Melatonin and its derivatives cyclic 3-hydroxymelatonin, N1-acetyl-N2-formyl-5-methoxykynuramine and 6-methoxymelatonin reduce oxidative DNA damage induced by Fenton reagents

Authors


Address reprint requests to Russel J. Reiter, Department of Cellular and Structural Biology, The University of Texas Health Science Center, 7703 Floyd Curl Drive, San Antonio, TX 78229-3900, USA. E-mail: reiter@uthscsa.edu

Abstract

Abstract:  Free radicals are generated in vivo and they oxidatively damage DNA because of their high reactivities. In the last several years, hundreds of publications have confirmed that melatonin is a potent endogenous free radical scavenger. Some of the metabolites produced as a result of these scavenging actions have been identified using pure chemical systems. This is the case with both N1-acetyl-N2-formyl-5-methoxykynuramine (AFMK), identified as a product of the scavenging reaction of H2O2 by melatonin, and cyclic 3-hydroxymelatonin (C-3-OHM) which results when melatonin detoxifies two hydroxyl radicals (ḃOH). In the present in vitro study, we investigated the potential of two different derivatives of melatonin to scavenger free radicals. One of these derivatives is C-3-OHM, while the other is 6-methoxymelatonin (6-MthM). We also examined the effect of two solvents, i.e., methanol and acetonitrile, in this model system. As an endpoint, using high-performance liquid chromatography we measured the formation of 8-hydroxy-2′-deoxyguanosine (8-OH-dG) in purified calf thymus DNA treated with the Fenton reagents, chromium(III) [Cr(III)] plus H2O2, in the presence and in the absence of these molecules. The 8-OH-dG is considered a biomarker of oxidative DNA damage. Increasing concentrations of Cr(III) (as CrCl3) and H2O2 was earlier found to induce progressively greater levels of 8-OH-dG in isolated calf thymus DNA because of the generation of ḃOH via the Fenton-type reaction. We found that C-3-OHM reduces ḃOH-mediated damage in a dose-dependent manner, with an IC50 = 5.0 ± 0.2 μm; melatonin has an IC50 = 3.6 ± 0.1 μm. These values differ statistically significantly with P < 0.05. In these studies, AFMK had an IC50 = 17.8 ± 0.7 μm (P < 0.01). The 6-MthM also reduced DNA damage in a dose-dependent manner, with an IC50 = 4.2 ± 0.2 μm; this value does not differ from the ICs for melatonin and C-3-OHM. We propose a hypothetical reaction pathway in which a mole of C-3-OHM scavenges 2 mol of ḃOH yielding AFMK as a final product. As AFMK is also a free radical scavenger, the action of melatonin as a free radical scavenger is a sequence of scavenging reactions in which the products are themselves scavengers, resulting in a cascade of protective reactions.

Introduction

It is widely accepted that antioxidants help to maintain good health by decreasing oxidative damage to key biomolecules such as DNA. Thus, biomarkers of oxidative damage are considered useful in establishing which antioxidants are effective. Free radicals and other reactive species are generated in vivo and many of them cause oxidative damage to DNA which is implicated in mutagenesis, carcinogenesis and aging [1]. One of the most reactive oxygen-derived free radicals is the hydroxyl radical (ḃOH) which generates multiple modifications in DNA including both base damage and sugar lesions [2]. In recent years, it has been accepted that one of the best analytical methods to estimate oxidative DNA damage is the determination of 8-hydroxy-2′-deoxyguanosine (8-OH-dG) by high-performance liquid chromatography (HPLC) [1–3].

In the present study, we investigated the ability of three molecules, i.e., N1-acetyl-N2-formyl-5-methoxykynuramine (AFMK), cyclic 3-hydroxymelatonin (C-3-OHM), and 6-methoxymelatonin (6-MthM) to inhibit ḃOH-induced oxidative DNA damage and also compared their efficacies with that of melatonin (N-acetyl-5-methoxytryptamine). With this aim we measured, using HPLC with electrochemical (EC) detection, the formation of 8-OH-dG in purified calf thymus DNA treated with the Fenton reagents chromium(III) [Cr(III)] (as CrCl3) plus hydrogen peroxide (H2O2), in the presence and in the absence of the above cited molecules.

