- Email: [email protected]
In vivo copper-mediated free radical production: an ESR spin-trapping study
Spectrochimica Acta Part A 58 (2002) 1227– 1239 www.elsevier.com/locate/saa
In vivo copper-mediated free radical production: an ESR spin-trapping study Maria B. Kadiiska *, Ronald P. Mason National Institutes of Health, National Institute of En6ironmental Health Sciences, PO Box 12233, Research Triangle Park, NC 27709 -2233, USA Received 1 July 2001; accepted 1 October 2001
Abstract Copper has been suggested to facilitate oxidative tissue injury through a free radical-mediated pathway analogous to the Fenton reaction. By applying the electron spin resonance (ESR) spin-trapping technique, evidence for hydroxyl radical formation in vivo was obtained in rats treated simultaneously with copper and ascorbic acid or paraquat. A secondary radical spin-trapping technique was used in which the hydroxyl radical formed the methyl radical upon reaction with dimethylsulfoxide. The methyl radical was then detected by ESR spectroscopy as its adduct with the spin trap phenyl-N-t-butyl- nitrone (PBN). In contrast, lipid derived radical was detected in vivo in copper-challenged, vitamin E and selenium-deficient rats. These findings support the proposal that dietary selenium and vitamin E can protect against lipid peroxidation and copper toxicity. Since copper excreted into the bile from treated animals is expected to be maintained in the Cu(I) state (by ascorbic acid or glutathione), a chelating agent that would redox-stablilize it in the Cu(I) state was used to prevent ex vivo redox chemistry. Bile samples were collected directly into solutions of bathocuproinedisulfonic acid, a Cu(I)-stabilizing agent, and 2,2%-dipyridyl, a Fe(II)-stabilizing agent. If these precautions were not taken, radical adducts generated ex vivo could be mistaken for radical adducts produced in vivo and excreted into the bile. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Rats; Spin trap; ESR; Free radicals
1. Introduction Although copper is a nutrient essential to health, the pathological processes associated with the various forms of copper overload, demonstrate that the metal can also be toxic. Copper poisoning may be either acute or chronic. Acute * Corresponding author. Tel.: + 1-919-541-0201; fax: +1919-541-1043. E-mail address: [email protected] (M.B. Kadiiska).
copper poisoning due to the accidental ingestion of copper-contaminated materials or water supplied by Cu pipes is the single most frequent cause of ingestion fatalities [1,2]. Chronic copper overload, however, can have several causes, including excessive intake, genetic metabolic diseases, and carcinomas [1,3]. Hypercumia in humans occurs in Wilson’s disease (hepatolenticular degeneration) and Indian childhood cirrhosis [1]. Maldistribution of body copper is observed in Menke’s kinky hair disease [3].
1386-1425/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S1386-1425(01)00713-2
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
1228
More work has been conducted on the hepatic disposition of copper than on that of most other metals, because of the attempts to characterize the molecular basis of these disorders. In the late 1970s the catalytic role of transition metals, particularly iron and copper, was implicated in free radical reactions. In biological systems there may be concentrations of redox-active copper sufficient to catalyze free radical reactions and account for various deleterious processes. It has been shown that traces of soluble copper or iron can catalyze the transformation of a superoxide radical anion (O− 2 ) to the highly reactive hydroxyl radical (OH) via the metal-catalyzed Haber– Weiss reaction: O2− +Mn + M(n − 1) + M = Cu (n=2)
or
Fe (n =3)
M(n − 1) + +H2O2 Mn + +OH− +HO Alternatively, a metal-mediated site-specific mechanism for free radical-induced biological damage has also been proposed [4– 6], where HO− is formed site-specifically in the vicinity of the target molecule (Biol) and reacts at the site of its production. Biol +Mn + Biol – Mn + Biol –Mn + + O2− Biol – M(n − 1) + +O2 Biol –M(n − 1) + + H2O2 (Biol –Mn + …HO) + OH− Other reductants present in cells (such as ascorbate or glutathione) might be able to replace superoxide in the Haber–Weiss cycle and, therefore, promote the toxicity of a metal/hydrogen peroxide system. It has been previously demonstrated that the combination of ascorbate and Cu2 + causes damage to macromolecular structures such as polysaccharides and proteins through generation of reactive oxygen species [7– 10]. Accordingly, copper overload has been shown to be associated with increased generation of lipid peroxidation products and with a number of biochemical and pathological lesions in rats [11,12]. Spin trapping is the most direct method for the detection of highly reactive free radicals in vivo.
With this electron spin resonance (ESR) technique, a higher steady-state concentration of free radicals (as radical adducts) is achieved, which can overcome the sensitivity problem inherent in the detection of endogenous radicals in biological systems [13–16]. Since the concentration of endogenous radicals in biological tissues is generally near the sensitivity limit of ESR spectroscopy, the spin-trapping technique is not limited by background signals. The technique of spin trapping involves the addition of a primary free radical across the double bond of a diamagnetic compound (the spin trap) to form a radical adduct more stable that the primary free radical. This technique involves the indirect detection of primary free radicals that cannot be directly observed by conventional ESR due to low steady-state concentrations or to very short relaxation times, which lead to very broad lines. A more recent approach to in vivo spin trapping has been to examine biological fluids such as urine [17,18], bile [19–27], and blood [28– 31] (or plasma [32,33]) directly for spin adducts. For these studies, the TM110 ESR cavity and a 17 mm flat call can be used to achieve the largest possible aqueous sample size in the active region of the cavity ( 100 Fl). This approach yields both high molar sensitivity and high resolution. In the study of hydroxyl radical formation in vivo, we used the scavenging reaction in which the hydroxyl radical is converted into the methyl radical via its reaction with dimethyl sulfoxide. The methyl radical is then detected as its long-lived phenyl N-t-butylnitrone (PBN) adduct. Alone, DMSO is relatively non-toxic, with a 24-h LD50 in the rat (ip) of 13.7 g/kg, and is therefore an ideal reagent for the in vivo detection of the hydroxyl radical [34]. We usually assay untreated bile for radical adducts. Anesthesia is maintained throughout the experiments, which are initiated by ip injection of DMSO containing PBN, followed by intragastric injection of ferrous sulfate. The resulting PBN/ methyl radical adducts (PBN/CH3) is detected by ESR, because, the in vivo detection of the hydroxyl radical adduct of PBN is extremely unlikely, due to its rapid decay [35].
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
2. Materials and methods
2.1. Chemicals PBN, DMSO, FeSO4 ·7H2O, CuSO4 ·5H2O, and acid (sodium salt) were purchased from Aldrich. DP, 2,2%-biquinoline, and BC were from Sigma. [13C2]DMSO (minimum 99 atom % 13C) was obtained from Isotec Inc. (Miamisburg, OH).
1229
and selenium. All rats were maintained on their respective diets, with access to an unlimited quantity of the diet and water, for 30 days, at which time the rats weighed 1659 30 g. Each dietary group consisted of eight to ten rats.
L-ascorbic
2.2. In 6i6o animal treatment Studies used Sprague– Dawley male rats of 380– 450 g (Charles River Breeding Laboratories, Raleigh, NC). Rats were fed a standard chow mix (NIH Open Formula; Zeigler Brothers, Gardner, PA) and were allowed free access to both food and water. Rats were anesthetized with Nembutal (50 mg/ kg intraperitoneally) and anesthesia was maintained throughout the experiments. Bile ducts were cannulated with a segment of PE 10 tubing. All animals were given an intraperitoneal injection of the spin trap PBN (100 or 250 mg/kg of body weight) dissolved in DMSO (1 ml/kg of body weight) and an intragastric injection of 1 M CuSO4 ·5H2O (3 or 1 ml/kg) dissolved in distilled water. A 1 M solution of ascorbate was administered intragastrically, where indicated, at a dose of 0.5 or 3 ml/kg body weight. Paraquat was injected at a dose of 55 mg/kg intragastrically.
2.3. Selenium- and 6itamin E-deficient diets Twenty-one-day-old rats were fed purified diets (ICN Biochemicals, Cleveland, OH) containing the following vitamin E and selenium concentrations: the vitamin E- and selenium-deficient diet contained no significant amounts of either vitamin E (a-tocopherol) derivatives or selenium; the selenium-deficient diet was the same diet but was supplemented with 121 IU of a-tocopherol/kg of food; the vitamin E-deficient diet was the same diet as the doubly deficient diet as but was supplemented with 0.22 mg of sodium selenite/kg of food; the vitamin E- and selenium-adequate (control) diet was the same diet as the doubly deficient diet but was supplemented with both vitamin E
2.4. In 6itro experiments in bile For in vitro studies, bile from untreated rats was added to tube containing various combinations of PBN, DMSO, DP, BC, and ascorbic acid. Cu(I), Cu(II), or Fe(II) was added last to initiate the reaction. Scans were taken 3 min after mixing.
2.5. In 6i6o bile experiments For ESR measurements, bile samples were thawed and transferred to flat quartz cells within 5 min. All spectra shown are from bile collected 2 h after the administration of CuSO4 and ascorbic acid.
2.6. ESR measurements ESR spectra were obtained using a Varian E109 spectrometer equipped with a TM110 cavity and operating at 9.3 GHz, 20-mW power, and 100-kHz modulation frequency. Spectra were recorded on an IBM-compatible computer interfaced to the spectrometer. Hyperfine coupling constants were determined from a spectral simulation program, SIMEPR. Student’s t test was used for statistical analysis of copper and iron levels from groups of animals. Values of P of B 0.05 were taken to indicated significant differences.
3. Results and discussion
3.1. Hydroxyl radical-mediated toxicity of paraquat and copper in rats Since the Fenton reaction requires hydrogen peroxide, we thought that a substance which catalyzes hydrogen peroxide formation would increase the signal. The activity of the herbicide paraquat (PQ2 + ) is attributed to its ability to catalyze the
1230
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
formation of superoxide and, subsequently, hydrogen peroxide. The proposed mechanism streams from the assumption that copper or iron and PQ2 + undergo cyclic reduction and reoxidation. The enzymatic reduction of PQ2 + to the stable monocation radical (PQ+) has been demonstrated [36]. PQ+ reacts very quickly with O2 to generate the parent PQ2 + and O2− [37], which rapidly dismutates to form O2 and H2O2. The Cu1 + and Fe2 + may subsequently react with the hydrogen peroxide in a Fenton-type reaction to form the highly reactive hydroxyl radical (HO), which would then cause extensive biological damage [4,38]. Therefore, the extreme toxicity of PQ2 + is thought to result from heavy intracellular fluxes of superoxide and hydrogen peroxide in the vicinity of redox-active metal ions, which are available for the promotion of the Fenton reaction, such that normal physiological defense mechanisms are overwhelmed. Considering all the evidence that PQ2 + toxicity is mediated by transition metals, we have investigated the role of Cu2 + in PQ2 + toxicity in vivo using the rat as an animal model. To date, free radical generation due to copper-mediated PQ2 + toxicity in whole animals has not been unequivocally demonstrated. PBN radical adducts were detected by ESR spectroscopy in the bile of rats 2 h after they had been given CuSO4 (3 mmol/kg, ig), PQ2 + (55 mg/kg, ig), and PBN (250 mg/kg, ip) dissolved in DMSO (1 ml/kg) (Fig. 1A). When the experiment was repeated in the absence of Cu2 + (Fig. 1B), PQ2 + (Fig. 1C), or both Cu2 + and PQ2 + (Fig. 1D), radical adducts were not detected, thereby, confirming the dependence of radical formation in the coadministration of both Cu2 + and PQ2 + . The signals in Fig. 1B– D are believed to be from a mixture of radical adducts, but were too weak to characterize [27]. The fact that Cu2 + or PQ2 + alone caused little detectable radical adduct formation may be attributed to their inability to form hydroxyl radicals at detectable concentrations due to strong defense systems against oxidative stress in living organisms. Minimum concentrations of 2 mmol/kg CuSO4 and 27.5 mg/kg PQ2 + given together were necessary for ESR-detectable quantities of radical ad-
ducts to occur in the bile. The spectrum shown in Fig. 1E demonstrates that the radical species observed in the bile was also PBN-dependent. (The doublet is from the ascorbate radical formed from the oxidation of endogenous ascorbate). To prevent the occurrence of reactions between copper or iron ions, O2, PBN, DMSO, and bile components during sample collection, bile must be collected directly into a solution of the Cu1 + stabilizing agent BC and the Fe2 + -stabilizing agent DP [27]. In order to identify the radical species detected in bile of animals which were treated with Cu2 + , PQ2 + , PBN, and DMSO, the experiment shown
Fig. 1. (A) The ESR spectrum of radical adducts detected in the bile of rats 2 h after CuSO4 (3 mmol/kg) and PQ2 + (55 mg/kg) were administered intragastrically, and PBN (250 mg/ kg) dissolved in DMSO (1 ml/kg) was administered intraperitoneally. Bile samples were collected into 25 ml of a solution containing 300 mM BC and 30 mM DP per 100 g body wt. (B) Same as in (A), but rats were not given CuSO4. (C) Same as in (A), but rates were not given PQ2 + . (D) Same as in (A), but rats were not given Cu(II) or PQ2 + . (E) Same as in (A), but rats were not given PBN and DMSO. Spectrometer settings: modulation amplitude, 1.3 G; microwave power, 20 mW; scan time, 16 min; time constant, 1 s.
