470850 Hemica-Hydrogel Encapsylated Micro-Channel Array in Cancer Metastasis
Methods: Silicone wafers with the desired micro-channel design are created via photolithography and used as a mold for PA. Access holes to the PA imprinted design are punched out. Glass cover-slips are activated and PA is cured on them creating a thin (60-100um) layer. Both gel layers are allowed to reach their respective maximum swelling. The upper layer with the design on its lower surface is then sandwiched to the bottom gel/glass via a NHS-ester ligand. The encapsulated micro-channels are coated with collagen type I using Sulfo-SANPAH as a ligand. The device is seeded with cells and subsequently submerged into media for live microscopy. To quantify cell speed, persistence and cell morphology we used FIJI and a custom MATLAB code. Statistical significance was assessed with two-way ANOVA test.
Results and Discussion: MDA-MB-231 breast cancer cells were seeded in HEMICA and studied in a wide range of microenvironments spanning full confinement (WxH=3x10μm2) to 2D spaces (WxH=50x10μm2 or 200x50μm2). Cell speed shows a biphasic behavior with maximum speed in 10um width (similar to the dimension of the nucleus) and intermediate stiffness 13kPa (similar to the stiffness of human breast tissue). Cell persistence is lower in the softest gels with higher significance on the larger width channels, suggesting a correlation between stiffness and directed migration. Although large widths allow cells to adhere preferentially to one channel wall (contact guidance), we show that low stiffness (7kPa) reduces contact guidance based on the frequency cells cross the channel center-line and the probability of following a path within the central regions of the channels. In addition, intermediate stiffnesses increase cell polarization, especially on small widths, while low stiffnesses increase cell circularity indicating reduced cell-matrix adhesions. On the other hand, cell projected area increases with increasing channel width independent of stiffness. Moreover, the cell variance of area and circularity is higher in large widths of low stiffness, indicating major morphodynamic changes in the process of migration. In contrast, the variance of solidity (how protrusive cells are) is higher in confinement with intermediate stiffness, which implies that cells in full confinement change the number of their protrusions frequently in order to achieve an optimum mode of migration. Through our ongoing experiments we are identifying the mechanobiology of cancer cell migration by establishing the link between stiffness, myosin II, microtubules and traction forces in confinement.
Conclusion: We have created a novel and versatile device which mimics in vivo microenvironments. HEMICA can be used to study collective migration, cell-cell communication, cell-matrix interactions and cell population dynamics, offers organ-on-a-chip applications and allows post-migration studies. Our results demonstrate that the geometry of the cell microenvironment and tissue compliance promote different modes of migration ultimately affecting all stages of cancer metastasis.