Metalloporphyrin Synergizes with Ascorbic Acid to Inhibit Cancer Cell Growth Through Fenton Chemistry

Abstract

Ascorbic acid (AA) has been reported to inhibit tumor cell growth through the generation of extracellular hydrogen peroxide (H₂O₂). However, the clinical utility of AA has been limited by relatively low potency and in vivo efficacy. This study reports that the metalloporphyrin, Mn(III) tetrakis(N-methylpyridinium-2-yl)porphyrin⁵⁺ (MnTMPyP), has a potent synergistic cytotoxic effect when combined with AA in a variety of cancer cell lines. In the presence of MnTMPyP, the concentration of AA required to inhibit cancer cell growth was markedly reduced. In vitro (cell-free) experiments demonstrated that AA alone enhanced the Fenton reaction that produces cytotoxic hydroxyl radical (HO^•); however, this reaction was limited by the low rate by which AA generates H₂O₂ (Fenton reaction substrate) from O₂. MnTMPyP catalyzed H₂O₂ generation through the AA-facilitated Mn(II ⇔ III)TMPyP redox cycle and thereby markedly potentiated the Fenton reaction. Accordingly, MnTMPyP and AA resulted in increased cellular levels of H₂O₂ and HO^• in cancer cells, which mediate the synergistic cytotoxicity of this combined treatment. This effect was inhibited by cellular enzymes that metabolize H₂O₂, such as catalase and glutathione peroxidase, suggesting that selective killing of cancer cells deficient in such enzymes can be achieved in vivo.

Introduction

Many cancer cells have abnormalities in the relative level of expression of important intracellular antioxidant enzymes.¹ For instance, prostatic and pancreatic cancer cells have relatively low levels of superoxide dismutase (SOD) and catalase compared with normal tissue.^2
–4 The disturbance of redox balance in such cancers provides a novel opportunity for the development of redox-dependent therapeutic strategies.

Ascorbic acid (AA) is a redox-active agent reported to selectively inhibit the growth of multiple cancer (but not normal) cell lines.^5,6 The underlying mechanism of action of this effect may be the ability of AA to reduce redox-active transitional metals (such as iron and copper), hence facilitating the redox cycling of these metals, which generates various reactive oxygen species (ROS). It has been shown that hydrogen peroxide (H₂O₂) is generated by AA and is essential for AA-mediated tumor cell killing.⁵ AA may also induce the Fenton reaction [Fe(II) + H₂O₂ → Fe(III) +HO^• + OH⁻] subsequent to H₂O₂ generation through the Fe(III ⇔ II) redox cycle, which produces the highly reactive and hence toxic hydroxyl radical (HO^•). However, the use of iron chelators to block the reaction has revealed inconsistent effects on AA-mediated toxicity,^5,7 suggesting that the Fenton reaction induced by AA is limited. Given that large doses of AA (4g/kg i.p. twice daily) had limited antitumor effects in mouse models,⁶ increasing the Fenton reaction may improve the in vivo efficacy of AA as an antitumor agent. The present study hypothesizes that the rate of AA-induced Fenton reaction is limited by the rate of H₂O₂ generation, because of the absence (or low efficiency) of a metallocatalyst(s), and/or the varying levels of cellular H₂O₂-metabolizing enzymes (catalase and glutathione peroxidase [GPX]). The experiments described here were designed to test this hypothesis that metalloporphyrin Mn(III) tetrakis(N-methylpyridinium-2-yl)porphyrin⁵⁺ (MnTMPyP) would enhance AA-induced H₂O₂ production and, subsequently, potentiate AA-mediated cytotoxicity via the Fenton reaction.

Known for its ability to eliminate O₂ ^•−, MnTMPyP has traditionally been considered to be an SOD mimic. However, compared with SOD, its in vitro (cell-free) activity is 2 orders of magnitude lower except when in the presence of reducing agents (such as AA, thioredoxin reductase/NADPH, or GSH) that reduce MnTMPyP from its stable state Mn(III)TMPyP to Mn(II)TMPyP.^8,9 Mn(II)TMPyP is then readily oxidized to Mn(III)TMPyP by reacting with O₂ ^•−, to form the Mn(III ⇔ II)TMPyP redox cycle. MnTMPyP is therefore considered to be an oxidoreductase instead of a dismutase. The product of this oxidoreductase reaction has not been documented, but the present study hypothesizes that O₂ ^•− is reduced to H₂O₂. Other enzymatic properties of MnTMPyP have been reported, including the reduction of O₂ to O₂ ^•− under hyperoxic conditions.⁹ This effect under normoxia is unknown, but may be potentiated in the presence of AA by enhancing the Mn(III ⇔ II)TMPyP redox cycle. As a result, MnTMPyP may also enhance H₂O₂ generation by increasing the substrate (O₂ ^•−) concentration. Because some cancer cells have decreased levels of SOD,^2
–4 the anticancer effect of MnTMPyP has been studied both in vitro and in vivo. ^10,11 However, in these studies, MnTMPyP alone had no significant cytotoxic effect. The present study hypothesizes that this was due to the absence of metal redox cyclers, such as AA, to generate O₂ ^•−, reduce Mn(III)TMPyP, and enhance Fenton reaction. In the present study, this hypothesis was tested and the potent synergistic antitumor cytotoxicity of MnTMPyP and AA was demonstrated.