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
in Fig. 1A was repeated using 13C-labeled DMSO ([13C]2-DMSO). The presence of hyperfine splittings in the ESR spectrum from 13 C(I =1/2) would allow an unambiguous assignment of the PBN/CH3 radical adduct formed in vivo. In addition, the use of deuterated PBN (d14-PBN) would significantly reduce inhomogenous broadening of the ESR signal from PBN protons and thereby, enhance spectral resolution. The spectrum obtained from this experiment is shown in Fig. 2A, and the complete spectral simulation is shown in Fig. 2B. The four radical species used to calculate the spectrum in Fig. 2B are shown in Fig. 2C– F, and the hyperfine coupling constants are given in Table 1. Only the assignments of the PBN/ CH3 and ascorbate radical species, however, can be considered unique. The appearance of 13C-hyperfine splittings, observed in the wings of each of the lines of the nitroxide triplet (Fig. 2A, cf. Fig. 1A) is proof that the PBN/CH3 radical adduct was formed in vivo during PQ2 + and Cu2 + intoxication. The other two PBN adducts (Fig. 2D and E) remain unidentified, but are comparable with non-DMSO-dependent PBN radical species with similar hyperfine coupling constants detected in bile of rats which were administered Fe2 + and PQ2 + [39]. In that study, the radical comparable with that in Fig. 2D was tentatively assigned to a carbonyl radical adduct of PBN based on its hyperfine coupling constants. The radical species, which gave, rise to the spectrum in Fig. 2E must be the result of complex PBN chemistry [39]. The analysis of radical adducts excreted via the biliary route provide strong ESR evidence for the generation of the hydroxyl radical as a result of the known futile enzymatic redox cycling of PQ2 + , with copper playing an essential mediatory role. No radical adducts were detected when either CuSO4 or PQ2 + was excluded. From a different perspective, in vivo copper-dependent hydroxyl radical generation could be said to be promoted by PQ2 + . This is the first report of ESR evidence for this synergetic hydroxyl radical generation by copper and PQ2 + in a whole animal.
1231
Fig. 2. (A) The ESR spectrum of radical adducts detected in the bile of rats 2 h after CuSO4 (3 mmol/kg) and PQ2 + (55 mg/kg) were administered intragastrically, and d14-PBN (250 mg/kg) dissolved in [13C]2-DMSO (1 ml/kg) was administered intraperitoneally. Bile samples were collected into a 25 ml of a solution containing 300 mM BC and 30 mM DP per 100 g body wt. Spectrometer settings: modulation amplitude, 0.67 G; microwave power, 20 mW; scan time, 60 min; time constant, 4 s. (B) Complete computer simulation of the spectrum in (A). (C) Simulation of the relative contribution of the PBN/13CH3 radical adduct to the spectrum in (A). (D) Simulation of the relative contribution of the unidentified PBN/C radical adduct to the spectrum in (A). (E) Simulation of the relative contribution of the PBN/Y radical adduct to the spectrum in (A). (F) Simulation of the relative contribution of the ascorbate radical (Asc−) to the spectrum in (A). The hyperfine coupling constants obtained from these simulations are given in Table 1.
3.2. Hydroxyl radical formation after acute copper and ascorbic acid intake To produce the highly reactive hydroxyl radical, Cu2 + must be reduced to Cu1 + . It is assumed that Cu1 + is oxidized by H2O2 to yield HO, although, the reaction between Cu1 + and H2O2 may not be a simple one-electron oxida-
1232
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
tion of Cu1 + to Cu2 + with concomitant formation of the free hydroxyl radical [5,40– 42]. We found that, in ascorbic acid-treated rats, hydroxyl radical (or a closely related species such as a high-valence metal species) is generated during acute copper overload. The suggested mechanism involves the oxidation of ascorbate to the ascorbate radical, the reduction of Cu2 + (either spuriously protein bound or non-protein bound) to Cu1 + , and the production of superoxide, hydrogen peroxide, and hydroxal radical.
were not able to show in vivo hydroxyl radical generation from rats treated only with CuSO4 and PBN/DMSO (Fig. 3D). Since copper excreted into the bile from untreated animals is expected to be maintained in the Cu1 + state (by ascorbic acid, glutathione, or other biological reductants), the chelating agent BC, which stabilizes copper in the Cu1 + state, was used. A chelating agent for redox-stabilizing Fe2 + , DP, was also used. We found that, if these precautions were not taken, signals from radical adduct generates ex vivo might be mistaken for the adducts generated in vivo and excreted into the bile. Having demonstrated in vivo copper- and ascorbic acid-dependent free radical formation, we next sought to identify the radical adducts detected. The incorporation of a magnetic nucleus (typically 13C) into a radical adduct, can permit identification of a variety of radical trapped in complex biological systems [16]. Using this approach, experiments were performed with 13Csubstituted DMSO, which were expected to confirm the formation of PBN/CH3 in vivo. The ESR signal detected in the bile of rats given an intraperitoneal dose of [13C2]DMSO/PBN and an intragastric dose of CuSO4 and ascorbic acid is shown in Fig. 4A. This spectrum clearly contains a signal from the PBN/CH3 radical adduct, which is the dominant species and is characterized by the hyperfine coupling constants of a N = 16.30 C 13 G, a H b = 3.63 G, and a b ( C) =3.81 G (Fig. 4B).
Ascorbate+Cu2 + Ascorbate radical+Cu+ +H+ Cu1 + +O2 Cu1 + + O2− 2O2− H2O2 +O2 H2O2 +Cu1 + Cu2 + +OH− +HO After an intragstric dose of CuSO4 and ascorbic acid and intraperitoneal administration of PBN in DMSO to rats, the ESR analysis of bile samples collected into aqueous solutions of BC and DP gave six-line spectrum (Fig. 3A). The formation of this radical adduct was dependent on the administration of CuSO4 (Fig. 3B), PBN/DMSO (Fig. 3C), and ascorbate (Fig. 3D) and was apparently due to the in vivo formation of the PBN-methyl adduct, PBN/CH3 from hydroxyl radical. The ESR spectrum of bile from untreated rats showed only the doublet of the ascorbate radical, which was variable in intensity (data not shown). We
Table 1 Hyperfine coupling constants obtained from the simulation of Fig. 3A Radical detected
PBN/13CH3 PBN/13CH3 PBN/C PBN/COR PBN/Y PBN/Y Asc−
Hyperfine coupling constants (G) aN
aN b
a 13C b
16.30 16.31 15.76 15.61 15.30 15.20
3.21 3.06 3.68 3.36
4.39 4.00
a N/a H b
Relative concentration
Source
5.08 5.33 4.28 4.66 12.45 12.67
23
Fig. [39] Fig. [39] Fig. [39] Fig.
aH
1.23 1.20 1.81
32 43 2
3C 3D 3E 3F
Note: the four radical species used to simulate the experimental spectrum are shown in Fig. 3C–F and are presented at the relative concentrations determined from the composite spectrum in Fig. 3B. The radical assignments listed above are discussed in the text.
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
1233
Clearly in our system of copper-intoxicated rats, copper alone does not cause the formation of detectable PBN/CH3, unlike what had been demonstrated for iron [22]. Either the binding of copper to its physiological ligands prevents it from catalyzing OH formation [42,46,47] or OH participates in site-specific reactions of the bound copper and does not have the opportunity to react with DMSO to form the methyl radical in detectable concentrations [6,40,44,45]. Free radical production from copper in vivo, however, can be stimulated by sufficient quantities of ascorbic acid to produce radical adduct similar to those observed from iron overload. Although the antagonistic effects due to ascorbic acid supplementation include inhibition of copper absorption, signs of copper deficiency [48], and lowered tissue copper and ceruloplasmin lev-
Fig. 3. (A) ESR spectrum of radical adducts detected in the bile from rats 2 h after CuSO4 (1 mmol/kg) and ascorbic acid (Asc) (0.5 mmol/kg) were administered intragastrically and PBN (100 mg/kg) dissolved in DMSO (1 ml/kg) was administered intraperitoneally. Bile samples were collected into a 25 ml of a solution containing 300 mM BC and 30 mM DP per 100 g of body weight. (B) Same as in (A), but rats were not given CuSO4; (C) same as in (A), but rats were not given PBN/ DMSO; (D) same as in (A), but rats were not given ascorbic acid. Spectrometer settings: modulation amplitude, 1.3 G; microwave power, 20 mW; scan time, 16 min; time constant, 1 s.
The results in this study support a free radical mechanism that attributes an important role to ascorbic acid and copper in vivo in the induction of biological damage already demonstrated in numerous in vitro experiments [4,7,9,43– 45]. From our studies here it is clear that generation of detectable OH during acute copper poisoning of rats is dependent on high intragastric concentrations of ascorbic acid. Conceivably, radical production may occur in the digestive tract, where the spin-trapped adduct is absorbed and transported directly to the liver via the portal vein. However, we could not detect any free radical in blood samples drawn from the portal vein (data not shown).
Fig. 4. (A) ESR spectrum of radical adducts detected in the bile from rats 2 h after CuSO4 (3 mmol/kg, intragastrically), ascorbic acid (0.5 mmol/kg, intragastrically) and PBN (250 mg/kg) dissolved in [13C2]DMSO (1 ml/kg, intraperitoneally) were administered. (B) Computer simulation of spectrum in (A). Hyperfine coupling constants are given in Table 1.
1234
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
els [49,50], the cytotoxicity of copper and ascorbate was attributed to OH radicals repeatedly formed by recycling Cu2 + back to Cu1 + [43,44]. However, under the same experimental conditions, investigators have noted both pro- and antioxidant effects of ascorbate, respectively, [51,52]. Since ascorbic acid is the single most heavily consumed nutrient supplement, it is important that more information be gathered on the effects of ascorbic acid intake on copper bioavailability and utilization. In the present study we show in vivo formation of hydroxyl radicals produced in rats during acute copper overload and ascorbic acid intake. We used a secondary trapping technique [22] in which the hydroxyl radical forms the methyl radical upon reaction with DMSO. The methyl radical is then detected by ESR spectroscopy as its adduct with the spin trap PBN. Finally we showed that, if bile samples from copper-treated rats were not collected directly into solutions of chelating agents that would redoxstabilize copper and traces of iron, signals from radicals adducts generated ex vivo might be mistaken for adducts generated in vivo and then excreted into the bile. In summary, the findings presented here indicate that radicals detected in the bile are formed in vivo due to the combination of acute copper and ascorbic acid intake.