Reagent Cr(III), although an essential trace element involved in glucose and lipid metabolism [4], is also toxic and causes genotoxicity when high exposures to this substance occur in occupational and environmental settings [5, 6]. The carcinogenic mechanism of Cr(III) relates to its ability to generate ḃOH in the presence of H2O2 via the Fenton-type reaction [7]. The ḃOH then targets DNA, resulting in oxidative DNA damage including the formation of 8-OH-dG. The 8-OH-dG is a key biomarker relevant to carcinogenesis in that the formation of 8-OH-dG in DNA leads to G → T transversions during the replication step of DNA [8].

Melatonin is endogenously produced in all organisms including bacteria [9], plants and animals [10]. It is an ancient and well-conserved molecule, which implies that melatonin may play a basic and crucial role in supporting life. Melatonin, under a variety of in vitro and in vivo conditions, has been shown to be a potent endogenous free radical scavenger [10–25]. Efficacy of melatonin in protecting against oxidative damage has been documented in experiments conducted on animals, in tissue culture and in cell-free systems [12–14]. Moreover, being highly lipophilic [26] as well as somewhat hydrophilic [27], melatonin easily penetrates all known morphophysiological barriers, including the blood–brain barrier [28] and the placenta [29], and readily accumulates in all tissues and subcellular compartments where it may be bound to proteins [30–32]. Relatively little is known concerning the intracellular concentrations of melatonin, although some studies suggest that nuclear levels of melatonin can be five times higher than concurrent concentrations in the blood [33, 34].

Several metabolites of melatonin including AFMK, N-acetyl-5-methoxykynuramine, and 6-hydroxymelatonin (6-OHM) have also been shown to have the capacity to scavenger free radicals [16, 18, 19]. AFMK is an enzymatic metabolite of melatonin in vivo. Furthermore, AFMK was also identified as the product formed when melatonin directly scavenges H2O2 [19].

Another metabolite of melatonin, C-3-OHM, is physiologically generated in vivo and is excreted in the urine [20, 35, 36]. This urinary metabolite is identical to the compound generated in the in vitro chemical reaction between ḃOH and melatonin [17, 20, 24, 35]. In this reaction, melatonin's scavenging stiochiometry of the ḃOH is 1 mol melatonin for 2 mol ḃOH. Some of the products that are produced when melatonin detoxifies reactive species are also efficient scavengers, e.g., AFMK [16, 37] and 6-OHM [15]. One aim of this study was to determine whether C-3-OHM and 6-MthM (Fig. 1) also function as antioxidants.

Figure 1.

Chemical structures of melatonin and its derivatives used in this study.

Materials and methods

Reagents

Calf thymus DNA, CrCl3·6H2O and H2O2 were purchased from Sigma (St Louis, MO, USA). Chromatographically pure melatonin was a gift from Helsinn Chemical Co. (Biasca, Switzerland). Nuclease P1 and alkaline phosphatase were from Boehringer Mannheim (Indianapolis, IN, USA). High purity solvents were used for HPLC solutions. MilliQ-purified H2O (Honeywell, Muskegon, MI, USA) was used for preparing the stock solutions. All other chemicals used were of the highest quality available and were obtained from commercial sources.

The AFMK, C-3-OHM and 6-MthM are not commercially available. They were synthesized in our laboratory. The synthesis methods for AFMK and C-3-OHM are described in previous reports [19, 20].

Treatment

In the presence of 0.5 mm H2O2, calf thymus DNA (500 μg) was treated as described in an earlier report [15] with 500 μm Cr(III) in 10 mm potassium phosphate buffer (pH 7.4) at a final volume of 0.5 mL. The reaction was carried out at 37°C for 1 hr in a water bath with shaking.

Several concentrations (μm) of melatonin (0.5, 1, 2, 5, 10, 20, 40 and 100), C-3-OHM (0.5, 1, 2, 5, 10, 20, 40 and 50), AFMK (5.0, 1, 2, 5, 10, 20 and 40) and 6-MthM (0.5, 1, 2, 5, 10, 20 and 40) were used in combination with the reaction solution to test the efficacy of these antioxidants in altering the level of oxidative DNA damage. For each dose–response curve, calf thymus DNA was either untreated (controls) or treated in the absence of any antioxidant (‘0’ dose). There was 0.4% acetonitrile in all experimental conditions.

