Ice nucleation and freezing in undercooled cells.

DSC has been employed to study the effect of cooling on a range of cells under exclusion of extracellular ice and in the absence of chemical cryoprotectants. In contrast to earlier reports, all the cells studied were found to freeze at temperatures above that indicated for homogeneous nucleation of ice in undercooled liquid water. In the case of human erythrocytes this temperature difference was only 0.5 degrees, but for yeast cells and cells of plant origin the difference amounted to congruent to 9 degrees. Nucleation of ice within the cell (or at the cell wall/membrane) must therefore be initiated by a heterogeneous mechanism. A kinetic analysis of the temperature dependence of nucleation shows the rates to be consistent with the dimensions of the plant cells (or organelles), if these were to be the active nucleators. However, the nucleation kinetics of human erythrocytes are extremely temperature sensitive, and the kinetic parameters only differ by small, though significant, extents from those of the suspension medium. Possible nucleation mechanisms are discussed in terms of the experimental data and the cell dimensions. Finally, one of the underlying assumptions of the kinetic analysis, i.e., that ice growth must be rapid compared to nucleation, has been tested and validated by freeze-fracture electron microscopy.