Adenine oxidation by pyrite-generated hydroxyl radicals
© Cohn et al; licensee BioMed Central Ltd. 2010
Received: 12 December 2009
Accepted: 26 April 2010
Published: 26 April 2010
Skip to main content
© Cohn et al; licensee BioMed Central Ltd. 2010
Received: 12 December 2009
Accepted: 26 April 2010
Published: 26 April 2010
Cellular exposure to particulate matter with concomitant formation of reactive oxygen species (ROS) and oxidization of biomolecules may lead to negative health outcomes. Evaluating the particle-induced formation of ROS and the oxidation products from reaction of ROS with biomolecules is useful for gaining a mechanistic understanding of particle-induced oxidative stress. Aqueous suspensions of pyrite particles have been shown to form hydroxyl radicals and degrade nucleic acids. Reactions between pyrite-induced hydroxyl radicals and nucleic acid bases, however, remain to be determined. Here, we compared the oxidation of adenine by Fenton-generated (i.e., ferrous iron and hydrogen peroxide) hydroxyl radicals to adenine oxidation by hydroxyl radicals generated in pyrite aqueous suspensions. Results show that adenine oxidizes in the presence of pyrite (without the addition of hydrogen peroxide) and that the rate of oxidation is dependent on the pyrite loading. Adenine oxidation was prevented by addition of either catalase or ethanol to the pyrite/adenine suspensions, which implies that hydrogen peroxide and hydroxyl radicals are causing the adenine oxidation. The adenine oxidation products, 8-oxoadenine and 2-hydroxyadenine, were the same whether hydroxyl radicals were generated by Fenton or pyrite-initiated reactions. Although nucleic acid bases are unlikely to be directly exposed to pyrite particles, the formation of ROS in the vicinity of cells may lead to oxidative stress.
The formation of reactive oxygen species (ROS) such as H2O2 and •OH is significant because of their reactivity; •OH will typically react with nearly all molecules in aqueous solution at diffusion-limited rates . Their extreme reactivity has been implicated in causing or contributing to disease and aging in humans [6–10]. Particles other than pyrite such as asbestos  and quartz  have also been shown to induce the formation of •OH in lung cells that have been exposed to the particles. The particulate-induced formation of •OH has been linked with oxidative stress [12, 13] and genotoxicity [14, 13]. Hence, •OH formation in vitro and in vivo has been used as an indicator for mineral-induced toxicity potential [14, 13, 12, 15, 16, 6].
The extremely short half-life of •OH hinders detection and quantification of •OH concentrations directly . Instead, detection requires the reaction of •OH with a target molecule. Upon reaction, characteristics of the target molecule such as light absorption , fluorescence [17–19], or electron spin resonance [20–23] may change. The detection of these changes is then used to determine the presence and concentration of •OH and other ROS. In the presence of cells or in tissue, the products of particle-induced radical oxidation include DNA strand-breaks [24, 14], RNA degradation , and nucleobase oxidation [25–27]. Nucleic acids react with •OH by hydrogen abstraction at the sugar or addition to the bases, both resulting in radical moieties and de-polymerization [28, 29, 24]. Oxidized base reaction products are typically detected using chromatography and mass spectroscopy [30–33]. Reaction of •OH with the purine bases guanine or adenine leads to common persistent products containing an additional single oxygen in the molecule (M+16). Examples of oxidation products generated by reaction of purine bases with •OH include 8-hydroxyguanine and 8-oxoadenine, in equilibrium with its less stable tautomer 8-hydroxyadenine (see [28, 34, 32, 35–37] for reviews). The reported M+16 products from reaction of adenine with ROS include 8-oxoadenine, 2-hydroxyadenine (isoguanine), and 6-N-hydroxyaminopurine (HAP) [38, 39].