Materials and Methods

Cell culture, reagents, and equipments

Hormone-dependent and -independent (subline) LAPC-4 cells were obtained from Dr. Charles Sawyers at UCLA. All other cells were obtained from ATCC (Manassas, VA). Cells were cultured and treated in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS) unless specified otherwise. MnTMPyP pentachloride was purchased from Alexis Biochemical (San Diego, CA). All other reagents were obtained from Sigma-Aldrich (St. Louis, MO) unless otherwise specified. Equipment used in the experiments described below included fluorescent (SpectraMax, Gemini XS) and spectroscopic (SpectraMax 190) microplate reader (Molecular Device, Ramsey, MN), fluorescent microscope (ECLIPSE TS100; Nikon, Melville, NY), image capture system (MicroPublisher 3.3RTV; QIMAGING, Surrey, BC, Canada), ultrapure water system (NANOpure; Barnstead, Waltham, MA), hypoxia chamber (INVIVO₂400; Ruskinn/Biotrace, St. Paul, MN), and flow cytometer (FACScalibur; Becton Dickinson, Franklin Lakes, NJ).

Cytotoxicity assay

Cells were plated in a 96-well microplate at a density of 1000 cells/well. After overnight incubation, the cells were treated as indicated in the figure legends and kept in fresh medium for 3 days before staining with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT).

Cell survival assay

Cell survival was determined using the clonogenic assay as previously described.¹² Briefly, subconfluent cells (250–1500 per 60-mm dish) were plated and incubated overnight before treatment. Colonies with over 50 cells were counted at 1–3 weeks after treatment.

Mitochondrial membrane potential determination

Mitochondrial membrane potential (MMP) was measured using rhodamine 123 staining, as previously described.¹¹ Briefly, 300,000 cells/well were plated the day prior to treatment in a six-well plate. After the treatment, the cells were stained with 5 μg/mL rhodamine 123 dissolved in medium for 1 hour and then photographed, or analyzed using flow cytometry.

Cell death measurement

Cells (300,000 per well) were plated the day prior to treatment in a six-well plate. After the treatment, the cells (including the floating and attached cells) were collected. Cell death was measured using the trypan blue (0.04%; Invitrogen, Carlsbad, CA) exclusion assay and expressed as the percentage of positively stained/total number of cells.

H₂O₂ measurement

To measure H₂O₂ generation in vitro (cell-free), a specific and sensitive marker, Amplex-Red (10-acetyl-3,7-dihydroxyphenoxazine; Biotium, Hayward, CA), was used, which is converted to a red fluorescent product (resorufin) upon reaction with H₂O₂.¹³ The Amplex-Red working solution contains Amplex-Red (100 μM) and horseradish peroxidase (HRP II, 1 U/mL) dissolved in PBS or DMEM. The rate of H₂O₂ production was monitored as ΔRFU (relative fluorescence unit) over 10 minutes (excitation, 540 nm; emission, 590 nm) after mixing the reagents of the reaction.

O₂ ^•− measurement

O₂ ^•− concentration in the reactions was determined using nitro blue tetrozolium (NBT) colorimetry and expressed as the Δ milli optical density (mOD)/minute at 570 nm over 20 minutes of incubation. The reaction mixture contained 0.75 mM NBT.

Cellular ROS measurement

Cellular ROS was measured using a fluorescent probe 5-(and-6)-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate, acetyl ester (CM-H₂DCFDA; Molecular Probe, Carlsbad, CA). Cells (10,000 per well) were plated in a 96-well plate, incubated the next day with 50 μL PBS containing 10 μM CM-H₂DCFDA for 30 minutes at 37°C, and then treated as described in the figure legends. The ΔRFU (excitation, 492 nm; emission, 520 nm) over time was measured using a microplate reader.

Catalase and GPX activity assay

The catalase activity was determined using Amplex^®Red Catalase Assay kit (Invitrogen, Carlsbad, CA), following the manufacturer's instructions. GPX activity was determined using UV spectrometry of NADPH that is oxidized to NADP⁻ in the presence of GSH, H₂O₂, and glutathione reductase.¹⁴ The results were normalized using protein concentration and expressed as U/mg protein.

HO^• measurement

In vitro (cell-free) HO^• generation was determined using the highly efficient and stable terephthalic acid (TA) assay.¹⁵ The reaction reagents were dissolved in calcium- and magnesium-free PBS containing 60 mM TA. The rate of HO^• generation was determined as the ΔRFU (excitation, 326 nm; emission, 420 nm) after 1 hour of incubation at 37°C in a fluorescent microplate reader.

Results

AA and GSH synergized with MnTMPyP to inhibit cancer cell growth

The synergistic effect of MnTMPyP and AA in the hormone-dependent (LNCaP and LAPC-4) and hormone-independent (PC-3, DU-145, LAPC-4) prostatic, pancreatic (PANC-1), and hepatic (Hep G2) cancer cell lines was tested. The treatment with MnTMPyP (up to 1 μM) or AA (up to 1 mM) as a single agent for 2 hours did not inhibit cell proliferation (measured using the clonogenic survival and MTT assays), whereas the combination of MnTMPyP and AA resulted in significantly decreased cell proliferation in a dose-dependent manner in all cell lines studied (Tables 1 and 2). Notably, both assays demonstrated that AA in combination with MnTMPyP significantly inhibited cell growth at a concentration of 0.2 mM, which is achievable in vivo with oral administration of AA.¹⁶

Table 1.