Assay for 8-OH-dG

The method for the assay of 8-OH-dG was similar to that described earlier [15]. After incubation, 50 μL sodium acetate (3 m, pH 5.3) and 3 mL of 100% ethanol at −20°C were added to each sample to terminate the reaction. DNA precipitate was collected and washed once with 80% ethanol. After 5 min centrifugation at 3000 g, the DNA samples were dried and dissolved in 200 μL of 30 mm sodium acetate (pH 5.3). The samples were denatured by heating at 65°C for 10 min. The DNA samples were digested to nucleotides by incubation with 8 U of nuclease P1 at 65°C for 15 min. The pH was then adjusted by adding 20 μL of 1 m Tris–HCl (pH 8.5) and the samples were treated with 4 U of alkaline phosphatase at 37°C for 1 hr. After incubation, 20 μL of 3 m sodium acetate (pH 5.3) and 20 μL of 50 mm ethylenediaminetetraacetic acid were added to each sample to terminate the reaction. The resulting deoxynucleoside mixture was filtered through a Millipore filter (0.22 μm) (Chelmsford, MA, USA) and analyzed by HPLC with a EC detection system. An ESA HPLC system equipped with an eight channel CoulArray 5600 EC detector was used: column, YMC-BD column (3 μm, 150 × 4.6 mm i.d.); eluent, 10% aqueous methanol containing 12.5 mm citric acid, 25 mm sodium acetate, 30 mm sodium hydroxide and 10 mm acetic acid, pH 5.2 at a flow rate of 1 mL/min. The quantities of 8-OH-dG and 2-deoxyguanosine (2-dG) were measured using different channels and two oxidative potentials (300 and 800 mV, respectively). The level of 8-OH-dG in each sample is expressed as the ratio of 8-OH-dG to 105 2-dG.

Statistical analysis

All data are expressed as mean ± S.E. (n = 6) and analyzed using a one-way ANOVA followed by the Student–Newman–Keul's test.

Results

Following their synthesis and purification, both AFMK and C-3-OHM (Fig. 1) were extracted with methanol. Methanol itself, even at very low concentrations, inhibited Cr(III)-induced formation of 8-OH-dG (Fig. 2). This effect obscured the inhibitory effect of AFMK and C-3-OHM. To avoid this circumstance, both AFMK and C-3-OHM purified samples were dried under vacuum and the residue was redissolved in acetonitrile at a final concentration of 10 mm. Fig. 3 shows that acetonitrile also had a slight dose-dependent inhibitory effect in Cr(III)-induced formation of 8-OH-dG. A concentration of 0.4% acetonitrile was selected for subsequent studies as it was the lowest concentration required for total solubility of melatonin and related molecules in the incubation solution; this concentration of acetonitrile reduced 8-OH-dG accumulation by 30 ± 2% of the treatment effect (Fig. 4). A 0.4% acetonitrile was present in all experimental conditions including both controls (Con) and treatment (‘0’) samples.

Figure 2.

The effect of different concentrations of methanol on Cr(III) (0.5 mm) plus H2O2 (0.5 mm)-induced 8-OH-dG levels in the DNA. The incubation time was 60 min. Results are given as mean ± S.E.M. (n = 6). Inset: Percent–inhibition curve.

Figure 3.

The effect of different concentrations of acetonitrile on Cr(III) (0.5 mm) plus H2O2 (0.5 mm)-induced 8-OH-dG levels in the DNA. The incubation time was 60 min. Results are given as mean ± S.E.M. (n = 6). *P < 0.01 compared with Cr(III) plus H2O2 group (‘T’).

Figure 4.

Percent–inhibition curve of different concentrations of acetonitrile in reducing Cr(III)-induced 8-OH-dG formation in DNA. The incubation time was 60 min. Percentages are given as mean ± S.E.M. (n = 6).