3.3. Lipid radical generation in copper-treated 6itamin E- and selenium-deficient rats Free radical-induced peroxidative damage to membrane lipids has long been regarded as a critical initiating event leading to cell injury [53– 57]. It is generally accepted that susceptibility to lipid peroxidation is influenced by the level of vitamin E and selenium in tissue [58–61]. Depletion of dietary vitamin E and selenium, for example, has been shown to be linked to the increased formation of lipid peroxidation products and increased hepatic peroxidative injury in different species, including rats [62– 64]. Therefore, the tissues of animals deficient in vitamin E and selenium have a greater tendency to undergo lipid peroxidation both in vitro [65] and in vivo [66]
than do tissues of animals with a diet adequate in these nutrients. Other studies also showed that rats whose diets were deficient in vitamin E and selenium were much more vulnerable to copper toxicity. Dougherty and Hoekstra [67] found that intraperitoneal administration of CuSO4 to rats deficient in both vitamin E and selenium caused dose-dependent increases in expired ethane (an indirect index of lipid peroxidation) and acute mortality. Selenium and vitamin E supplementation of the diet prevented the increase of ethane production and mortality. Similarly, Dillard and Tappel [12] found increased pentane production and other products of lipid peroxidation in copper-treated, vitamin E-deficient rats. Smaller increases of the same lipid peroxidation products were found in copper-treated rats fed a vitamin E-supplemented diet. Despite all the evidence for free radical-mediated lipid peroxidation from copper intoxication, the free radicals themselves have not been detected in vivo. Since it has been observed that the toxicity of copper is considerably higher in vitamin E- and/or selenium-deficient rats, the present experiments were designed to determine whether copper overload in vivo promotes radical generation in vitamin E- and/or selenium-deficient rats and whether the radical species generated are the same as in our previous findings. Fig. 5A shows the radical adduct signal in bile from rats fed for 10 days with a vitamin E- and selenium-deficient diet, as measured 2 h after the rats were challenged with CuSO4 (2 mmol/kg, intragastrically). A solution of the spin trap PBN (250 mg/kg) in DMSO (1 ml/kg, intraperitoneally) was also administered. The intensity of the signals was dose dependent with respect to both CuSO4 and PBN (data not shown). In the absence of copper (Fig. 5B) or PBN (Fig. 5C), no significant amount of radical adduct was detected. The small doublet observed is due to the ascorbate radical ion (a H = 1.7 G) formed from the one-electron oxidation of endogenous ascorbate. When the experiment was repeated with control rats fed a vitamin E- and selenium-deficient diet, only a residual PBN radical adduct signal was detectable in either copper-treated (Fig. 5D) or nontreated (Fig. 5E) rats.
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
1235
tutions and careful comparisons with ESR parameters of known PBN adducts. When [13C]DMSO was used instead of 12 [ C]DMSO, no additional hyperfine splitting from 13C (I=1/2) was observed (Fig. 7D), again indicating that PBN/CH3 was not present. Values of a N = 15.40 G and a H b = 2.54 G, with a 100% Lorentzian lineshape, were obtained from the simulation shown in Fig. 1A. These results indicated that the radical species trapped by PBN in vivo in copper-challenged, vitamin E- and selenium-deficient rats was formed probably from an endoge-
Fig. 5. ESR detection of a free radical formed in vivo and spin-trapped with PBN, from a copper-challenged rat deficient in both vitamin E and selenium. All spectra are from bile samples collected 2 h after administration of copper. (A) A vitamin E- and selenium-deficient rat was administered 2 mmol/kg CuSO4, intragastrically, and 250 mg/kg PBN as a DMSO solution (1 ml/kg), intraperitoneally. (B) As in (A), but without administration of CuSO4. (C) As in (A), but without administration of PBN. (D) As in (A), but the experiment was done with a rat fed a vitamin E- and selenium-adequate diet as in control. (E) As in (D), but without administration of CuSO4. Instrumental conditions: microwave power, 20 mW; modulation amplitude, 1.33 G; time constant, 1 s; scan rate, 5 G/min.
To investigate further the individual roles of vitamin E and selenium in protecting against radical generation in copper-challenged rats, additional experiments were carried out. If rats were fed diets adequate in vitamin E and/or selenium, little or no radical adduct could be detected (Fig. 6). Although a positive identification of radical adducts formed in vivo is very difficult, tentative assignments can be made through isotope substi-
Fig. 6. ESR detection of a free radical formed in vivo and spin-trapped with PBN in copper-challenged rats with various combinations of vitamin E and selenium deficiencies. All spectra are from bile samples collected 2 h after administration of copper. (A) A vitamin E- and selenium-deficient rat was administered 2 mmol/kg CuSO4, intragastrically, and 250 mg/ kg PBN as a DMSO solution (1 ml/kg), intraperitoneally. (B) As in (A), but with a vitamin E-deficient rat. (C) As in (A), but with a selenium-deficient rat. (D) As in (A), but the experiment was done with a rat fed a vitamin E- and selenium-adequate diet as a control. Instrumental conditions: microwave power, 20 mW; modulation amplitude, 1.33 G; time constant, 1 s; scan rate, 5 G/min.
1236
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
selenium- and vitamin E-deficient rats [12,67–69]. Therefore, the mechanism of copper toxicity apparently involves lipid peroxidation. According to the current hypothesis, transition metal ions are involved in lipid peroxidation by decomposing lipid peroxides into their peroxyl and alkoxyl radicals, which can then abstract hydrogen and perpetuate a chain reaction of lipid peroxidation [70,71]: LOOH +Mn + LO + M(n + 1) + OH− LOOH +M(n + 1) LOO + Mn + + H+ LO +LH LOH + L LOO + LH LOOH +L L +O2 LOO
Fig. 7. Effect of isotope substitutions on the ESR spectra of the PBN radical adduct formed in vivo in copper-challenged rats deficient in both vitamin E and selenium. Spectra are from bile samples collected 2 h after administration of copper. (A) A vitamin E- and selenium-deficient rat was administered 2 mmol/kg CuSO4, intragastrically, and 250 mg/kg d14-PBN in a DMSO solution (1 ml/kg), intraperitoneally. (B) Computer simulation of (A) with a N = 15.36 G, aH b = 2.50 G, and a 90% Lorentizian lineshape. (C) Residual of (A) minus (B), indicating the precision of the simulated spectrum. (D) A vitamin Eand selenium-deficient rat was administered 2 mmol/kg CuSO4, intragastrically, and 250 mg/kg d14-PBN in a [13C]DMSO solution (1 ml/kg), intraperitoneally. Additional carbon-12 (I =1/2) hyperfine splittings are not observed. (E) Computer simulation of (D) with a N = 15.40 G, aH b = 2.54 G, and a 100% Lorentzian lineshape. (F) Residual of (D) minus (E), indicating the precision of the simulated spectrum. Instrumental conditions: microwave power, 20 mW; modulation amplitude, 0.67 G; time constant, 4 s; scan rate, 1.33 G/min.
nous molecule such as a poly-unsaturated fatty acid. It is known from other studies that Cu2 + increases the lipid peroxidation products formed in
where L represents a lipid molecule and M represents a redox-active metal ion. Although no direct evidence has been presented for the generation of lipid-derived radicals in our animal model, we provide conclusive evidence that the radical adduct is formed from an endogenous molecule. Had the generation of hydroxyl radical been the primary toxic action of the copper in these experiments, then the DMSO-dependent PBN/CH3 adduct should have been detected [27]. From the hyperfine splitting constants obtained for the radical adduct shown in Fig. 1A and A and the lack of a carbon-13 isotope splitting when [13C]DMSO was used, it is apparent that the radical adduct is not a product of hydroxyl radical scavenging by DMSO. It should be noted, however, that an inability to detect radical adducts in bile by ESR is not proof that it is not formed in vivo. A number of factors, such as the relative rates of radical adduct formation, destruction, and excretion, can all affect the detection of a radical adduct. The fact that the radical adduct could be detected only in rats that were deficient in both vitamin E and selenium strongly suggests that its formation is associated with lipid peroxidation. In this respect, the primary role of vitamin E in preventing free radical-initiated peroxidative tissue damage is widely accepted, and the functional interrelation between vitamin E and selenium has
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
been recognized [72]. In a variety of experimental systems, studies have shown that selenium may complement the antioxidant function of vitamin E via selenoenzymes [73,74]. Indeed, in this role they are a part of a complex system of antioxidant protection in vivo that includes free radical scavengers and several families of enzymes. In conclusion, the present study provides the most direct ESR evidence yet of lipid peroxidation-associated free radical formation in the toxicity of copper.
4. Conclusions
1. The study provided strong ESR evidence for the generation of hydroxyl radical as a result of the futile enzymatic redox cycling of PQ2 + , with Cu playing an essential mediatory role. From a different perspective, in vivo Cu2 + -dependent hydroxyl radical generation could be said to be promoted by PQ2 + . This is the first report of ESR evidence for the synergetic radical generation by Cu and PQ2 + . 2. Cu2 + alone does not cause the formation of detectable radicals unlike what had been demonstrated for iron. However, free radical production from Cu in vivo can be stimulated by vitamin C. This is the first study to show ESR evidence for pro-oxidant activity of vitamin C. 3. The study provided the most direct ESR evidence yet of LPO-associated free radical formation in the toxicity of Cu. The data that Cu causes no radical generation in rats fed vitamin E- and Se-supplemented diet suggest that these nutrients can protect against lipid peroxidation and use copper toxicity.
References [1] I.H. Scheinberg, I. Sternleib, Copper toxicity and Wilson’s disease, in: A.S. Prasad (Ed.), Trace Elements in Human Health and Disease, Academic Press, New York, 1976, pp. 415 – 438. [2] E.D. Harris, Copper in human and animal health, in: J. Rose (Ed.), Trace Elements in Health, Butterworths, London, 1983, pp. 44 – 73.
1237
[3] M.J. Ettinger, H.M. Darwish, R.C. Schmitt, Mechanism of copper transport from plasma to hepatocytes, Fed. Proc. 45 (1986) 2800 – 2804. [4] M. Chevion, A site specific mechanism for free radical induced biological damage: the essential role of redox active transition metals, Free Radic. Biol. Med. 5 (1988) 27 – 37. [5] H.C. Sutton, C.C. Winterbourn, On the participation of higher oxidation states of iron and copper in Fenton reactions, Free Radic. Biol. Med. 6 (1989) 53 – 60. [6] S. Goldstein, G. Czapski, Transition metal ions and oxygen radicals, Int. Rev. Exp. Pathol. 31 (1990) 133 – 164. [7] S.-H. Chiou, DNA- and protein-scission activities of ascorbate in the presence of copper ion and a copper – peptide complex, J. Biol. Chem. 94 (1983) 1259 – 1267. [8] K. Uchida, S. Kawakishi, Oxidative depolymerization of polysaccharides induced by the ascorbic acid – copper ion systems, Agric. Biol. Chem. 50 (1986) 2579 – 2583. [9] E. Shinar, T. Navok, M. Chevion, The analogous mechanisms of enzymatic inactivation induced by ascorbate and superoxide in the presence of copper, J. Biol. Chem. 258 (1983) 14778 – 14783. [10] G. Marx, M. Chevion, Site-specific modification of albumin by free radicals: reaction with copper(II) and ascorbate, Biochem. J. 236 (1985) 397 – 400. [11] S.D. Aust, L.A. Morehouse, C.E. Thomes, Role of metals in oxygen radical reactions, J. Free Radic. Biol. Med. 1 (1985) 3 – 25. [12] C.J. Dillard, A.L. Tappel, Lipid peroxidation and copper toxicity in rats, Drug Chem. Toxicol. 7 (1984) 477 – 487. [13] G.R. Buettner, R.P. Mason, Spin-trapping methods for detecting superoxide and hydroxyl free radicals in vitro and in vivo, Methods Enzymol. 186 (1990) 127 – 133. [14] E.G. Janzen, R.A. Towner, D.L. Haire, Detection of free radicals generated from the in vitro metabolism of carbon tetrachloride using improved ESR spin trapping techniques, Free Radic. Res. Commun. 3 (1987) 357 – 364. [15] K.T. Knecht, R.P. Mason, In vivo spin trapping of xenobiotic free radical metabolites, Arch. Biochem. Biophys. 303 (1990) 185 – 194. [16] C. Mottley, R.P. Mason, Nitroxide radical adducts in biology: chemistry, applications and pitfalls, in: L.J. Berliner, J. Reuben (Eds.), Biological Magnetic Resonance, vol. 8, Plenum Press, New York, 1989, pp. 489 – 546. [17] H.D. Connor, R.G. Thurman, M.D. Galizi, R.P. Mason, The formation of a novel free radical metabolite from CCl4 in the perfused rat liver and in vivo, J. Biol. Chem. 261 (1986) 4542 – 4548. [18] L.B. LaCagnin, H.D. Connor, R.P. Mason, R.G. Thurman, The carbon dioxide anion radical adduct in the perfused rat liver: relationship to halocarbon-induced toxicity, Mol. Pharmacol. 33 (1988) 351 – 357. [19] K.T. Knecht, R.P. Mason, In vivo radical trapping and biliary secretion of radical adducts of carbon tetrachloride-derived free radical metabolites, Drug Metab. Dispos. 16 (1988) 813 – 817.
