Melatonin protects against copper-mediatedfree radical damage

Introduction

Copper is involved in the neuropathology of variousneurodegenerative disorders including Wilson’s, Menke’s,

Parkinson’s, Alzheimer’s, amyotrophic lateral sclerosis, andPrion diseases [1, 2]. Copper overload results in manypathological conditions that are consistent with oxidative

damage to membranes or molecules. Cu1+ is known togenerate free radicals by a Fenton type reaction. Copperions are able to catalyze the formation of hydroxyl radicalsvia the Haber-Weiss reaction (in vitro):

O�2 � þCu2þ ! O2 þ Cu1þ ð1Þ

Cu1þ þ H2O2 ! Cu2þ þ OH� þ �OH ð2Þ

Cupric chloride can also react with hydrogen peroxide toyield Cu1+ and the superoxide radical (O2

)Æ). Cu1+ formed

can then react with hydrogen peroxide (H2O2) according toEquation (2), increasing the potential for free radicaldamage.

Lipid peroxidation in biological membranes causes

impairment of membrane functioning, decreased fluidity,inactivation of membrane-bound receptors and enzymes,and an increased non-specific permeability to ions such as

Ca2+. Free, ligand-bound and vesicular copper is presentthroughout the brain. Copper is a rapidly acting neurotoxinat physiological concentrations (10–100 lM) [3]. Upon

exposure to iron or copper ions, lipid hydroperoxidesdecompose, resulting in hydrocarbon gases, for example,ethane and pentane, radicals that can abstract further

hydrogen atoms from fatty acid side chains and cytotoxiccarbonyl compounds [4].

Lipid peroxidation is often a late event, accompanyingrather than causing final cell death [5]. Thus, a failure toprotect against tissue damage by chain-breaking antioxid-

ant inhibitors of lipid peroxidation does not rule out freeradical damage as an injury mechanism [6].

Transition metals promote lipid peroxidation in two

ways (1): by catalyzing the formation of oxygen free radicalspecies capable of initiating lipid peroxidation and (2) bycatalyzing the decomposition of preformed lipid peroxidesto propagate lipid peroxidation [7]. Elevated copper levels

induce a variety of changes in tissues. In the liver, copperoverload results in hypertrophy of hepatocytes, hepatitis,hepatocellular necrosis and eventually hepatocellular death

[8].Melatonin, the principle secretory product of the pineal

gland is a potent free radical scavenger and general

antioxidant [9]. It efficiently scavenges the hydroxyl radical(ÆOH) [10] and possibly the peroxyl radical (LOOÆ) [11].Utilizing adsorptive cathodic stripping voltammetry, anelectrochemical method which has been used with success

to examine metal–ligand complex formation at an elec-trode, Limson et al. reported that melatonin binds heavymetals including Fe3+, Cu2+, Pb2+, Cd2+ and Zn2+ [12].

This implies that melatonin, besides acting as an antioxid-ant, may bind these metals and prevent them frompartaking in free radical generation.

In vivo studies have shown that melatonin offers neuro-protection against mercury [Hg(II)] toxicity in rats [13].

Parmar P, Limson J, Nyokong T, Daya S. Melatonin protects against copper-

mediated free radical damage. J. Pineal Res. 2002; 32:237–242. � Blackwell

Munksgaard, 2002

Abstract: Copper is an essential trace element which forms an integral

component of many enzymes. While trace amounts of copper are needed to

sustain life, excess copper is extremely toxic. Copper has been implicated in

various neurodegenerative disorders, such as Wilson’s and Alzheimer’s

diseases. Previous studies showed that melatonin, the principle secretory

product of the pineal gland, binds Cupric chloride (Cu2+) and that this may

have implications in copper-induced neurodegenerative diseases. In the

present study, in vitro copper-mediated lipid peroxidation was induced.

Melatonin (5 mM) protected against copper-mediated lipid peroxidation in

liver homogenates. Electron micrographs of in vivo administered Cu2+ and

melatonin show that melatonin affords some protection to rat hepatocytes in

the presence of copper. Electrochemical studies performed show that

melatonin, in addition to binding Cu2+, may provide protection against

copper-mediated free radical damage by binding Cu1+. The findings of these

studies provide further evidence for the neuroprotective role of melatonin.

