Mechanisms of ascorbate-induced cytotoxicity in pancreatic cancer

Abstract

Purpose: Pharmacologic concentrations of ascorbate may be effective in cancer therapeutics. We hypothesized that ascorbate concentrations achievable with i.v. dosing would be cytotoxic in pancreatic cancer for which the 5-year survival is <3%.

Experimental design: Pancreatic cancer cell lines were treated with ascorbate (0, 5, or 10 mmol/L) for 1 hour, then viability and clonogenic survival were determined. Pancreatic tumor cells were delivered s.c. into the flank region of nude mice and allowed to grow at which time they were randomized to receive either ascorbate (4 g/kg) or osmotically equivalent saline (1 mol/L) i.p. for 2 weeks.

Results: There was a time- and dose-dependent increase in measured H(2)O(2) production with increased concentrations of ascorbate. Ascorbate decreased viability in all pancreatic cancer cell lines but had no effect on an immortalized pancreatic ductal epithelial cell line. Ascorbate decreased clonogenic survival of the pancreatic cancer cell lines, which was reversed by treatment of cells with scavengers of H(2)O(2). Treatment with ascorbate induced a caspase-independent cell death that was associated with autophagy. In vivo, treatment with ascorbate inhibited tumor growth and prolonged survival.

Conclusions: These results show that pharmacologic doses of ascorbate, easily achievable in humans, may have potential for therapy in pancreatic cancer.

Figure 1

A. Effects of pharmacologic ascorbic acid concentrations on pancreatic cancer and pancreatic ductal epithelial cells. All cells were treated with ascorbate (0, 5, 10 mM) for one hour. Cell viability determined by MTT assay. MIA PaCa-2, AsPC-1, BxPC-3 are pancreatic cancer cell lines. Immortalized pancreatic ductal epithelial cell line, H6c7 and its derivatives, and H6c7er-Kras (H6c7 cells expressing K-ras oncogene), also received ascorbate (0, 5, 10 mM) for one hour. Ascorbate decreased cell viability in all pancreatic cancer cell lines and in the H6c7 cell lines that express K-ras. B. MIA PaCa-2 pancreatic cancer cells were treated with ascorbate (0–20 mM) for one hour and clonogenic survival determined. Ascorbate caused a dose-dependent decrease in clonogenic survival. Arrow indicates that no colonies were formed when 20 mM ascorbate was given for one hour. C. AsPC-1 pancreatic cancer cells were treated with ascorbate (0–20 mM) for one hour and clonogenic survival determined. Ascorbate again caused a dose-dependent decrease in clonogenic survival.

Figure 2

Overexpression of extracellular and cytosolic catalase reverses ascorbate induced cytotoxicity. A. Catalase is overexpressed in MIA PaCa-2 cells transfected with adenoviruses containing human catalase cDNA. Transfection with AdCAT (100 MOI) targeted catalase expression to the cytosol while AdmitCAT (100 MOI) directed catalase to the mitochondria via an 80-bp MnSOD leader sequence. Western blotting confirmed the overexpression of intracellular catalase in cells transfected with AdCAT and AdmitCAT. MnSOD was used as a loading control. B. Catalase pretreatment reverses ascorbate-induced decreases in clonogenic survival. Ascorbate (2 mM) for one hour decreased clonogenic survival in MIA PaCa-2 human pancreatic cancer cells. Catalase (100 μg/mL) and PEG-catalase (200 u/mL) alone had little effect on clonogenic survival. Pretreatment of cells with catalase or PEG-catalase followed by ascorbate (2 mM) reversed ascorbate-induced cytotoxicity. *P < 0.05 vs control, means ± SEM, n = 3. C. Cytosolic-directed catalase, but not mitochondrial-directed catalase blocked ascorbate-induced cytotoxicity. MIA PaCa-2 cells were treated with either AdEmpty (100 MOI), AdCAT (100 MOI) or AdmitCAT (100 MOI) and treated with ascorbate (2 mM) for one hour. Ascorbate (2 mM) decreases clonogenic survival in cells infected with the AdEmpty and AdmitCAT vectors, but has no effect on cells infected with AdCAT. *P < 0.05 vs. - abscorbate, means ± SEM, n = 3.

Figure 3

Ascorbate decreases intracellular ATP levels. A. Catalase pretreatment blocks the ascorbate-induced dose-dependent decrease in ATP. ATP levels were significantly lower in MIA PaCa-2 cells treated with ascorbate (2 and 5 mM) when compared to the same cells pretreated with catalase (100 μg/mL). A luciferase-based somatic ATP assay kit was utilized in the measurement of intracellular ATP. *P < 0.05 vs. + catalase, means ± SEM, n = 3. B. MIA PaCa-2 cells transfected with AdCAT or AdmitCAT restores ATP following treatment with ascorbate (5mM for one hour) relative to their controls. Ascorbate (5 mM) decreases ATP levels in control and AdEmpty treated pancreatic cancer cells. However, the decreased ATP levels are reversed in MIA PaCa-2 cells with infection of the adenoviral vectors containing human catalase cDNA (AdCAT and AdmitCAT). Although AdCAT had reversed ascorbate-induced cytotoxicity, AdmitCAT did not, suggesting that mitochondrial depletion of ATP may not play a role in ascorbate-induced cytotoxicity. *P < 0.05 vs. ascorbate 0 mM, means ± SEM, n = 3. C. Ascorbate (0–5 mM) demonstrated significant decreases in clonogenic survival in both MIA PaCa-2 rho⁺ and rho° cells without pyruvate. However, ascorbate-induced cytotoxicity was reversed when pyruvate is added to the media of rho° cells to compensate for the respiratory metabolism deficit induced by the generation of mitochondrial deficient cells.

