Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Feb 1.
Published in final edited form as: Environ Sci Pollut Res Int. 2015 Sep 23;23(3):2138–2147. doi: 10.1007/s11356-015-5007-4

Breaking the dogma: PCB-derived semiquinone free radicals do not form covalent adducts with DNA, GSH, and amino acids

Orarat Wangpradit 1,2,7, Asif Rahaman 3, S V Santhana Mariappan 4, Garry R Buettner 1,5, Larry W Robertson 1,2, Gregor Luthe 1,6,8
PMCID: PMC4767158  NIHMSID: NIHMS760886  PMID: 26396011

Abstract

Covalent bond formations of free radical metabolites with biomolecules like DNA and proteins are thought to constitute a major mechanism of toxicity and carcinogenesis. Glutathione (GSH) is generally accepted as a radical scavenger protecting the cell. In the present study, we investigated a semiquinone radical (SQ•-) metabolite of the semivolatile 4-chlorobiphenyl, using electron paramagnetic resonance spectroscopy, and oxygen consumption. Proton nuclear magnetic resonance (1H NMR) and liquid chromatography–mass spectrometry (LC-MS) were also employed to elucidate the radical interaction with DNA, amino acids, and GSH. We found that DNA and oligonucleotides stabilized SQ•- by electron delocalization in the π-stacking system, resulting in persistent radical intercalated, rather than forming a covalent bond with SQ•-. This finding was strongly supported by the semiempirical calculation of the semioccupied molecular orbital and the linear combination of the atomic orbitals, indicating 9.8 kcal mol−1 energy gain. The insertion of SQ•- into the DNA strand may result in DNA strand breaks and interruption of DNA replication process or even activate radical mediated secondary reactions. The presence of amino acids resulted in a decrease of the electron paramagnetic resonance (EPR) signal of SQ•- and correlated with their isoelectric points. The pH shifts the equilibrium of the dianions of hydroquinone and influenced indirectly the formation of SQ•-. Similar findings were observed with GSH and Cys. GSH and Cys functioned as indirect radical scavengers; their activities depend on their chemical equilibria with the corresponding quinones, and their further reaction via Michael addition. The generally accepted role of GSH as radical scavenger in biological systems should be reconsidered based upon these findings, questioning the generally accepted view of radical interaction of semiquinones with biologically active compounds, like DNA, amino acids, proteins, and radical scavengers like GSH.

Keywords: Electron paramagnetic resonance, Glutathione, Prostaglandin H synthase-2, Polychlorinated biphenyls, Reactive oxygen species, Semiquinone free radicals, Carcinogenesis, DNA

Introduction

It is generally accepted that free radicals can cause carcinogenesis. The free radicals generated by a biotransformation of xenobiotics are of particular interest as proximate carcinogens, since they react with DNA and other biomolecules readily to form adducts (Myers 1982). Free radicals have a wide range of reactivity (Buettner 1993), with lifetimes ranging from nanoseconds to years (Nonhebel and Walton 1974; Mill et al. 1980). The decrease of the Gibbs free energy (ΔG), referred to as energy minimization, is the driving force for their reactivity. Free radicals can undergo diverse reaction pathways: (i) redox reactions by electron or hydrogen transfers, (ii) termination by binding covalently to another free radical forming a closed-shell molecule, (iii) covalent binding with a non-radical, forming a new radical species, (iv) intramolecular delocalization of the free electron forming a relatively stable species, or (v) intermolecular by a non-covalent, coordinate, or ligand bond (e.g., π-interaction with an aromatic ring). In the present study, we investigated the molecular interaction of DNA with semiquinones (SQ•-), a group of free radicals generated by comproportionation of hydroquinone (H2Q) and quinone (Q). SQ•- and Q have been extensively studied as the ultimate carcinogens causing cytotoxicity, mutagenicity, and DNA strand breaks (Morrison et al. 1985; Dwivedy et al. 1992; Shen et al. 1998; Bolton and Thatcher 2008; Lehmann et al 2007). In addition, we investigate the reaction with glutathione (GSH), a generally accepted radical scavenger, and amino acids as monomers of the proteins.

Our model compound is from the group of environmental pollutants, the polychlorinated biphenyls (PCBs) that persist in our environment in spite of manufacturing cessation in the 1970s. Semivolatile PCBs have triggered our research interest, as a number of studies have shown that lower chlorinated, so-called semivolatile PCBs are genotoxic (Oakley et al. 1996a, b; Srinivasan et al. 2001; Zettner et al. 2007; Ludewig et al. 2008). In our previous studies, we demonstrated that metabolites of 4-chlorobiphenyl (4-CB), the major congener determined in indoor air samples (Davis et al. 2002), function as reducing co-substrates in prostaglandin H synthase (PGHS) metabolism. This resulted in the generation of reactive Q metabolites, undergoing Michael additions with nucleophilic species (Wangpradit et al. 2009). The subsequent PGHS metabolism is correlated with elevated downstream production of prostaglandins which may disrupt the homeostasis and result in the inflammation responses in tissues (Kobayashi and Narumiya 2002). The SQ•- intermediates of 4-CB such as 4-chlorobiphenyl-2',5'-semiquinone (4-CB-2',5'-SQ•-) are formed transiently in situ. Therefore, it has been suggested that SQ•- binds covalently to DNA resulting in DNA damage as well (Oakley et al. 1996a, b; Arif et al. 2003). However, mechanisms of DNA-SQ•- interaction remain elusive. As it is thought that a mutation resulting from these reactions can adversely affect the cell cycle and progressively lead to malignancy (Harman 1962), it is particularly important to understand the chemical nature of the interaction to develop a prevention treatment or decelerate the adverse reactions. Amino acids as the other major target source of radical stress were also investigated. A major focus was given to the GSH thought to be the protector of the cell from radical attack.