Synergistic Cytotoxicity of MnTMPyP and Ascorbic Acid (AA) on Cancer Cell Lines, as Measured Using the Clonogenic Survival Assay

	M (0.2 μM)	A (0.2 mM)	M (0.2 μM) A (0.2 mM)	ER1	A (0.5 mM)	M (0.2 μM) A (0.5 mM)	ER2
DU-145	91 ± 9	95 ± 2	61 ± 3^a	1.6	95 ± 4	9 ± 1^b	10.5
HD-LAPC-4	94 ± 1	105 ± 1	68 ± 3^a	1.5	90 ± 16	21 ± 1^b	4.3
Hep-G2	102 ± 1	99 ± 1	74 ± 2^a	1.4	109 ± 0	12 ± 3^b	9.0
HI-LAPC-4	98 ± 0	100 ± 5	65 ± 11	1.5	94 ± 5	5 ± 1^b	17.9
LnCAP	109 ± 5	108 ± 10	66 ± 16	1.7	97 ± 11	36 ± 2^b	2.7
PANC-1	110 ± 6	105 ± 9	79 ± 3^a	1.3	110 ± 8	10 ± 4^b	11.3
PC-3	107 ± 6	110 ± 12	41 ± 2^a	2.7	91 ± 5	0.4 ± 0.0^b	211.0

Cells were treated for 1 hour with MnTMPyP (M) at 0.2 μM in the presence or absence of AA (A, 0, 0.2, and 0.5 mM), and survival was determined using the clonogenic survival assay. Data (mean [of triplicates] ± standard error of the mean) are expressed as the survival fraction (%) compared with control.

p < 0.05 versus control, MnTMPyP, or AA (0.2 mM).

p < 0.05 versus control, MnTMPyP, or AA (0.5 mM). The synergistic effect of MA is represented as the ER of MA over A alone. ER1 is the ER of MA/A (M [0.2 μM] and A [0.2 mM]/A [0.2 mM]), and ER2 is the ER of MA/A (M [0.2 μM] and A [0.5 mM]/A [0.5 mM]).

ER, enhancement ratio.

Table 2.

Synergistic Cytotoxicity of MnTMPyP and Ascorbic Acid (AA) on Cancer Cell Line Proliferation, as Measured Using the MTT Assay

	M (1 μM)	A (0.25 mM)	M (1 μM) A (0.25 mM)	ER1	A (1 mM)	M (1 μM) A (1 mM)	ER2
DU-145	97 ± 1	96 ± 5	60 ± 3^a	1.6	91 ± 0	34 ± 12^b	2.7
HD-LAPC-4	98 ± 1	101 ± 5	60 ± 3^a	1.7	104 ± 2	24 ± 5^b	4.3
Hep-G2	97 ± 3	105 ± 8	73 ± 0^a	1.5	87 ± 12	22 ± 11^b	4.0
HI-LAPC-4	91 ± 13	94 ± 4	64 ± 8	1.5	98 ± 0	15 ± 0^b	6.5
LnCAP	105 ± 11	114 ± 5	68 ± 2	1.7	109 ± 3	14 ± 1^b	7.8
PANC-1	98 ± 1	99 ± 2	74 ± 2^a	1.3	102 ± 4	29 ± 2^b	3.5
PC-3	101 ± 1	98 ± 4	62 ± 1^a	1.6	99 ± 0	33 ± 2^b	3.0

Cells were treated for 2 hours with AA (A, 0, 0.25, and 1 mM) in the presence or absence of MnTMPyP (M, 1 μM). Cell proliferation was measured using the MTT assay at 3 days after treatment. Data are expressed as the % of OD normalized by the corresponding control group.

p < 0.05 versus control, MnTMPyP, or AA (0.25 mM).

p < 0.05 versus control, MnTMPyP, or AA (1 mM). The synergistic effect of MA is represented as the ER of MA over A alone. ER1 is the ER of MA/A (M [1 μM] and A [0.25 mM]/A [0.25 mM]), and ER2 is the ER of MA/A (M [1 μM] and A [1 mM]/A [1 mM]).

ER, enhancement ratio.

Having demonstrated that the combination of MnTMPyP/AA was cytotoxic in a variety of cell lines, the PC-3 cells were used for further study and it was demonstrated that the effect occurred rapidly (within 0.5 hour) and increased over the 2 hours of treatment time (Supplemental Fig. S1A, available online at www.liebertonline.com). The morphological changes of AA/MnTMPyP-treated PC-3 cells are shown in Supplemental Figure S2 (available online at www.liebertonline.com). The treated cells became flat and less refractory, with numerous vacuoles. These changes occurred within 1–2 hours of the treatment and preceded cell death as measured using the trypan blue exclusion assay (Fig. 1C).

FIG. 1.

Mitochondrial membrane potential (MMP) and cell death in PC-3 cells treated with MnTMPyP and AA. PC-3 cells were exposed for 1.5 hours to AA (1 mM) and/or MnTMPyP (1 μM). (A) Immediately after treatment, cells were stained with rhodamine 123 and used for bright-field (BF) and fluorescent (RHO) microscopies and flow cytometry (Flow). Arrows indicate the cells seen in BF but missing in the micrographs of RHO staining. (B) The cells were stained either immediately (0 hour) or at 24 hours after treatment with rhodamine 123 and analyzed using flow cytometry. Well-stained cells (M1 peak, as indicated in A) were expressed as percentage of M1/total counts. (C) Dead cells were measured using the trypan blue staining immediately (0 hour) or at 24 hours after the treatment. Data in B and C are expressed as the mean (of triplicates) ± SEM. MnTMPyP, Mn(III) tetrakis(N-methylpyridinium-2-yl)porphyrin⁵⁺; AA, ascorbic acid.

Notably, GSH (a major cellular reductant) also synergized with MnTMPyP to inhibit PC-3 cell growth, albeit to a lesser extent (Supplemental Fig. S1B, available online at www.liebertonline.com). The inhibitory effect of GSH on MnTMPyP-treated PC-3 cells was only observed at concentrations of 0.32–2 mM. Concentrations either below or above this range were without effect. When both MnTMPyP (1 μM) and GSH (0.8 mM) were included in the medium, the AA cytotoxicity was maximally potentiated (Supplemental Fig. S1C, available online at www.liebertonline.com).

