Advertisement
Metabolism

Pharmacologic Ascorbate Primes Pancreatic Cancer Cells for Death by Rewiring Cellular Energetics and Inducing DNA Damage

Visarut Buranasudja, Claire M. Doskey, Adrienne R. Gibson, Brett A. Wagner, Juan Du, David J. Gordon, Stacia L. Koppenhafer, Joseph J. Cullen and Garry R. Buettner
DOI: 10.1158/1541-7786.MCR-19-0381 Published October 2019

Abstract

The clinical potential of pharmacologic ascorbate (P-AscH−; intravenous delivery achieving mmol/L concentrations in blood) as an adjuvant in cancer therapy is being reevaluated. At mmol/L concentrations, P-AscH− is thought to exhibit anticancer activity via generation of a flux of H2O2 in tumors, which leads to oxidative distress. Here, we use cell culture models of pancreatic cancer to examine the effects of P-AscH− on DNA damage, and downstream consequences, including changes in bioenergetics. We have found that the high flux of H2O2 produced by P-AscH− induces DNA damage. In response to this DNA damage, we observed that PARP1 is hyperactivated. Using our unique absolute quantitation, we found that P-AscH− mediated the overactivation of PARP1, which results in consumption of NAD+, and subsequently depletion of ATP leading to mitotic cell death. We have also found that Chk1 plays a major role in the maintenance of genomic integrity following treatment with P-AscH−. Hyperactivation of PARP1 and DNA repair are ATP-consuming processes. Using a Seahorse XF96 analyzer, we demonstrated that the severe decrease in ATP after challenging with P-AscH− is because of increased demand, not changes in the rate of production. Genetic deletion and pharmacologic inhibition of PARP1 preserved both NAD+ and ATP; however, the toxicity of P-AscH− remained. These data indicate that disruption of bioenergetics is a secondary factor in the toxicity of P-AscH−; damage to DNA appears to be the primary factor.

Implications: Efforts to leverage P-AscH− in cancer therapy should first focus on DNA damage.

Introduction

High dose ascorbate given intravenously, that is, pharmacologic ascorbate (P-AscH−), is currently being investigated as an adjuvant to certain cancer therapies. Ascorbate, functioning as vitamin C, is an essential reducing agent required by a broad range of biological processes. At physiologic levels, ∼40–80 μmol/L in human plasma, AscH− also functions as an endogenous antioxidant. Thermodynamically and kinetically, it is able to readily donate an electron to neutralize detrimental, oxidizing free radicals, resulting in formation of the ascorbate free radical (1). The ability of ascorbate to donate one electron is also key to serving as a co-antioxidant with vitamin E to inhibit lipid peroxidation (1–3). In addition, ascorbate acts as an excellent reducing agent, required by a range of biochemical processes. For example, AscH− is involved in various enzymatic reactions that are catalyzed by members of Fe2+-2-oxoglutarate–dependent dioxygenase family (4). Hence, as a vitamin, AscH− is essential as a donor antioxidant as well as a cofactor for a variety of enzymes in cells and tissues.

When used at high concentrations (plasma ascorbate ∼10–20 mmol/L), achievable only by intravenous infusion, ascorbate can exert prooxidant properties via formation of high fluxes of H2O2 (5–9). Several preclinical studies on a broad range of preclinical models of human cancer have demonstrated that the generation of H2O2 is a key mechanism for the anticancer activity of P-AscH− (5–7, 9–19). In the presence of catalytic metals (e.g., Fe3+/Fe2+), P-AscH− readily oxidizes producing high fluxes of H2O2 in the extracellular space of tumors. This H2O2 can readily enter cells via peroxiporins in the plasma membrane and then attack vital intracellular molecules, including DNA (10, 20). Hence, P-AscH− functions as a prodrug to deliver a high flux of H2O2 to tumor cells. These unique properties of P-AscH− are being extensively investigated in many clinical trials as an adjuvant to the standard of care in the treatment of cancer.

DNA can be a major oxidative target of H2O2. Hydrogen peroxide is believed to react with Fe2+ associated with DNA through the Fenton reaction to produce the site-specific, very reactive hydroxyl radical, leading to oxidative damage to DNA (21). In neuroblastoma, pancreatic, and ovarian cancer cells, exposure to P-AscH− resulted in DNA double-strand breaks via generation of extracellular H2O2 (13, 14, 22). We have demonstrated that the frequency of lesions on DNA is increased upon treatment of P-AscH−; this DNA damage was prevented by extracellular catalase, consistent with involvement of H2O2 in the genotoxicity of P-AscH− (10). Collectively, DNA damage could be a major factor for the anticancer activity of P-AscH−.

Disruption of bioenergetics could be another potential mechanism for cytotoxicity of P-AscH− in the treatment of cancer. Exposure to P-AscH− causes loss of cellular ATP in several types of cancer cells in vitro (9, 10, 12–15, 23, 24). However, the underlying mechanism leading to this consequence is not yet understood. The response to DNA damage requires increased consumption of ATP (25). PARP1 is a cellular mediator that marks DNA damage, but in so doing can influence cellular bioenergetics. In response to DNA damage, PARP1 functions as a cellular sensor that directs cells to specific fates (DNA repair vs. cell death), based on types and extent of genotoxic stress. With mild to moderate levels of genotoxic stress, activation of PARP1 facilitates the repair of DNA damage promoting cell survival. In contrast, severe DNA damage can hyperactivate PARP1, leading to extended poly (ADP-ribose) synthesis that results in extensive consumption of NAD+. Subsequently, cellular ATP is heavily utilized in an effort to replenish the pool of NAD+, resulting in severe loss of ATP and induction of cell death (26, 27). PARP1 is intimately tied to DNA damage, NAD+ metabolism, cellular energetics, and subsequently to cell survival or cell death. Taken together, the depletion of energy following treatment with P-AscH− could be due to an increase in the demand for energy in response to DNA damage, for example, activation of PARP1.

In this study, we investigated the cytotoxic effects of P-AscH− on pancreatic cancer cells, with a principal focus on DNA damage and changes in cellular bioenergetics. We demonstrate that the H2O2 generated by the oxidation of P-AscH− causes DNA damage, which leads to activation of PARP1, consumption of NAD+, and subsequent depletion of intracellular ATP. We also show that severe depletion of intracellular ATP by P-AscH− is exclusively due to increased demand, not changes in rate of production. PARP1 appears to be a major contributor to this increased demand following treatment with P-AscH−. Our results suggest that damage to DNA is the predominant factor for cytotoxicity of P-AscH−, while disruption of bioenergetics is secondary.

Materials and Methods

Cell culture

MIA PaCa-2 and PANC-1 human pancreatic adenocarcinoma cells were purchased from ATCC. MIA PaCa-2 and PANC-1 cells were cultured in DMEM with high glucose, supplemented with 10% FBS and penicillin (100 U/mL)/streptomycin (100 μg/mL). The patient-derived pancreatic cancer cells, 339 cells, were obtained from the Medical College of Wisconsin Surgical Oncology Tissue Bank (28, 29). 339 cells were cultured in DMEM/F12 Medium (Thermo Fisher Scientific), supplemented with 6% FBS, 0.1 μg/mL recombinant human epidermal growth factor (EGF), 0.05 mg/mL bovine pituitary extract, 2 μg/mL hydrocortisone, 0.56 μg/mL recombinant human insulin, and 1× GlutaMAX. The human nontumorigenic, immortalized pancreatic ductal epithelial cells, H6c7 (30), were cultured in keratinocyte serum-free media, supplemented with EGF (5 ng/mL), bovine pituitary extract (50 μg/mL), and penicillin (100 U/mL)/streptomycin (100 μg/mL). All cells were cultured at 37°C in a humidified atmosphere of 95% air/5% CO2.

Generation of PARP1 CRISPR/Cas9-knockout cells

To knockout PARP1, CRISPR/Cas9 plasmid DNA, encoding PARP1-specific noncoding guide RNA, Cas9 nuclease, and GFP were used to transfect cells. With the GFP reporter system, the transfected cells were selected with FACS techniques.

Cells were seeded into 6-well plates at 150,000 cells in 3.0 mL media and were cultured at 37°C, 5% CO2 for 24 hours. Standard medium was replaced with 3.0 mL of antibiotic-free DMEM prior to transfection. After media replacement, cells were then transfected with PARP1 CRISPR/Cas9-knockout plasmid (Santa Cruz Biotechnology; catalog no. sc-400046) for 24 hours. After 24 hours of transfection, cells were grown in standard DMEM for an additional 48 hours. To create a single-monoclonal cell line with PARP1 knockout, the GFP-positive cells were selected by using BD FACSAria Fusion Flow Cytometer (Becton Dickinson); single cells were sorted into individual wells of a 96-well plate. The cell lines with biallelic or monoallelic PARP1 knockouts were determined with Western blot analysis. Using similar procedures, control cells were generated by transfecting parental cell lines with control CRISPR/Cas 9 plasmid (Santa Cruz Biotechnology; catalog no. sc-418922).