It is our hypothesis that SQ•-, such as 4-CB-2',5'-SQ•-, do not form direct covalent products by radical reactions with DNA, GSH, and amino acids. To test our hypothesis, we employed a direct approach observing the radical interaction with the DNA, GSH, and amino acids in electron paramagnetic resonance (EPR) spectroscopy, the-state-of-art technique to monitor real-time formation, and interactions of the radicals. In addition, we monitored the oxygen consumption of the adduct formation to support our findings. The adducts were identified by proton nuclear magnetic resonance (1H NMR) and high-performance liquid chromatography coupled to mass spectrometry (HPLC-MS). Semiempirical in silico computations were performed to visualize and explain the processes energetically. 4-CB-2',5'-SQ•- was generated by an enzymatic reaction catalyzed by human recombinant prostaglandin H synthase-2 (hPGHS-2) with deoxynucleosides, oligonucleotides, and DNA. The use of hPGHS-2 was shown in our previous study to be an excellent catalyst generating high concentration of the SQ•- over a time span of several hours in the medium (Wangpradit et al. 2010).

The present study provides fundamental findings relevant to the fields of toxicology, environmental science, medicinal and radical chemistry, and chemical carcinogens: (i) elucidating for the first time the guest-host interactions of in situ generated SQ•- with DNA and amino acids by a direct and real-time monitoring approach, and (ii) insights in the protection of the cells by GSH function as radical scavenger toward radical stress.

Materials and methods

Materials

4-Chlorobiphenyl-2',5'-hydroquinone (4-CB-2',5'-H2Q) was supplied by Dr. Hans-Joachim Lehmler, the University of Iowa. The corresponding quinone, 4-CB-2',5'-Q (4-chlorobiphenyl-2',5'-benzoquinone), was synthesized using the Meerwein arylation as described by Amaro et al. (1996). Human recombinant PGHS-2 (hPGHS-2) and arachidonic acid (AA) (Cayman Chemical Company, Ann Arbor, MI). Hematin (MP Biomedicals, Solon, OH); salmon sperm DNA (SA Biosciences, Frederick, MD); glutathione (GSH), cysteine (Cys), glycine (Gly), histidine (His), lysine (Lys), arginine (Arg), and methionine (Met) (Research Products International Corp., Mt. Prospect, IL); and deoxyadenosine (dA), deoxyguanosine (dG), deoxycytidine (dC), and thymidine (dT) (Acros Organic, Morris Plains, NJ) were purchased from the sources indicated. Four oligonucleotides with 20 bases each, (AAAA)5, (CCCC)5, (GGGG)5, and (TTTT)5, were supplied by Integrated DNA Technology, Iowa, IA. Dimethyl sulfoxide (DMSO) (Fisher Chemical, Chicago, IL) was used as a vehicle for 4-CB-2',5'-H2Q. The deuterated solvents, acetonitrile-d3 (ACN-d3) and deuterium oxide (D2O), were purchased from Acros organics, NJ. Each experiment contained less than 2 % DMSO. Potassium arachidonate (KAA) was prepared by adding equal molar amounts of arachidonic acid (AA) to potassium hydroxide in aqueous solution.

Methods

EPR determination of 4-CB-2',5'-SQ•-by hPGHS-2

EPR spectroscopy

EPR spectroscopy was determined using Bruker EMX spectrometer equipped with a high-sensitivity cavity and an Aqua-X sample system for EPR measurements. Typical parameters to obtain EPR spectra (room temperature) were as follows: 3510 G center field, 15 G scan width, 9.854-GHz microwave frequency, 20-mW power, 2×105 receiver gain, 100-kHz modulation frequency, and 1.0 G modulation amplitude, with the conversion time and time constant both being 40.96 ms with 5X-scans for each 1024-point spectrum. The spectra were recorded with the 20 scan accumulation mode for 30 min.

In the presence of salmon sperm DNA

Two hundred units of hPGHS-2 was added to the solutions containing 0.02, 0.2, 2, and 7.5 mg of salmon sperm DNA in 0.1 M phosphate buffer (pH 7.4). The reactions were initiated by addition of 4-CB-2', 5'-H2Q dissolved in DMSO to the concentration of 100 μM in 1 mL solution. The control experiment was conducted under the same condition without an addition of salmon sperm DNA. The reaction mixtures were immediately transferred into the EPR spectrometer and measured as previously described.

In the presence of deoxynucleosides and oligonucleotides

Two hundred units of hPGHS-2 was added to the phosphate buffer solutions (pH 7.4) containing 100 μM of dA, dC, dG, and dT for the deoxynucleosides, and 50 μM of (AAAA)5, (CCCC)5, (GGGG)5, and (TTTT)5 for the oligonuclotides. The reactions were initiated by addition of 4-CB-2',5'-H2Q dissolved in DMSO to the concentration of 100 μM in 1 mL solution.

In the presence of amino acids

Two hundred units of hPGHS-2 was added to the solutions containing 100 μM of Gly, Lys, Arg, Cys, His, and Met in 0.1 M phosphate buffer (pH 7.4). The reactions were initiated by addition of 4-CB-2', 5'-H2Q dissolved in DMSO to the concentration of 100 μM in 1 mL solution.

In the presence of GSH

Two hundred units of hPGHS-2 was added to the solutions containing 5, 10, 25, 50, and 100 GSH μM in 0.1 M phosphate buffer (pH 7.4). The reactions were initiated by addition of 4-CB-2',5'-H2Q dissolved in DMSO to the concentration of 100 μM in 1 mL solution.

Oxygen consumption

Oxygen uptake for each reaction was determined using Clark electrode YSI model 5300 Biological Oxygen Monitor (Yellow Springs Instrument Co., Yellow Springs, OH). All experiments were conducted at room temperature. One hundred millimolars of Tris-HCl buffer (pH 8), and 100 units of hPGHS-2 were loaded in the chamber of the oxygen monitor and aerated by stirring for 5 min. The electrode was then placed in contact with the solution, and all air bubbles were removed. After 5-min equilibration, 4-CB-2',5'-H2Q was introduced through the access slot to a final concentration of 50 μM. Amino acids, GSH, or DNA were added to a final concentration of 100 μM immediately after the addition of 4-CB-2',5'-H2Q. The total volume after addition of study compounds was 3.0 mL. Each experiment was monitored for 60–90 min and repeated two times. The solution without amino acids, GSH, or DNA was used as a control. The reaction mixtures after monitoring oxygen consumption were kept in −80 °C freezer for LC-MS quantification of the adducts.