1238
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
[20] K.T. Knecht, B.U. Bradford, R.P. Mason, R.G. Thurman, In vivo formation of a free radical metabolite of ethanol, Mol. Pharmacol. 38 (1990) 26 –30. [21] K.T. Knecht, R.P. Mason, The detection of halocarbonderived radical adducts in bile and liver of rats, Drug Metab. Dispos. 19 (1991) 325 –331. [22] M.J. Burkitt, R.P. Mason, Direct evidence for in vivo hydroxyl-radical generation in experimental iron overload: an ESR spin-trapping investigation, Proc. Natl. Acad. Sci. USA 88 (1991) 8440 –8444. [23] H.M. Hughes, I.M. George, J.C. Evans, C.C. Rowlands, G.M. Powell, C.G. Curtis, The role of the liver in the production of free radicals during halothane anaesthesia in the rat. Quantification of N-tert-butyl-alpha-(4- nitrophenyl)nitrone (PBN)-trapped adducts in bile from halothane as compared with carbon tetrachloride, Biochem. J. 277 (1991) 795 –800. [24] K.T. Knecht, J.A. DeGray, R.P. Mason, Free radical metabolism of halothane in vivo: radical adducts detected in bile, Mol. Pharmacol. 41 (1992) 943 –949. [25] M. Sentjurc, R.P. Mason, Inhibition of radical adduct reduction and reoxidation of the corresponding hydroxylamines in in vivo spin trapping of carbon tetrachloridederived radicals, Free Radic. Biol. Med. 13 (1992) 151 – 160. [26] W. Chamulitrat, S.J. Jordan, R.P. Mason, Fatty acid radical formation in rats administered oxidized fatty acids: in vivo spin trapping investigation, Arch. Biochem. Biophys. 299 (1992) 361 –367. [27] M.B. Kadiiska, P.M. Hanna, L. Hernandez, R.P. Mason, In vivo evidence of hydroxyl radical formation after acute copper and ascorbic acid intake: electron spin resonance spin-trapping investigation, Mol. Pharmacol. 42 (1992) 723 – 729. [28] K.R. Maples, S.J. Jordan, R.P. Mason, In vivo rat hemoglobin thiyl free radical formation following phenylhydrazine administration, Mol. Pharmacol. 33 (1998a) 344 – 350. [29] K.R. Maples, S.J. Jordan, R.P. Mason, In vivo rat hemoglobin thiyl free radical formation following administration of phenylhydrazine and hydrazine-based drugs, Drug Metab. Dispos. 16 (1988b) 799 – 803. [30] K.R. Maples, P. Eyer, R.P. Mason, Aniline-, phenylhydroxylamine-, nitrosobenzene-, and nitrobenzene-induced hemoglobin thiyl free radical formation in vivo and in vitro, Mol. Pharmacol. 37 (1990a) 311 –318. [31] K.R. Maples, C.H. Kennedy, S.J. Jordan, R.P. Mason, In vivo thiyl free radical formation from hemoglobin following administration of hydroperoxides, Arch. Biochem. Biophys. 277 (1990b) 402 –409. [32] L.A. Reinke, E.G. Janzen, Detection of spin adducts in blood after administration of carbon tetrachloride to rats, Chem. Biol. Interact. 78 (1991) 155 –165. [33] L.A. Reinke, R.A. Towner, E.G. Janzen, Spin trapping of free radical metabolites of carbon tetrachloride in vitro and in vivo: effect of acute ethanol administration, Toxicol. Appl. Pharmacol. 112 (1992) 17 –23.
[34] R.P. Mason, P.M. Hanna, M.J. Burkitt, M.B. Kadiiska, Detection of oxygen-derived radicals in biological systems using electron spin resonance, Environ. Health Perspect. 102 (Suppl. 10) (1994) 33 – 36. [35] Y. Kotake, E.G. Janzen, Decay and fate of the hydroxyl radical adduct of Ö-phenyl-N-tert-butylnitrone in aqueous media, J. Am. Chem. Soc. 113 (1991) 9503 – 9506. [36] R.P. Mason, J.L. Holtzman, The role of catalytic superoxide formation in the O2 inhibition of nitroreductase, Biochem. Biophys. Res. Commun. 67 (1975) 1267 – 1274. [37] J.A. Farrington, M. Ebert, E.J. Land, K. Fletcher, Bipyridylium quaternary salts and related compounds. V. Pulse radiolysis studies of the reaction of paraquat radical with oxygen. Implications for the mode of action of bipyridyl herbicides, Biochim. Biophys. Acta 314 (1973) 372 – 381. [38] H. Zer, J.H. Freedman, J. Peisach, M. Chevion, Inverse correlation between resistance towards copper and towards the redox-cycling compound paraquat: a study in copper-tolerant hepatocytes in tissue culture, Free Radic. Biol. Med. 11 (1991) 9 – 16. [39] M.J. Burkitt, M.B. Kadiiska, P.M. Hanna, S.J. Jordan, R.P. Mason, Electron spin resonance spin-trapping investigation into the effects of paraquat and desferrioxamine on hydroxyl radical generation during acute iron poisoning, Mol. Pharmacol. 43 (1993) 257 – 263. [40] A. Samuni, M. Chevion, G. Czapski, Unusual copper-induced sensitization of the biological damage due to superoxide radicals, J. Biol. Chem. 256 (1981) 12632 – 12635. [41] S. Goldstein, G. Czapski, Mechanism and reaction products of the oxidation of Cu1 + -phenantholine by H2O2, Free Radic. Biol. Med. 1 (1985) 373 – 380. [42] P.M. Hanna, R.P. Mason, Direct evidence for inhibition of free radical formation from Cu1 + and hydrogen peroxide by glutathione and other potential ligands using the EPR spin-trapping technique, Arch. Biochem. Biophys. 295 (1992) 205 – 213. [43] R. Stoewe, W.A. Prutz, Copper-catalyzed DNA damage by ascorbate and hydrogen peroxide: kinetics and yield, J. Free Radic. Biol. Med. 3 (1987) 97 – 105. [44] A. Samuni, J. Aronovtich, D. Godinger, M. Chevion, G. Czapski, On the cytotoxicity of vitamin C and metal ions: a site specific Fenton mechanism, Eur. J. Biochem. 137 (1983) 119 – 124. [45] D.A. Rowley, B. Halliwell, Superoxide-dependent and ascorbate-dependent formation of hydroxyl radicals in the presence of copper salts: a physiologically significant reaction, Arch. Biochem. Biophys. 225 (1983) 279 – 284. [46] M. Nakamura, I. Yamazaki, One-electron transfer reactions in biochemical systems. VI. Changes in electron transfer mechanism of lipoamide dehydrogenase by modification of sulfhydryl groups, Biochim. Biophys. Acta 267 (1972) 249 – 257. [47] J.H. Freedman, M.R. Ciriolo, J. Peisach, The role of glutathione in copper metabolism and toxicity, J. Biol. Chem. 264 (1989) 5598 – 5605.
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239 [48] C.E. Hunt, W.W. Carlton, P.M. Newberne, Interrelationships between copper deficiency and dietary ascorbic acid in the rabbit, Br. J. Nutr. 24 (1970) 61 –69. [49] D.B. Milne, S.T. Omaye, W.H. Amos, Effect of ascorbic acid on copper and cholesterol in adult cynomolgus monkeys fed a diet marginal in copper, Am. J. Clin. Nutr. 34 (1981) 2389 –2393. [50] C.H. Smith, W.R. Bidlack, Interrelationship of dietary ascorbic acid and iron on tissue distribution of ascorbic acid, iron and copper in female guinea pigs, J. Nutr. 110 (1980) 1398 – 1408. [51] A. Bendich, L.J. Machlin, O. Scandurra, G.W. Burton, D.D. Wayner, The antioxidant role of vitamin C, Adv. Free Radic. Biol. Med. 2 (1986) 419 –444. [52] D.C. Laudicina, L.J. Marnet, Enhancement of hydrogenperoxide-dependent lipid peroxidation in rat liver microsomes by ascorbic acid, Arch. Biochem. Biophys. 278 (1990) 73 – 80. [53] A.L. Tappel, Vitamin E and free radical peroxidation of lipids, Ann. New York Acad. Sci. 203 (1972) 12 –28. [54] J.F. Mead, Free radical mechanisms of lipid damage and consequences of cellular membranes, Free Radic. Biol. Med. 1 (1976) 51 –68. [55] R.O. Recknagel, E.A. Glende, R.L. Walker, K. Lowrey, Lipid peroxidation: biochemistry, measurement, and significance in liver cell injury, in: G. Plaa, W.R. Hweiit (Eds.), Toxicology of the Liver, Raven, New York, 1982, pp. 213 – 241. [56] A. Benedetti, M. Comporti, Formation, reactions, and toxicity of aldehydes produced in the course of lipid peroxidation in cell membranes, Bioelectrochem. Bioenerg. 18 (1987) 187 – 202. [57] F.R. Ungemach, Pathobiochemical mechanisms of hepatocellular damage following lipid peroxidation, Chem. Phys. Lipids 45 (1987) 171 –205. [58] G.F. Combs Jr., T. Noguchi, M.L. Scott, Mechanisms of action of selenium and vitamin E in protection of biological membranes, Fed. Proc. 34 (1975) 2090 – 2095. [59] A.L. Tappel, Vitamin E and selenium protection from in vivo lipid peroxidation, Ann. New York Acad. Sci. 355 (1980) 18 – 31. [60] A.K. De, R. Darad, Physiological antioxidants and antioxidative enzymes in vitamin E-deficient rats, Toxicol. Lett. 44 (1988) 47 – 54.
1239
[61] W.H. Saypil, H. Shimaski, N. Ueta, Free radical-induced liver injury. I. Effects of dietary vitamin E deficiency on triaglycerol level and its fatty acid profile in rat liver, Free Radic. Res. Commun. 14 (1991) 315 – 322. [62] C.K. Chow, C.J. Chen, Dietary selenium and age-related susceptibility of rat erythrocytes to oxidative damage, J. Nutr. 110 (1980) 2460 – 2466. [63] C.K. Chow, Effect of dietary vitamin E and selenium on rats: pyruvate kinase, glutathione peroxidase, and oxidative damage, Nutr. Res. 10 (1990) 183 – 194. [64] C.B. Hong, C.K. Chow, Induction of eosinophilic enteritis and eosinophilia in rats by vitamin E and selenium deficiency, Exp. Mol. Pathol. 48 (1988) 182 – 192. [65] T. Noguchi, A.H. Cantor, M.L. Scott, Mode of action of selenium and vitamin E in prevention of exudative diathesis in chicks, J. Nutr. 103 (1973) 1502 – 1511. [66] D.G. Hafeman, W.G. Hoekstra, Lipid peroxidation in vivo during vitamin E and selenium deficiency in the rat as monitored by ethane evolution, J. Nutr. 107 (1977) 666 – 672. [67] J.J. Dougherty, W.G. Hoesktra, Effects of vitamin E and selenium on copper-induced lipid peroxidation in vivo and on acute copper toxicity, Proc. Soc. Exp. Biol. Med. 169 (1982) 208 – 210. [68] P.C. Chen, O.G. Peller, L. Kesner, Copper(II)-catalyzed lipid peroxidation in liposomes and erythrocyte membranes, Biochem. Biophys. Res. Commun. 12 (1982) 388 – 394. [69] F.W. Sunderman Jr., Metals and lipid peroxidation, Acta Pharmacol. Toxicol. 59 (1986) 248 – 255. [70] J.S. Bus J.E. Gibson, Lipid peroxidation and its role in toxicology, Rev. Biochem. Toxicol., (1979) 125 – 149. [71] B. Halliwell, Reactive 1: Oxygen species living in systems: source biochemistry and role in human disease, Am. J. Med. 91 (1991) 14S – 22S. [72] C.K. Chow, Vitamin E and oxidative stress, Free Radic. Biol. Med. 11 (1991) 215 – 232. [73] L. Flohe, W.A. Gu¨ nzler, H.H. Schock, Glutathione peroxidase: a selenoenzyme, FEBS Lett. 32 (1973) 132 – 134. [74] F. Ursini, M. Maiorino, C. Gregolin, The seleoenzyme phospholipid hydroperoxide glutathione peroxidase, Biochim. Biophys. Acta 839 (1985) 62 –70.