Melatonin inhibited Cr(III)-induced formation of 8-OH-dG in a dose-dependent manner (Fig. 5). The minimal effective concentration of melatonin that significantly reduced 8-OH-dG formation induced by 0.5 mm Cr(III) plus 0.5 mm H2O2 was 1 μm (P < 0.01). At this concentration of melatonin, the inhibition of 8-OH-dG formation was already 24 ± 1% (Fig. 6). All concentrations of melatonin greater than 1 μm caused progressively greater reductions in 8-OH-dG levels in DNA such that, at a concentration of 100 μm, melatonin reduced 8-OH-dG accumulation by 80 ± 3%.

Figure 5.

The effect of different concentrations of melatonin on Cr(III) (0.5 mm) plus H2O2 (0.5 mm)-induced 8-OH-dG levels in the DNA in the presence of 0.4% of acetonitrile. The incubation time was 60 min. Results are given as mean ± S.E.M. (n = 6). *P < 0.01 compared with Cr(III) plus H2O2 and acetonitrile group (‘0’).

Figure 6.

Percent–inhibition curve of different concentrations of melatonin, cyclic 3-hydroxymelatonin (C-3-OHM), N1-acetyl-N2-formyl-5-methoxykynuramine (AFMK) and 6-methoxymelatonin (6-MthM) in reducing Cr(III)-induced 8-OH-dG formation in DNA. The incubation time was 60 min. Results are given as mean ± S.E.M. (n = 6). The error bars are not shown because they are smaller than the symbols.

Fig. 7 shows that C-3-OHM, like melatonin, dose-dependently inhibited Cr(III)-induced formation of 8-OH-dG. The minimal effective concentration of C-3-OHM that significantly reduced (P < 0.01) Cr(III)-induced formation of 8-OH-dG in DNA was 1 μm. At this concentration C-3-OHM reduced 8-OH-dG accumulation by 15 ± 1%. The maximal concentration C-3-OHM used, 50 μm, reduced DNA damage by 71 ± 3% (Fig. 6).

Figure 7.

The effect of different concentrations of cyclic 3-hydroxymelatonin (C-3-OHM) on Cr(III) (0.5 mm) plus H2O2 (0.5 mM)-induced 8-OH-dG levels in the DNA in the presence of 0.4% of acetonitrile. The incubation time was 60 min. Results are given as mean ± S.E.M. (n = 6). *P < 0.01 compared with Cr(III) plus H2O2 and acetonitrile group (‘0’).

The AFMK also reduced Cr(III)-induced formation of 8-OH-dG in a dose-dependent manner (Fig. 8). The effective concentrations of AFMK were 5 μm and greater. At 5 μm concentration, AFMK reduced 8-OH-dG accumulation by 25 ± 2%. AFMK concentrations of 2 μm or lower had no significant effect on the accumulated DNA damage. At the maximal concentration of AFMK used, 40 μm, 8-OH-dG levels were reduced by 61 ± 4% (Fig. 6).

Figure 8.

The effect of different concentrations of N1-acetyl-N2-formyl-5-methoxykynuramine (AFMK) on Cr(III) (0.5 mm) plus H2O2 (0.5 mm)-induced 8-OH-dG levels in the DNA in the presence of 0.4% of acetonitrile. The incubation time was 60 min. Results are given as mean ± S.E.M. (n = 6). *P < 0.01 compared with Cr(III) plus H2O2 and acetonitrile group (‘0’).

Fig. 9 shows that the inhibitory effect of 6-MthM on Cr(III)-induced formation of 8-OH-dG is dose-dependent. The minimal effective concentration of 6-MthM was 2 μm which lowered Cr(III)-induced formation of 8-OH-dG by 27 ± 2% (Fig. 6). Concentrations of 6-MthM at 1 μm or less had no significant effect on 8-OH-dG accumulation. The maximal dose of 6-MthM tested i.e., 40 μm, inhibited DNA damage by 83 ± 6% (Fig. 6).

Figure 9.

The effect of different concentrations of 6-methoxymelatonin (6-MthM) on Cr(III) (0.5 mm) plus H2O2 (0.5 mm)-induced 8-OH-dG levels in the DNA in the presence of 0.4% of acetonitrile. The incubation time was 60 min. Results are given as mean ± S.E.M. (n = 6). *P < 0.01 compared with Cr(III) plus H2O2 and acetonitrile group (‘0’).