Quantification of the GS adducts

Quantification of the GS-adducts was performed on a Thermo LCQ deca mass spectrometer (Thermo Scientific, CA) interfaced with Dionex Ultimate 3000 LC (Dionex Corp., Sunnyvale, CA). Chromatographic separations were performed at 5 °C on a 2.1 mm i.d.×15 cm, 5 μM, Supelco Discovery C-18 column coupled to Supelguard column, 2.1 mm i.d.×2 cm, 5 μM (Sigma-Aldrich, MO). The mobile phases used were 0.1 % formic acid in water (A) and 0.1 % formic acid in acetonitrile (B) delivered at a total flow rate of 200 μL/min. The gradient profile was programmed as follows: 90 % A/10 % B isocratic held from 0 to 2 min, and gradient B up to 100 % from 2 to 30 min. The solutions were injected using a 200 vial autosampler equipped with a 20-μL loop. The LC was coupled with a photodiode array (PDA) detector followed by mass spectrometer (MS) operated in positive electrospray ionization (ESI) mode. The detection wave-length was 280 nm. MS data were acquired over single ion monitoring (SIM) mode at m/z ranges of 523.5–528.5 (mono-GS adducts), 828.5–833.5 (di-GS adducts), and 1133.5–1138.5 (tri-GS adducts).

NMR determination of adduct formation

4-CB-2',5'-Q with dG

The reaction was carried out by addition of dG dissolved in D2O to a solution of 4-CB-2',5'-Q dissolved in ACN-d3. The final concentrations of dG and 4-CB-2',5-Q were 600 μM and 4 mM, respectively. The solvent ratio of ACN-d3:D2O was 2:3. This reaction was incubated at room temperature for 24 h before NMR determination. The reaction of 4-CB-2',5'-SQ•- with dG was prepared by addition of 600 μM dG to the mixture of 2 mM of 4-CB-2',5'-Q and 4-CB-2',5'-H2Q.

4-CB-2',5'-Q with GSH

The reaction was carried out by addition of GSH dissolved in D2O to a solution of 4-CB-2', 5'-Q dissolved in ACN-d3. The final concentrations of GSH and 4-CB-2',5-Q were 30 mM and 4 mM, respectively. The solvent ratio of ACN-d3/D2O was 2:3. This reaction was incubated at room temperature for 5 min before NMR determination. The reaction of 4-CB-2',5'-SQ•- with GSH was prepared by the addition of 30 mM GSH to the mixture of 2 mM of 4-CB-2',5'-Q and 4-CB-2',5'-H2Q.

NMR spectroscopy

All products were characterized by the Avance-600 Bruker NMR spectrometer (Billerica, MA) operating at 600 MHz using a mixed solvent system, ACN-d3/D2O (2:3 by volume). Among the solvent systems researched for these studies (DMSO-d6/CD3OD, DMSO-d6/Dioxane-d4, Dioxane-d4/CD3OD, and ACN-d3/D2O), ACN-d3/D2O was found to be the most optimal in terms of solubility of the component molecules and solution conditions for NMR studies. A 1.7-mm inverse microprobe was used for all the one dimensional 1H NMR experiments. The residual HOD signal was saturated with a weak radio frequency field (rf-field) to improve the signal-to-noise-ratio of the resonances of interest. The NMR data was collected with a 64k time domain data points, with 32 scans and a relaxation delay of 5 s. All the NMR data were processed with TOPSPIN 1.3 suite of software programs. One-dimensional 1H NMR data were processed with zero filling to 128k data points and 0.2-Hz exponential line broadening.

Semiempirical computations of adduct formations

Semiempirical studies were carried out to determine the structures and energies of the complexes formed when 4-CB-2',5'-SQ•- was added to the different deoxynucleosides, dA, dG, dC, and dT as well as with dinucleotides d(ApA), d(GpG), d(CpC), and d(TpT). For efficient computational purposes, we deleted one of the phosphate groups from the end of the dinucleotides which we assumed have little effect on the binding. Initial structures of these deoxynucleosides and dinucleotides were taken from the protein data bank (PDB) ID 3OOR.1 Subsequent 4-CB-2',5'-SQ•- was orientated forming complexes between nucleic acid base pairs, so-called sandwich complexes. These complexes were then energetically minimized using molecular mechanics followed by semiempirical Austin Model 1 (AM1) calculations with standard VSTO-3G (5d,7f) basis functions using Gaussian 03 (Frisch et al. 2004). The optimized structures and binding enthalpies of the semioccupied molecular orbitals (SOMO) were obtained from these studies by a linear combination of the atomic orbitals (LCAO) of the hypothetical complexes formed. The complexation energy was determined by subtraction of the energy of the free bases and semiquinone from that of the complex (Ecomplexation=Ecomplex−(Efree bases+Efree SQ)). The SOMO were presented as closed spheres and mesh images using Gaussview (Gaussian, Inc.).

Results and discussion

Formation of 4-CB-2',5'-SQ•- can be accomplished by the autoxidation of 4-CB-2',5'-H2Q and/or comproportionation of the 4-CB-2',5'-Q and 4-CB-2',5'-H2Q (20–22). However, the signal intensity of 4-CB-2',5'-SQ•- from these reactions was considerably lower than the one-electron oxidation of 4-CB-2',5'-H2Q by hPGHS-2 (20). Hence, hPGHS-2 was used as a catalyst to oxidize 4-CB-2',5'-H2Q for all the EPR studies.

