Fig. 1: Transition between super- and sub-diffusive cell migration is driven by substrate stress relaxation and cannot be explained by conventional motor clutch models. | Nature Communications

Fig. 1: Transition between super- and sub-diffusive cell migration is driven by substrate stress relaxation and cannot be explained by conventional motor clutch models.

From: Glassy adhesion dynamics govern transitions between sub-diffusive and super-diffusive cancer cell migration on viscoelastic substrates

Fig. 1: Transition between super- and sub-diffusive cell migration is driven by substrate stress relaxation and cannot be explained by conventional motor clutch models.The alternative text for this image may have been generated using AI.

A Schematic representation of a cancer cell seeded on an alginate-rBM IPN gel. B Normalized stress curve for fast and slow-relaxing IPNs. C Mean squared displacement (MSD) for fast and slow-relaxing IPNs (each data point represents the average of multiple cells measured on a single gel). D MSD \(\alpha\) value showing super-diffusive migration for fast-relaxing substrates and sub-diffusive migration for slow-relaxing substrates. The unpaired two-tailed Student’s t test was used for data analysis: P = 5.13E−05; n = 6 (fast) and 5 (slow) fits/experiments, each for over 1000 cells. E, F Track straightness and Normalized velocity autocorrelation showing higher persistency for fast relaxing substrates compared to slow relaxing substrates. In E, the unpaired two-tailed Student’s t test was used for data analysis: P = 2.09E−21; n = 203 (fast) and 87 (slow) over three independent samples. G Constant \({\tau }_{off}\) model. (i) MSD-time fit shows similar behavior/slope obtained from conventional model simulations for both fast and slow relaxing substrates (non-dimensionalised using time scale \(1/{r}_{{on}}\) and length scale \({v}_{p}/{r}_{{on}}\)). (ii) Diffusivity exponent values confirm that the conventional models capture only purely diffusive modes. The unpaired two-tailed Student’s t test was used for data analysis: P = 0.94; n = 5 (fast) and 5 (slow) fits, each for 100 independent simulations. (iii) The model with a constant off-rate fails to capture the sub-diffusive migration mode and the differences in migration persistence. The unpaired two-tailed Student’s t test was used for data analysis: P = 0.72; n = 200 (fast) and 200 (slow) independent simulations. (iv) Normalized velocity autocorrelation (VAC) function fails to capture the exponential decay of velocity due to the absence of distributed trapping times and step-sizes. (Black lines represent mean in the violin plots. ‘ns’ represents not significant p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Back to article page