To compare the relative efficiencies of melatonin, C-3-OHM, AFMK and 6-MthM, the percentage inhibition relative to specific concentrations was calculated as shown in Fig. 6. The IC50 is the concentration of a particular molecule that inhibits the formation of 8-OH-dG in DNA by 50%. The IC50 values (μm) for melatonin, C-3-OHM, AFMK and 6-MthM were 3.6 ± 0.1, 5.0 ± 0.2, 17.8 ± 0.7 and 4.2 ± 0.2, respectively. The difference between the IC50 for AFMK and the IC50 for the other indole derivatives shown in this report is statistically significant with P < 0.01. The difference between the IC50 for melatonin and that for C-3-OHM is statistically significant with P < 0.05.

Discussion

Chromium compounds are widely used in industrial chemicals which can cause serious health problems when they accumulate in tissues of humans or animals as a consequence of their dietary intake [5,6, 38–40]. The carcinogenic potential of chromium may relate to its ability to reduce H2O2 to form the extremely cytotoxic ḃOH via the Fenton reaction, i.e., Cr(III) + H2O2 → Cr(IV) + ḃOH + OH−− [7]. ḃOH initiates macromolecular damage in proteins, lipids and DNA. Cr(III) and H2O2 was earlier found to induce oxidative DNA damage as documented by examining the levels of 8-OH-dG in isolated calf thymus DNA [7, 15, 16]. The 8-OH-dG is one of the many lesions generated in DNA by oxidative processes. It is the most extensively investigated DNA lesion and it is generally accepted as a key biomarker of carcinogenesis because the presence of 8-OH-dG in DNA leads to G → T and A → C transversions during the replication step of DNA [2, 8].

The aim of this study was to investigate the ability of two derivatives of melatonin, C-3-OHM and 6-MthM, to protect DNA against oxidative damage induced by ḃOH and to compare their effects with those of melatonin and AFMK [15, 16]. The end-product in this study was 8-OH-dG. Endogenous levels of 8-OH-dG measured in various laboratories differ significantly from one another, suggesting a possible laboratory-dependent variability in the measurements [1, 3]. Therefore the first step in this study was to ensure that the experimental conditions were optimal and that they did not generate artifacts.

Both melatonin metabolites, C-3-OHM and AFMK, were extracted with methanol at the end of the purification process [19, 20] and it is been reported that, in microsomes, methanol reacts with ḃOH to produce formaldehyde [41], i.e., methanol is a good ḃOH scavenger. In others experimental conditions, methanol acts as ḃOH scavenger via different mechanisms [42–44]. Because of this, we investigated the effect of methanol in reducing DNA damage before measuring the protective effects of melatonin and related compounds. Under these conditions the lowest concentration of methanol tested (0.1%) inhibited 8-OH-dG levels by more than 90%, thereby making it impossible to detect the effect of any other solute present in the reaction mixture.

Conversely, although acetonitrile lowered 8-OH-dG formation in a dose-dependent manner, it allowed for measuring the effects of other solutes because 0.4% acetonitrile only reduced formation of 8-OH-dG by 30%; this value is consistent with that reported in the literature [1, 3]. Because of this we chose acetonitrile as the solvent in the current studies.

Melatonin has been widely documented as a potent, endogenous antioxidant that scavenges several free radicals including the highly toxic ḃOH [10–18, 21, 25, 37, 45–47]. From a thermodynamic point of view, the electron-rich structure of melatonin permits it to directly scavenge reactive species at a high rate [24]. By measuring a variety of oxidative indices (including levels of 8-OH-dG), earlier in vivo and in vitro studies have shown that melatonin effectively protects DNA from oxidative damage-induced by a variety of free radical generating agents and processes including chromium [15–17].

In the present experiment, we found that melatonin was highly effective in reducing Cr(III)-induced formation of 8-OH-dG with an IC50 = 3.6 ± 0.1 μm. Besides scavenging the ḃOH [20, 25, 35, 47], melatonin directly detoxifies H2O2 as well [19]. Additionally, melatonin reportedly is a metal chelator [48]. Each of these three processes may have been involved in melatonin's reduction of 8-OH-dG induced by Cr(III) plus H2O2.