In vivo copper-mediated free radical production: an ESR spin-trapping study Maria B. Kadiiska *, Ronald P. Mason National Institutes of Health, National Institute of En6ironmental Health Sciences, PO Box 12233, Research Triangle Park, NC 27709 -2233, USA Received 1 July 2001; accepted 1 October 2001
Abstract Copper has been suggested to facilitate oxidative tissue injury through a free radical-mediated pathway analogous to the Fenton reaction. By applying the electron spin resonance (ESR) spin-trapping technique, evidence for hydroxyl radical formation in vivo was obtained in rats treated simultaneously with copper and ascorbic acid or paraquat. A secondary radical spin-trapping technique was used in which the hydroxyl radical formed the methyl radical upon reaction with dimethylsulfoxide. The methyl radical was then detected by ESR spectroscopy as its adduct with the spin trap phenyl-N-t-butyl- nitrone (PBN). In contrast, lipid derived radical was detected in vivo in copper-challenged, vitamin E and selenium-deficient rats. These findings support the proposal that dietary selenium and vitamin E can protect against lipid peroxidation and copper toxicity. Since copper excreted into the bile from treated animals is expected to be maintained in the Cu(I) state (by ascorbic acid or glutathione), a chelating agent that would redox-stablilize it in the Cu(I) state was used to prevent ex vivo redox chemistry. Bile samples were collected directly into solutions of bathocuproinedisulfonic acid, a Cu(I)-stabilizing agent, and 2,2%-dipyridyl, a Fe(II)-stabilizing agent. If these precautions were not taken, radical adducts generated ex vivo could be mistaken for radical adducts produced in vivo and excreted into the bile. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Rats; Spin trap; ESR; Free radicals
1. Introduction Although copper is a nutrient essential to health, the pathological processes associated with the various forms of copper overload, demonstrate that the metal can also be toxic. Copper poisoning may be either acute or chronic. Acute * Corresponding author. Tel.: + 1-919-541-0201; fax: +1919-541-1043. E-mail address: [email protected] (M.B. Kadiiska).
copper poisoning due to the accidental ingestion of copper-contaminated materials or water supplied by Cu pipes is the single most frequent cause of ingestion fatalities [1,2]. Chronic copper overload, however, can have several causes, including excessive intake, genetic metabolic diseases, and carcinomas [1,3]. Hypercumia in humans occurs in Wilson’s disease (hepatolenticular degeneration) and Indian childhood cirrhosis [1]. Maldistribution of body copper is observed in Menke’s kinky hair disease [3].
1386-1425/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S1386-1425(01)00713-2
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
1228
More work has been conducted on the hepatic disposition of copper than on that of most other metals, because of the attempts to characterize the molecular basis of these disorders. In the late 1970s the catalytic role of transition metals, particularly iron and copper, was implicated in free radical reactions. In biological systems there may be concentrations of redox-active copper sufficient to catalyze free radical reactions and account for various deleterious processes. It has been shown that traces of soluble copper or iron can catalyze the transformation of a superoxide radical anion (O− 2 ) to the highly reactive hydroxyl radical (OH) via the metal-catalyzed Haber– Weiss reaction: O2− +Mn + M(n − 1) + M = Cu (n=2)
or
Fe (n =3)
M(n − 1) + +H2O2 Mn + +OH− +HO Alternatively, a metal-mediated site-specific mechanism for free radical-induced biological damage has also been proposed [4– 6], where HO− is formed site-specifically in the vicinity of the target molecule (Biol) and reacts at the site of its production. Biol +Mn + Biol – Mn + Biol –Mn + + O2− Biol – M(n − 1) + +O2 Biol –M(n − 1) + + H2O2 (Biol –Mn + …HO) + OH− Other reductants present in cells (such as ascorbate or glutathione) might be able to replace superoxide in the Haber–Weiss cycle and, therefore, promote the toxicity of a metal/hydrogen peroxide system. It has been previously demonstrated that the combination of ascorbate and Cu2 + causes damage to macromolecular structures such as polysaccharides and proteins through generation of reactive oxygen species [7– 10]. Accordingly, copper overload has been shown to be associated with increased generation of lipid peroxidation products and with a number of biochemical and pathological lesions in rats [11,12]. Spin trapping is the most direct method for the detection of highly reactive free radicals in vivo.
With this electron spin resonance (ESR) technique, a higher steady-state concentration of free radicals (as radical adducts) is achieved, which can overcome the sensitivity problem inherent in the detection of endogenous radicals in biological systems [13–16]. Since the concentration of endogenous radicals in biological tissues is generally near the sensitivity limit of ESR spectroscopy, the spin-trapping technique is not limited by background signals. The technique of spin trapping involves the addition of a primary free radical across the double bond of a diamagnetic compound (the spin trap) to form a radical adduct more stable that the primary free radical. This technique involves the indirect detection of primary free radicals that cannot be directly observed by conventional ESR due to low steady-state concentrations or to very short relaxation times, which lead to very broad lines. A more recent approach to in vivo spin trapping has been to examine biological fluids such as urine [17,18], bile [19–27], and blood [28– 31] (or plasma [32,33]) directly for spin adducts. For these studies, the TM110 ESR cavity and a 17 mm flat call can be used to achieve the largest possible aqueous sample size in the active region of the cavity ( 100 Fl). This approach yields both high molar sensitivity and high resolution. In the study of hydroxyl radical formation in vivo, we used the scavenging reaction in which the hydroxyl radical is converted into the methyl radical via its reaction with dimethyl sulfoxide. The methyl radical is then detected as its long-lived phenyl N-t-butylnitrone (PBN) adduct. Alone, DMSO is relatively non-toxic, with a 24-h LD50 in the rat (ip) of 13.7 g/kg, and is therefore an ideal reagent for the in vivo detection of the hydroxyl radical [34]. We usually assay untreated bile for radical adducts. Anesthesia is maintained throughout the experiments, which are initiated by ip injection of DMSO containing PBN, followed by intragastric injection of ferrous sulfate. The resulting PBN/ methyl radical adducts (PBN/CH3) is detected by ESR, because, the in vivo detection of the hydroxyl radical adduct of PBN is extremely unlikely, due to its rapid decay [35].
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
2. Materials and methods
2.1. Chemicals PBN, DMSO, FeSO4 ·7H2O, CuSO4 ·5H2O, and acid (sodium salt) were purchased from Aldrich. DP, 2,2%-biquinoline, and BC were from Sigma. [13C2]DMSO (minimum 99 atom % 13C) was obtained from Isotec Inc. (Miamisburg, OH).
1229
and selenium. All rats were maintained on their respective diets, with access to an unlimited quantity of the diet and water, for 30 days, at which time the rats weighed 1659 30 g. Each dietary group consisted of eight to ten rats.
L-ascorbic
2.2. In 6i6o animal treatment Studies used Sprague– Dawley male rats of 380– 450 g (Charles River Breeding Laboratories, Raleigh, NC). Rats were fed a standard chow mix (NIH Open Formula; Zeigler Brothers, Gardner, PA) and were allowed free access to both food and water. Rats were anesthetized with Nembutal (50 mg/ kg intraperitoneally) and anesthesia was maintained throughout the experiments. Bile ducts were cannulated with a segment of PE 10 tubing. All animals were given an intraperitoneal injection of the spin trap PBN (100 or 250 mg/kg of body weight) dissolved in DMSO (1 ml/kg of body weight) and an intragastric injection of 1 M CuSO4 ·5H2O (3 or 1 ml/kg) dissolved in distilled water. A 1 M solution of ascorbate was administered intragastrically, where indicated, at a dose of 0.5 or 3 ml/kg body weight. Paraquat was injected at a dose of 55 mg/kg intragastrically.
2.3. Selenium- and 6itamin E-deficient diets Twenty-one-day-old rats were fed purified diets (ICN Biochemicals, Cleveland, OH) containing the following vitamin E and selenium concentrations: the vitamin E- and selenium-deficient diet contained no significant amounts of either vitamin E (a-tocopherol) derivatives or selenium; the selenium-deficient diet was the same diet but was supplemented with 121 IU of a-tocopherol/kg of food; the vitamin E-deficient diet was the same diet as the doubly deficient diet as but was supplemented with 0.22 mg of sodium selenite/kg of food; the vitamin E- and selenium-adequate (control) diet was the same diet as the doubly deficient diet but was supplemented with both vitamin E
2.4. In 6itro experiments in bile For in vitro studies, bile from untreated rats was added to tube containing various combinations of PBN, DMSO, DP, BC, and ascorbic acid. Cu(I), Cu(II), or Fe(II) was added last to initiate the reaction. Scans were taken 3 min after mixing.
2.5. In 6i6o bile experiments For ESR measurements, bile samples were thawed and transferred to flat quartz cells within 5 min. All spectra shown are from bile collected 2 h after the administration of CuSO4 and ascorbic acid.
2.6. ESR measurements ESR spectra were obtained using a Varian E109 spectrometer equipped with a TM110 cavity and operating at 9.3 GHz, 20-mW power, and 100-kHz modulation frequency. Spectra were recorded on an IBM-compatible computer interfaced to the spectrometer. Hyperfine coupling constants were determined from a spectral simulation program, SIMEPR. Student’s t test was used for statistical analysis of copper and iron levels from groups of animals. Values of P of B 0.05 were taken to indicated significant differences.
3. Results and discussion
3.1. Hydroxyl radical-mediated toxicity of paraquat and copper in rats Since the Fenton reaction requires hydrogen peroxide, we thought that a substance which catalyzes hydrogen peroxide formation would increase the signal. The activity of the herbicide paraquat (PQ2 + ) is attributed to its ability to catalyze the
1230
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
formation of superoxide and, subsequently, hydrogen peroxide. The proposed mechanism streams from the assumption that copper or iron and PQ2 + undergo cyclic reduction and reoxidation. The enzymatic reduction of PQ2 + to the stable monocation radical (PQ+) has been demonstrated [36]. PQ+ reacts very quickly with O2 to generate the parent PQ2 + and O2− [37], which rapidly dismutates to form O2 and H2O2. The Cu1 + and Fe2 + may subsequently react with the hydrogen peroxide in a Fenton-type reaction to form the highly reactive hydroxyl radical (HO), which would then cause extensive biological damage [4,38]. Therefore, the extreme toxicity of PQ2 + is thought to result from heavy intracellular fluxes of superoxide and hydrogen peroxide in the vicinity of redox-active metal ions, which are available for the promotion of the Fenton reaction, such that normal physiological defense mechanisms are overwhelmed. Considering all the evidence that PQ2 + toxicity is mediated by transition metals, we have investigated the role of Cu2 + in PQ2 + toxicity in vivo using the rat as an animal model. To date, free radical generation due to copper-mediated PQ2 + toxicity in whole animals has not been unequivocally demonstrated. PBN radical adducts were detected by ESR spectroscopy in the bile of rats 2 h after they had been given CuSO4 (3 mmol/kg, ig), PQ2 + (55 mg/kg, ig), and PBN (250 mg/kg, ip) dissolved in DMSO (1 ml/kg) (Fig. 1A). When the experiment was repeated in the absence of Cu2 + (Fig. 1B), PQ2 + (Fig. 1C), or both Cu2 + and PQ2 + (Fig. 1D), radical adducts were not detected, thereby, confirming the dependence of radical formation in the coadministration of both Cu2 + and PQ2 + . The signals in Fig. 1B– D are believed to be from a mixture of radical adducts, but were too weak to characterize [27]. The fact that Cu2 + or PQ2 + alone caused little detectable radical adduct formation may be attributed to their inability to form hydroxyl radicals at detectable concentrations due to strong defense systems against oxidative stress in living organisms. Minimum concentrations of 2 mmol/kg CuSO4 and 27.5 mg/kg PQ2 + given together were necessary for ESR-detectable quantities of radical ad-
ducts to occur in the bile. The spectrum shown in Fig. 1E demonstrates that the radical species observed in the bile was also PBN-dependent. (The doublet is from the ascorbate radical formed from the oxidation of endogenous ascorbate). To prevent the occurrence of reactions between copper or iron ions, O2, PBN, DMSO, and bile components during sample collection, bile must be collected directly into a solution of the Cu1 + stabilizing agent BC and the Fe2 + -stabilizing agent DP [27]. In order to identify the radical species detected in bile of animals which were treated with Cu2 + , PQ2 + , PBN, and DMSO, the experiment shown
Fig. 1. (A) The ESR spectrum of radical adducts detected in the bile of rats 2 h after CuSO4 (3 mmol/kg) and PQ2 + (55 mg/kg) were administered intragastrically, and PBN (250 mg/ kg) dissolved in DMSO (1 ml/kg) was administered intraperitoneally. Bile samples were collected into 25 ml of a solution containing 300 mM BC and 30 mM DP per 100 g body wt. (B) Same as in (A), but rats were not given CuSO4. (C) Same as in (A), but rates were not given PQ2 + . (D) Same as in (A), but rats were not given Cu(II) or PQ2 + . (E) Same as in (A), but rats were not given PBN and DMSO. Spectrometer settings: modulation amplitude, 1.3 G; microwave power, 20 mW; scan time, 16 min; time constant, 1 s.
