(DAHK) Block Oxidant-Induced Neuronal Death
Elizabeth T. Gum, MS; Raymond A. Swanson, MD; Conrad Alano, PhD; Jialing Liu, PhD;
Shwuhuey Hong, BS; Philip R. Weinstein, MD; S. Scott Panter, PhD
Background and Purpose—Studies using animal models of stroke have shown that human serum albumin (HSA) significantly ameliorates cerebral ischemic injury after both transient and permanent ischemia, even when administered after the onset of ischemia or reperfusion. The mechanism of this effect remains uncertain, and prior studies suggest both indirect hemodynamic and direct cytoprotective effects. HSA is a potent antioxidant, in part because of its strong copper-binding capacity. Here we examined the effect of HSA on oxidant-induced neuronal death in a cortical cell culture system.
Methods—Murine cortical cultures were exposed to oxidative stress generated by hydrogen peroxide and by a mixture of copper plus ascorbic acid. We examined the ability of HSA and a tetrapeptide occupying its N-terminus (DAHK) to prevent neuronal death after these challenges.
Results—H2O2 and CuCl2/ascorbic acid were used at concentrations that, in the absence of HSA, killed ⬎90% of the
neurons. HSA provided complete protection at a concentration of 37.5mol/L and 50% protection at 3.75 mol/L. The copper-binding tetrapeptide DAHK had nearly identical potency and efficacy. HSA and DAHK were also equally effective in preventing neuronal death induced by CuCl2/ascorbic acid.
Conclusions—HSA has potent antioxidant properties, probably due to binding of copper and other transition metals. HSA extravasation into ischemic brain may provide neuroprotection by limiting metal-catalyzed oxidant stress. The tetrapeptide DAHK may be an effective, small-molecular-weight alternative to HSA as a therapeutic agent for stroke. (Stroke. 2004;35:590-595.)
Key Words: albumins 䡲 antioxidants 䡲 copper 䡲 ischemia 䡲 neurons
A
recent series of studies has explored the efficacy of human serum albumin (HSA) as a therapeutic agent in experimental models of stroke.1–7 HSA was administered intravenously to treat permanent or transient cerebral ische-mia and was more effective in transient ischeische-mia. HSA significantly reduced infarct size, edema, and sodium accu-mulation and improved neurological outcome, even when administered up to 2 hours after onset of ischemia. While these studies demonstrated significant efficacy for HSA in the treatment of stroke, the mechanism for this robust neuropro-tection remains undetermined. A number of possible mecha-nisms have been examined, including the effect of HSA on local cerebral perfusion, blood-brain barrier disruption, sys-temic fatty acid responses, and microvascular patency. While many of these mechanisms probably contribute, none appear robust enough to account for the large neuroprotective effect of HSA.2–5,8,9HSA is a unique molecule. It maintains colloidal osmotic pressure in the vasculature and has a number of important functional properties.10 It strongly binds fatty acids, some
drugs, and drug metabolites, and it has a number of cation and anion binding sites.8,10 HSA is also a potent antioxidant, acting both as a free radical scavenger and as a chelator of transition metals and heme.10A metal binding site on HSA, the 4 –amino acid terminal sequence of the molecule, is unique to human albumin and may play a prominent role in its neuroprotective effects.11–15
The 4 amino acids occupying the N-terminus of HSA— aspartate, alanine, histidine, and lysine (DAHK)— constitute a relatively high-affinity binding site for a number of cations, specifically nickel, cobalt, and copper.11In recent series of reports, it was demonstrated that a synthetic peptide com-posed of these 4 amino acids could inhibit copper-induced oxidative DNA double-strand breaks and telomere shortening in cell cultures.12 It was also determined that DAHK pre-vented lipid oxidation in a copper-catalyzed oxidant system.13 The protective effects of DAHK may also be due in part to the fact that, in addition to acting as a copper chelator, the copper-DAHK complex is a potent superoxide dismutase mimetic, thereby increasing the antioxidant potential of the
Received July 23, 2003; final revision received September 24, 2003; accepted October 14, 2003.
From the Departments of Neurology (E.T.G., R.A.S., C.A.) and Neurosurgery (J.L., S.H., P.R.W., S.S.P.), San Francisco Veterans Affairs Medical Center and University of California at San Francisco.
Correspondence to Dr S. Scott Panter, Veterans Affairs Medical Center, Neurology (127), 4150 Clement St, San Francisco, CA 94121. E-mail sspanter@hotmail.com
© 2004 American Heart Association, Inc.