Reagents and exposure of xenobiotics in mol/cell

l-Ascorbic acid was obtained from Macron Fine Chemicals. Stock solutions of l-ascorbic acid (1.00 mol/L, pH 7.0) were prepared and stored as described previously (31). Bovine catalase (Millipore Sigma) was prepared in water to produce a stock solution of 100 U/μL. Olaparib (Cayman Chemical) was dissolved in DMSO to produce a 30 mmol/L stock solution. Prexasertib (MedChem Express) was dissolved in DMSO to produce 10–100 μmol/L stock solutions.

Specifying dose as moles of a substance per cell (mol/cell) has been demonstrated to improve reproducibility and translatability of cell culture experiments (10, 17, 32, 33). For all in vitro experiments, when possible, doses of xenobiotic are expressed as mol/cell. Cells were seeded into appropriate culture containers at indicated cell densities; a set of dishes for various treatments while reserving one dish for determining cell number. Prior to exposure to xenobiotic, the number of cells in each set of dishes/wells was determined using a hemocytometer. This cell number and volume of medium present during treatment were used to determine the initial dose in units of mol/cell (10, 32). The formation of H2O2 due to oxidation of ascorbate is dependent on the pH of growth media. Hence, an exchange of media was done prior to treatment with ascorbate to minimize differences in pH and thereby differences in the flux of H2O2 between experiments. For olaparib and prexasertib, control cells were incubated with an equivalent amount of DMSO (vehicle control).

Clonogenic survival assay

Cells were seeded into multiple 60-mm culture dishes at 100,000 cells per dish in 4.0 mL of medium. Cells were cultured in their respective media for 48 hours at 37°C, 5% CO2. Following treatment, exposure media were removed; cells were trypsinized, counted using a Moxi Z Mini Automated Cell Counter (ORFLO Technologies), and seeded in triplicate into 6-well plates at 300 cells per well in 4.0 mL of their respective medium. For H6c7 cells, irradiated MIA PaCa-2 cells (30 Gy; 2 × 105 cells/well) were used as feeder layer to support clonogenicity. Plates were then incubated for 7–14 days at 37°C, 5% CO2. Surviving colonies were fixed with 70% ethanol for 5 minutes and stained with Coomassie Blue for 30 minutes. Colonies were counted as a cluster of at least 50 cells. Plating efficiency and surviving fraction were calculated (34): plating efficiency (PE) = [(average of number of colonies counted)/(number of cells plated)] × 100; clonogenic survival fraction = [(PE of treated sample)/(PE of untreated control)]. The effective dose of P-AscH− that caused 50% clonogenic cell death (ED50) for each cell line was calculated from dose–response curves using GraphPad Prism 7.0d.

Measurement of intracellular ATP

Cells were seeded into multiple culture dishes (60 mm) at 250,000 cells per dish in 4.0 mL of medium. Cells were cultured in their respective media for 48 hours at 37°C, 5% CO2 before exposure to experimental conditions. After treatment, cells were trypsinized, and resuspended with PBS. Intracellular ATP was then determined with CellTiter-Glo Luminescent Cell Viability Assay (Promega; catalog no. G7571) as described previously (10, 32). The effective dose of P-AscH− that resulted in a 50% reduction of intracellular ATP for each cell line was calculated from dose–response curves using GraphPad Prism 7.0d.

Measurement of intracellular NAD+ and NADH

Cell culture conditions and treatments were similar to those used for the measurement of ATP. After treatment, levels of NAD+ and NADH in samples were measured with the Luminescence-based-NAD/NADH-Glo Assay (Promega; catalog no. G9071) according to the manufacturer's protocol; 12,500 cells were used per measurement for either NAD+ or NADH. The concentrations of NAD+ and NADH of samples were calculated from the corresponding standard curve and then further transformed to an average intracellular concentration by using the cell number, as determined using a hemocytometer, and cell volume, as measured with a Moxi Z Mini Automated Cell Counter. The effective dose of P-AscH− that resulted in a 50% depletion of NAD+ for each cell line was calculated from dose–response curves using GraphPad Prism 7.0d.

Genomic isolation and measurement of DNA damage with quantitative PCR–based approach

Cells were seeded into multiple culture dishes (100 mm) at 500,000 cells per dish in 10.0 mL of medium. Cells were cultured in their respective media for 48 hours at 37°C, 5% CO2. After treatment, genomic DNA was isolated using Blood & Cell Culture DNA Mini Kit (Qiagen; catalog no. 13323). Isolation of genomic DNA by this technique has been demonstrated to be suitable for quantitative PCR (qPCR)-based measurement of both nuclear DNA (nDNA) and mitochondrial DNA (mtDNA) damage without a separate step for mtDNA purification (35). The qPCR method to quantitate DNA damage is based on the principle that various types of lesions on DNA can impede progression of DNA polymerase. If equal amounts of DNA from different biological samples are amplified under identical PCR conditions, DNA with a greater degree of damage will amplify to a lesser extent when compared with DNA with less damage (35). Hence, the amount of PCR amplification is inversely proportional to lesion frequency within DNA samples.

qPCR was performed in a 2720 Thermal Cycle (Applied Biosystems) with LA PCR Kit, Version 2.1 (Clontech Laboratories; catalog no. RR013B). The conditions for PCR were identical to our previous study (10). The sequences of primer nucleotide (Integrated DNA Technologies) are listed in Supplementary Table S3. The PCR amplicons were quantified by fluorescence measurement with Quant-iT PicoGreen dsDNA Assay Kit (Invitrogen; catalog no. P7589) according to the manufacturer's instructions. DNA lesion frequencies were calculated as described previously (35–37), by the formula λ = −ln (AD/ACt), where λ = lesion frequency per fragment, AD = amplification of treatment, and ACt = amplification of control.

Measurement of cellular metabolic flux

Cells were seeded into Seahorse 96-well XF Cell Culture Microplate V3-PET (Agilent) at 10,000 cells per well in 200 μL media. Cells were cultured in their respective media for 48 hours at 37°C, 5% CO2. After a 48-hour incubation period, cells were exposed to ascorbate (50 pmol/cell) for 1 hour. Following treatment, cells were washed gently with media; prewarmed XF Base Medium Minimal DMEM (Agilent), containing 25 mmol/L glucose, 2 mmol/L Glutamax (Thermo Fisher Scientific), and 1 mmol/L sodium pyruvate, with the pH of media being approximately 7.4 and buffer capacity determined prior to each assay; oxygen consumption rate (OCR) and rate of extracellular acidification (ECAR) were then measured using the Seahorse XF96 analyzer with the standard mitochondrial stress test (38). At the end of experiments, cell density (cells/L) from representative wells was determined with a hemocytometer. Using average of cell density, OCR and ECAR were then normalized per cell for control and treatment. The basal OCR was calculated from the parameters obtained from the mitochondrial stress test; OCRbasal = OCRATP + OCRproton leak. The basal ECAR was calculated from an average of first three readings of ECAR.

Immunofluorescence

Cells were seeded onto sterile round coverslips (15 mm) in 12-well plates at a density of 100,000 cells per well, in duplicate. Cells were cultured in their respective media for 48 hours at 37°C, 5% CO2. After treatment, cells were washed twice with culture media, fixed on coverslips with 4% paraformaldehyde in PBS (30 minutes), and incubated with 0.2% Triton X-100 in PBS (15 minutes). Following permeabilization, coverslips were washed with PBS (3×, 5 minutes), blocked with blocking solution (5% normal goat serum + 5% BSA in PBS) for 30 minutes, incubated with anti-PAR (1:100 in PBS; Santa Cruz Biotechnology; catalog no. sc-56198) at 4°C for 18 hours, and washed with PBS (3×, 5 minutes). Coverslips were subsequently incubated in the dark with Alexa Fluor 488–conjugated secondary antibody (1:200 in PBS; Thermo Fisher Scientific; catalog no. A-11017) for 2 hours. Fixed and stained cells were then washed with PBS (3×, 5 minutes) and incubated with TO-PRO-3 (1:1000 in PBS; 15 minutes; Thermo Fisher Scientific; catalog no. T3605) for nuclear staining. After final washing with PBS, coverslips were mounted face-down onto microscope slides using Vectashield antifade mounting medium (Vector Laboratories). Images were acquired with laser scanning microscope Zeiss LSM 510 with 63 × objective lens (Carl Zeiss AG). All microscopic parameters were kept constant across samples. At least three different areas were imaged per sample.

