Lowering Etoposide Doses Shifts Cell Demise From Caspase-Dependent to Differentiation and Caspase-3-Independent Apoptosis via DNA Damage Response, Inducing AML Culture Extinction

Abstract

Cytotoxic chemotherapy, still the most widely adopted anticancer treatment, aims at eliminating cancer cells inducing apoptosis with DNA damaging agents, exploiting the differential replication rate of cancer vs. normal cells; efficiency is evaluated in terms of extent of induced apoptosis, which depends on the individual cell sensitivity to a given drug, and on the dose. In this in vitro study, we report that the concentration of etoposide, a topoisomerase II poison widely used in clinics, determines both the kinetics of cell death, and the type of apoptosis induced. We observed that on a set of myeloid leukemia cell lines, etoposide at high (50 uM) dose promoted a rapid caspase-3-mediated apoptosis, whereas at low (0.5 uM) dose, it induced morphological and functional granulocytic differentiation and caspase-2-dependent, but caspase-3-independent, cell death, displaying features consistent with apoptosis. Both differentiation and caspase-2- (but not 3)-mediated apoptosis were contrasted by caffeine, a well-known inhibitor of the cellular DNA damage response (DDR), which maintained cell viability and cycling, indicating that the effects of low etoposide dose are not the immediate consequence of damage, but the result of a signaling pathway. DDR may be thus the mediator responsible for translating a mere dosage-effect into different signal transduction pathways, highlighting a strategic action in regulating timing and mode of cell death according to the severity of induced damage. The evidence of different molecular pathways induced by high vs. low drug doses may possibly contribute to explain the different effects of cytotoxic vs. metronomic therapy, the latter achieving durable clinical responses by treating cancer patients with stable, low doses of otherwise canonical cytotoxic drugs; intriguingly caspase-3, a major promoter of wounded tissue regeneration, is also a key factor of post-therapy cancer repopulation. All this suggests that cancer control in response to cytotoxic drugs arises from complex reprogramming mechanisms in tumor tissue, recently described as anakoinosis.

Keywords: AML; apoptosis; caspase-2; caspase-3; differentiation; metronomic chemotherapy.

FIGURE 1

0.5 uM etoposide induces caspase-independent apoptosis. (A) Fraction of apoptotic U937 cells after treatment with 50 uM or 0.5 uM etoposide ± pan caspase inhibitor (Z-VAD-fmk). Effect of caspase-3 (B) and caspase-2 (C) inhibitors on 50 uM and 0.5 uM etoposide-induced apoptosis (measured as nuclear fragmentation). Results are the average of 3 independent measurements ± SD: ^∗p < 0.05. (D) Caspase activation evaluated by western blotting analysis at the indicated times of etoposide (50 uM and 0.5 uM) treatments. Results are representative of 2 independent experiments.

FIGURE 2

TEM analysis of cells induced to apoptosis by etoposide. (A) U937 treated with 50 uM etoposide showing canonical apoptotic figures, including chromatin condensation at the nuclear margin. (B–D) Cells treated with 0.5 uM etoposide. (B) shows recognizable apoptotic figures, but with atypical membrane features, e.g., two nuclear fragments juxtaposed and separated by pore-enriched double nuclear membrane (white arrow) or interruptions of the nuclear membrane (arrowhead). ^∗ indicates nucleolar alterations. (C) Condensed chromatin spherical masses without nuclear membrane (arrowhead). (D) Fourfold membrane network connecting dense chromatin masses in late apoptosis, with the outer sheet wrapped by ribosomes (arrows).

FIGURE 3

0.5 uM etoposide induces nuclear blebbing. Untreated U937 cells (A) and cells treated with 0.5 uM etoposide (B–E). (C) is the magnification of the area indicated by the square in (B). ^∗ marks nucleoli in transition from ring-shaped nucleolus and nucleolus with nucleolonemas. Results are representative of 2 independent experiments.

FIGURE 4

0.5 uM etoposide promotes granulocytic-like morphology and differentiation-related apoptosis. (A) Cells with peculiar fragmentation of the nuclei in 2–3 segments with irregular shape, strongly resembling polymorphonuclear cell nuclei. Nuclei show dilated perinuclear membrane blank (arrowhead). (B) Cells presenting apoptotic nuclei and granulocyte-like characteristics, including electrondense granules at the cell periphery or in the process of being extruded.

FIGURE 5

0.5 uM etoposide promotes intracellular granularity in U937 cells. (A) Cytofluorimetric dot plot analysis of forward (size) vs. side (granularity) scatter in U937 cells untreated or after 24 or 48 h of 0.5 uM etoposide, before (top line) or at 30 min after addition of 200 ng/mL PMA (bottom line). Representative histogram of 3 experiments performed with similar results; the % values indicate the fraction of cells falling in the gated area. (B) Cytofluorimetric analysis of superoxide production (DHE signal) in U937 untreated or after 24–48 h of 0.5 uM etoposide, before (left bars) or at 30 min after addition of 200 ng/mL PMA (right bars). Results are provided as the mean value, and are the average of 3 independent measurements ± SD: ^∗p < 0.05 and ^∗∗p < 0.01. (C) Cytofluorimetric profile of DHE stained U937 at 48 h after 0.5 uM etoposide (left); the right graph shows the side vs. forward scatter dot plot of the cells falling in the two DHE peaks.

FIGURE 6

0.5 uM etoposide induces a caffeine-sensitive DNA damage response determining apoptosis and differentiation. (A) Cell cycle analysis of U937 cells treated with 0.5 uM etoposide ± caffeine (caff). Representative histograms of 4 experiments performed with similar results. (B) Apoptosis (nuclear fragmentation) in U937 cells treated with 50 uM or 0.5 uM etoposide ± caffeine (caff). Results are representative of 3 independent experiments ± SD: ^∗p < 0.05. (C) Cytofluorimetric dot plot analysis of forward (size) vs. side (granularity) scatter in U937 cells untreated or after 24 h of 0.5 uM etoposide ± caffeine (caff). Representative dot plot of 3 experiments performed with similar results. (D) Correlation between G2 arrest and granulation induced by 0.5 uM etoposide at 24 h. G2/G1 ratio was calculated on the basis of quantification of cells in each phase of the cell cycle by means of specific markers (X-axis); the % fraction of cells with higher side scatter (SSC) was calculated as described for panel B (Y-axis). R² = 0.6513 indicates good linear correlation.