Stroke is available at http://www.strokeaha.org DOI: 10.1161/01.STR.0000110790.05859.DA
590
tetrapeptide or HSA itself.13It was recently demonstrated that a novel synthetic analogue of DAHK could improve recovery of rat hearts from an ischemia/reperfusion insult,16suggesting that this approach may similarly be useful in the treatment of stroke.
We hypothesized that the antioxidant properties of this 4 –amino acid terminus of HSA might explain, in part, its neuroprotective effects after ischemia/reperfusion injury in brain. To test this hypothesis, we utilized HSA and DAHK in a neuronal cell culture model of oxidative injury. Both HSA and DAHK were neuroprotective in cell culture models of hydrogen peroxide– and copper ascorbate–induced neurotoxicity.
Materials and Methods
HSA was obtained from Centeon L.L.C. as a 25% solution, and DAHK was obtained from Genemed Synthesis Incorporated. All other reagents were obtained from Sigma-Aldrich, except where otherwise noted. Adventitious metals were removed from HSA by applying it to a column of iminodiacetic acid attached to cross-linked polystyrene, after which it was reconcentrated to a 25% solution. Cell Cultures
The studies were conducted in accordance with National Institutes of Health guidelines and with the use of protocols approved by the local institutional committee on animal studies. Mice were anesthetized with isoflurane before being killed for harvesting of brains. Cortical cell cultures were prepared in a 2-step process, which has previously been described in detail.17,18Astrocytes were plated at 1.5⫻105cells
per well. Neurons were plated on the astrocyte layers at a density of 6⫻105cells per well.
Experimental Procedures
Experiments were begun by replacing the culture medium with a balanced salt solution (BSS), as previously described.17,18The pH of
the BSS was adjusted to pH 7.2 and during equilibration in a 5% CO2
atmosphere. Osmolarity was measured with a Wescor vapor pressure osmometer and adjusted with H2O or NaCl when necessary to
achieve 280 to 320 mOsm. Test compounds were prepared as⫻100 stock solutions in distilled deionized water and were diluted to working concentrations in BSS before use. The test compounds were added to the cultures in BSS, and the cultures were then replaced in the 37°C, 5% CO2incubator. In studies in which H2O2was used, the
medium was replaced with BSS after 60 minutes of H2O2exposure.
All other treatment combinations were maintained for 20 to 22 hours. Control wells received only medium exchanges. In each experiment, all comparisons were made with the use of sister cultures derived from single plating.
Assessment of Neuronal Survival
Neuronal survival was assessed by the propidium iodide exclusion method.18,19 Propidium iodide was added at 0.03 mg/mL to each
well. Dead (fluorescent) and live (nonfluorescent) neurons were counted in 4 optical fields chosen randomly in each well, with the use of a Nikon fluorescence microscope with phase-contrast optics. Neurons were easily distinguished from the underlying astrocyte layer by their phase-bright, process-bearing morphology (Figure 3). Results from each well were expressed as percent neuronal survival, calculated as (live cells⫻100)/(live cells⫹dead cells). In a subset of the experiments, cell counts were also performed by a second observer in a blinded fashion to exclude observer bias.
Statistical Analysis
Data are presented as mean⫾SEM. Statistical significance was assessed with the use of 1-way ANOVA followed by the Dunnett post hoc test for multiple comparisons against a control group. Differences were considered significant at P⬍0.05.
Results
To determine the antioxidant capacity of HSA and DAHK, we first used H2O2 to generate oxidative stress. Cortical cultures were exposed for 1 hour to a range of H2O2 concentrations, and neuronal survival was assessed by pro-pidium iodide staining after an additional 22 to 24 hours in BSS. The resulting concentration-response curve showed a threshold effect on neuronal survival (Figure 1), similar to previous reports.18,20Exposure to 100 mol/L caused near-complete neuronal death, and this concentration was used for all subsequent studies.
We evaluated the efficacy of HSA as a neuroprotective agent by adding various concentrations of HSA to the culture simultaneously with H2O2. HSA had a dose-dependent neu-roprotective effect, with neuronal death reduced to a level comparable to control conditions at concentrations of ⱖ15
mol/L (Figure 2A). To determine whether the chelating
tetrapeptide DAHK would also prevent neuronal death result-ing from H2O2, cultures were incubated with 100 mol/L H2O2 in the presence of a range of DAHK concentrations. These studies showed a potent, dose-dependent effect of DAHK on H2O2-induced neuronal death (Figure 2B). Pho-tomicrographs showing the effect of HSA on H2O2-induced neuronal death are shown in Figure 3.
