Abstract
Cell migration, an essential process for normal cell development and cancer metastasis, differs from a simple random walk: the mean-square displacement (〈(Δr)2(t)〉) of cells sometimes shows non-Fickian behavior, and the spatiotemporal correlation function (G(r, t)) of cells is often non-Gaussian. We find that this intriguing cell migration should be attributed to heterogeneity in a cell population, even one with a homogeneous genetic background. There are two limiting types of heterogeneity in a cell population: cellular heterogeneity and temporal heterogeneity. Cellular heterogeneity accounts for the cell-to-cell variation in migration capacity, while temporal heterogeneity arises from the temporal noise in the migration capacity of single cells. We illustrate that both cellular and temporal heterogeneity need to be taken into account simultaneously to elucidate cell migration. We investigate the two-dimensional migration of A549 lung cancer cells using time-lapse microscopy and find that the migration of A549 cells is Fickian but has a non-Gaussian spatiotemporal correlation. We find that when a theoretical model considers both cellular and temporal heterogeneity, the model reproduces all of the anomalous behaviors of cancer cell migration.
Highlights
Cell migration is essential to normal cell development[1,2,3], cancer metastasis[4,5,6] and wound healing[7,8,9]
We find that only the CTH model may reproduce the experimental results for G(r, t) and the spatiotemporal correlation function gi(r, t) of each single cell of A549 lung cancer cells, and that both cellular and temporal heterogeneity need to be taken into account to elucidate single-cell migration
We investigate the migration of A549 cells that possess intermediate characteristics between epithelial and mesenchymal states
Summary
Cell migration is essential to normal cell development[1,2,3], cancer metastasis[4,5,6] and wound healing[7,8,9]. Metzner et al.[36] proposed a statistical framework to model and analyze heterogeneous cell migration of the breast carcinoma cell line MDA-MB-231 They found that in the presence of such temporal heterogeneity, the spatiotemporal correlation function could be non-Gaussian, which a simple random walk model can not elucidate. We find that only the CTH model may reproduce the experimental results for G(r, t) and the spatiotemporal correlation function gi(r, t) of each single cell (not averaged over the population of cells) of A549 lung cancer cells, and that both cellular and temporal heterogeneity need to be taken into account to elucidate single-cell migration
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