The intensity of 4-CB-2',5'-SQ•- spectrum observed in EPR depends on both formation and loss of the 4-CB-2',5'-SQ•- by its reaction with other molecules. Distortion of the redox equilibrium where 4-CB-2',5'-SQ•- is an intermediate can lead to the decrease of 4-CB-2',5'-SQ•- intensity as well. The latter case generally results from adduct formation of 4-CB-2',5'-Q, the oxidized form of 4-CB-2',5'-SQ•-, with a nucleophile. This reaction can initially shift the equilibrium to the formation of 4-CB-2',5'-Q in order to compensate for a shift in the redox system.

Salmon sperm DNA, oligonucleotides, and deoxynucleosides

In the presence of salmon sperm DNA, the intensities of 4-CB-2',5'-SQ•- spectra increased with increasing concentrations of salmon sperm DNA compared to the control (Fig. 1a). Addition of salmon sperm DNA to the oxygen monitor chamber did not consume more oxygen compared to the control (Fig. 1b). These findings indicated no adduct formation of 4-CB-2',5'-SQ•- and 4-CB-2',5'-Q with DNA.

Fig. 1.

Fig. 1

Open in a new tab

a EPR spectra of 4-CB-2',5'-SQ•- formed in the incubation of 100 μM 4-CB-2',5'-H2Q with 200 units of hPGHS-2 in the presence of salmon sperm DNA (SSDNA). b Oxygen consumption of the incubation of 50 μM 4-CB-2',5'-H2Q with hPGHS-2 in the presence of SSDNA and GSH

As an increased signal intensity of 4-CB-2',5'-SQ•- in the presence of DNA was not predicted from previous observations, we investigated the effects of each base on the EPR signal intensity of the semiquinone radical, by the addition of (AAAA)5, (GGGG)5, (TTTT)5, and (CCCC)5 to the incubation mixture. Figure 2a demonstrates that EPR signal of 4-CB-2',5'-SQ•- is enhanced in the presence of the oligonucleotides compared to control. Interestingly, addition of pyrimidine oligonucleotides, (TTTT)5 and (CCCC)5, resulted in greater enhancements. In addition, the purine oligonucleotides, (AAAA)5 and (GGGG)5, showed slightly increased signal intensities compared to the control. However, there was no significant difference in the 4-CB-2',5'-SQ•- intensities observed for single deoxynucleosides, dA, dC, dG, and dT, compared to the control (Fig. 2b).

Fig. 2.

Fig. 2

Open in a new tab

a EPR spectra of 4-CB-2',5'-SQ•- formed in the incubation of 100 μM 4-CB-2',5'-H2Q with 200 units of hPGHS-2 in the presence of 50 μM oligonucleotides (20 bases): (AAAA)5, (CCCC)5, (GGGG)5, and (TTTT)5. b EPR spectra of 4-CB-2',5'-SQ•- formed in the incubation of 100 μM 4-CB-2',5'-H2Q with 200 units of hPGHS-2 in the presence of 100 μM deoxynucleosides: dA, dC, dG, and dT

DNA is a non-radical compound, structured by a double helix phosphate backbone attached to four nucleosides connected as base pairs by hydrogen bonds. Within these nucleoside base pairs, planar molecules can reversibly insert, a process called intercalation. This host-guest interaction with aromatic guest molecules is a 3D π-stacking association that gains a lower Gibbs enthalpy by the extension or delocalization of the aromatic systems. We elucidated this by computations of the SOMO simulating the complex formations by LCAO. Figure 3 shows the optimized structures and the closed-shell models of the complexes of 4-CB-2',5'-SQ•- and the dinucleotides, d(ApA), d(GpG), d(TpT), and d(CpC). Computations of the SOMOs of the complexes of 4-CB-2', 5'-SQ•- with two single deoxynucleosides, dA, dG, dT and dC, were also carried out. The LCAO of the single nucleotides resulted in all cases in very little π-stacking or sandwiching and this even while they were artificially brought into close proximity allowing orbital interactions (−384.0, −182.6, and −107.9 kcal/mol for dT, dG, and dA, respectively). The LCAO of the dinucleosides with the 4-CB-2',5'-SQ•- on the other hand resulted in strong π-stacking interactions (−588.6, −317.5, and −158.8 kcal/mol for d(TpT), d(GpG), and d(ApA), respectively). The coordination by the phosphate group allows a stronger interaction; even the alignment of the axis is less favorable compared to free base pairs, due to the twist. Our computations demonstrate a successful LCAO resulting in an energy gain of 9.8 and 9.5 kcal mol−1 by delocalization for both purine and pyrimidine base pairs. This explains the increase of the 4-CB-2',5'-SQ•- intensity; half-life was increased due to stabilization. In addition, the computations demonstrated that the complexes formed by the pyrimidine bases (T and C) show a stronger LCAO compared to purine (A and G). This is in agreement with the observation of a stronger increase in EPR signal intensity for the 4-CB-2',5'-SQ•- in the presence of pyrimidine base pairs. We explain this finding on the basis of the differences in the stacking interaction between purine and pyrimidine oligonucleotides; steric hindrance results in a lower percentage of effective π-stacking in purine systems. This is in agreement with previous studies that demonstrated a stabilization in the alignment of an intercalator with the axes of the base pairs (Li et al. 2009). The twist of the DNA backbone affects the intercalation due to the difference in the alignment axis of the upper and the lower base pairs. Therefore, the stacking cannot uniquely determine the configuration of the final intercalation and by that the Gibbs energy (ΔG). Computations support our hypothesis that the stability of the non-covalent adducts is enhanced by strong π-stacking effects and delocalization of the free electron due to the formation of new molecular orbitals (MOs) in the formed complexes by LCAO. The energy gain of 9.8 and 9.5 kcal mol−1 by π-stacking is small compared to the formation of a C–O (78 kcal/mol) or a C–C (80 kcal/mol) covalent bond. However, the decreased steric hindrance of the aromatic rings by approaching with their planes is favorable and results in stacking. This is why π-stacking energy contribution to the stability of the nucleic acids in general is larger than the hydrogen bonding between the nucleic acid base pairs (Blanksby and Ellison 2003). This explains the signal increase of the SQ•- in the presence of DNA in general. We would like to emphasize that these delocalized guest radicals inside the DNA may initiate a variety of reactions themselves, or promote them.

