Mechanisms of the pH- and Oxygen-Dependent Oxidation Activities of Artesunate

Abstract

Artemisinin was discovered in 1971 as a constituent of the wormwood genus plant (Artemisia annua). This plant has been used as an herbal medicine to treat malaria since ancient times. The compound artemisinin has a sesquiterpene lactone bearing a peroxide group that offers its biological activity. In addition to anti-malarial activity, artemisinin derivatives have been reported to exert antitumor activity in cancer cells, and have attracted attention as potential anti-cancer drugs. Mechanisms that might explain the antitumor activities of artemisinin derivatives reportedly induction of apoptosis, angiogenesis inhibitory effects, inhibition of hypoxia-inducible factor-1α (HIF-1α) activation, and direct DNA injury. Reactive oxygen species (ROS) generation is involved in many cases. However, little is known about the mechanism of ROS formation from artemisinin derivatives and what types of ROS are produced. Therefore, we investigated the iron-induced ROS formation mechanism by using artesunate, a water-soluble artemisinin derivative, which is thought to be the underlying mechanism involved in artesunate-mediated cell death. The ROS generated by the coexistence of iron(II), artesunate, and molecular oxygen was a hydroxyl radical or hydroxyl radical-like ROS. Artesunate can reduce iron(III) to iron(II), which enables generation of ROS irrespective of the iron valence. We found that reduction from iron(III) to iron(II) was activated in the acidic rather than the neutral region and was proportional to the hydrogen ion concentration.

Malaria is caused by a protozoan, Plasmodium falciparum,¹⁾ which is transmitted by mosquitos. When a protozoan enters the body from the salivary gland of a mosquito, it is taken into the liver where it reproduces. The protozoan is released from hepatocytes into the blood, subsequently invades red blood cells, and then destroys them, which causes anemia. In severe cases, the disease can be fatal. Malaria is prevalent in over 100 countries, and according to the estimates of the WHO, there are more than 212 million sufferers and 429000 deaths in 2015.²⁾ There have been no malaria outbreaks in Japan, but it is thought that the number of Japanese who travel to countries that have malaria epidemics is increasing every year, which leads to so-called imported malaria-infected patients of which 100 to 150 cases are reported every year. The WHO recommends artemisinin-based combination therapies (ACTs) for the treatment of uncomplicated malaria caused by the P. falciparum parasite.³⁾ In this ACT treatment, artemisinin derivatives, such as artesunate, artemether, and lumefantrine complex drugs are used. In Japan, artemether and lumefantrine combined tablets were released in 2016.

Artemisinin was discovered in 1971 in the wormwood genus plant (Artemisia annua), which is an herbal medicine that has been used in the treatment of malaria since ancient times. Artemisinin has a sesquiterpene lactone bearing an peroxide group,⁴⁾ which confers its biological activity.⁵⁾ Although the mechanism of action of these artemisinin derivatives on malarial parasites has not been elucidated, the injury caused by malarial parasites is presumed to be caused by free radicals generated by a reaction with heme produced in malarial phagocyte cells.^6–10)

Recently, it has been reported that artemisinin derivatives also exert anti-tumor activity in cancer cells,^11–18) which has attracted attention for use in anti-cancer drugs. The reported mechanisms that might explain the antitumor activity of artemisinin derivatives include induction of apoptosis through reactive oxygen species (ROS) generation,¹¹⁾ angiogenesis inhibitory effects,¹⁹⁾ inhibition of hypoxia-inducible factor-1α (HIF-1α) activation by ROS generation,²⁰⁾ and direct DNA injury.²¹⁾ ROS generation is involved in many cases. However, little is known about the mechanism of ROS formation from artemisinin derivatives and of the types of ROS produced. Artesunate is a water-soluble derivative of artemisinin. In this study, we investigated the mechanism of iron-induced ROS formation, which is thought to be the underlying mechanism involved in artesunate-mediated cell death.

MATERIALS AND METHODS

Reagents and Cell Culture

We purchased 2,2,5,5-tetramethyl-3-pyrroline-3-carboxamide (TPC) and dimethyl sulfoxide-¹³C₂ from Sigma-Aldrich (St. Louis, MO, U.S.A.), and 5,5-dimethyl-1-pyrrorine-N-oxide (DMPO) was obtained from Labotec (Tokyo, Japan). Edaravone was a kind gift from Mitsubishi Tanabe Pharma Corporation (Osaka, Japan). All other reagents were analytical grade and obtained from Wako Pure Chemical Industries, Ltd. (Osaka, Japan), Tokyo Chemical Industry (Tokyo, Japan), or Kanto Chemical (Tokyo, Japan) unless otherwise stated.

A human hepatoma cell line, HepG2, was obtained from American Type Culture Collection (Manassas, VA, U.S.A.) and cultured in Dulbecco’s Modified Eagle Medium (Life Technologies Japan Ltd., Tokyo, Japan) supplemented with 10% fetal bovine serum (GIBCO, Life Technologies Japan Ltd., Tokyo, Japan). The cells were inoculated in the culture plate (Iwaki Glass, Tokyo, Japan) and incubated in a humidified atmosphere with 5% CO₂ and 95% air for 24 h, then incubated in an atmosphere with 5% CO₂ and 1% (ca. 7.6 mmHg O₂), 6% (ca. 45.6 mmHg O₂), or 20% (ca. 152 mmHg O₂) O₂ and filled with inert N₂ for 3 d in an automatic multi-gas incubator (IncuSafe, MCO-5M; Sanyo Electric, Osaka, Japan). Culture media were equilibrated in a hypoxic atmosphere ahead of stimulation for 30 min to achieve hypoxic stimulation.

Artesunate and bipyridine were dissolved in dimethyl sulfoxide (DMSO) to make a stock solution. The final concentration of DMSO in the cell culture medium was <0.5%, and the control groups had the same concentration of DMSO as that in the test preparation in all experiments. Stock solutions of iron(II) (ferrous ammonium sulfate hexahydrate) and iron(III) (ferric chloride hexahydrate) were dissolved in 0.2% HCl before use. Edaravone was dissolved in distilled water and filtered for sterilization.