Western blot analysis

Equal amounts of protein lysates (20.0–50.0 μg) were separated by 4%–20% precast polyacrylamide gel (Bio-Rad) and then electrotransferred to Immobilon PVDF Membrane (Millipore Sigma). Membranes were blocked in 5% nonfat milk in PBST for 1 hour, and then incubated with primary antibody overnight at 4°C; anti-catalase (1:1,000; Cell Signaling Technology; catalog no. 12980), anti-PAR (1:500; EMD Millipore; catalog no. AM80), anti-PARP1 (1:1,000; Cell Signaling Technology; catalog no. 9542S), anti-Chk1 (1:1,000; Cell Signaling Technology; catalog no. 2360), anti phosphor-Chk1 (Ser345) (1:1,000; Cell Signaling Technology; catalog no. 2348), and anti-phospho-histone H2AX (Ser139) (1:2,000; Cell Signaling Technology; catalog no. 9718). Anti-actin (1:10,000; Millipore Sigma; catalog no. A2103), and anti-alpha tubulin (1:50,000; ProteinTech; catalog no. 66031-1) were used as loading control. Horseradish peroxidase–conjugated goat anti-rabbit (1:10,000; Millipore Sigma; catalog no. AP307P) or goat anti-mouse (1:10,000; Millipore Sigma; catalog no. AP308P) was used as a secondary antibody. The signal was developed using SuperSignal West Pico Chemiluminescent Substrate (Thermo Fisher Scientific).

Murine xenograft models

Thirty-day-old athymic nude mice were obtained from Envigo. The nude mice protocol was reviewed and approved by the Animal Care and Use Committee of The University of Iowa. Animals were allowed to acclimate in the unit for 1 week before any manipulations were performed. MIA PaCa-2 human pancreatic tumor cells [(2–3) × 106] were delivered subcutaneously into each flank region. Tumors were allowed to grow until they reached between 3 mm and 4 mm in greatest dimension before experimental treatment began. Once tumors were established, mice were treated intraperitoneally with ascorbate (4 g/kg) or saline (NaCl at 1 mol/L) twice daily for 5.5 days. On day 5.5, mice were euthanized by CO2 asphyxiation. Tumors were then harvested for experimental analyses.

Statistical analysis

One-way ANOVA followed by Tukey post hoc was used to examine statistical differences between means for multiple comparisons. All means were calculated from at least three independent experiments, with error bars representing SEM. Western blot analyses were performed with at least two biological replicates. Data are expressed as mean ± SEM. Tests of statistical significance were performed using GraphPad Prism 7.0d.

Results

Pancreatic cancer cells are more sensitive to P-AscH− than nontumorigenic pancreatic ductal epithelial cells

Cancer cells appear to be more susceptible than their normal cell counterparts to the toxic species generated by P-AscH− (7, 9–11, 15, 17, 39). To investigate the selective cytotoxicity of P-AscH− on pancreatic cancer cells, we performed clonogenic survival assays using pancreatic ductal adenocarcinoma cells (MIA PaCa-2 and PANC-1), as well as patient-derived pancreatic cancer cells (339), and compared with nontumorigenic pancreatic ductal epithelial cells (H6c7). We observed differential sensitivities to P-AscH− across cell lines, as seen in the ED50 for clonogenic survival (Supplementary Fig. S1; Supplementary Table S1). For pancreatic cancer cells used in this study, MIA PaCa-2 cells appear to be the most sensitive, while PANC-1 and 339 cells are relatively resistance to treatment with P-AscH−. In comparison with pancreatic cancer cells, we found that H6c7, that is, normal pancreatic cells, are more resistant to P-AscH−. These results support the differential toxicity of P-AscH− in the treatment of pancreatic cancer.

P-AscH− induces DNA damage via formation of H2O2

Production of high-fluxes of H2O2 from the oxidation of P-AscH− has been proposed to be the principal mechanism for anticancer activities of P-AscH− (5–7, 9, 10, 40). Because there are numerous targets for H2O2 in biological systems, it has been challenging to identify a general molecular mechanism on how the H2O2 generated by P-AscH− elicits its cytotoxic effects against cancer cells. Results from different cancer cells as model systems have suggested that DNA could be a vulnerable target for the H2O2 generated by P-AscH− (9, 10, 13, 14, 23). Here, we observed substantial increases in markers for DNA damage (γH2AX; phosphorylation of histone H2AX) and DNA replication stress (P-Chk1; phosphorylation of Chk1 at serine 345) following exposure of MIA PaCa-2 and PANC-1 cells to P-AscH−. P-AscH− triggered phosphorylation of histone H2AX as well as Chk1 on both MIA PaCa-2 and PANC-1 cells (Fig. 1A), indicating that exposure to P-AscH− results in DNA damage. Cotreatment with extracellular catalase prevented formation of γH2AX and P-Chk1 in both cell lines (Fig. 1A). These results suggest that extracellular H2O2 is the primary contributor to the genotoxicity of P-AscH−.

Figure 1.
Figure 1.

Through formation of H2O2, P-AscH− activates Chk1 in response to DNA damage. A, Western blot analyses show the marker of DNA damage (phosphorylation of histone H2AX, i.e., γH2AX) and DNA replication stress (phosphorylation at S345 of Chk1; P-Chk1) at 60 minutes following treatment with P-AscH−. Prexasertib was used to determine the contribution of Chk1 in DNA damage response after exposure to P-AscH−, while extracellular catalase (200 U/mL) was used to investigate the involvement of H2O2. MIA PaCa-2 or PANC-1 cells were pretreated with prexasertib for 18 hours then incubated with P-AscH− + prexasertib for 1 hour. The amounts of prexasertib were 50 amol/cell (MIA PaCa-2) and 150 amol/cell (PANC-1), while the amounts of ascorbate were 7 pmol/cell (MIA PaCa-2) and 15 pmol/cell (PANC-1). Results are representative of three independent experiments. B and C, Following treatment of P-AscH− + prexasertib, cell viabilities of MIA PaCa-2 and PANC-1 were evaluated by clonogenic survival assay. The protocol for treatment is similar to A (n = 3, mean ± SEM; *, P < 0.05, ****, P < 0.0001 vs. untreated control; †, P < 0.05; †††, P < 0.001; ††††, P < 0.0001 vs. prexasertib + P-AscH−).

Chk1 is required for cell survival following treatment of P-AscH−

Chk1 is an essential component of checkpoint pathways of DNA damage. In response to replication stress and/or DNA damage, Chk1 is activated by autophosphorylation at serine 296 and by ATR via phosphorylation on serines 317 and 345. Following activation, phosphorylated Chk1 in turn activates numerous downstream effectors to further propagate signals that induce cell-cycle arrest and facilitate DNA repair (41). As presented above, the accumulation of P-Chk1 at S345 was observed in both MIA PaCa-2 and PANC-1 cells after treatment with P-AscH−, suggesting an involvement of Chk1 in response to DNA damage caused by P-AscH−. In addition, cotreatment with extracellular catalase inhibited these phosphorylations of Chk1 (Fig. 1A). These findings suggest that P-AscH− activates Chk1 through formation of extracellular H2O2.

The activity of Chk1 is essential to ensure the fidelity of DNA repair checkpoints in response to genotoxic stress and DNA damage, such as produced by H2O2 (42). We demonstrated that inhibition of Chk1 by prexasertib significantly enhanced cytotoxicity of P-AscH− in both cells (Fig. 1B and C). These data suggest that activation of Chk1 is crucial for cell survival following treatment with P-AscH−. Because of the role of Chk1 in DNA repair, we postulated that the increased cytotoxicity of prexasertib + P-AscH− is due to an accumulation of DNA damage. Consistent with our hypothesis, γH2AX and P-Chk1 were present in cells treated with P-AscH− + prexasertib (Fig. 1A). It is important to note that phosphorylation of Chk1 at serine 345 can also observed in cells treated only with prexasertib (Fig. 1A); prexasertib blocks the autophosphorylation of S296, but not the phosphorylation of S317 or S345 of Chk1. These data are parallel to previous observations demonstrating that inhibition of Chk1 by prexasertib can lead to DNA replication stress, which subsequently results in phosphorylation of Chk1 itself at serine 296 (43). Experiments designed to evaluate possible synergy between P-AscH− and prexasertib found that the combination index was consistently <1, indicating synergy (Supplementary Fig. S5). Collectively, our preclinical findings suggest that Chk1 plays a critical role in facilitating survival of cancer cells following treatment with P-AscH−.