Fig. 3.

Fig. 3

Open in a new tab

Semiempirical computations of the semioccupied molecular ortibals (SOMO) of the complexes of 4-CB-2',5'-SQ•- with the dinucleotides, d(ApA), d(GpG), d(TpT), and d(CpC), presenting as the optimized structures (left), and the closed orbital shell (right)

Glutathione (GSH)

The signal intensities of 4-CB-2',5'-SQ•- decreased with increasing concentrations of GSH (Fig. 4a). The lineshapes of 4-CB-2',5'-SQ•- spectra varied as a consequence of the combination of different 4-CB-2',5'-SQ•- species, a quartet of an original 4-CB-2',5'-SQ•-, triplets of the mono-SG adducts, and doublets of a di-SG adducts. The covalent products GS-4-CB-2',5'-Q and GS2-4-CB-2',5'-Q were characterized using 1H NMR spectroscopy. Figure 5 shows the differences of 1H NMR spectral signatures of the adducts from the reactants. A tri-SG or tri-NAC adduct was not observed either in EPR, NMR, or MS (15). Nevertheless, the 4-CB-2',5'-SQ•- signal was totally diminished in the presence of 100 μM GSH (Fig. 4a). The formation of GS substituted 4-CB-2',5'-SQ•- may derive from the following pathways: (i) the direct covalent binding between 4-CB-2',5'-SQ•- and a glutathiyl radical (GS•) generated from an oxidation of GSH by PGHS-2 (Eling et al. 1986), and (ii) Michael addition of 4-CB-2',5'-Q with GSH. Comparable to the comproportionation of 4-CB-2',5'-Q with the 4-CB-2',5'-H2Q to form 4-CB-2',5'-SQ•- (Song et al. 2008), the GS-4-CB-2',5'-Q and GS2-4-CB-2',5'-Q undergo comproportionation to form GS-4-CB-2',5'-SQ•- and GS2-4-CB-2',5'-SQ•- or undergo another hPGHS-catalyzed oxidation to the semiquione metabolites of the GS-adducts.

Fig. 4.

Fig. 4

Open in a new tab

a EPR spectra of 4-CB-2',5'-SQ•- formed in the incubation of 100 μM 4-CB-2',5'-H2Q with 200 units of hPGHS-2 in the presence of GSH. b Oxygen consumption of the incubations of 50 μM 4-CB-2',5'-H2Q in the presence of GSH (no hPGHS-2 added)

Fig. 5.

Fig. 5

Open in a new tab

1H NMR spectra of the adducts from incubations of 4-CB-2',5'-SQ•-/Q with dG, and GSH. The 4-CB-2',5'-SQ•- was generated from the comproportionation of an equal molar of 4-CB-2',5'-H2Q and 4-CB-2',5'-Q

To gain a deeper understanding of the observations, we investigated the reaction pathway by monitoring the formation of the adducts by oxygen consumption. The consumption of O2 was conducted without an addition of hPGHS-2 in order to avoid an enzymatic oxidation of GSH to GS• by hPGHS-2 that may distort the stoichiometry of the adduct formation. The consumption of oxygen was expected to result solely from the autoxidation of 4-CB-2',5'-H2Q, and reoxidation of the GS-4-CB-2',5'-H2Q adducts. Figure 4b depicts two major findings: (i) the increased O2 consumption with increasing concentrations of GSH, and (ii) an increased lag phase with increasing GSH concentration.

The first stage of O2 consumption began with autoxidation of 4-CB-2',5'-H2Q. The second stage indicated adduct formation of 4-CB-2',5'-Q with GSH, a process that pulled the equilibrium to the right, thus expediting the consumption of O2. Autoxidation of 4-CB-2',5'-H2Q (50 μM) consumed 22 μmol of O2 in Tris-HCl buffer pH 8, at 80 min. The oxidation of 4-CB-2',5'-H2Q consumed 1 mol of O2 for approximately 1.3, 1.4, and 1.7 mol of GSH in the presence of 50, 100, and 200 μM GSH, respectively. The overall stoichiometry of this reaction could be explained by Scheme 1. The mono-SG adduct formed in this reaction may undergo further oxidation to form the di-SG adduct. However, LC-MS data revealed that mono-SG adducts were the major adducts formed in all incubations, approximately 97 % in the presence of 50 and 100 μM GSH, and 90 % in the presence of 200 μM GSH. Therefore, the stoichiometry of 4-CB-2',5'-H2Q oxidation is likely to be 1:1:1 ratio of 4-CB-2',5'-H2Q/O2/GSH as shown in Fig. 6. The rest of GSH consumed in the oxidation may have resulted from the oxidation of GSH by H2O2.

Fig. 6.

Fig. 6

Open in a new tab

Mechanism of the adduct formation of 4-CB-2',5'-Q and GSH

The increased lag phase with increasing concentration of GSH is due to the formation of GS-4-CB-2',5'-H2Q adducts via the Michael addition of 4-CB-2',5'-Q with GSH. This reaction is favored, thus diminishing the concentration of 4-CB-2',5'-Q. The GS-4-CB-2',5'-H2Q can undergo a second round of autoxidation and Michael addition with GSH, resulting in the formation of GS2-4-CB-2',5'-H2Q. Lower level of 4-CB-2',5'-Q or GS-4-CB-2',5'-Q leads to a lower concentration of 4-CB-2',5'-SQ•- and GS-4-CB-2',5'-SQ•- as a consequence of the comproportionation reaction with the 4-CB-2',5'-H2Q or GS-4-CB-2',5'-H2Q. The formation of 4-CB-2',5'-SQ•- via the 4-CB-2',5'-Q2− (dianion) is the rate limiting reaction since the concentration of the dianion is relatively low in equilibrium.