Cytotoxicity Assay at Various Oxygen Concentrations

The ratio of living HepG2 cells for the hypoxia and edaravone treatment in the presence of artesunate with iron was assessed by a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay, as previously described with modifications.²²⁾ Briefly, 1×10⁴ cells were incubated in 96-well plates (Falcon, BD, Franklin Lakes, NJ, U.S.A.) for 24 h and then cultured under normoxic or hypoxic conditions as indicated for 72 h. Stimulants (artesunate and iron(II)) were then added and further cultured for 28 h, which was followed by a reaction for 2 h with MTT at a concentration of 200 µg/mL. The formazan formed in the cells was dissolved in 20% sodium dodecyl sulfate with 0.5 mM HCl, and the absorbance at 570/630 nm was measured by using a microplate spectrophotometer (Varioskan Flash; Thermo Scientific, Wilmington, DE, U.S.A.). The viable cells were expressed as a ratio to the control cells under each oxygen condition.

To investigate the effects of edaravone on the cytotoxicity induced by artesunate and iron, HepG2 cells (1×10⁵ cells/mL) were incubated 48 h in a 24-well tissue culture plate under 20% O₂, and then the cell culture medium was replaced. Next, 30 min after edaravone (10, 30, and 100 µM) treatment, artesunate (250 µM) and iron(III)–8 hydroxyquinoline (8HQ) complex (iron 100 µM) were added and incubated for 4 h. Cell viability was then assessed by a MTT assay. The iron(III)–8HQ complex was introduced to gain the solubility of iron(III) at neutral pH according to the method of Lehnen-Beyel et al.²³⁾

Free-Radical Analysis Using Electron Paramagnetic Resonance (EPR) Spectroscopy

Radical generation from artesunate was confirmed by using EPR, and ROS generation from artesunate was examined by using an EPR-spin trapping method and salicylic acid trapping method.²⁴⁾ The measurement of EPR was performed at room temperature. The sample was transferred to three sections of glass capillaries (10 µL, Drummond Co., Broomall, PA, U.S.A.) and set into the EPR cavity for the measurements. All solutions were mixed prior to EPR measurement to provide the final concentrations indicated in the figure legends, and the reactions were initiated by the addition of iron. For anaerobic measurements, the reaction solution was de-aerated beforehand, mixed in the anaerobic bag, placed in the capillary, and both ends were sealed. A Bruker EMXPlus EPR spectrometer (Bruker Biospin, Osaka Japan) with an X-band cavity (ER 4103TM) was used to collect the EPR signal of TPC or DMPO spin adducts. The typical instrumental conditions were: 10 mW microwave power, 2.0-Gauss modulation amplitude, 0.08-s time constant, 120-s scan time, and 100-Gauss scan range.

Hyperfine coupling constants and radical concentrations were obtained by using the computer program Winsim.²⁵⁾ Other individual conditions are shown in the figure legends.

Reduction of Iron(III) to Iron(II) by Artesunate

The amount of iron(II) formed by the addition of artesunate was determined from the amount of iron(II)–bipyridine formed by the reaction of bipyridine (BP) with iron(II).²⁶⁾ The sample was evacuated by passing argon gas for 15 min and then was mixed in a spectroscopic cell, and the absorbance of iron(II)–(BP)₃ formed was measured at 520 nm. The amount of iron(II)–(BP)₃ was determined by using the molar absorption coefficient at 520 nm (ε=8600 M⁻¹ cm⁻¹).²⁷⁾ The reduction rate was calculated from the slope of the line by using a Microsoft Excel.

Determination of Hydroxyl Radicals by DMSO Trapping

To clarify the hydroxyl radical production, we introduced the DMSO trapping method of Tai et al.,²⁸⁾ with a slight modification. The reaction mixture contained 100 mM DMSO, 1.6 mM Fe²⁺, 8 mM artesunate, and 0.24 mM 2,4-dinitrophenylhydrazine in 10 mM phosphate buffer (pH 4.0), which was maintained at room temperature for 150 min and analyzed by HPLC. HPLC was performed by using an EiCOM EP-300 pump (Eicom Co., Kyoto, Japan) and a manual injector equipped with a 50-µL loop. The separation was performed on a GL Sciences Inertsil ODS-SP (4.6×150 mm) column. The mobile phase consisted of 27.7 mM acetate buffer containing 30 mM sodium citrate (pH 4.75) and methanol (1 : 3), filtered through a 0.45-µm filter (Millipore, Bedford, MA, U.S.A.). The HPLC analysis was performed at 40°C by using a column oven (EiCOM ATC-300) under isocratic conditions with a flow rate of 1.0 mL/min and then monitored on an ECL detector (EiCOM ECD-300) at +800 mV with the full-scale current set at 0.5 nA AUFS.²⁹⁾

Statistical Analysis

The statistical significance of differences was determined by Dunnett’s or Tukey–Kramer’s test, independent-sample t-test, paired-sample t test, or Wilcoxon’s matched-pairs test. p<0.05 was defined as indicating statistical significance.

RESULTS

Artesunate-Stimulated Cytotoxicity at Various Oxygen Concentrations

It has been reported that artesunate exhibited cytotoxicity in various cells in the presence of iron. When human cultured hepatoma HepG2 cells were mixed with artesunate for 28 h, cytotoxicity occurred in a concentration-dependent manner. (Fig. 1). The oxygen partial pressure has been reported to be approximately 5% in many body tissues,^30,31) whereas it is 0–2% in the periphery of cancer cells.³²⁾ Therefore, we performed the same experiment by changing the culture condition to these oxygen partial pressures. As shown in Fig. 1, we found that low oxygen partial pressure (1% O₂) significantly attenuated the cytotoxicity of artesunate relative to the culture condition of 20% O₂ at all artesunate concentrations tested.

Fig. 1. Effect of Oxygen Concentration on Artesunate Plus Iron(II)-Induced Cell Death

HepG2 cells (1×10⁴ cells/mL) were incubated 24 h in a 96-well tissue culture plate under 20% O₂, then further incubated under 20% O₂ (open column), 5% O₂ (hatched column), and 1% O₂ (closed column) for 72 h. The cell culture medium was replaced and the indicated concentrations of artesunate with 0.1 mM iron(II) were added; the medium was then further incubated for 28 h under the indicated oxygen concentration. Thereafter, cell viability was assessed by an MTT assay, which was performed as described in Materials and Methods. The viable cells are shown as a percentage, and the level of formazan from HepG2 cells incubated without artesunate at each oxygen concentration was assigned the value of 100%. The values are the means±standard error (S.E.) (n=8). *, **, and *** indicate p<0.05, p<0.01, and p<0.001, respectively.