MnTMPyP and AA synergistically induced mitochondrial damage

Given the rapidity with which morphological changes occurred, and the sensitivity of mitochondria to oxidative damage, the MMP of cells incubated with MnTMPyP/AA was measured using rhodamine 123 staining. Comparing fluorescent and bright-field micrographs immediately after the treatment (1.5 hours), it was observed that some of the cells treated with MnTMPyP/AA were not stained with rhodamine 123 (Fig. 1A). This change was quantitated using flow cytometry, which demonstrated the loss of about half of the rhodamine 123-stained population compared with either untreated control cells or cells treated with AA or MnTMPyP alone (Fig. 1B, 0 hour). The MMP change was not due to cell death, because there was no difference in trypan blue staining between the AA/MnTMPyP-treated cells and those in other treatment groups at this time point (Fig. 1C, 0 hour). However, at 24 hours after the treatment, rhodamine 123 staining was similar in all treatment groups, but there was a slight increase (3.7%) in dead cells in the AA/MnTMPyP-treated group (Figs. 1B and 2C, 24 hours), suggesting that MMP-damaged cells became nonviable over time.

FIG. 2.

In vitro (cell-free) generation of H₂O₂ and O₂ ^•− by MnTMPyP and AA. (A) H₂O₂ standard curves in the Amplex-Red assay, measured alone or in the presence of AA (1 mM), MnTMPyP (1 μM), or GSH (1 mM). (B) Rate of H₂O₂ generation (ΔRFU/second) was determined using Amplex-Red in reactions of AA (1 mM), MnTMPyP (5 μM), catalase (10 kU/mL), DHA (1 mM), GSH (1 mM), and GSSG (1 mM). (C) H₂O₂ generation was determined using Amplex-Red in reactions of different concentrations of AA and MnTMPyP. (D) O₂ ^•− concentration in the reaction of AA (1 mM), GSH (1 mM), and MnTMPyP (5 μM) in purified water (H₂O), PBS, or DMEM (with 10% FBS). Data are expressed as the mean (of triplicates) ± SEM. Cont, control; A, AA; M, MnTMPyP; C, catalase; D, DHA; G, GSH; H₂O₂, hydrogen peroxide; MnTMPyP, Mn(III) tetrakis(N-methylpyridinium-2-yl)porphyrin⁵⁺; AA, ascorbic acid; DMEM, Dulbecco's modified Eagle's medium. Cont and G are distinguished by dark and gray tints.

MnTMPyP and AA generated H₂O₂ through the Mn(III ↔ II)TMPyP redox cycle

The enzymatic activity of MnTMPyP as an oxidoreductase requires the presence of reductants.⁸ To determine if MnTMPyP generates H₂O₂ by this mechanism, a simplified in vitro (cell-free) system that contained MnTMPyP in its stable state as Mn(III)TMPyP and AA (or GSH), which are known to reduce Mn(III)TMPyP to Mn(II)TMPyP, was used.^8,9,17,18 The agents were dissolved in PBS at concentrations that were found to be synergistically cytotoxic as shown in Tables 1 and 2. The reduction of Mn(III)TMPyP to Mn(II)TMPyP in this system was confirmed using spectrophotometry (Supplemental Fig. S3, available online at www.liebertonline.com). H₂O₂ generation was measured using the Amplex-Red assay (resorufin fluorescence formation) and expressed as ΔRFU as described in the Materials and Methods section. AA and GSH suppressed resorufin formation, as evidenced by a smaller (4–5-fold) slope of the H₂O₂ standard curve (Fig. 2A). Nevertheless, AA and GSH alone increased ΔRFU at a rate higher than in untreated control cells (Fig. 2), which is consistent with previous reports of AA generating H₂O₂.⁵ Importantly, the combination of AA and MnTMPyP resulted in a much higher (∼ 60-fold) rate of ΔRFU than AA alone (Fig. 2B). This increased rate was dependent on the dose of both AA and MnTMPyP (Fig. 2C) and was suppressed by catalase (Fig. 2B), confirming that this assay is H₂O₂ specific. Further, no H₂O₂ was detected when AA was replaced with dehydroascorbic acid (DHA, the oxidized form of AA) (Fig. 2B), confirming the importance of AA-mediated reduction of MnTMPyP for H₂O₂ generation. When MnTMPyP and AA were mixed in DMEM with 10% FBS, H₂O₂ generation was also observed, but was slightly reduced (1–2-fold) when compared with that in PBS (Supplemental Fig. S4, available online at www.liebertonline.com). H₂O₂ was also produced when GSH (but not the oxidized form, GSSG) and MnTMPyP were combined, although at a lower rate than that observed with AA/MnTMPyP (Fig. 2B).

It has been reported that O₂ ^•− is the active intermediate in the reduction of O₂ to H₂O₂ by AA,¹⁹ and DMEM and unpurified buffers such as PBS contain sufficient amounts of transition metals (such as iron and copper) to catalyze AA oxidation and generate O₂ ^•−.^19,20 Accordingly, it was observed that addition of AA in both PBS and DMEM (with 10% FBS) caused a significant increase of O₂ ^•−, whereas in purified (deionized) water, no O₂ ^•− or H₂O₂ was generated by AA (Fig. 2D, and Supplemental Fig. S5, available online at www.liebertonline.com). To test if the Mn(II ⇔ III)TMPyP cycle driven by AA would enhance O₂ ^•− and subsequent H₂O₂ generation, MnTMPyP and AA were mixed in purified water. Subsequent O₂ ^•− and H₂O₂ generation was not detected under these conditions (Fig. 2D and Supplemental Fig. S5).