Figure 4

Poly(ADP-ribose) polymerase-1 (PARP-1) activation is not associated with ascorbate-induced cell death. Effect of 3AB (10 mM) on ascorbate-induced PARP activation, intracellular ATP depletion, and cell death. MIA PaCa-2 cells were treated with 5 mM ascorbate for 30 min with or without 1 h pretreatment of 10 mM 3 AB (Tocris Bioscience, Ellisville, MO), an inhibitor of PARP. A. Cells were collected and subjected to Western blot analysis. Cell lysate (40 μg) was probed with polyclonal rabbit anti-PAR antibody (BD Biosciences, San Jose, CA) at dilution of 1:3000. Polymerize ADP-ribose (PAR) are identified at 116–200 kDa. Ascorbate (5 mM) induces PAR formation which is reversed with 3AB. B. ATP levels were measured as described. ATP decreases with ascorbate treatment, which is not reversed with 3AB pretreatment. There was a significant decrease in ATP levels with ascorbate (5 mM) treatment. 3AB alone had little effect on ATP while 3AB did not reverse the ascorbate-induced ATP depletion. Means ± SEM, n = 3, *p < 0.05 vs. controls. C. Clonogenic survival was measured after ascorbate (5 mM for 1 h) with and without 3AB (10 mM). The ascorbate-induced decrease in clonogenic survival is not reversed with 3AB. Means ± SEM, n = 3, *p < 0.05 vs. control.

Figure 5

Ascorbate induces autophagy in MIA PaCa-2 cells. A. MIA PaCa-2 cells were treated with ascorbate (5 mM), harvested at various time points and western analysis was performed. Fortyμg of protein per well was loaded onto a 4 – 20 % Tris-HCl gel. Polyclonal rabbit anti-LC3 primary antibody (1:1000 dilution) was followed by horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG (1: 50000 dilution). The increase in LC3-I and LC3-II proteins observed 6 h following ascorbate treatment is inhibited by pretreatment with catalase. B. MIA PaCa-2 cells transduced with LC3-GFP were treated with the indicated doses of ascorbate or H₂O₂ for 1 h and allowed to recover in fresh media for 24 h. Cells were harvested, stained with propidium iodide (1 μg/mL) to discriminate live cells and then run on a BD Calibur flow cytometer. The top row shows the gating strategy for the live cells (PI negative). The percentage of live cells is shown in each plot with decreasing viability shown for both treatments (Tx). The bottom row shows the GFP fluorescence of the live cells. The GFP negative cells (WT) were used to adjust the gate for the LC3-GFP positive cells. The percentage of LC3-GFP positive cells is indicated within each plot and the GFP mean florescence intensity (MFI) of this population is indicated above each plot. Both ascorbate and H₂O₂ increase the percentage of positive cells as well as the MFI. C. The relative increase in the percentage of LC3-GFP cells (No Tx = 1.0) is shown for three independent experiments (mean +/− standard deviation). A One-way ANOVA with Tukey’s post-hoc test shows that all 4 treatment groups were significantly different than the control (*p < 0.05). D. Ascorbate and H₂O₂ induced punctated distribution of LC3-GFP and autophagosome maturation detected by selective increase in GFP fluorescence in cells expressing the LC3-GFP fusion protein. Cells were fixed, stained, and digitized images were analyzed by confocal microscopy and representative cells were selected and photographed. The percentage of LC3-GFP cells as evaluated by digital image analysis was increased in pancreatic cancer cells treated with ascorbate (1 and 2 mM) and H₂O₂ (200 μM) compared to no treatment. Cells were fixed, stained, and the digitized images were analyzed E. Pretreatment of MIA PaCa-2 LC3-GFP cells with catalase 100 μg/mL reverses the increase in mean fluorescence intensity (MFI) induced by ascorbate. Means ± SEM, n = 3, *p < 0.05.

Figure 6

A. Ascorbate treatment decreased MIA PaCa-2 tumor growth in nude mice. Animals that received ascorbate (4 g/kg, i.p., b.i.d. for 14 days) had significantly slower tumor growth when compared to animals that received saline (1 M, i.p., b.i.d., for 14 days) (Means ± SEM, p < 0.001, n = 13 control animals and 15 ascorbate-treated animals). On day 21 there was greater than a 3-fold decrease in tumor growth in animals receiving ascorbate when compared to treatment with saline. B. Kaplan-Meier plots of estimated survival in nude mice. There was a significant difference in survival between saline and ascorbate treated groups of mice (p < 0.001).