Amino acids

Similar to DNA, it was found that certain amino acids covalently bind with quinones (Kalyanaraman et al. 1987; O'Brien 1991; Bolton et al. 2000). Semiquinone radicals were also thought to be reactive toward nucleophilic moieties of proteins, resulting in the radical-mediated protein damage (Flowers-Geary et al. 1996). Five amino acids from each group, Met (non-polar, hydrophobic), Lys and His (polar, basic), and Gly and Cys (polar, uncharged), were chosen to investigate the adduct formation in this study. Addition of these amino acids, except for Cys, resulted in decreased intensities of 4-CB-2',5'-SQ•- (Fig. 7a). At a first glance, these observations could indicate the adduct formation between 4-CB-2',5'-SQ•-/Q and amino acids. However, the oxygen consumption in the presence of these amino acids was not different from the control (Fig. 7b). Hence, the nucleophilic addition of 4-CB-2',5'-SQ•-/Q with these nitrogen-containing amino acids can be excluded. A study by Amaro et al. (1996) illustrated the adduct formation of 4-CB-2',5'-Q with amino acids at very slow rates for His (k=0.75 min−1 M−1), Gly (k=0.62 min−1 M−1), and Lys (k=0.64 min−1 M−1) at pH 7.4, supporting our findings that the adduct formation between 4-CB-2',5'-SQ•-/Q and these amino acids was unlikely within the limited incubation time (30 min).

Fig. 7.

Fig. 7

Open in a new tab

a EPR spectra of 4-CB-2',5'-SQ•- formed in the incubation of 100 μM 4-CB-2',5'-H2Q with 200 units of hPGHS-2 in the presence of 100 μM amino acids: Gly, glycine; Lys, lysine; Cys, cysteine; His, histidine; Met, methionine. b Oxygen consumption of the incubation of 50 μM 4-CB-2',5'-H2Q with hPGHS-2 in the presence of amino acids

The formation of 4-CB-2',5'-SQ•- is pH dependent; true autooxidation is mediated via the dianion of the hydroquinone (Song and Buettner 2010). Individual amino acids appear to influence the formation of 4-CB-2',5'-SQ•- differently based upon their isoelectric points (pI). The EPR signal intensity of correlates with the pI: Lys (pI 7.22), Gly (pI 6.07), Met (pI 5.73), and His (pI 5.68). A higher EPR signal intensity of the 4-CB-2',5'-SQ•- corresponds to a higher isoelectric point of the amino acid, representing a lower acidity. Therefore, we can speculate that an increase in the level of amino acid cation assists in generating more of the dianion of the hydroquinone.

The incubation with Cys completely diminished the EPR signal of 4-CB-2',5'-SQ•- (Fig. 7a). In addition, the consumption of O2 increased in the incubation of 4-CB-2',5'-H2Q with Cys (Fig. 7b). These findings indicate the formation of covalent adducts and were consistent with the previous experiment with GSH.

Conclusions

We found that (i) additions of SSDNA, and oligonucleotides (TTTT)5, (AAAA)5, (GGGG)5, and (CCCC)5) to the incubation of 4-CB-2',5'-H2Q with hPGHS-2 resulted in an increase of the EPR signal intensity of 4-CB-2',5'-SQ•-, while the deoxynucleotides (dT, dA, dG, dC) did not change the intensity of 4-CB-2',5'-SQ•-; (ii) SSDNA, oligonucleotides, and deoxynucleotides did not influence the oxygen consumption; and (iii) SSDNA, oligonucleotides, and the deoxynucleotides did not form covalent bonds with 4-CB-2',5'-SQ•- observed by 1H NMR and LC-MS.

In a previous study by Zhao et al. (2004), the covalent product of 4-CB-2',5'-Q with dG was found in an incubation of 4-CB-2',5'-Q in the presence of calf thymus DNA. These inconsistent sets of data could be explained by (i) the different experimental setup and reaction conditions, and/or (ii) non-covalent adducts formed with a high stability comparable with the covalent bond. These non-covalent adducts may have similar behavior to the covalent adducts in chromatographic separation. The latter supports our hypothesis that the 4-CB-2',5'-SQ•- forms a non-covalent adduct by π-stacking or intercalation with the aromatic base pairs of DNA. The in silico LCAO modulations of the complexes between 4-CB-2',5'-SQ•- and dinucleotides were calculated and revealed that there are energy gains from these complexes. The π-stacking interaction results in a delocalization and stabilization of the free electron of the radical and subsequently decreased ΔG that favors the guest-host complex formation.

The biological implications of these persistent guest-host complexes formed by PCB-semiquinones are unknown and cannot be demonstrated clearly at this moment. It is a new insight that stable radicals can be located within the DNA by guest-host complexations; the non-covalent intercalation of the radical may change the DNA structure as well, the implications of which are also unknown. The change of the reactivity of DNA due to the intercalation of SQ•- with long half-lives is a focus of ongoing studies. Aside from the undesired damaging effects of these xenobiotics, this might provide intriguing insights into the role of biogenic SQ•- formed by tyrosinase. Current studies in our laboratory investigate the effect on the epigenetics in the presence of semiquinones on the specific methylations of CpG sites. A better understanding of the interactions of intercalating radicals with the DNA at the molecular level will certainly shed new insights into the carcinogenesis of xenobiotics and may result in novel intercalation-based therapeutics.

Studies with biological compounds containing thiol function, such as GSH and CYS, showed the following: (i) GSH and Cys diminish the EPR signal of 4-CB-2',5'-SQ•-, (ii) 1H NMR and LC-MS measurement confirmed that GSH and Cys form mono- and di-substituted but no tri-substituted analogues, and (iii) GSH and Cys show stochiometry of 1:1:1 with oxygen and 4-CB-2',5'-H2Q in the overall reaction. The oxidation of GSH to glutathione disulfide (GSSG) by H2O2 may play a role in additional consumption of oxygen. The lag phase of oxygen consumption in the presence of GSH also increased with increasing concentration of GSH. We could explain this finding on the basis of the “capturing” of the 4-CB-2',5'-SQ•- formed at relatively low concentrations in the metal-free buffer via the dianion of the hydroquinone. The resulting 4-CB-2',5'-Q readily reacted with GSH and Cys forming the covalent adducts of GS and Cys substituted 4-CB-2',5'-H2Q. By shifting the chemical equilibrium, the concentration of 4-CB-2',5'-Q was too low to form sufficient 4-CB-2',5'-SQ•- by comproportionation with 4-CB-2',5'-H2Q or its GS-substituted. By this, the potential of GSH and Cys as the radical scavengers is more of an indirect nature depending on the chemical equilibrium. The generally accepted role of GSH as radical scavenger in biological systems should be reconsidered based upon these findings.