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
in Fig. 1A was repeated using 13C-labeled DMSO ([13C]2-DMSO). The presence of hyperfine splittings in the ESR spectrum from 13 C(I =1/2) would allow an unambiguous assignment of the PBN/CH3 radical adduct formed in vivo. In addition, the use of deuterated PBN (d14-PBN) would significantly reduce inhomogenous broadening of the ESR signal from PBN protons and thereby, enhance spectral resolution. The spectrum obtained from this experiment is shown in Fig. 2A, and the complete spectral simulation is shown in Fig. 2B. The four radical species used to calculate the spectrum in Fig. 2B are shown in Fig. 2C– F, and the hyperfine coupling constants are given in Table 1. Only the assignments of the PBN/ CH3 and ascorbate radical species, however, can be considered unique. The appearance of 13C-hyperfine splittings, observed in the wings of each of the lines of the nitroxide triplet (Fig. 2A, cf. Fig. 1A) is proof that the PBN/CH3 radical adduct was formed in vivo during PQ2 + and Cu2 + intoxication. The other two PBN adducts (Fig. 2D and E) remain unidentified, but are comparable with non-DMSO-dependent PBN radical species with similar hyperfine coupling constants detected in bile of rats which were administered Fe2 + and PQ2 + [39]. In that study, the radical comparable with that in Fig. 2D was tentatively assigned to a carbonyl radical adduct of PBN based on its hyperfine coupling constants. The radical species, which gave, rise to the spectrum in Fig. 2E must be the result of complex PBN chemistry [39]. The analysis of radical adducts excreted via the biliary route provide strong ESR evidence for the generation of the hydroxyl radical as a result of the known futile enzymatic redox cycling of PQ2 + , with copper playing an essential mediatory role. No radical adducts were detected when either CuSO4 or PQ2 + was excluded. From a different perspective, in vivo copper-dependent hydroxyl radical generation could be said to be promoted by PQ2 + . This is the first report of ESR evidence for this synergetic hydroxyl radical generation by copper and PQ2 + in a whole animal.
1231
Fig. 2. (A) The ESR spectrum of radical adducts detected in the bile of rats 2 h after CuSO4 (3 mmol/kg) and PQ2 + (55 mg/kg) were administered intragastrically, and d14-PBN (250 mg/kg) dissolved in [13C]2-DMSO (1 ml/kg) was administered intraperitoneally. Bile samples were collected into a 25 ml of a solution containing 300 mM BC and 30 mM DP per 100 g body wt. Spectrometer settings: modulation amplitude, 0.67 G; microwave power, 20 mW; scan time, 60 min; time constant, 4 s. (B) Complete computer simulation of the spectrum in (A). (C) Simulation of the relative contribution of the PBN/13CH3 radical adduct to the spectrum in (A). (D) Simulation of the relative contribution of the unidentified PBN/C radical adduct to the spectrum in (A). (E) Simulation of the relative contribution of the PBN/Y radical adduct to the spectrum in (A). (F) Simulation of the relative contribution of the ascorbate radical (Asc−) to the spectrum in (A). The hyperfine coupling constants obtained from these simulations are given in Table 1.
3.2. Hydroxyl radical formation after acute copper and ascorbic acid intake To produce the highly reactive hydroxyl radical, Cu2 + must be reduced to Cu1 + . It is assumed that Cu1 + is oxidized by H2O2 to yield HO, although, the reaction between Cu1 + and H2O2 may not be a simple one-electron oxida-
1232
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
tion of Cu1 + to Cu2 + with concomitant formation of the free hydroxyl radical [5,40– 42]. We found that, in ascorbic acid-treated rats, hydroxyl radical (or a closely related species such as a high-valence metal species) is generated during acute copper overload. The suggested mechanism involves the oxidation of ascorbate to the ascorbate radical, the reduction of Cu2 + (either spuriously protein bound or non-protein bound) to Cu1 + , and the production of superoxide, hydrogen peroxide, and hydroxal radical.
were not able to show in vivo hydroxyl radical generation from rats treated only with CuSO4 and PBN/DMSO (Fig. 3D). Since copper excreted into the bile from untreated animals is expected to be maintained in the Cu1 + state (by ascorbic acid, glutathione, or other biological reductants), the chelating agent BC, which stabilizes copper in the Cu1 + state, was used. A chelating agent for redox-stabilizing Fe2 + , DP, was also used. We found that, if these precautions were not taken, signals from radical adduct generates ex vivo might be mistaken for the adducts generated in vivo and excreted into the bile. Having demonstrated in vivo copper- and ascorbic acid-dependent free radical formation, we next sought to identify the radical adducts detected. The incorporation of a magnetic nucleus (typically 13C) into a radical adduct, can permit identification of a variety of radical trapped in complex biological systems [16]. Using this approach, experiments were performed with 13Csubstituted DMSO, which were expected to confirm the formation of PBN/CH3 in vivo. The ESR signal detected in the bile of rats given an intraperitoneal dose of [13C2]DMSO/PBN and an intragastric dose of CuSO4 and ascorbic acid is shown in Fig. 4A. This spectrum clearly contains a signal from the PBN/CH3 radical adduct, which is the dominant species and is characterized by the hyperfine coupling constants of a N = 16.30 C 13 G, a H b = 3.63 G, and a b ( C) =3.81 G (Fig. 4B).
Ascorbate+Cu2 + Ascorbate radical+Cu+ +H+ Cu1 + +O2 Cu1 + + O2− 2O2− H2O2 +O2 H2O2 +Cu1 + Cu2 + +OH− +HO After an intragstric dose of CuSO4 and ascorbic acid and intraperitoneal administration of PBN in DMSO to rats, the ESR analysis of bile samples collected into aqueous solutions of BC and DP gave six-line spectrum (Fig. 3A). The formation of this radical adduct was dependent on the administration of CuSO4 (Fig. 3B), PBN/DMSO (Fig. 3C), and ascorbate (Fig. 3D) and was apparently due to the in vivo formation of the PBN-methyl adduct, PBN/CH3 from hydroxyl radical. The ESR spectrum of bile from untreated rats showed only the doublet of the ascorbate radical, which was variable in intensity (data not shown). We
Table 1 Hyperfine coupling constants obtained from the simulation of Fig. 3A Radical detected
PBN/13CH3 PBN/13CH3 PBN/C PBN/COR PBN/Y PBN/Y Asc−
Hyperfine coupling constants (G) aN
aN b
a 13C b
16.30 16.31 15.76 15.61 15.30 15.20
3.21 3.06 3.68 3.36
4.39 4.00
a N/a H b
Relative concentration
Source
5.08 5.33 4.28 4.66 12.45 12.67
23
Fig. [39] Fig. [39] Fig. [39] Fig.
aH
1.23 1.20 1.81
32 43 2
3C 3D 3E 3F
Note: the four radical species used to simulate the experimental spectrum are shown in Fig. 3C–F and are presented at the relative concentrations determined from the composite spectrum in Fig. 3B. The radical assignments listed above are discussed in the text.
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
1233
Clearly in our system of copper-intoxicated rats, copper alone does not cause the formation of detectable PBN/CH3, unlike what had been demonstrated for iron [22]. Either the binding of copper to its physiological ligands prevents it from catalyzing OH formation [42,46,47] or OH participates in site-specific reactions of the bound copper and does not have the opportunity to react with DMSO to form the methyl radical in detectable concentrations [6,40,44,45]. Free radical production from copper in vivo, however, can be stimulated by sufficient quantities of ascorbic acid to produce radical adduct similar to those observed from iron overload. Although the antagonistic effects due to ascorbic acid supplementation include inhibition of copper absorption, signs of copper deficiency [48], and lowered tissue copper and ceruloplasmin lev-
Fig. 3. (A) ESR spectrum of radical adducts detected in the bile from rats 2 h after CuSO4 (1 mmol/kg) and ascorbic acid (Asc) (0.5 mmol/kg) were administered intragastrically and PBN (100 mg/kg) dissolved in DMSO (1 ml/kg) was administered intraperitoneally. Bile samples were collected into a 25 ml of a solution containing 300 mM BC and 30 mM DP per 100 g of body weight. (B) Same as in (A), but rats were not given CuSO4; (C) same as in (A), but rats were not given PBN/ DMSO; (D) same as in (A), but rats were not given ascorbic acid. Spectrometer settings: modulation amplitude, 1.3 G; microwave power, 20 mW; scan time, 16 min; time constant, 1 s.
The results in this study support a free radical mechanism that attributes an important role to ascorbic acid and copper in vivo in the induction of biological damage already demonstrated in numerous in vitro experiments [4,7,9,43– 45]. From our studies here it is clear that generation of detectable OH during acute copper poisoning of rats is dependent on high intragastric concentrations of ascorbic acid. Conceivably, radical production may occur in the digestive tract, where the spin-trapped adduct is absorbed and transported directly to the liver via the portal vein. However, we could not detect any free radical in blood samples drawn from the portal vein (data not shown).
Fig. 4. (A) ESR spectrum of radical adducts detected in the bile from rats 2 h after CuSO4 (3 mmol/kg, intragastrically), ascorbic acid (0.5 mmol/kg, intragastrically) and PBN (250 mg/kg) dissolved in [13C2]DMSO (1 ml/kg, intraperitoneally) were administered. (B) Computer simulation of spectrum in (A). Hyperfine coupling constants are given in Table 1.
1234
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
els [49,50], the cytotoxicity of copper and ascorbate was attributed to OH radicals repeatedly formed by recycling Cu2 + back to Cu1 + [43,44]. However, under the same experimental conditions, investigators have noted both pro- and antioxidant effects of ascorbate, respectively, [51,52]. Since ascorbic acid is the single most heavily consumed nutrient supplement, it is important that more information be gathered on the effects of ascorbic acid intake on copper bioavailability and utilization. In the present study we show in vivo formation of hydroxyl radicals produced in rats during acute copper overload and ascorbic acid intake. We used a secondary trapping technique [22] in which the hydroxyl radical forms the methyl radical upon reaction with DMSO. The methyl radical is then detected by ESR spectroscopy as its adduct with the spin trap PBN. Finally we showed that, if bile samples from copper-treated rats were not collected directly into solutions of chelating agents that would redoxstabilize copper and traces of iron, signals from radicals adducts generated ex vivo might be mistaken for adducts generated in vivo and then excreted into the bile. In summary, the findings presented here indicate that radicals detected in the bile are formed in vivo due to the combination of acute copper and ascorbic acid intake.