EPR Measurement for Nitroxide Radicals through Oxidation of TPC

TPC itself does not have an EPR signal, but when it is oxidized by an oxidizing agent such as a hydroxyl radical or singlet oxygen, it produces a characteristic triplet EPR signal.³³⁾ Therefore, we measured radicals derived from artesunate by using TPC. When 4 mM copper, iron(II), and iron(III) were added to 50 mM TPC and 4 mM artesunate, TPC radicals (a^N=14.7 gauss) were confirmed with iron(II) alone under aerobic conditions (Fig. 2G), and this signal was augmented by the coexistence of artesunate (Fig. 2H). With or without artesunate, copper ion did not affect the formation of the TPC radicals (Figs. 2C, D). Iron(III) showed a slight increase in TPC radicals with a 2 min incubation when present (Figs. 2E, F). On the other hand, under anaerobic atmosphere, no TPC radical derived from iron(II) or iron(II)-artesunate was observed (Figs. 2J, K).

Fig. 2. The EPR Spectra of TPC Radicals Formed by Incubation of a Solution of 50 mM TPC with 10 mM Artesunate and 4 mM Metal in N,N-Dimethylformamide (DMF)

A: TPC alone; B: TPC+artesunate; C: TPC+CuSO₄; D: TPC+artesunate+CuSO₄; E: TPC+FeCl₃; F: TPC+artesunate + FeCl₃; G: TPC+FeSO₄; H: TPC+artesunate+FeSO₄; I: same as B, but anaerobic condition; J: same as G, but anaerobic condition; K: same as H, but anaerobic condition. Stock solutions of artesunate (200 mM in DMSO), TPC (1 M in distilled water), and metals (200 mM in 0.1% HCl) were prepared and introduced into DMF to give final concentrations of 10, 50, and 4 mM, respectively. After 2 min, the TPC radicals were measured by EPR spectroscopy. The EPR spectrometer settings were: 10 mW microwave power, 2.0 Gauss modulation amplitude, 100 kHz modulation frequency, 0.08 s time constant, 120 s scan time, 1×10⁴ receiver gain and 100 gauss scan range.

Reduction of Iron(III) by Artesunate and Production of TPC Radicals

The results shown in Figs. 2F, G, and H confirm that increase in TPC radicals were observed with the combination of iron(III) and artesunate, and because the TPC radical increased with iron(II), artesunate appeared to have the ability to reduce iron(III). Therefore, we examined the reducing ability of iron(III) by artesunate under anaerobic conditions by using bipyridine, which is an indicator for iron(II).²⁶⁾ As shown in Fig. 3, when artesunate and iron(III) were mixed under anaerobic conditions, the absorbance spectrum of iron(II)–(BP)₃ peaked at 522 nm, and the complex was gradually produced over time. No significant change in the absorption spectrum of bipyridine was observed (data not shown).

Fig. 3. Reduction of Iron(III) to Iron(II) by Artesunate

Change in the absorption spectra of the iron(II)–bipyridine complex by the time-dependent reduction of iron(III) in the presence of artesunate under anaerobic conditions. All sample solutions (200 mM artesunate in DMSO, 300 mM bipyridine in DMSO, 200 mM FeCl₃ and 50 mM FeSO₄ in 0.2% HCl,) were de-aerated with argon gas for 20 min. A 2.7-mL aliquot of distilled water was placed in a quartz spectrometric cell with a septum seal cap and degassed by bubbling argon gas for 20 min. Then the reagents (final concentrations were; 5 mM artesunate, 0.1 mM FeCl₃, and 30 mM bipyridine) were added under anaerobic conditions, and the absorption spectra of iron(II)–(BP)₃ were acquired at 0 (a), 30 (b), and 60 (c) min after mixing. The dashed line (d) represents an authentic spectrum of an iron(II)–(BP)₃ complex with 0.025 mM FeSO₄ instead of FeCl₃.

Next, we investigated the effect of hydrogen ion concentration on the reduction of iron(III) by artesunate. Under anaerobic conditions, when artesunate and iron(III) were reacted in various pH solutions, as shown in Fig. 4A, we found that the formation of iron(II) occurred in correlation with incubation time, and accelerated under acidic conditions rather than under neutral conditions. The rate of iron(II) production was a function of physiological hydrogen concentration and gave a linear relationship (r²=0.9959) (Fig. 4B). Iron(II) is known to generate oxidizing substances, such as ferryl species or hydroxyl radical then is oxidized to the iron(III) form.³⁴⁾ The results in Figs. 3 and 4 show that when iron was incubated with an excess amount of artesunate, iron(III) was reduced to iron(II), and thereafter oxidizing species were continuously generated by autoxidation. Therefore, when the formation of the oxidizing substances by the iron and artesunate was measured by TPC, as shown in Fig. 5, the TPC radicals persisted continuously even with iron(III) or iron(II) in comparison to iron alone.

Fig. 4. Effect of Physiological Hydrogen Concentration on the Rate of Iron(II) Formation from Iron(III) in the Presence of Artesunate

A: Effect of varying initial hydrogen concentration on the production of iron(II). The iron(II) concentration was calculated by using the molar extinction coefficient of iron(II)–(BP)₃ at 520 nm (ε=8600 M⁻¹ cm⁻¹).²⁷⁾ All sample solutions (200 mM artesunate in DMSO, 300 mM bipyridine in DMSO, and 200 mM FeCl₃ in 0.2% HCl) were de-aerated with argon gas for 20 min. A 2.7-mL aliquot of 1% HEPES was placed in a quartz spectrometric cell with a septum seal cap, degassed by bubbling argon gas for 20 min, and then the reagents (final concentrations were 5 mM artesunate, 0.2 mM FeCl₃, and 30 mM bipyridine) were added under anaerobic conditions, and absorption at 520 nm was measured at the indicated times after mixing (closed circle, pH 6.0; open circle, pH 6.2; closed triangle, pH 6.5; open triangle, pH 7.0; closed square, pH 7.5). Data from three individual experiments are expressed as means±S.E.; *, **, and *** indicate p<0.05, p<0.01 and p<0.001 compared with pH 7.5 respectively. B: Effect of initial hydrogen concentration on the rate of iron(II) production. The vertical axis shows the rate of formation of iron(II) in each pH solution from Fig. 4A and the horizontal axis shows the hydrogen ion concentration at each pH. The correlation coefficient was 0.9959.