We tested the ability of HSA to protect neurons against the mixture of 25 mol/L CuCl2 and 50mol/L ascorbic acid (Figure 4), which in oxygenated solutions generates oxygen-derived free radical species.21,22In the absence of HSA, this exposure killed⬎95% of the neurons, but in the presence of
150 mol/L HSA neuronal death was reduced to control
values. HSA has a high-affinity binding site for copper and other transition metals at its N-terminus in the form of a DAHK tetrapeptide.21–23Synthetic DAHK tetrapeptide alone also completely blocked Cu/ascorbic acid– dependent neuro-toxicity when added at concentrations of ⱖ50 mol/L, although the potency of DAHK was⬎100-fold greater than HSA when the concentrations of both were expressed as
Figure 1. Hydrogen peroxide produced neuronal death in a
dose-dependent manner. The cultures were treated with H2O2
for 60 minutes. Twenty-four hours after treatment, neuronal death was assessed by propidium iodide staining. n⫽16 to 20, pooled from 5 independent experiments. **P⬍0.001.
percent weight/volume. H2O2 requires interaction with a transition metal to produce reactive oxygen species.24
The antioxidant effect of HSA could, in principle, occur in either the extracellular or the intracellular compartments, since HSA can be internalized by neurons under some conditions.25As a possible way to determine the compartment in which the neuroprotective effect of HSA was exerted, we assessed the ability of HSA to prevent N-methyl-D-aspartate (NMDA)–induced neuronal death, a process that is mediated in part by intracellular production of oxygen free radicals and is blocked by cell-permeant oxygen free radical scaven-gers.26,27HSA had no effect on NMDA neurotoxicity. This suggests that the action of HSA occurs in the extracellular space (Figure 5).
To test whether it is the DAHK moiety of HSA that is primarily responsible for its antioxidant effects, we compared HSA with several other proteins that do not contain this tetrapeptide at the N-terminus: lactalbumin, ␥-globulin, bo-vine serum albumin, and casein. Somewhat surprisingly, each of these proteins also protected neurons against H2O2toxicity with potencies roughly similar to that of HSA (Figure 6).
Discussion
These results show that the neurotoxic oxidant stress induced by either hydrogen peroxide or copper/ascorbic acid can be blocked by HSA and by a tetrapeptide that is the same sequence as the N-terminus of HSA. These results may be relevant to the recent series of articles that describe the protective effect of HSA in animal models of stroke and traumatic brain injury.2–5,9
In vivo, the concentration of HSA in cerebrospinal fluid is approximately 3.7 mol/L.28 This concentration of HSA produced an approximately 40% reduction in H2O2-induced neuronal death under the conditions used in the cell culture studies described here. Additional HSA may enter the brain from the plasma compartment after stroke through an open blood-brain barrier.25 Since the concentration of HSA in serum is approximately 588 mol/L,29even a small move-ment of serum proteins across the blood-brain barrier could substantially raise HSA concentrations in the extracellular space surrounding postischemic neurons and increase resis-tance to oxygen free radicals in the extracellular space.
Figure 3. Photomicrographs show the protective effects of HSA
on H2O2neurotoxicity. The cultures were treated simultaneously
with 100mol/L H2O2plus various concentrations of HSA for 60
minutes. Twenty-four hours after treatment, the photomicro-graphs were taken with combined fluorescence and phase-contrast optics. Dead neurons were identified by propidium iodide fluorescence.
Figure 2. A, HSA decreased H2O2-induced neuronal death. The
cultures were simultaneously treated with 100mol/L H2O2plus
various concentrations of HSA for 60 minutes. Twenty-four hours after treatment, neuronal death was assessed by pro-pidium iodide staining. n⫽24 to 30, pooled from 6 independent experiments. **P⬍0.01. B, DAHK decreased H2O2-induced
neu-ronal death. Cultures were exposed to H2O2plus various
con-centrations of DAHK for 60 minutes. Neuronal death was assessed 24 hours after treatment. n⫽16 to 20, pooled from 4 independent experiments. **P⬍0.01.