P-AscH− depletes intracellular ATP and NAD+ via generation of extracellular H2O2

The response to DNA damage is closely associated to the changes in bioenergetics. We found that the loss of ATP and NAD+ occurred rapidly after 1-hour treatment with P-AscH− (Fig. 2A–D). In MIA PaCa-2, extracellular catalase preserved levels of intracellular ATP as well as NAD+ following treatment with P-AscH− (Fig. 2E and F). In addition, the effects of bolus H2O2 on bioenergetics were parallel to treatment with P-AscH−, that is, depletion in ATP and NAD+ (Supplementary Fig. S2). These results clearly indicate that H2O2 is a primary agent for the decrease of cellular ATP and NAD+ by P-AscH−. The effects of P-AscH− on NADH are parallel (Supplementary Fig. S3).

Figure 2.
Figure 2.

P-AscH− depletes cellular ATP and NAD+ via formation of H2O2. A–D, MIA PaCa-2, PanC-1, 339, or H6c7 cells were incubated with varying amounts of P-AscH− (0–40 pmol/cell) for 1 hour; levels of intracellular ATP and NAD+ were measured immediately after (n = 3; mean ± SEM). E and F, To demonstrate the role of H2O2, MIA PaCa-2 cells were treated with P-AscH− (14 pmol/cell) ± bovine catalase (200 U/mL) for 1 hour, then levels of intracellular ATP and NAD+ were determined (n = 3; mean ± SEM). G and H, The amount of ascorbate required (ED50) to reduce the level of cellular ATP or NAD+ by 50% was determined for MIA PaCa-2, PANC-1, 339, and H6c7 cells on the basis of the dose–response curves shown in A–D. The ED50's for ascorbate are plotted versus the rate constants for removal of extracellular H2O2 (kcell).

The differential sensitivity to depletion of ATP and NAD+ from exposure to P-AscH− across cell lines correlates with their capacity to remove H2O2.

The four cell lines used in this study, MIA PaCa-2, PANC-1, 339, and H6c7 cells, have different susceptibility to loss of ATP and NAD+ as seen by ED50 (dose of P-AscH− that leads to 50% depletion) for ATP and NAD+ (Fig. 2A–D). These cell lines have different abilities to remove extracellular H2O2 as quantitatively determined by kcell. The kcell and ED50 of P-AscH− for depletion of ATP and NAD+ are provided in Supplementary Table S2. Because H2O2 is a primary agent for depletion of intracellular ATP and NAD+ induced by P-AscH− (Fig. 2E and F), the difference in sensitivity to changes in ATP and NAD+ of these cell lines could be explained by their differential capacities to remove extracellular H2O2. We observed a strong, positive correlation between kcell and the sensitivity to depletion of ATP and NAD+ from P-AscH− across different cell lines (R2, ED50 for ATP vs. kcell = 0.96 and ED50 for NAD+ vs. kcell = 0.79); cells with higher ability to eliminate H2O2 (e.g., PANC-1 and 399 cells) are more resistance to changes in bioenergetics induced by P-AscH−, as compared with cells with lower kcell (e.g., MIA PaCa-2; Fig. 2G–H). These data strongly support the crucial role of H2O2-detoxifying systems in the resulting toxicity observed from exposure to P-AscH−.

The repair of DNA damage induced by P-AscH− concurrently occurs with the restoration of cellular bioenergetics

Exposure to P-AscH− can lead to DNA damage and disruption of bioenergetics as demonstrated in Figs. 1 and 2, respectively. We performed a series of time-course analyses with MIA PaCa-2 cells to further investigate the correlation between these two consequences. With these time-course experiments, MIA PaCa-2 cells were exposed to P-AscH− (14 pmol/cell) for 1 hour, and then: (i) frequency of lesions on DNA; (ii) level of NAD+; and (iii) intracellular ATP were determined over time. These time-course studies on DNA repair using a qPCR-based approach demonstrated that the lesions on nDNA and mtDNA were present immediately after treatment. P-AscH− generated greater numbers of lesions on mtDNA, compared with nDNA. The number of lesions on nDNA was returned to the level of the control, that is, no exposure to P-AscH−, in about 3 hours (Fig. 3A); in this assay the ordinate is the change in frequency of DNA lesions compared with untreated control; the control has a value of “0”. The negative values seen at about 12 hours post oxidative challenge indicate fewer lesions in the nDNA than in the control, a potential hormetic effect. However, even after 24 hours, the number of lesions on mtDNA did not return to that of the control (Fig. 3A). These results suggest that the machineries of DNA repair in mitochondria are less efficient than in the nucleus. In conjunction with the damage and repair of lesions on DNA, the size of the pools of NAD+ and ATP were reduced to approximately 20% of the control immediately after a 1-hour exposure to P-AscH−. The depletion of these bioenergetic species was not permanent; both NAD+ and ATP were restored to near basal levels within 12–24 hours (Fig. 3B–D). Restoration of these species to the levels of the control took considerably longer than the repair of the lesions induced by P-AscH− on DNA. Taken together, these results suggest that DNA repair is complete before the restoration of bioenergetics.

Figure 3.
Figure 3.

Repair of DNA occurs prior to restoration of cellular energetic state. A, nDNA lesions introduced by P-AscH− are fewer and repaired more efficiently than lesions in mtDNA. The ordinate is the change in frequency compared with untreated control,“0”, that is, lesion frequency per fragment = −ln (AD/ACt), where AD = amplification of treatment, ACt = amplification of control. B–D, Intracellular ATP as well as NAD+ and NADH are depleted by P-AscH−. Each is restored but more slowly than the repair of nDNA. MIA PaCa-2 cells were treated with P-AscH− (14 pmol/cell) for 1 hour and then were allowed to recover in fresh media. Responses were determined at indicated times (0–24 hours following treatment). The dashed line in B–D represents the untreated control. Lesions of DNA, n = 4; ATP, NAD+, and NADH, n = 3; mean ± SEM.

The reduction in energy stores induced by P-AscH− is not due to changes in rate of production

The severe depletion of energy stores demonstrated in Figs. 2 and 3B could be due to two possibilities: (i) disruptions in rate of energy production; and/or (ii) increases in the demand for energy. To investigate the effects of ascorbate on rate of production, MIA PaCa-2 or PANC-1 cells were exposed to lethal amounts of P-AscH− (50 pmol/cell; 20 mmol/L) for 1 hour; the OCR (surrogate for respiration), and the ECAR (marker for glycolysis), were then determined using Seahorse XF96 analyzer. Even at this very high concentration of ascorbate, we observed no change (or only a slight change) in the rate of respiration (OCR; Fig. 4A and B) as well as the rate of glycolysis (ECAR; Fig. 4C and D) following treatment with P-AscH− in both MIA PaCa-2 and PANC-1 cells. These studies of cellular bioenergetics strongly suggest that the rapid reduction in energy stores upon exposure to ascorbate is not caused by changes in the rate of production. Hence, the compromised availability of energy stores could be predominantly because of increased demand, for example, the need for energy to repair DNA.

Figure 4.
Figure 4.

Treatment with P-AscH− does not affect the rate of energy production in MIA PaCa-2 or PANC-1 cells. OCR (A and B), and ECAR (C and D) were determined immediately after treatment. MIA PaCa-2 or PANC-1 cells were treated with a lethal amount of ascorbate (50 pmol/cell; 20 mmol/L) for 1 hour. OCR and ECAR were determined using a Seahorse XF96 analyzer (n = 4, each biological replicate consists of 10 independent wells, mean ± SEM; *, P < 0.05 vs. untreated control).

P-AscH− rapidly activates PARP1 via formation of hydrogen peroxide

The high flux of H2O2 from the oxidation of P-AscH− caused DNA damage in cancer cells (Figs. 1 and 3A). Hence, the decrease in intracellular steady-state concentration of ATP could be because of an increased demand for energy in response to DNA damage. PARP1 is an NAD+-consuming enzyme that can link DNA damage to changes in cellular energy through its activity of posttranslational modification, poly(ADP-ribosyl)ation. Activation of PARP1 in response to DNA damage leads to poly(ADP-ribosyl)ation, extensive consumption of NAD+, depletion of ATP, and the possible induction of cell death (27, 44, 45).