Fig. 5. Time Course of TPC Radical Formation in the Presence of Artesunate and Iron

A 20 µL of 500 mM TPC (in H₂O), 4 µL of 200 mM iron (FeCl₃ as iron(III), and FeSO₄ as iron(II) in 0.2% HCl) with or without 10-µL of 200 mM artesunate (dissolved in DMF) were diluted with DMF to 200 µL, and then the TPC radicals were measured at the indicated time by EPR spectroscopy. Open circle; iron(II)+artesunate, closed circle; iron(III)+artesunate, open square; iron(II), and closed square; iron(III). The TPC radical intensity was defined as the signal intensity indicated by the double arrow in the EPR signal of the TPC radical shown in the figure. EPR measurement was performed as described in Materials and Methods. EPR spectrometer settings were: 10-mW microwave power, 2.0-Gauss modulation amplitude, 100-kHz modulation frequency, 0.08-s time constant, 120-s scan time, 1×10⁴ receiver gain, and 100-gauss scan range. Data from three experiments are expressed as means±S.E.

Identification of Oxidation Species Produced by Artesunate and Iron

Since the formation of hydroxyl radical can be thermodynamically estimated by using artesunate and iron(II),³⁵⁾ the formation of hydroxyl radical was confirmed by using the ¹³C-labeled DMSO trapping method.^28,36) When the hydroxyl radical reacts with DMSO, the methyl radical (^·CH₃) is formed very rapidly (k=4.5×10⁹ M ⁻¹ s⁻¹), and the origin of this ^·13CH₃ radical was proven to be DMSO.³⁷⁾   

(1)

Then, we used DMPO as a radical trapping agent, and the generated methyl radical was detected by EPR. Here we used 5 mM and 8 mM artesunate due to the sensitivity of EPR and HPLC methods, respectively.

When 500 mM ¹³C-DMSO was added to the reaction mixture containing 5 mM artesunate, 2 mM iron(II) and 500 mM DMPO in distilled water under aerobic conditions, three spin adducts were detected with hyperfine splitting constants of a^N=16.01 and a^H=22.64 gauss (Fig. 6B, open circle), a^N=14.81 and a^H=14.48 gauss (Fig. 6B, closed circle) and a^N=16.20, a^H_β=23.70 and a_β^13C=7.50 gauss (Fig. 6B, open triangle) and these were assigned as DMPO/Artesunate radical,^38,7) DMPO/^·OH³⁹⁾ and DMPO/^·13CH₃³⁶⁾ from their hyperfine splitting constants, respectively.

Fig. 6. EPR Spectra of Spin Adducts Formed by Incubation of DMPO with Artesunate and Iron(II) in DMSO Solution

A: 5 mM artesunate, 2 mM iron(II), and 1 M DMSO were introduced into a 500-mM DMPO solution in distilled water. B: computer simulation using the hyperfine coupling constants derived from spectrum A of a^N=16.01 and a^H=22.64 gauss for DMPO/Artesunate, and a^N=16.20, a^H=23.70 and a_β^13C=7.50 gauss for the DMPO/^•13CH₃ adduct. C: Iron(II) and DMSO were introduced into a 500-mM DMPO solution in distilled water to give 2 mM and 1 M, respectively. D: Artesunate and DMSO were introduced into a 500-mM DMPO solution in distilled water to give 5-mM and 1-M concentration, respectively. All EPR measurements were performed under aerobic conditions, and the EPR spectrometer settings were the same as those in Fig. 2.

  

(2)

  

(3)

  

(4)

  

(5)

Here we could not detect the DMPO/^·OH adduct in Fig. 6C from the reaction mixture containing iron(II) plus DMPO under aerobic condition because DMPO/^·OH adduct decays rapidly in the presence of high concentration of iron(II) without suitable iron chelating agent.^40,41)

In Fig. 6B, we show the computer simulation for the radical adduct (Fig. 6A) that used the hyperfine splitting constants above. On the other hand, the generated methyl radical reacts with oxygen molecules to form a CH₃OO radical, which in turn decomposes into formaldehyde.   

(6)

  

(7)

Because formaldehyde reacts with 2,4-dinitrophenylhydrazine under acidic conditions to form the corresponding hydrazone, formaldehyde, 2-(2,4-dinitrophenyl)hydrazone,^42,43) we measured the hydrazone compound by HPLC. As shown in Fig. 7, formaldehyde, 2-(2,4-dinitrophenyl)hydrazone was produced even with iron(II) only, but both iron(II) and iron(III) significantly increased its production due to the coexistence of artesunate.

Fig. 7. Effect of Artesunate on Formaldehyde Formation from DMSO

Either 1.6 mM iron or 8 mM artesunate or both were added to 10 mM phosphate buffer (pH 4.0) containing 100 mM DMSO and 0.2 mM DNPH and incubated at room temperature for 150 min, then introduced into the HPLC column as described in Materials and Methods. Data from three individual experiments are expressed as means±S.E. *** p<0.001.

Effects of Edaravone on Artesunate-Induced Cell Death

Edaravone, a potent hydroxyl radical scavenger,^39,44) has been used in acute ischemic stroke in clinical settings and exerts a neuroprotective effect on ischemic injured brain.⁴⁵⁾

Cell viability was unchanged by 100 µM edaravone or 250 µM artesunate, and slightly, but not significantly (p=0.326) decreased by 100 µM Fe-8HQ treatment. When HepG2 cells were incubated with 250 µM artesunate and 100 µM iron(III), cell viability was significantly reduced. When edaravone was allowed to coexist in this culture solution, cell death significantly improved in a concentration-dependent manner for edaravone (Fig. 8).

Fig. 8. Effect of Edaravone on Artesunate Plus Iron(III)-Induced Cell Death

The MTT assay was performed as described in Materials and Methods. The percentages of viable cells are shown, and the level of formazan from HepG2 cells incubated without artesunate and iron was assigned the value of 100%. The values are the means±S.E. (n=4). ** p<0.01 vs. control, and ^## p<0.01, ^### p<0.001 iron and artesunate vs. edaravone treated iron and artesunate.

DISCUSSION

In this study, we found that cytotoxicity by artesunate was dependent on oxygen partial pressure, artesunate reduced iron(III) to iron(II), and the reduction activity was promoted under acidic conditions. Artesunate, artemisinin, and their derivatives are now widely used as antimalarial drugs.^46–50) Iron(II)-induced cleavage of peroxide bonds in artesunate derivatives has been found to produce carbon-centered free radicals and to exert antimalarial activity as alkylated species against malaria parasites.⁵¹⁾ In addition to its use in the treatment of malaria, artesunate is being developed as an anticancer agent.⁵²⁾ The mechanisms by which artesunate exerts its anticancer activity are thought to be induction of apoptosis,^11,53) cell cycle arrest,⁵⁴⁾ and modulation of p38 and calcium signaling,⁵⁵⁾ although the detailed mechanisms remain to be elucidated, which may limit further development of this compound in preclinical and clinical settings.