MnTMPyP and AA synergistically increased ROS in PC-3 cells

Using an intracellular ROS probe, CM-H₂DCFDA, ROS generation (as the fluorescence change, ΔRFU) of cells exposed to MnTMPyP or H₂O₂ (positive control) was measured after loading with CM-H₂DCFDA. Addition of external H₂O₂ to the media resulted in a dose-dependent increase of ΔRFU after 1 hour of incubation (Fig. 3A). MnTMPyP (up to 25 μM) and AA (1 mM) alone did not change ΔRFU within the 1 hour of exposure. However, the two combined markedly increased cellular ΔRFU (Fig. 3B, C). Exogenous catalase inhibited the increase of ΔRFU, indicating that this effect was primarily mediated by extracellular generation of H₂O₂. When cells were exposed to MnTMPyP/AA in PBS, H₂O₂ generation was also observed (Supplemental Fig. S6, available online at www.liebertonline.com). However, MnTMPyP alone in PBS induced a more robust increase of fluorescence than in tissue culture media. This is likely due to the binding of MnTMPyP to proteins in the medium, without which more MnTMPyP enters the cells and generates H₂O₂ intracellularly after reduction by intracellular reductants.

FIG. 3.

ROS generation in cells treated with MnTMPyP and/or AA. PC-3 cells were incubated with CM-H₂DCFDA (10 μM, 30 minutes) and then placed in a medium with different concentrations of H₂O₂ (A) or MnTMPyP (B), each in the presence or absence of catalase (300 U/mL) or AA (1 mM). Cellular ROS was determined as the ΔRFU over 1-hour treatment and expressed as the mean (of triplicates) ± SEM. (C) Representative fluorescent micrographs of the cells treated in A and B by H₂O₂ (1 μM), MnTMPyP (5 μM), and AA (1 mM). Cat, catalase; M, MnTMPyP; ROS, reactive oxygen species; MnTMPyP, Mn(III) tetrakis(N-methylpyridinium-2-yl)porphyrin⁵⁺; AA, ascorbic acid; H₂O₂, hydrogen peroxide.

MnTMPyP and AA synergism was dependent on H₂O₂

Exogenous catalase completely blocked the synergistic effect of MnTMPyP/AA on PC-3 growth inhibition (Fig. 4A), whereas the catalase inhibitor aminotriazole (ATZ) enhanced this effect (Fig. 4B). Both catalase and ATZ alone had no effect on cell growth. AA synergized with external H₂O₂ to inhibit cell growth in the absence of MnTMPyP (Fig. 4C). This effect was further enhanced by ATZ, which again demonstrates the importance of H₂O₂ as a mediator of MnTMPyP effects. In a series of prostate cancer cell lines with different levels of activities of H₂O₂-metabolizing enzymes, the growth of cells treated with MnTMPyP/AA was associated with the catalase and GPX activity (Fig. 4D), suggesting that H₂O₂ generated by low concentrations of MnTMPyP/AA can be overcome by high activities of endogenous H₂O₂-metabolizing enzymes.

FIG. 4.

MnTMPyP and AA synergism was dependent on H₂O₂. PC-3 cells were exposed for 1 hour to (A) AA (A, 1 mM), MnTMPyP (M, 0.25 mM), and different concentrations of catalase; (B) AA (0.25 mM), MnTMPyP (1 mM), and catalase inhibitor ATZ (20 mM); and (C) AA (0.5 mM), ATZ (20 mM), and different concentrations of H₂O₂. Cell proliferation in A–C was measured using the MTT assay at 3 days after treatment. (D) Different prostate-cancer cell lines were treated for 1 hour with AA (1 mM) and MnTMPyP (0.25 mM) and measured for proliferation by using the MTT assay, or the cell lines were treated for 1 hour with AA (0.5 mM) and MnTMPyP (0.2 mM) and measured for survival using the clonogenic assay. The data are expressed as the ratio of optical density (OD; or PE) of the MnTMPyP/AA-treated cells over that of the untreated control and then normalized relative to DU-145. Catalase and GPX activities in D were determined in untreated control cells and expressed as U/mg protein. Cell line abbreviations: DU, DU-145; P3, PC-3; HI, HI-LAPC-4; HD, HD-LAPC-4; LN, LnCAP. Data presented are the mean (of triplicates) standard error of the mean. MnTMPyP, Mn(III) tetrakis(N-methylpyridinium-2-yl)porphyrin5+; AA, ascorbic acid; H₂O₂, hydrogen peroxide; ATZ, aminotriazole; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; GPX, glutathione peroxidase.

MnTMPyP and AA synergism was dependent on the Fenton reaction

The present study hypothesized that the H₂O₂ generated by MnTMPyP/AA results in the Fenton reaction [Fe(II) + H₂O₂ → Fe(III) + HO^• + OH⁻] in the presence of Fe(II), which is enhanced by AA that recycles Fe(III) to Fe(II). First, this hypothesis was tested in a simplified in vitro cell-free system that contained the above agents dissolved in PBS at concentrations similar to those used in the cytotoxicity studies (Tables 1 and 2). As a positive control, there was a large increase of HO^• within 1 hour of combining AA, Fe(II), and H₂O₂ (Fig. 5A1). In the absence of AA, the reaction rate was relatively low and only became detectable after 12 hours (Fig. 5A1 and Supplemental Fig. S7, available online at www.liebertonline.com). Because trace amounts of Fe were present in the system, when Fe was omitted from this mixture, HO^• generation was only partially decreased. However, adding the iron chelator desferrioxamine (DFO) completely blocked HO^• generation. As expected, when H₂O₂ in the mixture of AA, Fe(II), and H₂O₂ was replaced by MnTMPyP, HO^• generation was still observed and occurred in a MnTMPyP dose-dependent manner (Fig. 5A2), indicating that the H₂O₂ generated by MnTMPyP and AA enhanced the Fenton reaction. In the absence of MnTMPyP, HO^• generation was detectable but relatively low (Fig. 5A2).