Activation of PARP1 was investigated by detection of its end product, polymers of ADP-ribose (PAR). Representative confocal microscopic images from MIA PaCa-2 and PANC-1 cells demonstrated that the generation of PAR polymers occurred rapidly in the nucleus and cytoplasm, within 30 minutes of treatment with P-AscH− (Fig. 5A). Extracellular catalase completely abrogated the formation of PAR induced by P-AscH− in MIA PaCa-2 cells (Fig. 5A). These results imply that activation of PARP1 by P-AscH− is dependent on formation of extracellular H2O2.

Figure 5.
Figure 5.

P-AscH− activates PARP1, resulting in depletion of NAD+ and ATP. A, MIA PaCa-2 or PANC-1 cells were incubated with 10 or 30 pmol/cell of ascorbate, respectively, for 30 minutes; then activation of PARP1 was determined by generation of PAR (green) using immunofluorescence microscopy. Nuclei were counter stained with TO-PRO-3 (blue). Extracellular catalase (cat; 200 U/mL) was used to investigate the contribution of H2O2 in activation of PARP1 (n = 3; magnification, 63 ×; scale bar, 20 μm). B, MIA PaCa-2 cells were injected into the flank of mice to form tumors. Mice were treated with normal saline (control) or ascorbate (4 g/kg) twice daily for 5.5 days. After final treatment on day 5.5, mice were sacrificed and tumors harvested. The formation of PAR in tumor tissues was determined by Western blot analysis. Each lane represents tissue from individual tumors from different mice. C, MIA PaCa-2 cells were pretreated with olaparib (50 fmol/cell) for 18 hours then exposed to P-AscH− (7 pmol/cell) + olaparib (50 fmol/cell) for 30 minutes. Formation of PAR were determined immediately after treatment with Western blot analysis. Results are representative of three independent experiments. D and E, MIA PaCa-2 cells were incubated with olaparib (50 fmol/cell) for 18 hours following by treatment of P-AscH− (7 pmol/cell) + olaparib (50 fmol/cell) for 1 hour. Intracellular ATP and NAD+ were determined immediately after treatment (n = 3; mean ± SEM; ***, P < 0.001; ****, P < 0.0001 vs. untreated control). F, To verify the involvement of PARP1 in the depletion of ATP and NAD+, MIA PaCa-2 PARP1-knockout (KO) cells were generated with CRISPR/Cas9 technology. The Western blot analysis demonstrated a successful knockout of PARP1. G and H, CRISPR control (CRISPR Ctrl) or PARP1-knockout MIA PaCa-2 cells were exposed to P-AscH− (10 pmol/cell) for 1 hour. Intracellular ATP and NAD+ were observed immediately after treatment (n = 3; mean ± SEM; *, P < 0.05; ****, P < 0.0001 vs. untreated control; NS = no significant).

P-AscH− activates PARP1 in vivo

To determine whether activation of PARP1 by P-AscH− observed in cell culture experiments also occurs in vivo, a mouse xenograft model of human pancreatic cancer was used. MIA PaCa-2 xenografts were established in nude mice; then mice were treated with ascorbate (4 g/kg) twice daily for 5.5 days; in this model this dose slows tumor growth (9). The cellular response to P-AscH− in this murine model of pancreatic cancer paralleled the in vitro results; increased formation of poly (ADP-ribose) was observed in xenograft tumor tissues harvested from P-AscH−-treated mice (Fig. 5B). The results from these animal studies highlight the contribution of PARP1 in cytotoxicity of P-AscH−.

Activation of PARP1 by P-AscH− leads to increased utilization of NAD+ and depletion of ATP availability

To validate the contribution of PARP1 in the cytotoxic mechanism of P-AscH−, MIA PaCa-2 cells were treated with an inhibitor of PARP, olaparib, at clinically achievable doses (3 μmol/L; 50 fmol/cell) for 18 hours followed by exposure to P-AscH− (7 pmol/cell) in combination with olaparib (50 fmol/cell). Pretreatment with olaparib significantly suppressed P-AscH− induced formation of PAR (Fig. 5C). Moreover, inhibition of PARP1 by olaparib preserved the levels of intracellular NAD+ (Fig. 5D) as well as intracellular ATP (Fig. 5E) upon exposure to P-AscH−. These data suggest that activation of PARP1 by P-AscH− produces a dramatic loss of intracellular NAD+ and as well as depletion of ATP stores.

Specificity is a significant consideration in the use of olaparib as a tool to investigate the involvement of PARP1 in the cytotoxicity of P-AscH−. Olaparib inhibits not only PARP1, but also other members of the PARP family, including PARP2, and tankyrase1 (26). Similar to PARP1, both PARP2 and takyrase1 use NAD+ as a substrate to generate PAR in response to DNA damage (46, 47). Hence, these two enzymes may participate in changes in cellular bioenergetics following treatment with P-AscH−. To rule out the contribution of other PARP family members, we generated PARP1-knockout MIA PaCa-2 cells using CRISPR/Cas9 technology (Fig. 5F). The genetic deletion of PARP1 preserved both the intracellular NAD+ (Fig. 5G) and ATP (Fig. 5H) upon exposure to P-AscH−. Taken together, the results from pharmacologic inhibition and genetic deletion of PARP1 clearly indicate that PARP1 is a major mediator of the loss of NAD+ and ATP following treatment of P-AscH−.

Pharmacologic inhibition of PARP1 by olaparib did not prevent the cytotoxicity of P-AscH−

Although inhibition of PARP1 by olaparib prevented loss of intracellular NAD+ (Fig. 5D) as well as ATP (Fig. 5E), olaparib did not prevent the cytotoxicity of P-AscH− in MIA PaCa-2 cells (Fig. 6A). Increased cytotoxicity of P-AscH− was observed when MIA PaCa-2 cells were pretreated with a higher concentration of olaparib (25 μmol/L; 500 fmol/cell; Supplementary Fig. S4). The enhanced cytotoxicity of P-AscH− by olaparib was also found in PANC-1 cells (Fig. 6B). The results from Western blot analysis demonstrated that P-Chk1 is greatly enhanced in both MIA PaCa-2 and PANC-1 cells that were treated with olaparib + P-AscH− (Fig. 6C). These results suggest that the inhibition of PARP1 by olaparib may sensitize tumor cells to P-AscH− by increased accumulation of DNA damage, which ultimately leads to cell death, even when levels of NAD+ and ATP are preserved.

Figure 6.
Figure 6.

Disruption to bioenergetics is not the principal factor for anticancer activity of P-AscH−. A and B, Olaparib did not inhibit the cytotoxicity of P-AscH− in MIA PaCa-2 or PANC-1 cells. Cells were pretreated with olaparib for 18 hours then treated with P-AscH− + olaparib for 1 hour. Olaparib was at 50 fmol/cell (MIA PaCa-2) and 150 fmol/cell (PANC-1), while the amounts of ascorbate were 7 pmol/cell (MIA PaCa-2) and 15 pmol/cell (PANC-1). Cell viability was determined by clonogenic survival assay (n = 3; mean ± SEM; ****, P < 0.0001 vs. untreated control; ††††, P < 0.0001 vs. olaparib + P-AscH−). C, The activation of Chk1 and phosphorylation of histone H2AX were detected in both MIA PaCa-2 and PANC-1 cells at 1-hour following treatment. The protocol for treatment was as in A and B. D, CRISPR control (CRISPR Ctrl) or PARP1-knockout (PARP1 KO) MIA PaCa-2 cells were treated with different concentrations of P-AscH− (0–10 pmol/cell) for 1 hour. Cell viability was evaluated using a clonogenic survival assay (n = 3; mean ± SEM).

The cytotoxicity of P-AscH− remains in PARP1-knockout MIA PaCa-2 cells, even as levels of NAD+ and ATP are preserved

If disruption of bioenergetics is essential for the anticancer activity of P-AscH−, preservation of NAD+ (Fig. 5G) and ATP (Fig. 5H) in PARP1-knockout cells would result in prevention of ascorbate-induced cell death. However, results from clonogenic survival assays demonstrated that the cytotoxicity of P-AscH− was enhanced in PARP1-knockout cells (Fig. 6D), even though the levels of NAD+ and ATP were preserved. These data are parallel to the results from pharmacologic inhibition of PARP1 by olaparib (Fig. 6A and B). Taken together, data generated from genetic deletion and pharmacologic inhibition of PARP1 clearly indicate that disruption of bioenergetics is not the primary contributor for the anticancer activity of P-AscH−.