Interestingly, as shown in Fig. 1, the cytotoxicity of artesunate was significantly suppressed when the oxygen partial pressure was 1% relative to the cytotoxicity at 5 or 20%, which suggested that molecular oxygen concentration was a key to artesunate’s cytotoxicity. Similarly, Ohgami et al. found that the cytotoxicity of artemisinin to Molt-4 human leukemia cells was significantly increased under hyperbaric oxygen conditions, and suggested that the combination of hyperbaric oxygen therapy and artemisinin was an effective anticancer chemotherapeutic strategy.^11,13,56) Oxygen molecules produce ROS, such as superoxide anion radical, hydrogen peroxide, hydroxyl radical, and singlet oxygen, but it is unknown which active oxygen species are involved in artesunate cytotoxicity. Artesunate has an endoperoxide bridge moiety, so it was expected that artesunate would react with iron(II) to generate singlet oxygen.⁵⁷⁾ Therefore, to clarify this mechanism, we observed the EPR signal by using TPC, which is oxidized by singlet oxygen and hydroxyl radical to generate TPC radicals,³³⁾ and as shown in Fig. 2H, augmentation of TPC radicals was observed in the presence of both iron(II) and artesunate relative to artesunate (Fig. 2B) or iron(II) (Fig. 2G) alone under aerobic conditions. In Fig. 2G, generation of the TPC radical from the mixture of TPC and iron(II) was due to hydroxyl radical⁵⁸⁾ or hydroxyl-radical like species,⁵⁹⁾ such as ferryl³⁴⁾ generation by autoxidation of iron(II) under aerobic conditions.   

(8)

  

(9)

  

(10)

Since TPC radicals from artesunate and iron(II) could not be observed under anaerobic conditions (Fig. 2K), it was thought that endoperoxide bridge-derived singlet oxygen would not be generated from artesunate but that ROS from dissolved oxygen molecules would participate in the formation of TPC radicals.

In this study, we found that iron(III) slightly increased generation of TPC radicals (Fig. 2F) relative to that of iron(III) (Fig. 2E) or artesunate alone (Fig. 2B). It is known that some compounds having a carboxyl group are capable of reducing iron(III) to produce iron(II) without reducing agents.^26,60) Because artesunate has a carboxyl group, we hypothesized that this increase in TPC radicals was caused by reduction of iron(III) to iron(II). To better understand this phenomenon, the reducing activity of iron(III) by artesunate was examined by using bipyridine, which is an indicator for iron(II).²⁶⁾ As shown in Fig. 3, the absorption at 520 nm derived from the iron(II)–bipyridine complex appeared over time from the reaction mixture containing iron(III) and artesunate, which indicated that the formation of iron(II) was caused by artesunate. Furthermore, we found that this reducing activity depended on the pH of the solution and was accelerated under acidic conditions (Fig. 4A), and the rate of iron(II) production was first order in the concentration of hydrogen ion (Fig. 4B) (r²=0.9959). When we observed the formation of TPC radicals from the mixture of TPC, iron(III), and artesunate under aerobic conditions, only a very small amount of TPC radical was observed immediately after mixing, but it increased with time and reached half that of iron(II) at the end of the experimental period, which indicated that artesunate interacted not only with iron(II) but also with iron(III) and then produced ROS through reduction to iron(II) (reactions (8) to (10)). Oxidation of iron(II) proceeds rapidly under aerobic conditions,⁶¹⁾ whereas the reduction of iron(III) by artesunate was slow (Fig. 3), suggesting iron would exist as iron(III) regardless of the initial valence of it. Therefore, it was expected that the enhancement rate of TPC radical signal intensity became constant after 120 min despite the valence of iron with artesunate (Fig. 5).

Cancer cells are known to exhibit a characteristic metabolic phenomenon called the Warburg effect,⁶²⁾ which facilitates the production of lactic acid even in the presence of abundant oxygen and results in extracellular acidosis.^63–65)

Taking these factors into consideration, it is conceivable that reduction of iron(III) by artesunate may occur in low pH environments around cancer cells, and iron(II)-derived ROS with high cytotoxicity are likely to be generated.

The results of Figs. 2 and 5 show that oxygen molecules were involved in the TPC radical generation by artesunate and iron, and because the TPC radicals were generated by oxidation of TPC, we tried to identify the ROS involved in this reaction. It is known that iron(II) reacts with oxygen to generate ROS, such as superoxide anion radical (O₂^·−), hydroxyl radical (^·OH), and hydrogen peroxide (H₂O₂).^61,66) Therefore, we used DMPO as a radical trapping agent to investigate the production of ROS, which are formed during air oxidation from artesunate and iron(II).

On addition of excessive amount of DMSO into a reaction mixture containing artesunate, iron(II), with sufficient DMPO under aerobic conditions, DMPO/^·13CH₃, DMPO/Artesunate radical and DMPO/^·OH adducts were formed (Fig. 6A). The origin of DMPO/^·13CH₃ is considered to be ¹³C-DMSO.³⁶⁾

Only a few of these DMPO spin adducts were observed in the absence of artesunate (Fig. 6C) because DMPO/^·OH adduct decays rapidly in the presence of high concentration of iron(II).^40,41)

No EPR spectrum of DMPO/O₂^·− adduct was detected as an intermediate in this reaction because the efficiency of O₂^·− spin trapping by DMPO is low.⁶⁷⁾ Butler et al. reported that iron(II) interacted with the peroxide bond of artemisinin to give an alkoxyl radical or radicals and gave a DMPO adduct (a^N=15.7 and a^H=22.5 gauss) as DMPO/Artesunate radical in their experiment.⁷⁾

Therefore, to further clarify the assignment of the EPR signal of a^N=16.20, a^H=23.70 and a_β^13C=7.50 gauss in Fig. 6, the DMSO trapping method was performed in which the formic acid derivative was quantified by HPLC. As shown in Fig. 7, the formation of formaldehyde-2-(2,4-dinitrophenyl)hydrazone generated by the reaction of 2,4-dinitrophenylhydrazine (DNPH) with DMSO-derived formaldehyde was observed iron(II) alone due to the hydroxyl radical generation by autoxidation (reactions (8)–(10)), and it was significantly increased in the presence of artesunate relative to that for iron(II) alone, which indicated that artesunate and iron facilitated the generation of hydroxyl radical. Furthermore, the formation of formaldehyde-2-(2,4-dinitrophenyl)hydrazone in iron(III) and artesunate suggests that hydroxyl radicals are also formed by iron(III) and artesunate.