FIG. 5.

MnTMPyP and AA synergism was dependent on the Fenton reaction. (A) HO^• generation in cell-free reactions containing Fe (FeSO₄ · 7H₂O, 5 μM), AA (1 mM), DFO (10 mM), and DMSO (HO^• scavenger, as negative control, 0.67 M), in the presence of different concentrations of H₂O₂ (A1) or MnTMPyP (A2). The reaction rate was measured using the terephthalic acid assay and expressed as ΔRFU after 1 hour of incubation. (B) PC-3 cells were pretreated with dipyridyl (100 μM), DFO (100 μM), or EDTA (100 μM) for 2 hours and then treated with M (MnTMPyP, 1 μM) and the above chelators for 1 hour, in the presence of different concentrations of AA. (C) PC-3 cells were pretreated with mannitol (10 mM), melatonin (1 mM), or benzoic acid (5 mM) for 1 hour and then treated with MnTMPyP (1 μM)/AA (0.5 mM) and the above HO^• scavengers for 1 hour. In B and C, cell growth was measured using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay at 3 days after treatment and expressed as the mean of OD (B) or % OD relative to control (without AA treatment) (C). p < 0.01 (one-way ANOVA) for the difference among tested groups in C. *Significant difference (p < 0.01, Tukey post hoc test) versus MnTMPyP/AA alone. (D) Proposed mechanisms for the synergistic effect of AA and MnTMPyP. Me, metal; Mn, MnTMPyP; HA⁻, AA at pH = 7; MnTMPyP, Mn(III) tetrakis(N-methylpyridinium-2-yl)porphyrin⁵⁺; AA, ascorbic acid; DFO, desferrioxamine; HO^•, hydroxyl radical; H₂O₂, hydrogen peroxide.

To determine if MnTMPyP/AA cytotoxicity was mediated by the Fenton reaction, the iron chelators dipyridyl (DP), DFO, and EDTA were used to remove free iron from the cells. The table in Figure 5B summarizes and compares the membrane permeability of these chelators, as well as their binding affinity to Fe in different oxidation states and the redox activity (HO^• generation). Importantly, these chelators did not inhibit O₂ ^•− and H₂O₂ generation by MnTMPyP/AA (Supplemental Fig. S8, available online at www.liebertonline.com). None of the chelators alone affected cell growth when incubated with PC-3 cells for 3 hours at 100 μM (Fig. 5B). As a cell-permeable and Fe(II)-binding chelator, DP abolished the MnTMPyP/AA-induced growth inhibition. Less inhibition was observed with DFO, a less-permeable and Fe(III)-binding chelator. In comparison, EDTA, which is a cell-impermeable chelator known to enhance the Fenton reaction,²¹ did not affect the MnTMPyP/AA-induced growth inhibition. These data therefore indicate that the synergistic effect of MnTMPyP/AA is mediated by intracellular iron. To further confirm that MnTMPyP/AA acts through the Fenton reaction, HO^• scavengers (mannitol, melatonin, and benzoic acid) were added prior to and during the treatment. As expected, the growth inhibition by MnTMPyP/AA was significantly decreased by these scavengers (Fig. 5C). This inhibitory effect of scavengers was partial, rather than complete, which may be explained, at least in part, by the short half-life and site-specific damage mediated by HO^•.²²

As O₂ is required for HO^•-induced cell damage,²³ the effect of hypoxia on MnTMPyP/AA cytotoxicity was also assessed. As expected, hypoxia (0.5 and 2% O₂) significantly decreased the growth inhibition induced by MnTMPyP/AA (Supplemental Fig. S9A, available online at www.liebertonline.com). This decrease of cytotoxicity was not due to decreased ROS generation, because the CM-H₂DCFDA assay indicated no significant decrease of ΔRFU except under complete anoxia (Supplemental Fig. S9B, available online at www.liebertonline.com). Similarly, 0.5% O₂ did not affect HO^• generation by MnTMPyP/AA/Fe in the cell-free experiment (Supplemental Fig. S9C, available online at www.liebertonline.com).

Discussion

Clinical cancer trials in which AA was given orally have yielded inconsistent results.^24,25 More recent studies have demonstrated that AA has in vivo activity against cancer cells only at concentrations achievable with intravenous administration.^6,16 Yet, even with large doses (4 g/kg, twice daily) of intraperitoneal administration, the efficacy of AA was limited in different mouse tumor models.⁶ Therefore, alternative strategies are needed to make AA more potent and efficacious and to increase its potential clinical utility. When studying the phenotypes of a K-ras transformed prostate cell line (RWPE-2) overexpressing MnSOD or treated with MnTMPyP, Zhong et al. observed a synergistic inhibition of growth when the cells were treated with MnTMPyP (5 μM)/AA (2 mM) for 72 hours, and they concluded that MnTMPyP enhances the oxidative cellular damage.²⁶ In addition, Ye et al. recently reported that two other Mn(III) alkylpyridylporphyrins (MnTE-2-PyP⁵⁺ and MnTnHex-2-PyP) synergize with AA to induce cytotoxicity.²⁷ The underlying mechanism of this effect is not clear, but it has been suggested to be mediated by H₂O₂. Indeed, it has been proposed that this pro-oxidant mode of action of MnSOD and its mimetics may be responsible for their anticancer effects.^28,29 The present study reports a synergistic effect of MnTMPyP with AA, which is much more potent and rapid than that previously reported. The combined MnTMPyP/AA treatment for less than 2 hours resulted in a rapid increase of intracellular H₂O₂, with secondary mitochondrial injury and morphological changes (within the treatment period), cell death (within 24 hours after treatment), and growth inhibition (within 72 hours after treatment). Importantly, in this combination, the efficacious concentration of AA is markedly reduced to a level that can be achieved and maintained pharmacologically in human blood following oral administration.¹⁶ The pharmacokinetics of MnTMPyP has not been published; however, Spasojevic et al. recently reported the pharmacokinetics of Mn(III) tetrakis(N-ethylpyridinium-2-yl)porphyrin (MnTEPyP), a manganese porphyrin that has an ethyl in lieu of the methyl group in MnTMPyP that is linked with the porphyrin ring.³⁰ Plasma concentrations of MnTEPyP of 10 mM were reported following a single dose of 10 mg/kg i.p. in mice, which remained above the effective concentration of 0.2 μM in the MnTMPyP/AA regimen for several days. This favorable pharmacokinetic profile suggests that a reasonable dosing regimen may be possible in clinical trials.

