Figure 4.

(A) Normalized frequency distributions of times of cell death (onset times of the PI signal) are plotted against time for the indicated NP doses. Note that the distributions shift to earlier time points and get narrower with increasing NP dose. The distributions are fitted to log-normal functions (red curve); (B) The cumulative fraction of dead cells is plotted against time for NP doses of 0.1 $\mathsf{\mu }$g · mL${}^{-1}$ (light gray), 1 $\mathsf{\mu }$g · mL${}^{-1}$ (medium gray), 10 $\mathsf{\mu }$g · mL${}^{-1}$ (dark gray), and 100 $\mathsf{\mu }$g · mL${}^{-1}$ (black). At the lowest dose 20% of the cells die within 24 h, whereas at the highest dose all cells (100%) are dead by this point. The cumulative distributions are fitted to log-normal functions (red curve); (C) Standard dose-response curves with fraction of dead cells which can be extracted from the distribution of (B) for several different time points between 10 and 24 h (black to light gray). Data are fitted to dose-response functions (dashed lines). The ${\mathrm{EC}}_{50}$ values shift towards lower dose with increasing late time endpoints (black arrow); (D) The ${\mathrm{EC}}_{50}$ values are plotted in logarithmic scale against the endpoints. At an end-point of 18 h, the ${\mathrm{EC}}_{50}$ value approaches a constant value; (E, F) Time points ${\tau }_{50}$ (E) and a rate-dependent σ (F) for the cumulative distributions are plotted against dose. Both values exhibit dose-response behavior