And the reason why the formation of it became smaller than the combination of iron(II) and artesunate was expected that the reduction rate from iron(III) to iron(II) was slow as indicated in Fig. 3.

Consequently, we found that artesunate promoted auto-oxidation of iron(II) to produce hydroxyl radical or hydroxyl radical-like ROS in the presence of oxygen and facilitated this reaction by reducing iron(III) to iron(II). Moreover, it was expected that these reactions would be promoted as the acidity increased. If the cytotoxicity of artesunate and iron shown in Fig. 1 was due to hydroxyl radicals, it was expected that cell death by artesunate with iron would be suppressed by edaravone, a potent hydroxyl radical scavenger. As shown in Fig. 8, we found that edaravone recovered the percentage of viable cells in a concentration-dependent manner. This could partly explain how hydroxyl radical or hydroxyl radical-like ROS contributed to the cytotoxicity in the case of artesunate with iron.

CONCLUSION

In conclusion, these results showed that artesunate was capable of causing cytotoxicity not only with iron(II) but also with iron(III) because of their reducing activity, which was accelerated in acidic environments. In addition, we found that hydroxyl radical or hydroxyl radical-like ROS contributed to the cytotoxicity. Denison predicted the formation of hydroxyl radical by the reaction of the artemisinin derivative with divalent iron according to the enthalpy calculation and demonstrated a correlation between their physiological activities and the yield of hydroxyl radical,³⁵⁾ which is consistent with the results of this study. It has been reported that the antitumor activity of artesunate was attenuated by antioxidants, such as N-acetylcysteine^68,11) and ascorbic acid,¹⁰⁾ which suggests that the oxidative reaction was mediated by the onset of the pharmacological effect of artesunate. Because cytotoxicity induced by hydroxyl radical or hydroxyl radical-like compounds causes chemical damage to cells, drug resistance is unlikely to occur. This may explain why resistance has not been observed in either cancer or malaria against artemisinin and its derivatives.¹⁶⁾ From the viewpoint of hydroxyl radical or hydroxyl radical-like ROS generation, we obtained that the cytotoxicity of artesunate depended on iron (regardless of the iron valence, oxygen, and low pH). We believe that the results of this study will be useful for increasing the efficacy and reducing the side effects of artesunate when used as an antimalarial or anticancer drug.

