Figure 1.
Cytostatic/cytotoxic effects of 5-FU on cardiomyocytes and endothelial cells.
The cytostatic effect of 5-FU on HCMs (A) was not marked but statistically significant starting at 48 hours and from 10 μM 5-FU. The inhibitory effect of 5-FU was significant in endothelial cells (B) already after 24 hours at the intermediate concentration of 10 μM. The inhibition of cell viability was maintained up to 96 hours. Drug efficacy was also tested on the HT29 and HCT116 colorectal cancer cell lines (C, D). Mean ± S.E.M. of four different experiments for each cell type is shown. (E) MTT data at 72 hours were used to calculate the EC50 values using the non-linear regression function of GraphPad Prism. 5-FU concentrations are reported in μM on a Log(10) scale, DMSO: 0.2% DMSO (vehicle) negative control, 0.1% saponin, positive control.
Figure 2.
Effects of 5-FU on HCMs and HUVECs integrity and cycling ability.
Membrane damage was calculated as the ratio between released LDH and total LDH. The effect of 5-FU on LDH release was statistically significant at 5-FU concentrations higher that 10 μM, less marked on cardiomyocytes (*P<0.05; **P<0.001) (A) with respect to HUVECs (****P<0.0001) (B). Mean ± S.E.M. of three different experiments for each cell type is shown. After 96 hours of 5-FU treatment cells were collected and stained with Annexin-V/7AAD to determine induction of apoptosis. Cumulative graphs of three independent experiments showing Annexin V+/7AAD± cells are shown. Higher concentrations induced statistically significant amount of apoptotic cells as compared to negative control both in HCM (* P<0.05) (C) and HUVEC (**P<0.01; ***P<0.005) (D). Impairment of proliferative capacity was assessed using BrdU incorporation after 84 hours of treatment with 5-FU. BrdU-unlabeled cells (BrdU-) represent cells in G1 and G2/M phases, BrdU-labeled cells (BrdU+) represent replicating cells in S phase. Staining of HCMs (E) revealed a certain degree of replicating capability in the negative control and at the lower concentrations of 5-FU, with complete absence of cells in the BrdU+ fraction in doses of 10 μM or more. Endothelial cells (F) display a dose dependent accumulation of BrdU- cells in G0 in the presence of 5-FU. Mean ± S.E.M. of three different experiments for each cell type is shown. Negative controls (Ctrl-) consisted of cells treated with vehicle alone (0.2% DMSO), positive controls (Ctrl+) were cells treated with 0.1% saponin.
Figure 3.
Ultrastructural analysis by electron microscopy.
HCMs in culture without any treatment (A 1, 4) or treated with DMSO (A 2, 5) showed only very mild signs of distress. HCMs treated with 1 mM 5-FU for 72 and 96 hours (A 3, 6) showed signs of suffering including rupture of the mitochondrial christae, dilatation of reticulum cysternae and autophagic vacuoles. HUVEC in culture without any treatment (B 1, 4) or treated with DMSO (B 2, 5) did not show any signs of mitochondrial rupture. HUVEC treated with 1 mM 5-FU for both 72 and 96 hours (B 3, 6) showed several signs of injury (rupture of mitochondrial christae and dilatation of reticulum cysternae).
Figure 4.
Acidic Vesicular Organelles (AVOs) formation in HCM after 5-FU treatment.
AVOs formation is associated with the establishment of an autophagic process. Prolonged treatment with 5-FU significantly increased AVOs accumulation (*P<0.05; **P<0.01; ***P<0.005; ****P<0.0001) in HCM (A) as compared to control. Both 48 as well as 96 hours treatment of HUVEC (B) did not significantly influence the amount of acidic compartment staining over time. Mean ± S.E.M. of four different experiments per cell types are shown.
Figure 5.
LC-3 expression as autophagic marker induced by 5-FU treatment.
Staining with anti-LC3 showed punctae formation associated with the induction of the autophagic process. HCM (A) at all concentrations evaluated showed presence of LC-3 positive vesicles, while in HUVEC (B) the presence of these structures were not detectable unless using the highest 5-FU concentration. Cloroquine used as a positive control (Ctrl+) clearly induced vesicle formation. (C) Western blot analyses confirmed LC3 expression and increase on HCMs at the higher concentration of the drug.
Figure 6.
Induction of reactive oxygen species by 5-FU.
DCFH-DA stained cells revealed an increased ROS production after 48, 72 and 96 hours of treatment with increasing concentrations of 5-FU. The percentage of cells positive for ROS production was statistically significant at several concentrations examined for both cell types (A, B). *P<0.05; **P<0.01; ***P<0.005. 4-HPR and 0.2% DMSO (vehicle) were used as positive and negative controls. Mean ± S.E.M. of four different experiments for each cell type is shown.
Figure 7.
Effects of a ROS scavenger on ROS and AVOs formation.
HCM and HUVE cells treated with 5-FU at the indicated concentrations together with the ROS scavenger NAC (10 mM) resulted in a reduction of ROS production. Both relative ROS as percentage of negative control (A, B) and representative histograms of flow cytometric data (C, D) are shown; light grey histograms 10 μM 5-FU, dark grey histograms 100 μM 5-FU, black histograms 1000 μM 5-FU; grey shaded histograms = negative control (Ctrl-); solid lines = 5-FU; dotted lines = 5-FU+NAC. NAC also reduced formation of AVOs in HCM (E), suggesting that ROS is also involved in the autophagic response in these cells. Since endothelial cells did not show induction of AVOs, NAC has little effect on AVO production in these cells (F). *P<0.05; **P<0.01. Mean ± S.E.M. of three different experiments for each cell type is shown.
Figure 8.
After 72 hours of treatment with 5-FU (shown in μM), both cell types display a dose-dependent increase in the percentage of senescent cells. Representative images of β-GAL-stained blue-senescent cells are shown for 5-FU or DMSO treated HCMs (A) and endothelial cells (C). The ratio between the β-galactosidase positive blue-stained cells and total nuclei was calculated for each observed field and the results reported as mean of percentages of senescent cells per treatment (B, D). Five fields per conditions were observed; mean ± S.E.M. of three different experiments for each cell type is shown. (*P<0.05; **P<0.01; ***P<0.005; ****P<0.0001).
Figure 9.
The CT26 murine colon adenocarcinoma cell line was used to evaluate effect of 5-FU in vivo on BALB/c mice. Sensitivity to 5-FU was demonstrated both in vitro in MTT (A, inset) and in vivo (A, B). Both treatment schedules (1 mg/kg every day, 10 mg/kg every other day) significantly reduced tumor growth (A) and weight upon sacrifice (B) as compared to control (PBS). Hematoxylin/eosin staining of renal and cardiac tissues from PBS and 5-FU treated animals (C). Transmission electron microscopy of renal tissues showed alterations (cytoplasmic vacuolization, membrane breakage) of endothelial cells (insets) in both treated groups with respect to controls (D). Cardiac tissues showed intact cardiomyocytes (nuclei indicated by thick white arrows) and occasional minor alterations of endothelial nuclei (thin black arrows) at the 10 mg/kg dose.