In an in vivo environment, the majority of redox active metals, such as iron and copper, are tightly controlled by chelating protein. For instance, the nontransferrin-bound iron (NTBI) measured in body fluids are rarely greater than 1–2 μM and often undetectable in plasma from healthy humans. Given the findings that the AA/MnTMPyP-induced H₂O₂ generation requires other metal(s) in addition to the manganese in MnTMPyP, in future experiments it will be important to determine if such reaction and subsequent cytotoxicity can be achieved in vivo. Nonetheless, AA-induced H₂O₂ generation has been documented previously in mouse subcutaneous and tumor extracellular fluid.^6,31 The exact mechanism of the observed redox activity is not clear, but authors of the above articles argue that a putative metaloprotein(s) present in interstitial fluid is responsible for catalyzing the redox reactions.⁵

Conclusions

The present study demonstrated that the synergistic cytotoxicity of MnTMPyP and AA is mediated by a cascade of ROS generation (O₂ → O₂ ^•− → H₂O₂ → HO^•), which culminates in the generation of highly reactive and toxic HO^• radicals (Fig. 5D). AA facilitates each step of the reactions via metal-mediated redox cycling, including the Mn(II ⇔ III)TMPyP redox cycle that markedly enhanced H₂O₂ generation from O₂ ^•−. As a known reductant, GSH may induce cytotoxicity with MnTMPyP by the same mechanism. However, compared with AA, GSH is less able to reduce transitional metal ions at higher valence at neutral pH.³² As a result, the cytotoxicity of MnTMPyP/GSH is limited. In addition, at higher concentrations of GSH, the effect becomes less pronounced as H₂O₂ is more rapidly cleared. It has been reported that GSH, at concentrations (∼0.7 μM) similar to those observed in this experiment, inhibit clonogenic survival in Chinese hamster lung fibroblasts V79 in the presence of copper.³³

Importantly, it was observed that at low concentrations of MnTMPyP (<0.5 μM), the susceptibility of different cell lines to AA/MnTMPyP cytotoxicity is associated with the levels of endogenous catalase and GPX activity. It should therefore be theoretically possible to selectively kill cancer cells with relatively low basal levels of such enzymes within a range of MnTMPyP concentrations in the MnTMPyP/AA regimen that leaves normal cells relatively unaffected. As decreased levels of catalase activity have been reported in prostatic and pancreatic cancer cells,^2
–4 these tumors may be susceptible to this therapeutic approach with an acceptable therapeutic window.

Footnotes

Acknowledgments

This work was supported in part by a grant from the AUA Foundation Research Scholars Program.

Disclosure Statement

The authors claim no conflict of interest associated with this study.

References

Fry

, Jacob

. Sensor/effector drug design with potential relevance to cancer. Curr Pharm Des, 2006; 12:4479.

Baker

, Oberley

, Cohen

. Expression of antioxidant enzymes in human prostatic adenocarcinoma. Prostate, 1997; 32:229.

Bostwick

, Alexander

, Singh

et al. Antioxidant enzyme expression and reactive oxygen species damage in prostatic intraepithelial neoplasia and cancer. Cancer, 2000; 89:123.

Cullen

, Weydert

, Hinkhouse

et al. The role of manganese superoxide dismutase in the growth of pancreatic adenocarcinoma. Cancer Res, 2003; 63:1297.

Chen

, Espey

, Krishna

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 USA, 2005; 102:13604.

Chen

, Espey

, Sun

et al. Pharmacologic doses of ascorbate act as a prooxidant and decrease growth of aggressive tumor xenografts in mice. Proc Natl Acad Sci USA, 2008; 105:11105.

Giommarelli

, Corti

, Supino

et al. Cellular response to oxidative stress and ascorbic acid in melanoma cells overexpressing gamma-glutamyltransferase. Eur J Cancer, 2008; 44:750.

Faulkner

, Liochev

, Fridovich

. Stable Mn(III) porphyrins mimic superoxide dismutase in vitro and substitute for it in vivo. J Biol Chem, 1994; 269:23471.

Gardner

, Nguyen

, White

. Superoxide scavenging by Mn(II/III) tetrakis (1-methyl-4-pyridyl) porphyrin in mammalian cells. Arch Biochem Biophys, 1996; 325:20.

10.

Oberley

, Leuthauser

, Pasternack

et al. Anticancer activity of metal compounds with superoxide dismutase activity. Agents Actions, 1984; 15:535.

11.

Perez

, Cederbaum

. Antioxidant and pro-oxidant effects of a manganese porphyrin complex against CYP2E1-dependent toxicity. Free Radic Biol Med, 2002; 33:111.

12.

Husbeck

, Nonn

, Peehl

et al. Tumor-selective killing by selenite in patient-matched pairs of normal and malignant prostate cells. Prostate, 2006; 66:218.

13.