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1) Sherman IW. Biochemistry of Plasmodium (malarial parasites). Microbiol. Rev., 43, 453–495 (1979).
• 2) WHO. “World Malaria Report 2016.”: ‹http://www.who.int/malaria/publications/world-malaria-report-2016/report/en/›.
• 3) WHO. Guidelines for the treatment of malaria. Third edition. (2015).
• 4) Klayman DL. Qinghaosu (artemisinin): an antimalarial drug from China. Science, 228, 1049–1055 (1985).
• 5) Asano M, Iwahashi H. Determination of the structures of radicals formed in the reaction of antimalarial drug artemisinin with ferrous ions. Eur. J. Med. Chem., 127, 740–747 (2017).
• 6) Meshnick SR, Taylor TE, Kamchonwongpaisan S. Artemisinin and the antimalarial endoperoxides: from herbal remedy to targeted chemotherapy. Microbiol. Rev., 60, 301–315 (1996).
• 7) Butler AR, Gilbert BC, Hulme P, Irvine LR, Renton L, Whitwood AC. EPR evidence for the involvement of free radicals in the iron-catalysed decomposition of qinghaosu (artemisinin) and some derivatives; antimalarial action of some polycyclic endoperoxides. Free Radic. Res., 28, 471–476 (1998).
• 8) Avery MA, Gao F, Chong WK, Mehrotra S, Milhous WK. Structure–activity relationships of the antimalarial agent artemisinin. 1. Synthesis and comparative molecular field analysis of C-9 analogs of artemisinin and 10-deoxoartemisinin. J. Med. Chem., 36, 4264–4275 (1993).
• 9) Berman PA, Adams PA. Artemisinin enhances heme-catalysed oxidation of lipid membranes. Free Radic. Biol. Med., 22, 1283–1288 (1997).
• 10) Pradines B, Rolain JM, Ramiandrasoa F, Fusai T, Mosnier J, Rogier C, Daries W, Baret E, Kunesch G, Le Bras J, Parzy D. Iron chelators as antimalarial agents: in vitro activity of dicatecholate against Plasmodium falciparum. J. Antimicrob. Chemother., 50, 177–187 (2002).
• 11) Efferth T, Giaisi M, Merling A, Krammer PH, Li-Weber M. Artesunate induces ROS-mediated apoptosis in doxorubicin-resistant T leukemia cells. PLoS ONE, 2, e693 (2007).
• 12) Ooko E, Saeed MEM, Kadioglu O, Sarvi S, Colak M, Elmasaoudi K, Janah R, Greten HJ, Efferth T. Artemisinin derivatives induce iron-dependent cell death (ferroptosis) in tumor cells. Phytomedicine, 22, 1045–1054 (2015).
• 13) Eling N, Reuter L, Hazin J, Hamacher-Brady A, Brady NR. Identification of artesunate as a specific activator of ferroptosis in pancreatic cancer cells. Oncoscience, 2, 517–532 (2015).
• 14) Button RW, Lin F, Ercolano E, Vincent JH, Hu B, Hanemann CO, Luo S. Artesunate induces necrotic cell death in schwannoma cells. Cell Death Dis., 5, e1466 (2014).
• 15) Berdelle N, Nikolova T, Quiros S, Efferth T, Kaina B. Artesunate induces oxidative DNA damage, sustained DNA double-strand breaks, and the ATM/ATR damage response in cancer cells. Mol. Cancer Ther., 10, 2224–2233 (2011).
• 16) Efferth T. Mechanistic perspectives for 1,2,4-trioxanes in anti-cancer therapy. Drug Resistance Updates: Reviews and Commentaries in Antimicrobial and Anticancer Chemotherapy, 8, 85–97 (2005).
• 17) Tran TH, Nguyen TD, Poudel BK, Nguyen HT, Kim JO, Yong CS, Nguyen CN. Development and evaluation of artesunate-loaded chitosan-coated lipid nanocapsule as a potential drug delivery system against breast cancer. AAPS PharmSciTech, 16, 1307–1316 (2015).
• 18) Zhang YJ, Gallis B, Taya M, Wang S, Ho RJ, Sasaki T. pH-responsive artemisinin derivatives and lipid nanoparticle formulations inhibit growth of breast cancer cells in vitro and induce down-regulation of HER family members. PLOS ONE, 8, e59086 (2013).
• 19) Dell’Eva R, Pfeffer U, Vene R, Anfosso L, Forlani A, Albini A, Efferth T. Inhibition of angiogenesis in vivo and growth of Kaposi’s sarcoma xenograft tumors by the anti-malarial artesunate. Biochem. Pharmacol., 68, 2359–2366 (2004).
• 20) Huang XJ, Ma ZQ, Zhang WP, Lu YB, Wei EQ. Dihydroartemisinin exerts cytotoxic effects and inhibits hypoxia inducible factor-1alpha activation in C6 glioma cells. J. Pharm. Pharmacol., 59, 849–856 (2007).
• 21) Li PC, Lam E, Roos WP, Zdzienicka MZ, Kaina B, Efferth T. Artesunate derived from traditional Chinese medicine induces DNA damage and repair. Cancer Res., 68, 4347–4351 (2008).
• 22) Miyamoto L, Yagi Y, Hatano A, Kawazoe K, Ishizawa K, Minakuchi K, Tomita S, Tsuchiya K. Spontaneously hyperactive MEK-Erk pathway mediates paradoxical facilitation of cell proliferation in mild hypoxia. Biochim. Biophys. Acta, 1850, 640–646 (2015).
• 23) Lehnen-Beyel I, Groot HD, Rauen U. Enhancement of iron toxicity in L929 cells by D-glucose: accelerated (re-)reduction. Biochem. J., 368, 517–526 (2002).
• 24) Jen JF, Leu MF, Yang TC. Determination of hydroxyl radicals in an advanced oxidation process with salicylic acid trapping and liquid chromatography. J. Chromatogr. A, 796, 283–288 (1998).
• 25) Duling DR. Simulation of multiple isotropic spin-trap EPR spectra. J. Magn. Reson. B., 104, 105–110 (1994).
• 26) Tsuchiya K, Akai K, Tokumura A, Abe S, Tamaki T, Takiguchi Y, Fukuzawa K. Oxygen radicals photo-induced by ferric nitrilotriacetate complex. Biochim. Biophys. Acta, 1725, 111–119 (2005).
• 27) Eguchi LA, Saltman P. Kinetics and mechanisms of metal reduction by hemoglobin. 1. Reduction of iron(III) complexes. Inorg. Chem., 26, 3665–3669 (1987).
• 28) Tai C, Peng JF, Liu JF, Jiang GB, Zou H. Determination of hydroxyl radicals in advanced oxidation processes with dimethyl sulfoxide trapping and liquid chromatography. Anal. Chim. Acta, 527, 73–80 (2004).
• 29) Goldring C, Casini AF, Maellaro E, Del Bello B, Comporti M. Determination of 4-hydroxynonenal by high-performance liquid chromatography with electrochemical detection. Lipids, 28, 141–145 (1993).
• 30) Carreau A, El Hafny-Rahbi B, Matejuk A, Grillon C, Kieda C. Why is the partial oxygen pressure of human tissues a crucial parameter? Small molecules and hypoxia. J. Cell. Mol. Med., 15, 1239–1253 (2011).
• 31) Erecińska M, Silver IA. Tissue oxygen tension and brain sensitivity to hypoxia. Respir. Physiol., 128, 263–276 (2001).
• 32) Clavo B, Robaina F, Morera J, Ruiz-Egea E, Perez JL, Macias D, Carames MA, Catala L, Hernandez MA, Gunderoth M. Increase of brain tumor oxygenation during cervical spinal cord stimulation. Report of three cases. J. Neurosurg., 96 (Suppl.), 94–100 (2002).
• 33) Nakamura K, Ishiyama K, Ikai H, Kanno T, Sasaki K, Niwano Y, Kohno M. Reevaluation of analytical methods for photogenerated singlet oxygen. J. Clin. Biochem. Nutr., 49, 87–95 (2011).
• 34) Reinke LA, Rau JM, McCay PB. Characteristics of an oxidant formed during iron(II) autoxidation. Free Radic. Biol. Med., 16, 485–492 (1994).
• 35) Denisov E. An important role of intramolecular free radical reactions in antimalarial activity of artemisinin and its analogs. Org. Biomol. Chem., 9, 4219–4225 (2011).