The C-3-OHM is a metabolite of melatonin physiologically produced when melatonin detoxifies reactive species; it has been detected in the urine of both rats and humans [20, 35, 36]. In a pure chemical system, C-3-OHM has been shown to be generated by the chemical reaction between ḃOH and melatonin [17, 20, 24, 35] with a scavenging stiochiometry of 1 mol melatonin for 2 mol ḃOH. In the present study, C-3-OHM showed a high efficacy for reducing oxidative DNA damage. The C-3-OHM reduced Cr(III)-induced formation of 8-OH-dG in DNA in a dose-dependent manner. The relationship between the inhibition of DNA damage and doses of C-3-OHM fits a curve parallel to that for melatonin and slightly but significantly (P < 0.05) shifted to the right. The IC50 for C-3-OHM was 5.0 ± 0.2 μm; this IC50 is 1.6 times higher and significantly different (P < 0.05) from that for melatonin.

The AFMK is known to be an enzymatic metabolite of melatonin in vivo, and it is also a reported product of the interaction of melatonin with both the superoxide anion (Oinline image) and H2O2 [19]. To date there is little known concerning the physiological functions or in vivo levels of AFMK, although recent evidence has shown it to be an effective antioxidant and ḃOH scavenger in a pure chemical system [16, 37]. In the current study, we found that AFMK reduced 8-OH-dG formation induced by Cr(III) plus H2O2 in a dose-dependent manner. In our experimental conditions, AFMK had a calculated IC50 = 17.8 ± 0.7 μm which is statistically significantly different (P < 0.01) from that of both melatonin and C-3-OHM, and four and three times higher that them, respectively.

The 6-MthM is a chemical derivative of melatonin; it is not known to exist in nature. The present study shows that 6-MthM significantly reduces the Cr(III)-induced formation of 8-OH-dG in DNA in a dose-dependent manner; thus, 6-MthM is an antioxidant equivalent to melatonin at higher concentrations in protecting DNA from the action of ḃOH. 6-MthM is, however, less effective than melatonin at low concentrations, i.e., 1 μm melatonin inhibited 8-OH-dG formation by 24% while 6-MethM at this concentration had no inhibitory action in reducing 8-OH-dG. Conversely, 6-MthM is more active than melatonin at high concentrations.

Regarding the interaction between C-3-OHM and ḃOH, we propose that one of the final products would be AFMK. The proposed reaction pathway is shown in Fig. 10. The reaction could proceed via at least two steps: (A) first, ḃOH interacts with C-3-OHM at position C-2 producing a rearrangement of the electrons. In this process, the N2-formyl group is formed and two rings are opened; (B) secondly, ḃOH extracts a hydrogen atom to form water, and generates AFMK. Based on this scheme, two ḃOH are detoxified by C-3-OHM as previously reported for melatonin [20]. Thus, melatonin scavenges two ḃOH generating C-3-OHM which also has the capability to neutralize two ḃOH, generating AFMK which also is a ḃOH scavenger, in a sort of cascade of scavenging reactions. This reaction cascade may greatly increase the efficiency of melatonin in protecting against ḃOH toxicity.

Figure 10.

Possible reaction pathway of cyclic 3-hydroxymelatonin (C-3-OHM) with ḃOH The reaction may proceed in at least two steps: (A) first, ḃOH joins the molecule of C-3-OHM producing a rearrangement of the electrons, forming the N2-formyl group and a carbon centered radical. (B) secondly, ḃOH joins the intermediate produced in step A pulling off a hydrogen atom and combining with it to form water, leaving behind N1-acetyl-N2-formyl-5-methoxykynuramine (AFMK).

Acknowledgements

SLB was in a sabbatical year supported by the Programa de Movilidad del Profesorado from the Spanish Ministry of Education, Culture and Sports. SLB acknowledges the University of Valladolid, School of Medicine, and Department of Biochemistry, Molecular Biology and Physiology for allowing her to come to UTHSCSA, Texas, USA. JCM acknowledges a post-doctoral fellowship from Fondo de Investigación Científica y Desarrollo Tecnológico (FICYT), Principado de Asturias, Spain. RMS acknowledges support from Fulbright Grant and the financial sponsorship of the Spanish Ministry of Education, Culture and Sports. DTX was supported by NIH training grant T32AG00165-13.

Ancillary