Mechanical Memory in Cells Migrating Through Nanostructured Confining Microenvironments

The mechanical architecture of extracellular matrix (ECM) presents variable levels of stiffness and confinement to nearly all the cells in the body. Migrating cells are exposed to this confinement during diverse biological processes such as embryonic development, cancer metastasis, and immune cell homing, all of which involve cellular relocation from one mechanical niche to another. During this process, cells are also exposed to nanoscale cues in the form of matrix topography, which can be aligned, orthogonal, or random to the direction of migration. While many studies have explored the cellular response to externally-applied compression, it remains unclear why cells actively decide to enter highly confined spaces. We propose that cells retain mechanical memory of their microenvironmental stiffness, and that this memory plays an active role in their migratory potential. Multiple cell types that show non-permanent responses to substrate stiffness (e.g. cancer cells, fibroblasts) were cultured on soft (1 kPa) or stiff (34 kPa) polyacrylamide hydrogel substrates or glass. Cells were exposed to either 3 or 7 days of mechanical dosing prior to transfer to glass coverslips or into microchannel chips. Microchannel chips were fabricated via a two-step photolithography process that results in microchannels with heights of 10 µm, lengths of 150 µm, and widths between 3 and 10 µm. Nanostructuring of microchannel imprints on silicon wafers was performed to provide parallel, orthogonal, or random nanoscale cues. Cells were then assayed for their ability to spread and migrate on 2D surfaces, as well as interact with and enter confinement in diverse microchannel environments. Mechanically-dosed MDA-MB-231 cancer cells did not show any changes in cell or nuclear morphology after dosing on soft, stiff, or glass substrates. Migration speed and the ability to enter confinement also was unaffected by mechanical dosing. On the other hand, HFF-1 fibroblasts were able to retain a significant mechanical memory of their dosing. HFF cells were found to spread on glass significantly slower when they had been preconditioned on 1 kPa substrates compared to either 34 kPa substrates or glass. This effect was primarily in the initial spreading phase, as cells reached equivalent levels of spread area after four hours. Fibroblasts dosed on soft substrates were also found to migrate significantly faster than those on stiff or glass substrates. Interestingly, the amount of time cells were dosed enhanced this difference in migration speed. While dosing did not affect the overall number of cells interacting with confining microchannels, fibroblasts dosed on soft substrates were able to permeate through both moderate (10 µm) and extreme (3 µm) confinement faster than cells dosed on stiff or glass substrates. As a wide variety of cells must navigate diverse microenvironments, our results suggest that the stiffness of the ‘home’ niche may convey important mechanical advantages to cells migrating through confining interstitial spaces in vivo, and that this can influence subsequent migration on or through nanostructured surfaces. Future work on the precise cellular transcriptome and machinery found in dosed fibroblasts but not dosed cancer cells will likely reveal clinically relevant information for targeting metastasizing cancer cells.