For the purpose of testing the neuroprotective effects of HSA and DAHK in cultured neurons, 2 different models of oxidant-mediated neuronal injury were developed. The first uses hydrogen peroxide as the stressor, and the second uses a copper/ascorbic acid– driven free radical– generating system. Both oxidants were used under conditions that, in the absence of HSA or DAHK, killed nearly 100% of the neurons in the cultures. Copper-ascorbic acid– driven stress involves a tran-sition metal, and it is almost certain that the hydrogen
peroxide neurotoxicity also requires transition metals. In the absence of metals, hydrogen peroxide is extremely stable, with a calculated half-life for its uncatalyzed unimolecular homolysis at 30°C of 1011 years.24 H
2O2 interaction with organic molecules requires transition metals, and chelation of transition metals prevents this interaction.
Transition metals (in particular, iron and copper) are capable of cycling between their reduced and oxidized states and, in the process, generating an electron that can create a free radical. In many cases, transition metals themselves may be bound by a lipid, protein, or nucleic acid molecule, and a
Figure 4. DAHK decreased CuCl2/ascorbic acid–induced
neuro-nal death in a dose-dependent manner. Cultures were exposed to CuCl2/ascorbic acid plus various concentrations of DAHK or
HSA for 60 minutes. Neuronal death was assessed 24 hours after treatment. HSA and DAHK produced similar protection when expressed as micromoles per liter. n⫽16 to 20, pooled from 4 independent experiments. **P⬍0.01.
Figure 5. HSA did not decrease NMDA-induced excitotoxicity.
Cocultures were exposed to 3 different concentrations of NMDA for 20 minutes in the presence of 37.5mol/L HSA, which was added 1 hour before the NMDA. At 20 minutes, medium was removed and replaced with fresh medium plus 37.5mol/L HSA. Neuronal death was determined 24 hours later. n⫽16, pooled from 4 independent experiments. **P⬍0.01.
Figure 6. Other proteins also decreased H2O2-induced neuronal
death. The cocultures were exposed simultaneously to 100 mol/L H2O2plus lactalbumin,␥-globulin, bovine serum
albu-min, and casein for 60 minutes. Neuronal death was determined 24 hours later. n⫽12, pooled from 4 independent experiments. **P⬍0.01.
free radical causes site-specific damage at or near its binding site.21–23It is this type of activity that is most likely respon-sible for copper-dependent cell death in culture. If copper and/or iron is decompartmentalized by ischemia/reperfusion, the neurotoxic effects may be accelerated, and the presence of HSA or its N-terminal tetrapeptide may prevent its binding to sites where it could contribute to cellular injury. Alterna-tively, ambient, normal levels of transition metals in the extracellular space may have no deleterious effects under normal conditions but may become highly deleterious in the presence of H2O2formed by the superoxide generated during ischemia/reperfusion.
The N-terminus DAHK of HSA is not the only amino acid sequence that can bind transition metals or copper specifical-ly.30 –33 In the present studies DAHK was found to be equipotent with HSA in preventing H2O2or copper/ascorbic acid neurotoxicity, despite the fact that HSA is 126-fold larger than DAHK. This suggests that the DAHK N-terminal tetrapeptide is the primary locus of HSA antioxidant activity. However, the other proteins examined had effects similar to those of HSA, despite absence of the DAHK domain. It is possible that relatively weak interactions between sulfhydryl groups and/or amino acids (particularly tryptophan or histi-dine34,35) on these proteins and transition metals produce a chelating effect that, in aggregate, is comparable to that achieved with HSA or DAHK. The relatively high-affinity, nonspecific binding of copper to proteins has been previously reported to inhibit its capacity to generate hydroxyl radi-cals,34,35 which may explain in part the neuroprotective effects of proteins other than HSA. Regardless of the mech-anism of this effect, these results suggest that other proteins may also contribute to brain antioxidant effects during blood-brain barrier breakdown. This possibility has not been tested in vivo.
In summary, we have demonstrated that HSA and its N-terminal tetrapeptide DAHK can block oxidant-driven neuronal injury produced with the use of 2 different oxidant-generating systems: hydrogen peroxide and copper/ascorbic acid. The fact that the peptide and HSA can block the neurotoxicity of the latter generating system in a stoichiomet-ric fashion implies that they are both binding copper and can stop its redox cycling. The efficacy of DAHK in the hydrogen peroxide– driven system suggests that copper is also involved in oxidant-driven neurotoxicity in vitro and may be involved in tissue injury after ischemia and reperfusion in vivo. DAHK may be a useful alternative to HSA for the treatment of stroke.