The bio-available iron that is associated with pyrite in coal samples has been linked to the development of coal workers pneumoconiosis (CWP) in coal miners [40, 41]. Similarly, coal samples that contain pyrite have been shown to cause nucleic acid strand-breaks with an increasing degree of strand-breaks with greater pyrite content in the coal samples . While nucleic acid strand-breaks can occur in the presence of pyrite, the fate of the bases in the presence of pyrite-generated •OH has not been evaluated. The objective of this study was threefold: a) determine the effect of •OH concentration on the stability of the nucleobase adenine; b) determine if pyrite-generated • OH degrade adenine; and c) evaluate the adenine degradation products from reaction with pyrite.
In order to evaluate •OH-induced degradation of adenine, several experiments were performed exposing adenine solutions to various reactants and pyrite suspensions. The aqueous reactants included Fenton-generated •OH, the separate Fenton reagents [i.e., H2O2 and Fe(II)], and Fenton reagents with addition of catalase or ethanol. Catalase is an enzyme that reacts with H2O2 to form H2O and O2. When a high concentration of ethanol is added to a solution containing lower concentrations of adenine and Fenton reagents, ethanol will compete in scavenging •OH. Hence, addition of ethanol is expected to stabilize adenine in pyrite suspensions. Batch experiments were also performed by exposing adenine solutions to pyrite particles and with the addition of ethanol or catalase. The concentrations of adenine remaining after incubation with aqueous reactants or pyrite were determined using UV-Vis spectrophotometry and the reaction products were analyzed using high-pressure liquid chromatography time-of-flight mass spectroscopy (LCTOF-MS).
The aim of this study was to determine the effect of pyrite loading on the stability of adenine and to evaluate the adenine degradation products. The findings presented here show that adenine will react with both Fenton and pyrite-generated •OH by addition reactions to form 8-oxoadenine and, to a lesser extent, 2-hydroxyadenine. This is consistent with previous studies where •OH adds to one of the carbons in adenine followed by oxidation [28, 34]. In the experiments where adenine was exposed to Fenton and pyrite-generated •OH, an increase in either the Fenton reagent, H2O2 or pyrite particle loadings (i.e., greater pyrite surface area) led to higher rates of adenine degradation. The addition of either catalase or ethanol stabilized the adenine suggesting the role of H2O2 and •OH, respectively. In the Fenton reaction (eq. 3), H2O2 is added to a solution containing Fe(II) resulting in the formation of •OH. By adding catalase or ethanol, these species limit the availability of H2O2 and •OH to react with adenine. When pyrite particles are added to a solution, we hypothesize that pyrite first forms H2O2 through a reaction involving dissolved molecular oxygen and ferrous iron (eqs 1 and 2) either with Fe(II) located at the pyrite surface or dissolved into solution. Additional Fe(II) then reacts with the H2O2 to form •OH through the Fenton reaction . In the experiments where catalase is added to adenine solutions with either aqueous Fe(II) or pyrite suspensions, the concentration of adenine does not change over time suggesting that the kinetics of catalase-induced removal of H2O2 is faster than the reaction whereby •OH is generated via the Fenton reaction (eq. 3). Although not explored in this manuscript, it may be interesting to evaluate the independent roles of sulfur and iron oxidation in the generation of ROS and to investigate changes at mineral surface pre- and post-oxidation to evaluate the role of the mineral surface versus aqueous Fe(II) in the formation of ROS.
In addition to •OH, superoxide [(O2•)-] also forms in the Fe(II) oxidation reactions (eqs 1 to 3) that lead to the formation of •OH. While (O2•)- may also oxidize adenine, the role of (O2•)- compared to •OH in reaction with adenine from addition of pyrite has not been supported by other experiments performed by the authors. In an experiment where we evaluated the concentration of •OH in pyrite suspensions, the addition of superoxide dismutase did not have a substantial effect on the concentration of •OH while the addition of catalase did result in less •OH detection (data not shown).