• 36) Tsuchiya K, Jiang JJ, Yoshizumi M, Tamaki T, Houchi H, Minakuchi K, Fukuzawa K, Mason RP. Nitric oxide-forming reactions of the water-soluble nitric oxide spin-trapping agent, MGD. Free Radic. Biol. Med., 27, 347–355 (1999).
• 37) Bardouki H, da Rosa MB, Mihalopoulos N, Palm WU, Zetzsch C. Kinetics and mechanism of the oxidation of dimethylsulfoxide (DMSO) and methanesulfinate (MSI−) by OH radicals in aqueous medium. Atmos. Environ., 36, 4627–4634 (2002).
• 38) Meshnick SR, Yang YZ, Lima V, Kuypers F, Kamchonwongpaisan S, Yuthavong Y. Iron-dependent free radical generation from the antimalarial agent artemisinin (qinghaosu). Antimicrob. Agents Chemother., 37, 1108–1114 (1993).
• 39) Abe S, Kirima K, Tsuchiya K, Okamoto M, Hasegawa T, Houchi H, Yoshizumi M, Tamaki T. The reaction rate of edaravone (3-methyl-1-phenyl-2-pyrazolin-5-one (MCI-186)) with hydroxyl radical. Chem. Pharm. Bull., 52, 186–191 (2004).
• 40) Yamazaki I, Piette LH. ESR spin-trapping studies on the reaction of Fe²⁺ ions with H₂O₂-reactive species in oxygen toxicity in biology. J. Biol. Chem., 265, 13589–13594 (1990).
• 41) Makino K, Hagiwara T, Imaishi H, Nishi M, Fujii S, Ohya H, Murakami A. DMPO spin trapping in the presence of Fe ion. Free Radic. Res. Commun., 9, 233–240 (1990).
• 42) Oliva-Teles MT, Paíga P, Delerue-Matos CM, Alvim-Ferraz MCM. a PPı, Delerue-Matos CM, Alvim-Ferraz MCM. Determination of free formaldehyde in foundry resins as its 2,4-dinitrophenylhydrazone by liquid chromatography. Anal. Chim. Acta, 467, 97–103 (2002).
• 43) Zegota H. High-performance liquid chromatography of methanol released from pectins after its oxidation to formaldehyde and condensation with 2,4-dinitrophenylhydrazine. J. Chromatogr. A, 863, 227–233 (1999).
• 44) Yamamoto T, Yuki S, Watanabe T, Mitsuka M, Saito KI, Kogure K. Delayed neuronal death prevented by inhibition of increased hydroxyl radical formation in a transient cerebral ischemia. Brain Res., 762, 240–242 (1997).
• 45) Ren Y, Wei B, Song X, An N, Zhou Y, Jin X, Zhang Y. Edaravone’s free radical scavenging mechanisms of neuroprotection against cerebral ischemia: review of the literature. Int. J. Neurosci., 125, 555–565 (2015).
• 46) Li Y. Qinghaosu (artemisinin): chemistry and pharmacology. Acta Pharmacol. Sin., 33, 1141–1146 (2012).
• 47) Gomes MF, Faiz MA, Gyapong JO, Warsame M, Agbenyega T, Babiker A, Baiden F, Yunus EB, Binka F, Clerk C, Folb P, Hassan R, Hossain MA, Kimbute O, Kitua A, Krishna S, Makasi C, Mensah N, Mrango Z, Olliaro P, Peto R, Peto TJ, Rahman MR, Ribeiro I, Samad R, White NJ, Study 13 Research Group. Pre-referral rectal artesunate to prevent death and disability in severe malaria: a placebo-controlled trial. Lancet, 373, 557–566 (2009).
• 48) Kremsner PG, Krishna S. Antimalarial combinations. Lancet, 364, 285–294 (2004).
• 49) Kremsner PG, Taylor T, Issifou S, Kombila M, Chimalizeni Y, Kawaza K, Bouyou Akotet MK, Duscha M, Mordmuller B, Kosters K, Humberg A, Miller RS, Weina P, Duparc S, Mohrle J, Kun JF, Planche T, Teja-Isavadharm P, Simpson JA, Kohler C, Krishna S. A simplified intravenous artesunate regimen for severe malaria. J. Infect. Dis., 205, 312–319 (2012).
• 50) Nealon C, Dzeing A, Muller-Romer U, Planche T, Sinou V, Kombila M, Kremsner PG, Parzy D, Krishna S. Intramuscular bioavailability and clinical efficacy of artesunate in gabonese children with severe malaria. Antimicrob. Agents Chemother., 46, 3933–3939 (2002).
• 51) Krishna S, Uhlemann AC, Haynes RK. Artemisinins: mechanisms of action and potential for resistance. Drug Resistance Updates: Reviews and Commentaries in Antimicrobial and Anticancer Chemotherapy, 7, 233–244 (2004).
• 52) Krishna S, Ganapathi S, Ster IC, Saeed ME, Cowan M, Finlayson C, Kovacsevics H, Jansen H, Kremsner PG, Efferth T, Kumar DA. Randomised, double blind, placebo-controlled pilot study of oral artesunate therapy for colorectal cancer. EBioMedicine, 2, 82–90 (2015).
• 53) Hou J, Wang D, Zhang R, Wang H. Experimental therapy of hepatoma with artemisinin and its derivatives: in vitro and in vivo activity, chemosensitization, and mechanisms of action. Clin. Cancer Res., 14, 5519–5530 (2008).
• 54) Sen R, Bandyopadhyay S, Dutta A, Mandal G, Ganguly S, Saha P, Chatterjee M. Artemisinin triggers induction of cell-cycle arrest and apoptosis in Leishmania donovani promastigotes. J. Med. Microbiol., 56, 1213–1218 (2007).
• 55) Mu D, Zhang W, Chu D, Liu T, Xie Y, Fu E, Jin F. The role of calcium, P38 MAPK in dihydroartemisinin-induced apoptosis of lung cancer PC-14 cells. Cancer Chemother. Pharmacol., 61, 639–645 (2008).
• 56) Ohgami Y, Elstad CA, Chung E, Shirachi DY, Quock RM, Lai HC. Effect of hyperbaric oxygen on the anticancer effect of artemisinin on molt-4 human leukemia cells. Anticancer Res., 30, 4467–4470 (2010).
• 57) Cassileth BR, Yeung KS, Gubili J, Ebrary Inc. Herb–drug interactions in oncology. McGraw-Hill Medical; McGraw-Hill distributor, New York, London, xvi, p. 769 (2010).
• 58) Kachur AV, Tuttle SW, Biaglow JE. Autoxidation of ferrous ion complexes: a method for the generation of hydroxyl radicals. Radiat. Res., 150, 475–482 (1998).
• 59) Shen X, Tian J, Li J, Li X, Chen Y. Formation of the excited ferryl species following fenton reaction. Free Radic. Biol. Med., 13, 585–592 (1992).
• 60) Chao CC, Aust AE. Photochemical reduction of ferric iron by chelators results in DNA strand breaks. Arch. Biochem. Biophys., 300, 544–550 (1993).
• 61) Kosaka H, Katsuki Y, Shiga T. Spin trapping study on the kinetics of Fe²⁺ autoxidation: formation of spin adducts and their destruction by superoxide. Arch. Biochem. Biophys., 293, 401–408 (1992).
• 62) Warburg O. On the origin of cancer cells. Science, 123, 309–314 (1956).
• 63) Morishima H, Washio J, Kitamura J, Shinohara Y, Takahashi T, Takahashi N. Real-time monitoring system for evaluating the acid-producing activity of oral squamous cell carcinoma cells at different environmental pH. Sci. Rep., 7, 10092 (2017).
• 64) Gerweck LE, Seetharaman K. Cellular pH gradient in tumor versus normal tissue: potential exploitation for the treatment of cancer. Cancer Res., 56, 1194–1198 (1996).
• 65) Schornack PA, Gillies RJ. Contributions of cell metabolism and H+ diffusion to the acidic pH of tumors. Neoplasia, 5, 135–145 (2003).
• 66) Lloyd RV, Hanna PM, Mason RP. The origin of the hydroxyl radical oxygen in the Fenton reaction. Free Radic. Biol. Med., 22, 885–888 (1997).
• 67) Bull C, Fee JA, O’Neill P, Fielden EM. Iron-ethylenediaminetetraacetic acid (EDTA)-catalyzed superoxide dismutation revisited: an explanation of why the dismutase activity of Fe-EDTA cannot be detected in the cytochrome c/xanthine oxidase assay system. Arch. Biochem. Biophys., 215, 551–555 (1982).
• 68) Du JH, Zhang HD, Ma ZJ, Ji KM. Artesunate induces oncosis-like cell death in vitro and has antitumor activity against pancreatic cancer xenografts in vivo. Cancer Chemother. Pharmacol., 65, 895–902 (2010).