Pharmacologic Ascorbate and DNMT Inhibitors Increase DUOX Expression and Peroxide-Mediated Toxicity in Pancreatic Cancer

Abstract

Recent studies have demonstrated an important role for vitamin C in the epigenetic regulation of cancer-related genes via DNA demethylation by the ten-eleven translocation (TET) methylcytosine dioxygenase enzymes. DNA methyltransferase (DNMT) reverses this, increasing DNA methylation and decreasing gene expression. Dual oxidase (DUOX) enzymes produce hydrogen peroxide (H₂O₂) in normal pancreatic tissue but are silenced in pancreatic cancer (PDAC). Treatment of PDAC with pharmacologic ascorbate (P-AscH^-, intravenous, high dose vitamin C) increases DUOX expression. We hypothesized that inhibiting DNMT may act synergistically with P-AscH^- to further increase DUOX expression and cytotoxicity of PDAC. PDAC cells demonstrated dose-dependent increases in DUOX mRNA and protein expression when treated with DNMT inhibitors. PDAC cells treated with P-AscH^- + DNMT inhibitors demonstrated increased DUOX expression, increased intracellular oxidation, and increased cytotoxicity in vitro and in vivo compared to either treatment alone. These findings suggest a potential therapeutic, epigenetic mechanism to treat PDAC.

Keywords: DNA methyltransferase (DNMT); ascorbic acid; epigenetics; pancreatic cancer; pharmacologic ascorbate; ten-eleven translocation (TET) methylcytosine dioxygenase.

Figure 1

DUOX hypermethylation in PDAC. (A) DUOX1 methylation vs. overall survival in PDAC. The NIH Cancer Genome Atlas (TCGA) Pancreatic Cancer Database (PAAD) (n = 223) was accessed via the University of California Santa Cruz (UCSC) Xena Functional Genomics Explorer (n = 148; overall survival at 5 years 30% vs. 5%; * p = 0.03; Gehan–Breslow–Wilcoxon test). (B) DUOX1 mRNA expression is increased in a dose-dependent manner in the MIA PaCa-2, PANC-1, and PDX-339 PDAC cell lines after exposure to AZC (1–2 µM) for 5 days (means ± SEM; n = 3; * p < 0.05). (C) DUOX2 mRNA expression is increased in a dose-dependent manner in MIA PaCa-2, PANC-1, and PDX-339 PDAC cell lines after exposure to AZC (1–2 µM) for 5 days (means ± SEM; n = 3; * p < 0.05). (D) DUOX1 and DUOX2 immunoreactive protein was increased in PDX-339 cells after AZC (1–2 µM). Representative blots are shown.

Figure 2

DNMT inhibitors with P-AscH⁻ increase DUOX expression in PDAC cell lines in a dose-dependent manner. Treatment with a DNMT inhibitor also produces sustained increases in DUOX expression. (A) DUOX1 mRNA expression is increased after exposure to AZC (2 µM) for 5 days ± P-AscH⁻ (20 pmol/cell) for 1 h in PDX-339 cells. The combination group demonstrated a significant increase in expression compared to either treatment group alone (means ± SEM; n = 3; * p < 0.05). (B) DUOX1 mRNA expression increased in a dose-dependent manner after exposure to AZD (0.5–1 µM) for 3 days in MIA PaCa-2 cells. The addition of P-AscH⁻ (10 pmol/cell) for 1 h produces a significant increase in expression compared to either treatment group alone (means ± SEM; n = 3; * p < 0.05). (C) DUOX2 mRNA expression increased in a similar manner with the same treatments. (D) DUOX1 immunoreactive protein increased in MIA-PaCa-2 cells with AZD (0.5–1 µM) for 3 days and P-AscH⁻ (10 pmol/cell) for 1 h. (E) DUOX2 immunoreactive protein increased with the same treatments. Representative blots are shown. (F) DUOX1 mRNA expression is increased for up to 72 h after exposure to AZD (1 µM). (G) DUOX2 mRNA expression is increased immediately following exposure to AZD (1 µM) (means ± SEM; n = 4; * p < 0.05 vs. control).

Figure 3

DNMT inhibitors and P-AscH⁻ generate dose-dependent, H₂O₂-dependent cytotoxicity. (A) Mean fluorescence intensity (MFI) is increased following exposure to AZC (2 µM) for 5 days and P-AscH⁻ (20 pmol/cell) for 1 h in PDX-339 cells. Pretreatment with bovine catalase (100 µg/mL) reverses this effect demonstrating that the oxidation of DCFH-DA is mediated by H₂O₂ (means ± SEM; n = 3; * p < 0.05). (B) MIA PaCa-2, PANC-1 and PDX-339 cells were treated with AZC (0.5–2 µM) for 5 days ± P-AscH⁻ (20 pmol/cell) for 1 h demonstrating decreases in clonogenic survival with the combination treatments (means ± SEM; n = 3; * p < 0.05). (C) MIA PaCa-2 and PANC-1 cells were treated with AZD ± P-AscH⁻ demonstrating decreases in clonogenic survival with the combination treatments (means ± SEM; n = 3; * p < 0.05). (D) Clonogenic survival in MIA PaCa-2 treated with AZD (0.1 µM), P-AscH⁻ (10 pmol/cell), and catalase (100 µg/mL). Catalase reverses the decrease in clonogenic survival supporting a H₂O₂ mechanism (means ± SEM; n = 3; * p < 0.01).

Figure 4

The hypoxia-induced increase in DNMT1 expression is decreased with P-AscH⁻. (A) DNMT1 immunoreactive protein was increased after exposure to 4% O₂ for 6 h and decreased to baseline following exposure to P-AscH⁻ (10 pmol/cell). Representative blots are shown. (B) Quantification of densitometric evaluation of Western blots (mean ± SEM, values normalized to control; n = 3; * p < 0.05). (C) DNMT1 immunofluorescence staining was performed on xenograft tumor samples. Samples were visualized using a Zeiss Confocal Microscope 40× oil objective. Results show decreased DNMT1 immunofluorescence in the P-AscH⁻ treatment group compared to control. Green staining, DNMT1; blue staining, nuclear topoisomerase-3. Representative images are shown. (D) Quantification demonstrating MFI normalized to nuclear content (means ± SEM; n = 8; * p < 0.05).

Figure 5

DNMT inhibitors combined with P-AscH⁻ decrease tumor volume and increase DUOX1 expression in vivo. Athymic nude mice with heterotopic MIA PaCa-2 xenografts were treated with intraperitoneal normal saline (4 g/kg, 1 M, daily), AZD (1 g/kg, three times weekly), P-AscH⁻ (4 g/kg, daily), or a combination of P-AscH⁻ and AZD. Mice were treated for 21 days and tumor volume was measured twice weekly. (A) Tumor growth was significantly inhibited in the combination AZD + P-AscH⁻ group compared to the control group and compared to either treatment group alone. Tumor volumes (mm³) were normalized to their starting volumes on treatment day 1 to avoid heterogeneity in the starting tumor volumes. Data represent average tumor volume over 18 d (means ± SEM; * p < 0.05). (B) DUOX1 immunofluorescence staining was performed on xenograft tumor samples. Samples were visualized using a Zeiss Confocal Microscope 40× oil objective. Results show increased DUOX1 immunofluorescence in the AZD treatment groups compared to control. Green staining, DUOX1; blue staining, nuclear topoisomerase-3. Representative images are shown. (C) Quantification demonstrating MFI normalized to nuclear content (means ± SEM; n = 7; * p < 0.05).