Pyrite is the most common sulfide mineral and is present in mining waste and marine sediment. The inhalation of particles that are capable of generating •OH have been linked to biomolecular oxidation [12, 13] and genotoxicity [14, 13]. The formation of •OH by pyrite and its reaction with adenine as shown here may be relevant when pyrite particles are inhaled. For example, many sulfur-rich coals contain iron disulfide (FeS2) in the form of pyrite [44, 45] and there is a correlation between the pyritic sulfur content in coal samples and coal workers' pneumoconiosis prevalence . While this study shows that adenine is oxidized in the presence of pyrite, there are several limitations inherent in this study when extrapolating the data to other systems. This study was limited to only one of the bases, was executed with dissolved adenine instead of intact nucleic acids, and the experiments were performed in aqueous systems instead of in the presence of cells or tissues. Further experiments are necessary to determine the effect of pyrite-generated •OH on the bases in vivo.
Natural pyrite (Huanzala, Peru) obtained from Wards Natural Science (Rochester, NY) was crushed in an agate mill. After crushing, the pyrite was sieved so that the collected particles were <90 μm but did not traverse the sieve with 10 μm openings. The particles were then washed with hydrochloric acid to remove surface oxides using a protocol described in earlier work . A surface area of roughly 1.25 m2/g was determined using five-point N2 adsorption BET. The pyrite particles were stored in a vacuum desiccator until used in experiments.
The degradation of adenine in the presence of variable amounts of pyrite (1-100 g/L) was compared to adenine degradation in the presence of Fenton generated •OH. All experiments were performed in either polypropylene opaque 2-mL or 15-mL centrifuge tubes at room temperature (23 ± 2°C). Solutions of adenine were produced by dissolving adenine powder (Sigma, 99% pure) in water (Easy Pure 18.3 MΩ-cm, UV-irradiated, and ultra-filtered) followed by filtration (Millipore PVDF 0.2 μm) and quantification (260 nm, ε = 13300 ). An aqueous suspension of catalase (64 kUnits final concentration) was purchased (Sigma, C100 from bovine liver), diluted in water and added to some of the tubes. Ethanol was also added to some of the tubes so that the ethanol concentration in each tube was 50% by volume. The experiments were initiated with the addition of all reagents in the tubes. The tubes were set on an end-over-end rotator. Samples taken after 24 hour incubations were taken from 2-mL tubes and samples taken as a function of time were taken from 15-mL tubes. Samples with pyrite suspensions were filtered (Millipore 13 mm 0.45 μm PVDF) before quantification of remaining adenine by UV absorption at 260 nm.
Selected samples were further analyzed for identification of adenine oxidation products by HPLC-MS, utilizing electrospray ionization and a time-of-flight (TOF) mass detector (MicroMass LCT™). Separation of adenine and its oxidation products were achieved using a Waters Alliance 2695 HPLC on a Phenomonex reverse-phase C18 column (250 mm × 3.0 mm). LC-MS was conducted both in positive and negative ionization modes. Capillary and cone voltages were 2700 V and 25 V for positive ionization mode and 2200 V and 30 V in negative ion mode. The chromatography for negative ion analysis involved a solvent gradient of acetonitrile and water starting at 2% acetonitrile and increasing to 12% at 0.5% per minute for 20 minutes. For positive ion mode, the initial mobile phase composition was 5% methanol and 95% 25 μM ammonium formate solution, which after 3 minutes the percentage methanol was increased at 0.5% per minute for 30 minutes. Full spectra were continuously measured for ions between 100-800 m/z. Estimates of the accurate mass of adenine and product peaks were based on post column flow injection of the mass calibration standard leucine enkephalin and calculations performed using Masslynx™ v3.5 software (Reddy and Brownawell, 2005).
This work is supported by the Center for Environmental Molecular Science (NSF CHE 0221934), a seed grant from the Office of the Vice President for Research at Stony Brook University and by the NSF IGERT program (grant number DGE0549370).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.