Modeling and Analysis of Cancer Cell Motility and Drug Response Under 3D Confinement

Abstract

Introduction/Aim: Previous research has indicated that metastatic cancer cells utilize integrin-mediated surface adhesions to exert force on their surrounding substrate, a type of movement called mesenchymal motility. However, inhibiting mesenchymal motility to slow cancer metastasis has shown limited efficacy, possibly due to the development of bleb-based (amoeboid) motility. Amoeboid motility utilizes increased cellular contractility to produce blebs and possibly facilitate faster movement than mesenchymal motility. If our hypothesis is true, we propose that using drugs solely designed to inhibit mesenchymal motility may be harming patients. Methods: In this project we model the 3D confined environment of metastasizing cancer cells with novel microfluidics channels and seek to elucidate how their motility is altered by anti-metastatic drugs. A375-M2 melanoma cells introduced into our 3D confinement model were treated with drugs targeting mesenchymal motility (CK-666 and Dasatinib) or no drug control. Cell motility was observed over 5 hours by deconvolution microscopy and imaging data was analyzed for speed, directional persistence (directionality), and type of cell protrusions. Results: Our analysis showed that metastatic melanoma cells used two distinct phenotypes when moving through fibronectin coated microchannels. The first phenotype (phenotype 1) included cells that utilized small blebs and cell membrane protrusions while the other phenotype (phenotype 2) included cells that did not produce protrusions and had larger blebs. When the fibronectin was replaced by non-adherent BSA or treated with Dasatinib, cells adopted rounded, blebbing motility similar to phenotype 2. Surprisingly, cells treated with CK-666, an actin branching inhibitor, cells continued to utilize protrusion-dependent motility (phenotype 1). In both CK-666 and Dasatinib treatment groups movement speed almost doubled the speed of untreated cells and increased cell displacement compared to untreated cells. Conclusion: Our research indicates that cells can utilize distinct modes of motility that vary based on their dependence on blebbing and protrusions. In addition, treatment with mesenchymal motility inhibition-based cancer therapy may lead to phenotype switching which can increase cell speed, suggesting that it would be unwise to rely solely on these inhibitors for cancer therapy. However, we propose that these drugs may have higher efficacy when combined with amoeboid motility inhibitors. In the future, we hope to utilize this model to analyze cell motility changes with combinations of mesenchymal and amoeboid motility inhibitors and compare results with in vivo testing.