However, we could not confirm the semiquinone radical reaction with nucleophilic moieties of proteins. No change of the oxygen consumption compared to the control was observed in the presence of the amino acids from all selected groups: Met (non-polar, hydrophobic), Lys and His (polar, basic), and Gly and Cys (polar, uncharged), with the exception of Cys. The EPR investigations, on the other hand, showed a diminished 4-CB-2',5'-SQ•- concentration. We explain that the lower EPR signal intensity of the 4-CB-2',5'-SQ•- corresponds to a higher isoelectric point of the amino acid present in the solution, referring to a lower acidity, and higher concentration of the basic dianion.

The findings presented here could change the scientific view of radical interaction of semiquinones in general with biologically active compounds like DNA, amino acids, and proteins. The role of GSH as radical scavenger in the cell and the influence of persistent co-planar radicals intercalating in the DNA, and their potential effects on replication, catalysis of other reactions, or influence of epigenetic signaling need to be investigated in much more detail in future studies.

Acknowledgments

We would like to thank Brett A. Wagner and Joost van't Erve, the University of Iowa, for their assistance with EPR and oxygraph measurements. The University of Iowa EPR Facility provided invaluable support. We thank Dr. Lynn M. Teesch, and the high-resolution mass spectrometry facility for HPLC-MS support. This publication was made possible by NIH grant P42 ES 013661 and its training core from the National Institute of Environmental Health Sciences (NIEHS), and by The University of Iowa Environmental Health Sciences Research Center, P30 ES 05605. The project was supported by the Tech For Future fund, an initiative of the Saxion and Windesheim Universities of Applied Sciences and the regional government Overijsel, The Netherlands. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the funding agencies.