Acknowledgments
This work was supported by a Department of Veterans Affairs REAP grant and National Institutes of Health grant RO1-HL53040 (to Dr Panter). We gratefully acknowledge the technical assistance of Angelo Zegna and the advice of Dr Weihai Ying.
References
1. Belayev L, Khoutorova L, Deisher TA, Belayev A, Busto R, Zhang Y, Zhao W, Ginsberg MD. Neuroprotective effect of SolCD39, a novel platelet aggregation inhibitor, on transient middle cerebral artery occlusion in rats. Stroke. 2003;34:758 –763.
2. Belayev L, Zhao W, Pattany PM, Weaver RG, Huh PW, Lin B, Busto R, Ginsberg MD. Diffusion-weighted magnetic resonance imaging confirms marked neuroprotective efficacy of albumin therapy in focal cerebral ischemia. Stroke. 1998;29:2587–2599.
3. Belayev L, Saul I, Huh PW, Finotti N, Zhao W, Busto R, Ginsberg MD. Neuroprotective effect of high-dose albumin therapy against global is-chemic brain injury in rats. Brain Res. 1999;845:107–111.
4. Belayev L, Liu Y, Zhao W, Busto R, Ginsberg MD. Human albumin therapy of acute ischemic stroke: marked neuroprotective efficacy at moderate doses and with a broad therapeutic window. Stroke. 2001;32: 553–560.
5. Belayev L, Pinard E, Nallet H, Seylaz J, Liu Y, Riyamongkol P, Zhao W, Busto R, Ginsberg MD. Albumin therapy of transient focal cerebral ischemia: in vivo analysis of dynamic microvascular responses. Stroke. 2002;33:1077–1084.
6. Ginsberg MD. Adventures in the pathophysiology of brain ischemia: penumbra, gene expression, neuroprotection: the 2002 Thomas Willis lecture. Stroke. 2003;34:214 –223.
7. Matsui T, Sinyama H, Asano T. Beneficial effect of prolonged adminis-tration of albumin on ischemic cerebral edema and infarction after occlusion of middle cerebral artery in rats. Neurosurgery. 1993;33: 293–300; comment 300.
8. Rodriguez de Turco EB, Belayev L, Liu Y, Busto R, Parkins N, Bazan NG, Ginsberg MD. Systemic fatty acid responses to transient focal cerebral ischemia: influence of neuroprotectant therapy with human al-bumin. J Neurochem. 2002;83:515–524.
9. Liu Y, Belayev L, Zhao W, Busto R, Belayev A, Ginsberg MD. Neuro-protective effect of treatment with human albumin in permanent focal cerebral ischemia: histopathology and cortical perfusion studies. Eur J Pharmacol. 2001;428:193–201.
10. Emerson TE Jr. Unique features of albumin: a brief review. Crit Care Med. 1989;17:690 – 694.
11. Bar-Or D, Curtis G, Rao N, Bampos N, Lau E. Characterization of the CO(2⫹) and Ni(2⫹) binding amino-acid residues of the N-terminus of human albumin: an insight into the mechanism of a new assay for myocardial ischemia. Eur J Biochem. 2001;268:42– 47.
12. Bar-Or D, Thomas GW, Rael LT, Lau EP, Winkler JV. Asp-Ala-His-Lys (DAHK) inhibits copper-induced oxidative DNA double strand breaks and telomere shortening. Biochem Biophys Res Commun. 2001;282: 356 –360.
13. Bar-Or D, Rael LT, Lau EP, Rao NK, Thomas GW, Winkler JV, Yukl RL, Kingston RG, Curtis CG. An analog of the human albumin N-terminus (Asp-Ala-His-Lys) prevents formation of copper-induced reactive oxygen species. Biochem Biophys Res Commun. 2001;284: 856 – 862.
14. Bar-Or D, Winkler JV, Vanbenthuysen K, Harris L, Lau E, Hetzel FW. Reduced albumin-cobalt binding with transient myocardial ischemia after elective percutaneous transluminal coronary angioplasty: a preliminary comparison to creatine kinase-MB, myoglobin, and troponin I. Am Heart J. 2001;141:985–991.