3.3. Lipid radical generation in copper-treated 6itamin E- and selenium-deficient rats Free radical-induced peroxidative damage to membrane lipids has long been regarded as a critical initiating event leading to cell injury [53– 57]. It is generally accepted that susceptibility to lipid peroxidation is influenced by the level of vitamin E and selenium in tissue [58–61]. Depletion of dietary vitamin E and selenium, for example, has been shown to be linked to the increased formation of lipid peroxidation products and increased hepatic peroxidative injury in different species, including rats [62– 64]. Therefore, the tissues of animals deficient in vitamin E and selenium have a greater tendency to undergo lipid peroxidation both in vitro [65] and in vivo [66]
than do tissues of animals with a diet adequate in these nutrients. Other studies also showed that rats whose diets were deficient in vitamin E and selenium were much more vulnerable to copper toxicity. Dougherty and Hoekstra [67] found that intraperitoneal administration of CuSO4 to rats deficient in both vitamin E and selenium caused dose-dependent increases in expired ethane (an indirect index of lipid peroxidation) and acute mortality. Selenium and vitamin E supplementation of the diet prevented the increase of ethane production and mortality. Similarly, Dillard and Tappel [12] found increased pentane production and other products of lipid peroxidation in copper-treated, vitamin E-deficient rats. Smaller increases of the same lipid peroxidation products were found in copper-treated rats fed a vitamin E-supplemented diet. Despite all the evidence for free radical-mediated lipid peroxidation from copper intoxication, the free radicals themselves have not been detected in vivo. Since it has been observed that the toxicity of copper is considerably higher in vitamin E- and/or selenium-deficient rats, the present experiments were designed to determine whether copper overload in vivo promotes radical generation in vitamin E- and/or selenium-deficient rats and whether the radical species generated are the same as in our previous findings. Fig. 5A shows the radical adduct signal in bile from rats fed for 10 days with a vitamin E- and selenium-deficient diet, as measured 2 h after the rats were challenged with CuSO4 (2 mmol/kg, intragastrically). A solution of the spin trap PBN (250 mg/kg) in DMSO (1 ml/kg, intraperitoneally) was also administered. The intensity of the signals was dose dependent with respect to both CuSO4 and PBN (data not shown). In the absence of copper (Fig. 5B) or PBN (Fig. 5C), no significant amount of radical adduct was detected. The small doublet observed is due to the ascorbate radical ion (a H = 1.7 G) formed from the one-electron oxidation of endogenous ascorbate. When the experiment was repeated with control rats fed a vitamin E- and selenium-deficient diet, only a residual PBN radical adduct signal was detectable in either copper-treated (Fig. 5D) or nontreated (Fig. 5E) rats.
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
1235
tutions and careful comparisons with ESR parameters of known PBN adducts. When [13C]DMSO was used instead of 12 [ C]DMSO, no additional hyperfine splitting from 13C (I=1/2) was observed (Fig. 7D), again indicating that PBN/CH3 was not present. Values of a N = 15.40 G and a H b = 2.54 G, with a 100% Lorentzian lineshape, were obtained from the simulation shown in Fig. 1A. These results indicated that the radical species trapped by PBN in vivo in copper-challenged, vitamin E- and selenium-deficient rats was formed probably from an endoge-
Fig. 5. ESR detection of a free radical formed in vivo and spin-trapped with PBN, from a copper-challenged rat deficient in both vitamin E and selenium. All spectra are from bile samples collected 2 h after administration of copper. (A) A vitamin E- and selenium-deficient rat was administered 2 mmol/kg CuSO4, intragastrically, and 250 mg/kg PBN as a DMSO solution (1 ml/kg), intraperitoneally. (B) As in (A), but without administration of CuSO4. (C) As in (A), but without administration of PBN. (D) As in (A), but the experiment was done with a rat fed a vitamin E- and selenium-adequate diet as in control. (E) As in (D), but without administration of CuSO4. Instrumental conditions: microwave power, 20 mW; modulation amplitude, 1.33 G; time constant, 1 s; scan rate, 5 G/min.
To investigate further the individual roles of vitamin E and selenium in protecting against radical generation in copper-challenged rats, additional experiments were carried out. If rats were fed diets adequate in vitamin E and/or selenium, little or no radical adduct could be detected (Fig. 6). Although a positive identification of radical adducts formed in vivo is very difficult, tentative assignments can be made through isotope substi-
Fig. 6. ESR detection of a free radical formed in vivo and spin-trapped with PBN in copper-challenged rats with various combinations of vitamin E and selenium deficiencies. All spectra are from bile samples collected 2 h after administration of copper. (A) A vitamin E- and selenium-deficient rat was administered 2 mmol/kg CuSO4, intragastrically, and 250 mg/ kg PBN as a DMSO solution (1 ml/kg), intraperitoneally. (B) As in (A), but with a vitamin E-deficient rat. (C) As in (A), but with a selenium-deficient rat. (D) As in (A), but the experiment was done with a rat fed a vitamin E- and selenium-adequate diet as a control. Instrumental conditions: microwave power, 20 mW; modulation amplitude, 1.33 G; time constant, 1 s; scan rate, 5 G/min.
1236
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
selenium- and vitamin E-deficient rats [12,67–69]. Therefore, the mechanism of copper toxicity apparently involves lipid peroxidation. According to the current hypothesis, transition metal ions are involved in lipid peroxidation by decomposing lipid peroxides into their peroxyl and alkoxyl radicals, which can then abstract hydrogen and perpetuate a chain reaction of lipid peroxidation [70,71]: LOOH +Mn + LO + M(n + 1) + OH− LOOH +M(n + 1) LOO + Mn + + H+ LO +LH LOH + L LOO + LH LOOH +L L +O2 LOO
Fig. 7. Effect of isotope substitutions on the ESR spectra of the PBN radical adduct formed in vivo in copper-challenged rats deficient in both vitamin E and selenium. Spectra are from bile samples collected 2 h after administration of copper. (A) A vitamin E- and selenium-deficient rat was administered 2 mmol/kg CuSO4, intragastrically, and 250 mg/kg d14-PBN in a DMSO solution (1 ml/kg), intraperitoneally. (B) Computer simulation of (A) with a N = 15.36 G, aH b = 2.50 G, and a 90% Lorentizian lineshape. (C) Residual of (A) minus (B), indicating the precision of the simulated spectrum. (D) A vitamin Eand selenium-deficient rat was administered 2 mmol/kg CuSO4, intragastrically, and 250 mg/kg d14-PBN in a [13C]DMSO solution (1 ml/kg), intraperitoneally. Additional carbon-12 (I =1/2) hyperfine splittings are not observed. (E) Computer simulation of (D) with a N = 15.40 G, aH b = 2.54 G, and a 100% Lorentzian lineshape. (F) Residual of (D) minus (E), indicating the precision of the simulated spectrum. Instrumental conditions: microwave power, 20 mW; modulation amplitude, 0.67 G; time constant, 4 s; scan rate, 1.33 G/min.
nous molecule such as a poly-unsaturated fatty acid. It is known from other studies that Cu2 + increases the lipid peroxidation products formed in
where L represents a lipid molecule and M represents a redox-active metal ion. Although no direct evidence has been presented for the generation of lipid-derived radicals in our animal model, we provide conclusive evidence that the radical adduct is formed from an endogenous molecule. Had the generation of hydroxyl radical been the primary toxic action of the copper in these experiments, then the DMSO-dependent PBN/CH3 adduct should have been detected [27]. From the hyperfine splitting constants obtained for the radical adduct shown in Fig. 1A and A and the lack of a carbon-13 isotope splitting when [13C]DMSO was used, it is apparent that the radical adduct is not a product of hydroxyl radical scavenging by DMSO. It should be noted, however, that an inability to detect radical adducts in bile by ESR is not proof that it is not formed in vivo. A number of factors, such as the relative rates of radical adduct formation, destruction, and excretion, can all affect the detection of a radical adduct. The fact that the radical adduct could be detected only in rats that were deficient in both vitamin E and selenium strongly suggests that its formation is associated with lipid peroxidation. In this respect, the primary role of vitamin E in preventing free radical-initiated peroxidative tissue damage is widely accepted, and the functional interrelation between vitamin E and selenium has
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
been recognized [72]. In a variety of experimental systems, studies have shown that selenium may complement the antioxidant function of vitamin E via selenoenzymes [73,74]. Indeed, in this role they are a part of a complex system of antioxidant protection in vivo that includes free radical scavengers and several families of enzymes. In conclusion, the present study provides the most direct ESR evidence yet of lipid peroxidation-associated free radical formation in the toxicity of copper.
4. Conclusions
1. The study provided strong ESR evidence for the generation of hydroxyl radical as a result of the futile enzymatic redox cycling of PQ2 + , with Cu playing an essential mediatory role. From a different perspective, in vivo Cu2 + -dependent hydroxyl radical generation could be said to be promoted by PQ2 + . This is the first report of ESR evidence for the synergetic radical generation by Cu and PQ2 + . 2. Cu2 + alone does not cause the formation of detectable radicals unlike what had been demonstrated for iron. However, free radical production from Cu in vivo can be stimulated by vitamin C. This is the first study to show ESR evidence for pro-oxidant activity of vitamin C. 3. The study provided the most direct ESR evidence yet of LPO-associated free radical formation in the toxicity of Cu. The data that Cu causes no radical generation in rats fed vitamin E- and Se-supplemented diet suggest that these nutrients can protect against lipid peroxidation and use copper toxicity.
References [1] I.H. Scheinberg, I. Sternleib, Copper toxicity and Wilson’s disease, in: A.S. Prasad (Ed.), Trace Elements in Human Health and Disease, Academic Press, New York, 1976, pp. 415 – 438. [2] E.D. Harris, Copper in human and animal health, in: J. Rose (Ed.), Trace Elements in Health, Butterworths, London, 1983, pp. 44 – 73.
1237
[3] M.J. Ettinger, H.M. Darwish, R.C. Schmitt, Mechanism of copper transport from plasma to hepatocytes, Fed. Proc. 45 (1986) 2800 – 2804. [4] M. Chevion, A site specific mechanism for free radical induced biological damage: the essential role of redox active transition metals, Free Radic. Biol. Med. 5 (1988) 27 – 37. [5] H.C. Sutton, C.C. Winterbourn, On the participation of higher oxidation states of iron and copper in Fenton reactions, Free Radic. Biol. Med. 6 (1989) 53 – 60. [6] S. Goldstein, G. Czapski, Transition metal ions and oxygen radicals, Int. Rev. Exp. Pathol. 31 (1990) 133 – 164. [7] S.-H. Chiou, DNA- and protein-scission activities of ascorbate in the presence of copper ion and a copper – peptide complex, J. Biol. Chem. 94 (1983) 1259 – 1267. [8] K. Uchida, S. Kawakishi, Oxidative depolymerization of polysaccharides induced by the ascorbic acid – copper ion systems, Agric. Biol. Chem. 50 (1986) 2579 – 2583. [9] E. Shinar, T. Navok, M. Chevion, The analogous mechanisms of enzymatic inactivation induced by ascorbate and superoxide in the presence of copper, J. Biol. Chem. 258 (1983) 14778 – 14783. [10] G. Marx, M. Chevion, Site-specific modification of albumin by free radicals: reaction with copper(II) and ascorbate, Biochem. J. 236 (1985) 397 – 400. [11] S.D. Aust, L.A. Morehouse, C.E. Thomes, Role of metals in oxygen radical reactions, J. Free Radic. Biol. Med. 1 (1985) 3 – 25. [12] C.J. Dillard, A.L. Tappel, Lipid peroxidation and copper toxicity in rats, Drug Chem. Toxicol. 7 (1984) 477 – 487. [13] G.R. Buettner, R.P. Mason, Spin-trapping methods for detecting superoxide and hydroxyl free radicals in vitro and in vivo, Methods Enzymol. 186 (1990) 127 – 133. [14] E.G. Janzen, R.A. Towner, D.L. Haire, Detection of free radicals generated from the in vitro metabolism of carbon tetrachloride using improved ESR spin trapping techniques, Free Radic. Res. Commun. 3 (1987) 357 – 364. [15] K.T. Knecht, R.P. Mason, In vivo spin trapping of xenobiotic free radical metabolites, Arch. Biochem. Biophys. 303 (1990) 185 – 194. [16] C. Mottley, R.P. Mason, Nitroxide radical adducts in biology: chemistry, applications and pitfalls, in: L.J. Berliner, J. Reuben (Eds.), Biological Magnetic Resonance, vol. 8, Plenum Press, New York, 1989, pp. 489 – 546. [17] H.D. Connor, R.G. Thurman, M.D. Galizi, R.P. Mason, The formation of a novel free radical metabolite from CCl4 in the perfused rat liver and in vivo, J. Biol. Chem. 261 (1986) 4542 – 4548. [18] L.B. LaCagnin, H.D. Connor, R.P. Mason, R.G. Thurman, The carbon dioxide anion radical adduct in the perfused rat liver: relationship to halocarbon-induced toxicity, Mol. Pharmacol. 33 (1988) 351 – 357. [19] K.T. Knecht, R.P. Mason, In vivo radical trapping and biliary secretion of radical adducts of carbon tetrachloride-derived free radical metabolites, Drug Metab. Dispos. 16 (1988) 813 – 817.