Mohanty

, Jaffe

, Schulman

et al. A highly sensitive fluorescent micro-assay of H₂O₂ release from activated human leukocytes using a dihydroxyphenoxazine derivative. J Immunol Methods, 1997; 202:133.

14.

Paglia

, Valentine

. Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J Lab Clin Med, 1967; 70:158.

15.

Saran

. Assaying for hydroxyl radicals: Hydroxylated terephthalate is a superior fluorescence marker than hydroxylated benzoate. Free Radic Res, 1999; 31:429.

16.

Padayatty

, Sun

, Wang

et al. Vitamin C pharmacokinetics: Implications for oral and intravenous use. Ann Intern Med, 2004; 140:533.

17.

Spasojevic

, Batinic-Haberle

, Fridovich

. Nitrosylation of manganese(II) tetrakis(N-ethylpyridinium-2-yl)porphyrin: A simple and sensitive spectrophotometric assay for nitric oxide. Nitric Oxide, 2000; 4:526.

18.

Batinic-Haberle

, Spasojevic

, Fridovich

. Tetrahydrobiopterin rapidly reduces the SOD mimic Mn(III) ortho-tetrakis(N-ethylpyridinium-2-yl)porphyrin. Free Radic Biol Med, 2004; 37:367.

19.

Scarpa

, Stevanato

, Viglino

et al. Superoxide ion as active intermediate in the autoxidation of ascorbate by molecular oxygen. Effect of superoxide dismutase. J Biol Chem, 1983; 258:6695.

20.

Buettner

. 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.

21.

Gutteridge

, Maidt

, Poyer

. Superoxide dismutase and Fenton chemistry. Reaction of ferric-EDTA complex and ferric-bipyridyl complex with hydrogen peroxide without the apparent formation of iron(II) Biochem J, 1990; 269:169.

22.

Chevion

. A site-specific mechanism for free radical induced biological damage: The essential role of redox-active transition metals. Free Radic Biol Med, 1988; 5:27.

23.

Wardman

. The importance of radiation chemistry to radiation and free radical biology (The 2008 Silvanus Thompson Memorial Lecture) Br J Radiol, 2009; 82:89.

24.

Moertel

, Fleming

, Creagan

et al. High-dose vitamin C versus placebo in the treatment of patients with advanced cancer who have had no prior chemotherapy. A randomized double-blind comparison. N Engl J Med, 1985; 312:137.

25.

Shekelle

, Hardy

, Coulter

et al. Effect of the supplemental use of antioxidants vitamin C, vitamin E, and coenzyme Q10 for the prevention and treatment of cancer. Evid Rep Technol Assess (Summ), 2003; 75:1.

26.

Zhong

, Yan

, Webber

et al. Alteration of cellular phenotype and responses to oxidative stress by manganese superoxide dismutase and a superoxide dismutase mimic in RWPE-2 human prostate adenocarcinoma cells. Antioxid Redox Signal, 2004; 6:513.

27.

, Fels

, Dedeugd

et al. The in vitro Cytotoxic Effects of Mn(III) alkylpyridylporphyrin/ascorbate system on ° tumor cell lines. Free Radic Biol Med, 2009; 47:S137.

28.

Jaramillo

, Frye

, Crapo

et al. Increased manganese superoxide dismutase expression or treatment with manganese porphyrin potentiates dexamethasone-induced apoptosis in lymphoma cells. Cancer Res, 2009; 69:5450.

29.

Batinic Haberle

, Rebouças

, Spasojević

. Superoxide dismutase mimics: Chemistry, pharmacology and therapeutic potential. Antioxid Redox Signal, 2010; 13:877.

30.

Spasojevic

, Chen

, Noel

et al. Pharmacokinetics of the potent redox-modulating manganese porphyrin, MnTE-2-PyP(5+), in plasma and major organs of B6C3F1 mice. Free Radic Biol Med, 2008; 45:943.

31.

Chen

, Espey

, Sun

et al. Ascorbate in pharmacologic concentrations selectively generates ascorbate radical and hydrogen peroxide in extracellular fluid in vivo. Proc Natl Acad Sci USA, 2007; 104:8749.

32.

Ferrer-Sueta

, Batinic-Haberle

, Spasojevic

et al. Catalytic scavenging of peroxynitrite by isomeric Mn(III) N-methylpyridylporphyrins in the presence of reductants. Chem Res Toxicol, 1999; 12:442.

33.

Held

, Sylvester

, Hopcia

et al. Role of Fenton chemistry in thiol-induced toxicity and apoptosis. Radiat Res, 1996; 145:542.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

2.14 MB

Metalloporphyrin Synergizes with Ascorbic Acid to Inhibit Cancer Cell Growth Through Fenton Chemistry

Abstract

Introduction

Materials and Methods

Cell culture, reagents, and equipments

Cytotoxicity assay

Cell survival assay

Mitochondrial membrane potential determination

Cell death measurement

H2O2 measurement

O2 •− measurement

Cellular ROS measurement

Catalase and GPX activity assay

HO• measurement

Results

AA and GSH synergized with MnTMPyP to inhibit cancer cell growth

MnTMPyP and AA synergistically induced mitochondrial damage

MnTMPyP and AA generated H2O2 through the Mn(III ↔ II)TMPyP redox cycle

MnTMPyP and AA synergistically increased ROS in PC-3 cells

MnTMPyP and AA synergism was dependent on H2O2

MnTMPyP and AA synergism was dependent on the Fenton reaction

Discussion

Conclusions

Footnotes

Acknowledgments

Disclosure Statement

References

Supplementary Material

H₂O₂ measurement

O₂ ^•− measurement

HO^• measurement

MnTMPyP and AA generated H₂O₂ through the Mn(III ↔ II)TMPyP redox cycle

MnTMPyP and AA synergism was dependent on H₂O₂