Discussion

Our current study provides novel and useful insights into the biochemical mechanism underlying the anticancer activity of P-AscH−. The generation of high fluxes of H2O2 by the oxidation of P-AscH− is believed to be central to the mechanism of its antitumor activities (5–7, 9–17). Hydrogen peroxide can attack numerous macromolecules of cancer cells. Preclinical data presented here demonstrate that DNA appears to be a vulnerable target for P-AscH− in the treatment of cancer. Our data demonstrate that exposure to P-AscH− results in DNA damage through formation of extracellular H2O2 (Fig. 1A). These data support the previous observations that H2O2 is a primary factor for genotoxicity of P-AscH− (10, 13, 14). mtDNA is more susceptible to oxidative damage from treatment with P-AscH−, compared with nDNA (Fig. 3A; ref. 10). The results from our time-course study revealed that the differential sensitivity to P-AscH− of mtDNA versus nDNA may be due to differences in the efficiency of repair systems; the repair systems of mtDNA are less efficient in removing lesions compared with nDNA (Fig. 3A). We further investigated the response to DNA damage following treatment of pancreatic cancer cells with P-AscH−. We demonstrated that P-AscH− via formation of H2O2 activates Chk1 kinase (Fig. 1A). The activation of Chk1 appears to be an essential pathway for cell survival following treatment with P-AscH− (Fig. 1B and C).

In addition to DNA damage, treatment with P-AscH− can result in depletion of intracellular NAD+ and ATP (Fig. 2A–D). This depletion can be detected immediately after a 1-hour treatment with P-AscH−. We observed that the sensitivity to depletion of NAD+ and ATP is correlated directly with the capacity of cells to remove extracellular H2O2 (Fig. 2G and H). These data strongly support an essential role of the H2O2-detoxifying system in the depletion of the intracellular energy pool observed upon exposure to P-AscH−. Moreover, the decrease of NAD+ and ATP stores was prevented by extracellular catalase, consistent with the contribution of H2O2 in disruption of cellular bioenergetics upon treatment of P-AscH− (Fig. 2E and F).

We also examined whether the depletion of ATP upon treatment with P-AscH− was due to changes in its rate of production. Previous studies on cellular bioenergetics using Seahorse Bioscience XF instrumentation reported that upon treatment with P-AscH− the ECAR of cancer cells was reduced and the OCR was initially increased, then in a longer time frame OCR was reduced (13, 23). However, these studies did not consider the contribution of the oxidation of P-AscH− to these measurements. AscH− in medium readily oxidizes, leading to an increased OCR. Moreover, the oxidation of AscH− can lead to the oxidation of other components in the medium, leading to an apparent change in ECAR. Hence, these potential artifacts could have led to the misinterpretation on cellular bioenergetics data. To avoid these potential artifacts, we completely removed P-AscH− prior to initiation of XF-measurements. Our data on cellular bioenergetics demonstrate that immediately after treatment with P-AscH− there are no significant changes in either OCR or ECAR (Fig. 4). If P-AscH− did not disrupt the rate of energy production from either respiration or glycolysis, then the depletion of ATP by P-AscH− must be due to an increase in demand, for example, the demand for NAD+ and ATP to repair DNA.

Data from our time-course study suggest that the changes in cellular bioenergetics following treatment with P-AscH− are associated with increased demand because of DNA damage (Fig. 3). PARP1 is a central regulator that can sense the extent of DNA damage and initiate repair process. However, activation of this system comes at the cost of energy consumption through formation of PAR; formation of PAR directly consumes NAD+, which leads to consumption of ATP. We showed that the H2O2 generated from oxidation of P-AscH− activates PARP1, leading to formation of PAR (Fig. 5A and C), increased consumption of NAD+ (Fig. 2), and subsequently reduction of cellular ATP (Fig. 2) in pancreatic cancer cells.

Consistent with our cell culture experiments, the activation of PARP1 was observed in tumor xenografts from mice treated with infusions of P-AscH− (Fig. 5B). These observations provide the first evidence of in vivo activation of PARP1 following treatment of P-AscH−. Interestingly, pharmacologic inhibition and genetic deletion of PARP1 did not prevent cytotoxicity of P-AscH− on pancreatic cancer cells (Fig. 6A, B, and D), despite that olaparib preserved NAD+ and ATP availability upon exposure to P-AscH− (Fig. 5D, E, G, and H). Using absolute quantitation, Fig. 5E, we found that the level of intracellular ATP in MIA PaCa-2 dropped from 8 mmol/L to 3.6 mmol/L after exposure to P-AscH− (7 pmol/cell; ≅ ED50 of P-AscH− for clonogenic cell death for MIA PaCa-2); the KM values for ATP for most kinase enzymes are less than 1 mmol/L, usually much less (48). Hence, the level of ATP following treatment of P-AscH−, although lower, was still considerably greater than KM for ATP for a wide range of kinase enzymes. Hence, the function of cellular kinases and other enzymes that have ATP as a substrate should not be impaired by treatment with P-AscH−. Taken together, these data strongly suggest that the depletion of energy is not the predominant contributor to the anticancer activity of P-AscH−.

The major finding of this work is that DNA damage is a key aspect of the anticancer activity of P-AscH−. P-AscH− has been demonstrated to enhance the cytotoxicity of ionizing radiation (11, 49), gemcitabine (50), olaparib (Fig. 6B; Supplementary Fig. S4; ref. 13), and prexasertib (Fig. 1B and C; Supplementary Fig. S5); each targets DNA or DNA repair. Hence, using P-AscH− as an adjuvant with anticancer therapies that target DNA or DNA repair as part of their mechanisms could be an attractive strategy for treatment of cancer in the future.

In summary, the information from these preclinical studies support translational efforts by providing novel and useful insight into the biochemical mechanisms underlying the cytotoxicity of P-AscH− in cancer treatment (Fig. 7). This study provides definitive evidence that DNA damage, not disruption of cellular bioenergetics, is a predominant factor for the anticancer activity of P-AscH−. Thus, the effort to leverage P-AscH− as an adjuvant should first focus on DNA damage and repair processes; targeting of the production of ATP may be better as a secondary effort.

Figure 7.
Figure 7.

Proposed model for anticancer activity of ascorbate. P-AscH− functions as a prodrug for the delivery of a significant flux of H2O2 to tumors. In the presence of redox active metals, P-AscH− generates high fluxes of H2O2 in the extracellular space of tumors. This H2O2 is readily transported across the plasma membrane via peroxiporins; it attacks DNA through the site-specific Fenton reaction. This DNA damage activates PARP1, leading to depletion of NAD+ and ATP. To maintain genomic integrity, the DNA damage from this high flux of H2O2 leads to activation of Chk1. Our data show that the DNA damage caused by the H2O2 produced by P-AscH− appears to be the principal biochemical change responsible for the anticancer activity of P-AscH−. The disruption of bioenergetics is secondary. The combination of P-AscH− and agents that target DNA repair (e.g., olaparib and prexasertib) could be a promising approach for the treatment of pancreatic cancer.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Disclaimer

The content is solely the responsibility of the authors and does not represent views of the NIH.

Authors' Contributions

Conception and design: V. Buranasudja, C.M. Doskey, B.A. Wagner, G.R. Buettner

Development of methodology: V. Buranasudja, C.M. Doskey, B.A. Wagner, G.R. Buettner

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): V. Buranasudja, C.M. Doskey, A.R. Gibson, B.A. Wagner, J. Du, D.J. Gordon, S.L. Koppenhafer, J.J. Cullen, G.R. Buettner

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): V. Buranasudja, B.A. Wagner, D.J. Gordon, J.J. Cullen, G.R. Buettner

Writing, review, and/or revision of the manuscript: V. Buranasudja, C.M. Doskey, A.R. Gibson, B.A. Wagner, J. Du, D.J. Gordon, J.J. Cullen, G.R. Buettner

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): B.A. Wagner, J.J. Cullen, G.R. Buettner

Study supervision: G.R. Buettner

Acknowledgments

This work was supported by NIH grants R01 CA169046 (to G.R. Buettner), R01 GM073929 (to G.R. Buettner), R01 CA184051 (to J.J. Cullen), P01 CA217797 (to J.J. Cullen), R37 CA217910 (to D.J. Gordon), T32 CA078586 (to A.R. Gibson), and P42 ES013661 (to V. Buranasudja/G.R. Buettner). Core facilities were supported in part by the Carver College of Medicine and the Holden Comprehensive Cancer Center, NIH P30 CA086862. The ESR Facility at The University of Iowa provided invaluable support.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Footnotes

  • Note: Supplementary data for this article are available at Molecular Cancer Research Online (http://mcr.aacrjournals.org/).

  • Mol Cancer Res 2019;17:2102–14

  • Received April 10, 2019.
  • Revision received June 21, 2019.
  • Accepted July 17, 2019.
  • Published first July 23, 2019.