1238
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239
[20] K.T. Knecht, B.U. Bradford, R.P. Mason, R.G. Thurman, In vivo formation of a free radical metabolite of ethanol, Mol. Pharmacol. 38 (1990) 26 –30. [21] K.T. Knecht, R.P. Mason, The detection of halocarbonderived radical adducts in bile and liver of rats, Drug Metab. Dispos. 19 (1991) 325 –331. [22] M.J. Burkitt, R.P. Mason, Direct evidence for in vivo hydroxyl-radical generation in experimental iron overload: an ESR spin-trapping investigation, Proc. Natl. Acad. Sci. USA 88 (1991) 8440 –8444. [23] H.M. Hughes, I.M. George, J.C. Evans, C.C. Rowlands, G.M. Powell, C.G. Curtis, The role of the liver in the production of free radicals during halothane anaesthesia in the rat. Quantification of N-tert-butyl-alpha-(4- nitrophenyl)nitrone (PBN)-trapped adducts in bile from halothane as compared with carbon tetrachloride, Biochem. J. 277 (1991) 795 –800. [24] K.T. Knecht, J.A. DeGray, R.P. Mason, Free radical metabolism of halothane in vivo: radical adducts detected in bile, Mol. Pharmacol. 41 (1992) 943 –949. [25] M. Sentjurc, R.P. Mason, Inhibition of radical adduct reduction and reoxidation of the corresponding hydroxylamines in in vivo spin trapping of carbon tetrachloridederived radicals, Free Radic. Biol. Med. 13 (1992) 151 – 160. [26] W. Chamulitrat, S.J. Jordan, R.P. Mason, Fatty acid radical formation in rats administered oxidized fatty acids: in vivo spin trapping investigation, Arch. Biochem. Biophys. 299 (1992) 361 –367. [27] M.B. Kadiiska, P.M. Hanna, L. Hernandez, R.P. Mason, In vivo evidence of hydroxyl radical formation after acute copper and ascorbic acid intake: electron spin resonance spin-trapping investigation, Mol. Pharmacol. 42 (1992) 723 – 729. [28] K.R. Maples, S.J. Jordan, R.P. Mason, In vivo rat hemoglobin thiyl free radical formation following phenylhydrazine administration, Mol. Pharmacol. 33 (1998a) 344 – 350. [29] K.R. Maples, S.J. Jordan, R.P. Mason, In vivo rat hemoglobin thiyl free radical formation following administration of phenylhydrazine and hydrazine-based drugs, Drug Metab. Dispos. 16 (1988b) 799 – 803. [30] K.R. Maples, P. Eyer, R.P. Mason, Aniline-, phenylhydroxylamine-, nitrosobenzene-, and nitrobenzene-induced hemoglobin thiyl free radical formation in vivo and in vitro, Mol. Pharmacol. 37 (1990a) 311 –318. [31] K.R. Maples, C.H. Kennedy, S.J. Jordan, R.P. Mason, In vivo thiyl free radical formation from hemoglobin following administration of hydroperoxides, Arch. Biochem. Biophys. 277 (1990b) 402 –409. [32] L.A. Reinke, E.G. Janzen, Detection of spin adducts in blood after administration of carbon tetrachloride to rats, Chem. Biol. Interact. 78 (1991) 155 –165. [33] L.A. Reinke, R.A. Towner, E.G. Janzen, Spin trapping of free radical metabolites of carbon tetrachloride in vitro and in vivo: effect of acute ethanol administration, Toxicol. Appl. Pharmacol. 112 (1992) 17 –23.
[34] R.P. Mason, P.M. Hanna, M.J. Burkitt, M.B. Kadiiska, Detection of oxygen-derived radicals in biological systems using electron spin resonance, Environ. Health Perspect. 102 (Suppl. 10) (1994) 33 – 36. [35] Y. Kotake, E.G. Janzen, Decay and fate of the hydroxyl radical adduct of Ö-phenyl-N-tert-butylnitrone in aqueous media, J. Am. Chem. Soc. 113 (1991) 9503 – 9506. [36] R.P. Mason, J.L. Holtzman, The role of catalytic superoxide formation in the O2 inhibition of nitroreductase, Biochem. Biophys. Res. Commun. 67 (1975) 1267 – 1274. [37] J.A. Farrington, M. Ebert, E.J. Land, K. Fletcher, Bipyridylium quaternary salts and related compounds. V. Pulse radiolysis studies of the reaction of paraquat radical with oxygen. Implications for the mode of action of bipyridyl herbicides, Biochim. Biophys. Acta 314 (1973) 372 – 381. [38] H. Zer, J.H. Freedman, J. Peisach, M. Chevion, Inverse correlation between resistance towards copper and towards the redox-cycling compound paraquat: a study in copper-tolerant hepatocytes in tissue culture, Free Radic. Biol. Med. 11 (1991) 9 – 16. [39] M.J. Burkitt, M.B. Kadiiska, P.M. Hanna, S.J. Jordan, R.P. Mason, Electron spin resonance spin-trapping investigation into the effects of paraquat and desferrioxamine on hydroxyl radical generation during acute iron poisoning, Mol. Pharmacol. 43 (1993) 257 – 263. [40] A. Samuni, M. Chevion, G. Czapski, Unusual copper-induced sensitization of the biological damage due to superoxide radicals, J. Biol. Chem. 256 (1981) 12632 – 12635. [41] S. Goldstein, G. Czapski, Mechanism and reaction products of the oxidation of Cu1 + -phenantholine by H2O2, Free Radic. Biol. Med. 1 (1985) 373 – 380. [42] P.M. Hanna, R.P. Mason, Direct evidence for inhibition of free radical formation from Cu1 + and hydrogen peroxide by glutathione and other potential ligands using the EPR spin-trapping technique, Arch. Biochem. Biophys. 295 (1992) 205 – 213. [43] R. Stoewe, W.A. Prutz, Copper-catalyzed DNA damage by ascorbate and hydrogen peroxide: kinetics and yield, J. Free Radic. Biol. Med. 3 (1987) 97 – 105. [44] A. Samuni, J. Aronovtich, D. Godinger, M. Chevion, G. Czapski, On the cytotoxicity of vitamin C and metal ions: a site specific Fenton mechanism, Eur. J. Biochem. 137 (1983) 119 – 124. [45] D.A. Rowley, B. Halliwell, Superoxide-dependent and ascorbate-dependent formation of hydroxyl radicals in the presence of copper salts: a physiologically significant reaction, Arch. Biochem. Biophys. 225 (1983) 279 – 284. [46] M. Nakamura, I. Yamazaki, One-electron transfer reactions in biochemical systems. VI. Changes in electron transfer mechanism of lipoamide dehydrogenase by modification of sulfhydryl groups, Biochim. Biophys. Acta 267 (1972) 249 – 257. [47] J.H. Freedman, M.R. Ciriolo, J. Peisach, The role of glutathione in copper metabolism and toxicity, J. Biol. Chem. 264 (1989) 5598 – 5605.
M.B. Kadiiska, R.P. Mason / Spectrochimica Acta Part A 58 (2002) 1227–1239 [48] C.E. Hunt, W.W. Carlton, P.M. Newberne, Interrelationships between copper deficiency and dietary ascorbic acid in the rabbit, Br. J. Nutr. 24 (1970) 61 –69. [49] D.B. Milne, S.T. Omaye, W.H. Amos, Effect of ascorbic acid on copper and cholesterol in adult cynomolgus monkeys fed a diet marginal in copper, Am. J. Clin. Nutr. 34 (1981) 2389 –2393. [50] C.H. Smith, W.R. Bidlack, Interrelationship of dietary ascorbic acid and iron on tissue distribution of ascorbic acid, iron and copper in female guinea pigs, J. Nutr. 110 (1980) 1398 – 1408. [51] A. Bendich, L.J. Machlin, O. Scandurra, G.W. Burton, D.D. Wayner, The antioxidant role of vitamin C, Adv. Free Radic. Biol. Med. 2 (1986) 419 –444. [52] D.C. Laudicina, L.J. Marnet, Enhancement of hydrogenperoxide-dependent lipid peroxidation in rat liver microsomes by ascorbic acid, Arch. Biochem. Biophys. 278 (1990) 73 – 80. [53] A.L. Tappel, Vitamin E and free radical peroxidation of lipids, Ann. New York Acad. Sci. 203 (1972) 12 –28. [54] J.F. Mead, Free radical mechanisms of lipid damage and consequences of cellular membranes, Free Radic. Biol. Med. 1 (1976) 51 –68. [55] R.O. Recknagel, E.A. Glende, R.L. Walker, K. Lowrey, Lipid peroxidation: biochemistry, measurement, and significance in liver cell injury, in: G. Plaa, W.R. Hweiit (Eds.), Toxicology of the Liver, Raven, New York, 1982, pp. 213 – 241. [56] A. Benedetti, M. Comporti, Formation, reactions, and toxicity of aldehydes produced in the course of lipid peroxidation in cell membranes, Bioelectrochem. Bioenerg. 18 (1987) 187 – 202. [57] F.R. Ungemach, Pathobiochemical mechanisms of hepatocellular damage following lipid peroxidation, Chem. Phys. Lipids 45 (1987) 171 –205. [58] G.F. Combs Jr., T. Noguchi, M.L. Scott, Mechanisms of action of selenium and vitamin E in protection of biological membranes, Fed. Proc. 34 (1975) 2090 – 2095. [59] A.L. Tappel, Vitamin E and selenium protection from in vivo lipid peroxidation, Ann. New York Acad. Sci. 355 (1980) 18 – 31. [60] A.K. De, R. Darad, Physiological antioxidants and antioxidative enzymes in vitamin E-deficient rats, Toxicol. Lett. 44 (1988) 47 – 54.
1239
[61] W.H. Saypil, H. Shimaski, N. Ueta, Free radical-induced liver injury. I. Effects of dietary vitamin E deficiency on triaglycerol level and its fatty acid profile in rat liver, Free Radic. Res. Commun. 14 (1991) 315 – 322. [62] C.K. Chow, C.J. Chen, Dietary selenium and age-related susceptibility of rat erythrocytes to oxidative damage, J. Nutr. 110 (1980) 2460 – 2466. [63] C.K. Chow, Effect of dietary vitamin E and selenium on rats: pyruvate kinase, glutathione peroxidase, and oxidative damage, Nutr. Res. 10 (1990) 183 – 194. [64] C.B. Hong, C.K. Chow, Induction of eosinophilic enteritis and eosinophilia in rats by vitamin E and selenium deficiency, Exp. Mol. Pathol. 48 (1988) 182 – 192. [65] T. Noguchi, A.H. Cantor, M.L. Scott, Mode of action of selenium and vitamin E in prevention of exudative diathesis in chicks, J. Nutr. 103 (1973) 1502 – 1511. [66] D.G. Hafeman, W.G. Hoekstra, Lipid peroxidation in vivo during vitamin E and selenium deficiency in the rat as monitored by ethane evolution, J. Nutr. 107 (1977) 666 – 672. [67] J.J. Dougherty, W.G. Hoesktra, Effects of vitamin E and selenium on copper-induced lipid peroxidation in vivo and on acute copper toxicity, Proc. Soc. Exp. Biol. Med. 169 (1982) 208 – 210. [68] P.C. Chen, O.G. Peller, L. Kesner, Copper(II)-catalyzed lipid peroxidation in liposomes and erythrocyte membranes, Biochem. Biophys. Res. Commun. 12 (1982) 388 – 394. [69] F.W. Sunderman Jr., Metals and lipid peroxidation, Acta Pharmacol. Toxicol. 59 (1986) 248 – 255. [70] J.S. Bus J.E. Gibson, Lipid peroxidation and its role in toxicology, Rev. Biochem. Toxicol., (1979) 125 – 149. [71] B. Halliwell, Reactive 1: Oxygen species living in systems: source biochemistry and role in human disease, Am. J. Med. 91 (1991) 14S – 22S. [72] C.K. Chow, Vitamin E and oxidative stress, Free Radic. Biol. Med. 11 (1991) 215 – 232. [73] L. Flohe, W.A. Gu¨ nzler, H.H. Schock, Glutathione peroxidase: a selenoenzyme, FEBS Lett. 32 (1973) 132 – 134. [74] F. Ursini, M. Maiorino, C. Gregolin, The seleoenzyme phospholipid hydroperoxide glutathione peroxidase, Biochim. Biophys. Acta 839 (1985) 62 –70.