(91d) Fractional Re-Distribution Among Motility States during Ageing | AIChE

(91d) Fractional Re-Distribution Among Motility States during Ageing


Phillip, J. - Presenter, Johns Hopkins University
Zamponi, N., Weill Cornell Medicine
Phillip, M. P., Weill Cornell Medicine
Daya, J., Weill Cornell Medicine
McGovern, S., Lummus Technology, a CB&I Company
Williams, W., Weill Cornell Medicine
Jayatilaka, H., Johns Hopkins University
Wu, P. H., Johns Hopkins University
Walston, J., Weill Cornell Medicine
Wirtz, D., Johns Hopkins University
Ageing can be defined as the accumulation of dysfunctions with the passage of time that limits the ability of organisms, tissues and cellular systems to absorb and rebound after perturbations and stresses. As integrators of molecular signals, cells offer a meso-scale view of ageing, with cellular dysfunctions occurring prior to the manifestation of age-related diseases at the clinical level. Since cells typically display dynamic phenotypes both in health and disease, we postulated that analyzing dynamic cellular movements may provide significantly more mechanistic insights relative to the standard “snapshot” data from fixed cells.

Typically, cell motility is measured in bulk, and when measured at the single-cell level, results are reported as average changes in speed and persistence. Here, we develop a novel single-cell analysis approach that defines cell motility states, and classifies cells as a function of age-related spatio-temporal motility patterns. This analysis revealed an unforeseen shift in cellular heterogeneity with age. Using primary dermal fibroblasts derived from healthy donors, we measured the motility patterns for hundreds of single cells spanning an age range of 2-92 years. Here we show that the global age-dependent decrease in bulk motility is not due to decreased motility for all cells, but rather from the fractional re-distribution among motility states.

Using the anisotropic persistent random walk together with hierarchical clustering analysis, we first defined spatial clusters that describe spatial movement patterns of cells. From this analysis we found that: (1) Age does not solely define spatial clusters, but each cluster harbors a mixture of cells from each age, (2) donors separate into polarized spatial clusters as a function of age, and (3) there is a significant decrease in the cellular heterogeneity (measures by the Shannon Entropy) for older adults.

Secondly, to define the cellular activity we converted the x-y trajectories into one-dimensional profiles to define temporal patterns of motion. This analysis identified four activity clusters, that describe the bursty vs. consistent movements of cells. We found that cells from young donors were highly enriched for activity clusters showing consistent movement with infrequent, short movement bursts. In contrast, cells from older adults were enriched for clusters showing longer, frequent bursts of movements.

Because of the complimentary information provided by the spatial and activity clusters, we co-classified each cell and computed the abundance of cells within each of the 32 motility states (8 spatial x 4 activity). Analysis of the abundances per age group showed not only significant enrichments and depletions for various motility states, but also a significant decrease in the heterogeneity of cells derived from older adults. This work reveals that ageing information is encoded within the intrinsic cellular variations in motility.

Together, our study presents a general framework to analyze single-cell motility based on spatial- and activity-patterns of cells. This analysis provides novel insights into the ageing process, with potential applications for precision health via the development of robust biomarkers for stratification in health and disease.