References

  1. 1.↵
    1. Buettner GR
    . The pecking order of free radicals and antioxidants: lipid peroxidation, alpha-tocopherol, and ascorbate. Arch Biochem Biophys 1993;300:535–43.
    OpenUrlCrossRefPubMed
  2. 2.↵
    1. Bruno RS,
    2. Ramakrishnan R,
    3. Montine TJ,
    4. Bray TM,
    5. Traber MG
    . {alpha}-Tocopherol disappearance is faster in cigarette smokers and is inversely related to their ascorbic acid status. Am J Clin Nutr 2005;81:95–103.
    OpenUrlAbstract/FREE Full Text
  3. 3.↵
    1. Buettner GR,
    2. Jurkiewicz BA
    . Catalytic metals, ascorbate and free radicals: combinations to avoid. Radiat Res 1996;145:532–41.
    OpenUrlCrossRefPubMed
  4. 4.↵
    1. Du J,
    2. Cullen JJ,
    3. Buettner GR
    . Ascorbic acid: chemistry, biology and the treatment of cancer. Biochim Biophys Acta 2012;1826:443–57.
    OpenUrlCrossRefPubMed
  5. 5.↵
    1. Chen Q,
    2. Espey MG,
    3. Krishna MC,
    4. Mitchell JB,
    5. Corpe CP,
    6. Buettner GR,
    7. et al.
    Pharmacologic ascorbic acid concentrations selectively kill cancer cells: action as a pro-drug to deliver hydrogen peroxide to tissues. Proc Natl Acad Sci U S A 2005;102:13604–9.
    OpenUrlAbstract/FREE Full Text
  6. 6.↵
    1. Chen Q,
    2. Espey MG,
    3. Sun AY,
    4. Lee JH,
    5. Krishna MC,
    6. Shacter E,
    7. et al.
    Ascorbate in pharmacologic concentrations selectively generates ascorbate radical and hydrogen peroxide in extracellular fluid in vivo. Proc Natl Acad Sci U S A 2007;104:8749–54.
    OpenUrlAbstract/FREE Full Text
  7. 7.↵
    1. Chen Q,
    2. Espey MG,
    3. Sun AY,
    4. Pooput C,
    5. Kirk KL,
    6. Krishna MC,
    7. et al.
    Pharmacologic doses of ascorbate act as a prooxidant and decrease growth of aggressive tumor xenografts in mice. Proc Natl Acad Sci U S A 2008;105:11105–9.
    OpenUrlAbstract/FREE Full Text
  8. 8.↵
    1. Padayatty SJ,
    2. Sun H,
    3. Wang Y,
    4. Riordan HD,
    5. Hewitt SM,
    6. Katz A,
    7. et al.
    Vitamin C pharmacokinetics: implications for oral and intravenous use. Ann Intern Med 2004;140:533–7.
    OpenUrlCrossRefPubMed
  9. 9.↵
    1. Du J,
    2. Martin SM,
    3. Levine M,
    4. Wagner BA,
    5. Buettner GR,
    6. Wang SH,
    7. et al.
    Mechanisms of ascorbate-induced cytotoxicity in pancreatic cancer. Clin Cancer Res 2010;16:509–20.
    OpenUrlAbstract/FREE Full Text
  10. 10.↵
    1. Doskey CM,
    2. Buranasudja V,
    3. Wagner BA,
    4. Wilkes JG,
    5. Du J,
    6. Cullen JJ,
    7. et al.
    Tumor cells have decreased ability to metabolize H2O2: Implications for pharmacological ascorbate in cancer therapy. Redox Biol 2016;10:274–84.
    OpenUrl
  11. 11.↵
    1. Du J,
    2. Cieslak JA III.,
    3. Welsh JL,
    4. Sibenaller ZA,
    5. Allen BG,
    6. Wagner BA,
    7. et al.
    Pharmacological ascorbate radiosensitizes pancreatic cancer. Cancer Res 2015;75:3314–26.
    OpenUrlAbstract/FREE Full Text
  12. 12.↵
    1. Chen P,
    2. Yu J,
    3. Chalmers B,
    4. Drisko J,
    5. Yang J,
    6. Li B,
    7. et al.
    Pharmacological ascorbate induces cytotoxicity in prostate cancer cells through ATP depletion and induction of autophagy. Anticancer Drugs 2012;23:437–44.
    OpenUrlCrossRefPubMed
  13. 13.↵
    1. Ma E,
    2. Chen P,
    3. Wilkins HM,
    4. Wang T,
    5. Swerdlow RH,
    6. Chen Q
    . Pharmacologic ascorbate induces neuroblastoma cell death by hydrogen peroxide mediated DNA damage and reduction in cancer cell glycolysis. Free Radic Biol Med 2017;113:36–47.
    OpenUrl
  14. 14.↵
    1. Ma Y,
    2. Chapman J,
    3. Levine M,
    4. Polireddy K,
    5. Drisko J,
    6. Chen Q
    . High-dose parenteral ascorbate enhanced chemosensitivity of ovarian cancer and reduced toxicity of chemotherapy. Sci Transl Med 2014;6:222ra218.
    OpenUrlCrossRef
  15. 15.↵
    1. Polireddy K,
    2. Dong R,
    3. Reed G,
    4. Yu J,
    5. Chen P,
    6. Williamson S,
    7. et al.
    High dose parenteral ascorbate inhibited pancreatic cancer growth and metastasis: mechanisms and a Phase I/IIa study. Sci Rep 2017;7:17188.
    OpenUrl
  16. 16.↵
    1. Schoenfeld JD,
    2. Sibenaller ZA,
    3. Mapuskar KA,
    4. Bradley MD,
    5. Wagner BA,
    6. Buettner GR,
    7. et al.
    Redox active metals and H2O2 mediate the increased efficacy of pharmacological ascorbate in combination with gemcitabine or radiation in pre-clinical sarcoma models. Redox Biol 2018;14:417–22.
    OpenUrl
  17. 17.↵
    1. Schoenfeld JD,
    2. Sibenaller ZA,
    3. Mapuskar KA,
    4. Wagner BA,
    5. Cramer-Morales KL,
    6. Furqan M,
    7. et al.
    O2.- and H2O2-mediated disruption of Fe metabolism causes the differential susceptibility of NSCLC and GBM cancer cells to pharmacological ascorbate. Cancer Cell 2017;32:268.
    OpenUrl
  18. 18.↵
    1. Du J,
    2. Wagner BA,
    3. Buettner GR,
    4. Cullen JJ
    . Role of labile iron in the toxicity of pharmacological ascorbate. Free Radic Biol Med 2015;84:289–95.
    OpenUrl
  19. 19.↵
    1. Erudaitius D,
    2. Mantooth J,
    3. Huang A,
    4. Soliman J,
    5. Doskey CM,
    6. Buettner GR,
    7. et al.
    Calculated cell-specific intracellular hydrogen peroxide concentration: Relevance in cancer cell susceptibility during ascorbate therapy. Free Radic Biol Med 2018;120:356–67.
    OpenUrl
  20. 20.↵
    1. Erudaitius D,
    2. Huang A,
    3. Kazmi S,
    4. Buettner GR,
    5. Rodgers VG
    . Peroxiporin expression is an important factor for cancer cell susceptibility to therapeutic H2O2: Implications for pharmacological ascorbate therapy. PLoS One 2017;12:e0170442.
    OpenUrl
  21. 21.↵
    1. Henle ES,
    2. Linn S
    . Formation, prevention, and repair of DNA damage by iron/hydrogen peroxide. J Biol Chem 1997;272:19095–8.
    OpenUrlFREE Full Text
  22. 22.↵
    1. Cieslak JA,
    2. Strother RK,
    3. Rawal M,
    4. Du J,
    5. Doskey CM,
    6. Schroeder SR,
    7. et al.
    Manganoporphyrins and ascorbate enhance gemcitabine cytotoxicity in pancreatic cancer. Free Radic Biol Med 2015;83:227–37.
    OpenUrl
  23. 23.↵
    1. Yun J,
    2. Mullarky E,
    3. Lu C,
    4. Bosch KN,
    5. Kavalier A,
    6. Rivera K,
    7. et al.
    Vitamin C selectively kills KRAS and BRAF mutant colorectal cancer cells by targeting GAPDH. Science 2015;350:1391–6.
    OpenUrlAbstract/FREE Full Text
  24. 24.↵
    1. Castro ML,
    2. Carson GM,
    3. McConnell MJ,
    4. Herst PM
    . High dose ascorbate causes both genotoxic and metabolic stress in glioma cells. Antioxidants 2017;6:pii:E58.
    OpenUrl
  25. 25.↵
    1. Zong WX,
    2. Ditsworth D,
