Figure 1.
Comparison of dose dependent effects of phenformin and metformin in cancer cell lines.
Cells were treated for 2(A) or the number of live cells (B–F) was determined. (A) E6E7Ras cells, a mouse model of HPV+ head and neck squamous cell carcinoma, (B) B16F10 mouse melanoma cells, (C) A549 human lung adenocarcinoma cells, (D) MCF7 human breast cancer cells, (E) CT26 mouse colon cancer cells, and (F) DU145 human prostate cancer cells. *: P<0.05.
Figure 2.
Synergism between phenformin and oxamate in mediating cancer cell death.
(A) E6E7Ras cells were treated for 2 days with oxamate at the indicated concentrations (0–80 mM) and then dead cells were counted by flow cytometry. (B, C) The indicated cells lines were treated with varying concentrations of phenformin, oxamate, or combinations of the two drugs. In (B) cells were treated for 1, 2, or 3 days prior to counting dead cells. In (C) cells were treated for 24 hours before determining number of dead cells. C: control, P: phenformin, O: oxamate, PO: phenformin+oxamate. In (C) the numbers below each bar indicate concentrations of each drug in mM (e.g., P0.5O20 means P 0.5 mM+O 20 mM). * indicates a synergistic effect in the group PO compared with the other groups.
Figure 3.
Changes in lactate and pH of the medium in cells treated with phenformin and oxamate.
CT26 cells were treated with the indicated compounds for 1, 2, or 3(A) or medium pH (B) was determined. P: phenformin 1 mM, O: oxamate 40 mM, PO: phenformin 1 mM+oxamate 40 mM, C: untreated control. *: P<0.05 compared with the other groups. †: P<0.05 compared with the group C and P.
Figure 4.
Complex I inhibition by phenformin.
(A) CT26 cells were treated with or without phenformin for 24 hours and then extracts were prepared to measure complex I activity as described in Materials and Methods. The Y axis is % of complex I activity when the activity of complex I in the control group is regarded as 100%. (B) Effects of the indicated compounds on oxygen consumption by CT26 cells were determined as an indicator of mitochondrial oxidative metabolism. (C) Cells were treated with or without phenformin in the presence or absence of methyl succinate, which bypasses complex I of the electron transport chain. After 24 hours the number of live cells in the cultures was determined. MS: methyl succinate. C: control, P: phenformin 1 mM, O: oxamate 40 mM, PO: phenformin 1 mM+oxamate 40 Mm. *: P<0.05.
Figure 5.
Role of LDH inhibition in enhancing phenformin cytotoxicity.
(A) CT26 cells were treated with compounds, as indicated below each bar, for 24 days. Extracts were then prepared for determination of LDH enzyme activity. The Y axis is % of LDH activity when the activity of LDH of the control group is regarded as 100%. (B) Extracellular acidification rate in the presence of the indicated drugs was measured using a Seahorse XF24 instrument as described in Materials and Methods. (C) LDH expression in CT26 cells was repressed using siRNA transfection (groups labelled Ckd and Pkd). A scrambled siRNA transfection was used for groups C, P and PO. The cells were then treated for 24 hours with the indicated drugs and then the number of dead cells in each culture was determined. A western blot for LDHA is shown at the bottom, along with a blot for β-actin as a loading control. C: control, P: phenformin 1 mM, O: oxamate 40 mM, PO: phenformin 1 mM+oxamate 40 Mm, kd: LDH knock down by siRNA. *: P<0.05 compared with the other groups. †: P<0.05 compared with the group C and P.
Figure 6.
Effects of phenformin and oxamate on ROS, ATP levels, and DNA damage.
(A) CT26 cells were treated with compounds as indicated on the left. Eight hours after drug treatment MitoSOX staining was used to examine cellular levels of superoxide by confocal imaging. Mitotracker Green was used to label mitochondria. Magnification 100X. (B) CT26 cells were treated with the indicated compounds in the presence or absence of the ROS scavenger NAC (N-acetyl-cysteine). NAC was added to cultures 6 hours prior to adding phenformin or oxamate. Live cell number was then determined 24 hours after drug treatment. (C) Cells were treated with phenformin, oxamate, or both for 24 hours and then cellular ATP levels were measured. (D) Cells were treated as indicated for 24 hours and then the medium was collected and the cells fractionated into nuclear and mitochondria enriched fractions. In each compartment the level of oxidative damage to DNA was estimated using an ELISA to detect 8-OHdG. C: control, P: phenformin 1 mM, O: oxamate 40 mM, PO: phenformin 1 mM+oxamate 40 mM. *: P<0.05 compared with the other groups. †: P<0.05 compared with the group C and PO.
Figure 7.
Cell death pathways induced by phenformin and oxamate.
(A) CT26 cells were treated as indicated at the bottom of each lane. Experiments were performed after either 1 day (left) or 2 days (right) of treatment. Western blot analysis of cPARP in total cell extracts was used as an estimate of apoptotic cell death. Western blot analysis of nuclear AIF was used as an estimate of PARP-dependent cell death. β-actin and SP1 were used for protein loading controls. (B) AIF (red) was detected by immunofluorescence in cells that had been treated with the compounds indicated on the left and for the time indicated at the top. DAPI was use to stain nuclei (blue). (C) Cells were treated with phenformin or phenformin plus oxamate in the presence or absence of either a pan-caspase inhibitor or a PARP inhibitor. The percentage of dead cells was determined 24 hours after treatment in the P group and 12 hours after treatment in the PO group. C: control, P: phenformin 1 mM, PO: phenformin 1 mM+oxamate 40 mM. *: p<0.05 compared with the control group. †: P<0.05 compared with PO+PARP inhibitor.
Figure 8.
Effects of phenformin and oxamate on tumors in vivo.
(A) CT26 tumors were developed in syngeneic host mice. Three days after cell injection the mice were treated with oxamate, phenformin, or both daily for 21 days. Average tumor size for each group on day 21 of treatment is shown. Group PO tumors were significantly smaller compared to the other groups (P<0.05). There was no significant difference in tumor sizes between groups C, O, and P. (B, C) Tumor samples were processed to examine TUNEL positive cells as a measure of apoptosis. Cells which showed strong TUNEL positive were counted in three sections (304 µm×304 µm) in each mouse at 20X by confocal microscopy. The PO group showed significantly higher apoptosis than group C (apoptotic cells: 42.8±23.5 vs. 18.9±11.1) (P = 0.001). (D, E) Tumor bearing mice were subjected to PET/CT scanning to determine the effect of phenformin plus oxamate on glucose uptake. Group C showed significantly higher glucose uptake compared to the PO group (SUVavg: 2.0±0.6 vs. 1.6±0.3) (P = 0.033).
Figure 9.
Model of phenformin and oxamate activity in tumor cells.
We propose that the two drugs act synergistically by simultaneous inhibition of complex I and LDH. Phenformin increases ROS production by inhibiting mitochondria complex I. Inhibition of LDH by oxamate results in decreased ATP levels and elevated ROS production in the presence of phenformin because of increased flow of electrons through complex I.