose-dependent manner.