    3. Bauer DE,
    4. Wang ZQ,
    5. Thompson CB
    . Alkylating DNA damage stimulates a regulated form of necrotic cell death. Genes Dev 2004;18:1272–82.
    OpenUrlAbstract/FREE Full Text
  26. 26.↵
    1. Rouleau M,
    2. Patel A,
    3. Hendzel MJ,
    4. Kaufmann SH,
    5. Poirier GG
    . PARP inhibition: PARP1 and beyond. Nat Rev Cancer 2010;10:293–301.
    OpenUrlCrossRefPubMed
  27. 27.↵
    1. Luo X,
    2. Kraus WL
    . On PAR with PARP: cellular stress signaling through poly(ADP-ribose) and PARP-1. Genes Dev 2012;26:417–32.
    OpenUrlAbstract/FREE Full Text
  28. 28.↵
    1. Roy I,
    2. Zimmerman NP,
    3. Mackinnon AC,
    4. Tsai S,
    5. Evans DB,
    6. Dwinell MB
    . CXCL12 chemokine expression suppresses human pancreatic cancer growth and metastasis. PLoS One 2014;9:e90400.
    OpenUrlCrossRefPubMed
  29. 29.↵
    1. Kim MP,
    2. Evans DB,
    3. Wang H,
    4. Abbruzzese JL,
    5. Fleming JB,
    6. Gallick GE
    . Generation of orthotopic and heterotopic human pancreatic cancer xenografts in immunodeficient mice. Nat Protoc 2009;4:1670–80.
    OpenUrlCrossRefPubMed
  30. 30.↵
    1. Qian J,
    2. Niu J,
    3. Li M,
    4. Chiao PJ,
    5. Tsao MS
    . In vitro modeling of human pancreatic duct epithelial cell transformation defines gene expression changes induced by K-ras oncogenic activation in pancreatic carcinogenesis. Cancer Res 2005;65:5045–53.
    OpenUrlAbstract/FREE Full Text
  31. 31.↵
    1. Buettner GR
    . In the absence of catalytic metals ascorbate does not autoxidize at pH 7: ascorbate as a test for catalytic metals. J Biochem Biophys Methods 1988;16:27–40.
    OpenUrlCrossRefPubMed
  32. 32.↵
    1. Doskey CM,
    2. van't Erve TJ,
    3. Wagner BA,
    4. Buettner GR
    . Moles of a substance per cell is a highly informative dosing metric in cell culture. PLoS One 2015;10:e0132572.
    OpenUrl
  33. 33.↵
    1. Alexander MS,
    2. Wilkes JG,
    3. Schroeder SR,
    4. Buettner GR
    . Pharmacologic ascorbate reduces radiation-induced normal tissue toxicity and enhances tumor radiosensitization in pancreatic cancer. Cancer Res 2018;78:6838–51.
    OpenUrlAbstract/FREE Full Text
  34. 34.↵
    1. Munshi A,
    2. Hobbs M,
    3. Meyn RE
    . Clonogenic cell survival assay. Methods Mol Med 2005;110:21–8.
    OpenUrlCrossRefPubMed
  35. 35.↵
    1. Furda A,
    2. Santos JH,
    3. Meyer JN,
    4. Van Houten B
    . Quantitative PCR-based measurement of nuclear and mitochondrial DNA damage and repair in mammalian cells. Methods Mol Biol 2014;1105:419–37.
    OpenUrlCrossRefPubMed
  36. 36.↵
    1. Van Houten B,
    2. Cheng S,
    3. Chen Y
    . Measuring gene-specific nucleotide excision repair in human cells using quantitative amplification of long targets from nanogram quantities of DNA. Mutat Res 2000;460:81–94.
    OpenUrlPubMed
  37. 37.↵
    1. Salazar JJ,
    2. Van Houten B
    . Preferential mitochondrial DNA injury caused by glucose oxidase as a steady generator of hydrogen peroxide in human fibroblasts. Mutat Res 1997;385:139–49.
    OpenUrlCrossRefPubMed
  38. 38.↵
    1. Divakaruni AS,
    2. Paradyse A,
    3. Ferrick DA,
    4. Murphy AN,
    5. Jastroch M
    . Analysis and interpretation of microplate-based oxygen consumption and pH data. Methods Enzymol 2014;547:309–54.
    OpenUrlCrossRefPubMed
  39. 39.↵
    1. Ranzato E,
    2. Biffo S,
    3. Burlando B
    . Selective ascorbate toxicity in malignant mesothelioma: a redox Trojan mechanism. Am J Respir Cell Mol Biol 2011;44:108–17.
    OpenUrlCrossRefPubMed
  40. 40.↵
    1. Rawal M,
    2. Schroeder SR,
    3. Wagner BA,
    4. Cushing CM,
    5. Welsh JL,
    6. Button AM,
    7. et al.
    Manganoporphyrins increase ascorbate-induced cytotoxicity by enhancing H2O2 generation. Cancer Res 2013;73:5232–41.
    OpenUrlAbstract/FREE Full Text
  41. 41.↵
    1. Bartek J,
    2. Lukas J
    . Chk1 and Chk2 kinases in checkpoint control and cancer. Cancer Cell 2003;3:421–9.
    OpenUrlCrossRefPubMed
  42. 42.↵
    1. Willis J,
    2. Patel Y,
    3. Lentz BL,
    4. Yan S
    . APE2 is required for ATR-Chk1 checkpoint activation in response to oxidative stress. Proc Natl Acad Sci U S A 2013;110:10592–7.
    OpenUrlAbstract/FREE Full Text
  43. 43.↵
    1. Koppenhafer SL,
    2. Goss KL,
    3. Terry WW,
    4. Gordon DJ
    . mTORC1/2 and protein translation regulate levels of CHK1 and the sensitivity to CHK1 inhibitors in Ewing sarcoma cells. Mol Cancer Ther 2018;17:2676–88
    OpenUrlAbstract/FREE Full Text
  44. 44.↵
    1. Gibson BA,
    2. Kraus WL
    . New insights into the molecular and cellular functions of poly(ADP-ribose) and PARPs. Nat Rev Mol Cell Biol 2012;13:411–24.
    OpenUrlCrossRefPubMed
  45. 45.↵
    1. Jagtap P,
    2. Szabo C
    . Poly(ADP-ribose) polymerase and the therapeutic effects of its inhibitors. Nat Rev Drug Discov 2005;4:421–40.
    OpenUrlCrossRefPubMed
  46. 46.↵
    1. Lehtio L,
    2. Chi NW,
    3. Krauss S
    . Tankyrases as drug targets. FEBS J 2013;280:3576–93.
    OpenUrlCrossRefPubMed
  47. 47.↵
    1. Jungmichel S,
    2. Rosenthal F,
    3. Altmeyer M,
    4. Lukas J,
    5. Hottiger MO,
    6. Nielsen ML
    . Proteome-wide identification of poly(ADP-Ribosyl)ation targets in different genotoxic stress responses. Mol Cell 2013;52:272–85.
    OpenUrlCrossRefPubMed
  48. 48.↵
    1. Knight ZA,
    2. Shokat KM
    . Features of selective kinase inhibitors. Chem Biol 2005;12:621–37.
    OpenUrlCrossRefPubMed
  49. 49.↵
    1. Schoenfeld JD,
    2. Alexander MS,
    3. Waldron TJ,
    4. Sibenaller ZA,
    5. Spitz DR,
    6. Buettner GR,
    7. et al.
    Pharmacological ascorbate as a means of sensitizing cancer cells to radio-chemotherapy while protecting normal tissue. Semin Radiat Oncol 2019;29:25–32.
    OpenUrl
  50. 50.↵
    1. Espey MG,
    2. Chen P,
    3. Chalmers B,
    4. Drisko J,
    5. Sun AY,
    6. Levine M,
    7. et al.
    Pharmacologic ascorbate synergizes with gemcitabine in preclinical models of pancreatic cancer. Free Radic Biol Med 2011;50:1610–19.
    OpenUrlCrossRefPubMed
Back to top
View this article with LENS

Open full page PDF
Article Alerts
Email Article
Citation Tools
Pharmacologic Ascorbate Primes Pancreatic Cancer Cells for Death by Rewiring Cellular Energetics and Inducing DNA Damage
Visarut Buranasudja, Claire M. Doskey, Adrienne R. Gibson, Brett A. Wagner, Juan Du, David J. Gordon, Stacia L. Koppenhafer, Joseph J. Cullen and Garry R. Buettner
Mol Cancer Res October 1 2019 (17) (10) 2102-2114; DOI: 10.1158/1541-7786.MCR-19-0381
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Advertisement