Abbreviations

4-CB-2',5'-Q

4-Chlorobiphenyl-2',5'-benzoquinone

4-CB-2',5'-H2Q

4-Chlorobiphenyl-2',5'-hydroquinone

4-CB-2',5'-SQ•-

4-Chlorobiphenyl-2',5'-semiquinone radical

AA

Arachidonic acid

DMSO

Dimethyl sulfoxide

EPR

Electron paramagnetic resonance

GSH

Glutathione, reduced form

H2Q

Hydroquinones

hPGHS-2

Human recombinant prostaglandin H synthase-2

KAA

Potassium arachidonate

LCAO

Linear combination of the atomic orbitals

PCBs

Polychlorinated biphenyls

Q

Quinones

ROS

Reactive oxygen species

SOMO

Semioccupied molecular orbital

SQ•-

Semiquinone free radicals

Footnotes

References

  1. Amaro AR, Oakley GG, Bauer U, Spielmann HP, Robertson LW. Metabolic activation of PCBs to quinones: reactivity toward nitrogen and sulfur nucleophiles and influence of superoxide dismutase. Chem Res Toxicol. 1996;9(3):623–629. doi: 10.1021/tx950117e. [DOI] [PubMed] [Google Scholar]
  2. Arif JM, Lehmler HJ, Robertson LW, Gupta RC. Interaction of benzoquinone- and hydroquinone-derivatives of lower chlorinated biphenyls with DNA and nucleotides in vitro. Chem Biol Interact. 2003;142(3):307–316. doi: 10.1016/s0009-2797(02)00141-2. [DOI] [PubMed] [Google Scholar]
  3. Blanksby SJ, Ellison GB. Bond dissociation energies of organic molecules. Acc Chem Res. 2003;36(4):255–263. doi: 10.1021/ar020230d. [DOI] [PubMed] [Google Scholar]
  4. Bolton JL, Thatcher GR. Potential mechanisms of estrogen quinone carcinogenesis. Chem Res Toxicol. 2008;21(1):93–101. doi: 10.1021/tx700191p. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bolton JL, Trush MA, Penning TM, Dryhurst G, Monks TJ. Role of quinones in toxicology. Chem Res Toxicol. 2000;13(3):135–160. doi: 10.1021/tx9902082. [DOI] [PubMed] [Google Scholar]
  6. Buettner GR. The pecking order of free radicals and antioxidants: lipid peroxidation, alpha-tocopherol, and ascorbate. Arch Biochem Biophys. 1993;300(2):535–543. doi: 10.1006/abbi.1993.1074. [DOI] [PubMed] [Google Scholar]
  7. Davis BK, Beach JK, Wade MJ, Klein AK, Hock K. Risk Assessment of Polychlorinated Biphenyls (PCBs) in Indoor Air. 2002. [Google Scholar]
  8. Dwivedy I, Devanesan P, Cremonesi P, Rogan E, Cavalieri E. Synthesis and characterization of estrogen 2,3- and 3,4-quinones. Comparison of DNA adducts formed by the quinones versus horseradish peroxidase-activated catechol estrogens. Chem Res Toxicol. 1992;5(6):828–833. doi: 10.1021/tx00030a016. [DOI] [PubMed] [Google Scholar]
  9. Eling TE, Curtis JF, Harman LS, Mason RP. Oxidation of glutathione to its thiyl free radical metabolite by prostaglandin H synthase. A potential endogenous substrate for the hydroperoxidase. J Biol Chem. 1986;261(11):5023–5028. [PubMed] [Google Scholar]
  10. Flowers-Geary L, Bleczinki W, Harvey RG, Penning TM. Cytotoxicity and mutagenicity of polycyclic aromatic hydrocarbon ortho-quinones produced by dihydrodiol dehydrogenase. Chem Biol Interact. 1996;99(1–3):55–72. doi: 10.1016/0009-2797(95)03660-1. [DOI] [PubMed] [Google Scholar]
  11. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, et al. Gaussian 03. Gaussian, Inc.; Wallingford, CT: 2004. [Google Scholar]
  12. Harman D. Role of free radicals in mutation, cancer, aging, and the maintenance of life. Radiat Res. 1962;16:753–763. [PubMed] [Google Scholar]
  13. Kalyanaraman B, Premovic PI, Sealy RC. Semiquinone anion radicals from addition of amino acids, peptides, and proteins to quinones derived from oxidation of catechols and catecholamines. An ESR spin stabilization study. J Biol Chem. 1987;262(23):11080–11087. [PubMed] [Google Scholar]
  14. Kobayashi T, Narumiya S. Function of prostanoid receptors: studies on knockout mice. Prostaglandins Other Lipid Mediat. 2002;68–69:557–573. doi: 10.1016/s0090-6980(02)00055-2. [DOI] [PubMed] [Google Scholar]
  15. Lehmann L, Esch HL, Kirby PA, Robertson LW, Ludewig G. 4-monochlorobiphenyl (PCB3) induces mutations in the livers of transgenic Fisher 344 rats. Carcinogenesis. 2007;28(2):471–478. doi: 10.1093/carcin/bgl157. [DOI] [PubMed] [Google Scholar]
  16. Li S, Cooper VR, Thonhauser T, Lundqvist BI, Langreth DC. Stacking interactions and DNA intercalation. J Phys Chem B. 2009;113(32):11166–11172. doi: 10.1021/jp905765c. [DOI] [PubMed] [Google Scholar]
  17. Ludewig G, Lehmann L, Esch H, Robertson LW. Metabolic Activation of PCBs to Carcinogens in Vivo - A Review. Environ Toxicol Pharmacol. 2008;25(2):241–246. doi: 10.1016/j.etap.2007.10.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Mill T, Hendry DG, Richardson H. Free-radical oxidants in natural waters. Science. 1980;207(4433):886–887. doi: 10.1126/science.207.4433.886. [DOI] [PubMed] [Google Scholar]
  19. Morrison H, Di Monte D, Nordenskjold M, Jernstrom B. Induction of cell damage by menadione and benzo(a)pyrene-3,6-quinone in cultures of adult rat hepatocytes and human fibroblasts. Toxicol Lett. 1985;28(1):37–47. doi: 10.1016/0378-4274(85)90007-4. [DOI] [PubMed] [Google Scholar]
  20. Myers LS. Free radical damages of nucleic acids and their components. In: Pryor WA, editor. Free Radicals in Biology. IV. Academic Press; New York: 1982. pp. 95–115. [Google Scholar]
  21. Nonhebel DC, Walton JC. Free-radical Chemistry. Cambridge University Press; London: 1974. [Google Scholar]
  22. Oakley GG, Devanaboyina U, Robertson LW, Gupta RC. Oxidative DNA damage induced by activation of polychlorinated biphenyls (PCBs): implications for PCB-induced oxidative stress in breast cancer. Chem Res Toxicol. 1996a;9(8):1285–1292. doi: 10.1021/tx960103o. [DOI] [PubMed] [Google Scholar]
  23. Oakley GG, Robertson LW, Gupta RC. Analysis of polychlorinated biphenyl-DNA adducts by 32P-postlabeling. Carcinogenesis. 1996b;17(1):109–114. doi: 10.1093/carcin/17.1.109. [DOI] [PubMed] [Google Scholar]
  24. O'Brien PJ. Molecular mechanisms of quinone cytotoxicity. Chem Biol Interact. 1991;80(1):1–41. doi: 10.1016/0009-2797(91)90029-7. [DOI] [PubMed] [Google Scholar]
  25. Shen L, et al. Alkylation of 2'-deoxynucleosides and DNA by the Premarin metabolite 4-hydroxyequilenin semiquinone radical. Chem Res Toxicol. 1998;11(2):94–101. doi: 10.1021/tx970181r. [DOI] [PubMed] [Google Scholar]
  26. Song Y, Buettner GR. Thermodynamic and kinetic considerations for the reaction of semiquinone radicals to form superoxide and hydrogen peroxide. Free Radic Biol Med. 2010;49(6):919–962. doi: 10.1016/j.freeradbiomed.2010.05.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Song Y, Wagner BA, Lehmler HJ, Buettner GR. Semiquinone radicals from oxygenated polychlorinated biphenyls: electron paramagnetic resonance studies. Chem Res Toxicol. 2008;21(7):1359–1367. doi: 10.1021/tx8000175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Srinivasan A, Lehmler HJ, Robertson LW, Ludewig G. Production of DNA strand breaks in vitro and reactive oxygen species in vitro and in HL-60 cells by PCB metabolites. Toxicol Sci. 2001;60(1):92–102. doi: 10.1093/toxsci/60.1.92. [DOI] [PubMed] [Google Scholar]
  29. Wangpradit O, et al. Oxidation of 4-chlorobiphenyl metabolites to electrophilic species by prostaglandin H synthase. Chem Res Toxicol. 2009;22(1):64–71. doi: 10.1021/tx800300t. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Wangpradit O, et al. Observation of an unusual electronically distorted semiquinone radical of PCB metabolites in the active site of prostaglandin H synthase-2. Chemosphere. 2010;81(11):1501–1508. doi: 10.1016/j.chemosphere.2010.08.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Zettner MA, et al. Quinoid metabolites of 4-monochlorobiphenyl induce gene mutations in cultured Chinese hamster v79 cells. Toxicol Sci. 2007;100(1):88–98. doi: 10.1093/toxsci/kfm204. [DOI] [PubMed] [Google Scholar]
  32. Zhao S, Narang A, Ding X, Eadon G. Characterization and quantitative analysis of DNA adducts formed from lower chlorinated PCB-derived quinones. Chem Res Toxicol. 2004;17(4):502–511. doi: 10.1021/tx034245b. [DOI] [PubMed] [Google Scholar]

RESOURCES