Automated three-dimensional single cell phenotyping of spindle dynamics, cell shape, and volume

Abstract

We present feature finding and tracking algorithms in 3D in living cells, and demonstrate their utility to measure metrics important in cell biological processes. We developed a computational imaging hybrid approach that combines automated three-dimensional tracking of point-like features, with surface reconstruction from which cell (or nuclear) volume, shape, and planes of interest can be extracted. After validation, we applied the technique to real space context-rich dynamics of the mitotic spindle, and cell volume and its relationship to spindle length, in dividing living cells.These methods are additionally useful for automated segregation of pre-anaphase and anaphase spindle populations. We found that genetic deletion of the yeast kinesin-5 mitotic motor cin8 leads to abnormally large mother and daughter cells that were indistinguishable based on size, and that in those cells the spindle length becomes uncorrelated with cell size. The technique can be used to visualize and quantify tracked feature coordinates relative to cell bounding surface landmarks, such as the plane of cell division. By virtue of enriching the information content of subcellular dynamics, these tools will help advance studies of non-equilibrium processes in living cells.

Nonthermal fluctuations of the mitotic spindle

“…1(B)), obtaining a spatial resolution of o10 nm in 3D for each pole. 21 All analysis was carried out using custom scripts in MATLAB (MathWorks).…”