15. Bar-Or D, Rael LT, Winkler JV, Yukl RL, Thomas GW, Shimonkevitz RP. Copper inhibits activated protein C: Protective effect of human albumin and an analogue of its high-affinity copper-binding site, D-DAHK. Biochem Biophys Res Commun. 2002;290:1388 –1392. 16. Sutherland FJ, Bar-Or D, Curtis CG, Hearse DJ. Attenuation of trace
element-mediated injury during ischemia and reperfusion by an N-terminus analogue of human albumin (h4dus60131). J Cardiovasc Pharmacol. 2002;39:722–728.
17. Anderson CM, Norquist BA, Vesce S, Nicholls DG, Soine WH, Duan S, Swanson RA. Barbiturates induce mitochondrial depolarization and potentiate excitotoxic neuronal death. J Neurosci. 2002;22:9203–9209. 18. Ying W, Han SK, Miller JW, Swanson RA. Acidosis potentiates
oxi-dative neuronal death by multiple mechanisms. J Neurochem. 1999;73: 1549 –1556.
19. Edidin M. A rapid, quantitative fluorescence assay for cell damage by cytotoxic antibodies. J Immunol. 1970;104:1303–1306.
20. Mischel RE, Kim YS, Sheldon RA, Ferriero DM. Hydrogen peroxide is selectively toxic to immature murine neurons in vitro. Neurosci Lett. 1997;231:17–20.
21. Lovstad RA. A kinetic study on the copper-albumin catalyzed oxidation of ascorbate. Biometals. 2002;15:351–355.
22. Simpson JA, Cheeseman KH, Smith SE, Dean RT. Free-radical gen-eration by copper ions and hydrogen peroxide: stimulation by HEPES buffer. Biochem J. 1988;254:519 –523.
23. Frenkel K, Blum F, Troll W. Copper ions and hydrogen peroxide form hypochlorite from NaCl thereby mimicking myeloperoxidase. J Cell Biochem. 1986;30:181–193.
24. Tien M, Aust SD. Comparative aspects of several model lipid peroxi-dation systems. In: Yagi K, ed. Lipid Peroxides in Biology and Medicine. New York, NY: Academic Press; 1982:23–39.
25. Remmers M, Schmidt-Kastner R, Belayev L, Lin B, Busto R, Ginsberg MD. Protein extravasation and cellular uptake after high-dose human-al-bumin treatment of transient focal cerebral ischemia in rats. Brain Res. 1999;827:237–242.
26. Dugan LL, Gabrielsen JK, Yu SP, Lin TS, Choi DW. Buckminsterfullerenol free radical scavengers reduce excitotoxic and apoptotic death of cultured cortical neurons. Neurobiol Dis. 1996;3:129–135.
27. Lipton P. Ischemic cell death in brain neurons. Physiol Rev. 1999;79: 1431–1568.
28. Reiber H. Dynamics of brain-derived proteins in cerebrospinal fluid. Clin Chim Acta. 2001;310:173–186.
29. Scientific Tables. Ardsley, NY: Geigy Pharmaceuticals Division of Ciba-Geigy Corporation; 1970.
30. Chruscinska E, Dyba M, Micera G, Ambroziak W, Olczak J, Zabrocki J, Kozlowski H. Binding ability of Cu2⫹ ions by opiate-like fragments of bovine casein. J Inorg Biochem. 1997;66:19 –22.
31. Predki PF, Harford C, Brar P, Sarkar B. Further characterization of the N-terminal copper(II)- and nickel(II)-binding motif of proteins: studies of metal binding to chicken serum albumin and the native sequence peptide. Biochem J. 1992;287(pt 1):211–215.
32. Tieghem E, Van Dael H, Van Cauwelaert F. Comparison of the Cu2⫹ binding to bovine, goat and human alpha-lactalbumin. Biochem Int. 1991; 23:119 –126.
33. Zamora PO, Mercer-Smith JA, Marek MJ, Schulte LD, Rhodes BA. Similarity of copper and technetium binding sites in human IgG. Int J Rad Appl Instrum B. 1992;19:797– 802.
34. Gutteridge JM, Wilkins S. Copper salt-dependent hydroxyl radical for-mation: damage to proteins acting as antioxidants. Biochim Biophys Acta. 1983;759:38 – 41.
35. Rowley DA, Halliwell B. Superoxide-dependent and ascorbate-dependent formation of hydroxyl radicals in the presence of copper salts: a physi-ologically significant reaction? Arch Biochem Biophys. 1983;225: 279 –284.