Paresh Parmar1, Janice Limson1,Tebello Nyokong2 and SantyDaya1

1Faculty of Pharmacy and 2Department of

Chemistry, Rhodes University, Box 94,

Grahamstown, South Africa

Keywords: copper, lipid peroxidation,

melatonin, voltammetry

Address reprint requests to Santy Daya,

Faculty of Pharmacy, Rhodes University, Box

94, Grahamstown, 6140 South Africa.

E-mail: s.daya@ru.ac.za

Received August 1, 2001;

accepted October 24, 2001.

J. Pineal Res. 2002; 32:237–242 Copyright � Blackwell Munksgaard, 2002

Journal of Pineal ResearchISSN 0742-3098

237

Melatonin may offer this protection by forming a metal–ligand complex with mercury. Further electrochemicalstudies have also shown an interaction of lithium and

aluminium with melatonin [14]. These findings suggest anew role for melatonin in sequestering and removing toxicmetals from the central nervous system. Since melatonin issuggested to bind Fe3+, melatonin may function as an

antioxidant by binding this metal ion. In the present study,electrochemical methods were employed to gauge theaffinity of melatonin for the reduced form of copper,

Cu1+. As glutathione is known to prevent copper toxicityby binding Cu1+, adsorptive stripping experiments werealso conducted with glutathione.

Materials and methods

Chemicals and reagents

Hexafluorophosphoric acid (HPF6), butylated hydroxy-

toluene (BHT), melatonin, bovine serum albumin (BSA)and 2-thiobarbituric acid (TBA) were purchased fromSigma Chemical Co., St Louis, MO, USA. Cupric chloride

(Cu2+), cuprous oxide (Cu1+), glutaraldehyde, sodiumphosphate, osmium tetroxide, trichloroacetic acid (TCA),Folin and Ciocalteu’s reagent and copper chloride (CuCl2)

were purchased from Saarchem (Pty) Ltd, Krugersdorp,South Africa. Tetrakis (acetonitrile) copper (I) hexafluoro-phosphate [Cu(CH3CN)4]PF6 was synthesized from Cu2O

and HPF6 [15]. 1.1.3.3-Tetramethoxypropane (MDA) waspurchased from Fluka AG, Switzerland. Cu(I) solution(1 mM) was prepared by dissolving [Cu(CH3CN)4]PF6 in5 mL 0.2% HNO3 before being made to volume with

water. Melatonin (Sigma) was dissolved in 95% ethanol.Glutathione (Sigma) solutions were prepared daily withwater. All aqueous solutions were prepared from Millipore

water. Sodium acetate buffer (0.2 M) of pH 4.4 wasprepared from stock solutions of 0.2 M acetic acid and0.2 M sodium acetate and used as the electrolyte solution

for adsorptive stripping voltammograms. 0.2 M Tris-HClbuffer (pH 7.3) was used as the electrolyte in all cyclicvoltammetry experiments. All other reagents were obtainedfrom local sources and were of the highest purity available.

In vitro lipid peroxidation

Male Wistar rats weighing 250–300 g were used in theexperiments. The rats were killed by cervical dislocationand rapidly decapitated. The liver was removed and a 10%

w/v homogenate was made with 0.1 M phosphate bufferedsaline (PBS) pH 7.4. The experiment included three groups,viz. a control group, a copper group and a copper/

melatonin group. The control group consisted of 0.9 mLhomogenate and 0.1 mL of drug vehicle (50:50 eth-anol:water). A stock solution of copper chloride (2 mM)was prepared. In the copper group, 0.1 mL of the copper

stock solution was added, plus 0.1 mL distilled water to0.8 mL homogenate to give a final copper concentration of1 mM. The copper/melatonin group consisted of 0.8 mL

homogenate, plus 0.1 mL copper of the various copperstock solutions and 0.1 mL melatonin (10 mM). This gave afinal melatonin concentration of 5 mM. Melatonin was

dissolved in 50:50 ethanol:water.

The lipid peroxidation assay was performed according tothe modified method of Anoopkumar-Dukie et al. [16].Aliquots of 1 mL of the homogenate were incubated at

37�C for 60 min. 0.5 mL BHT (0.5 mg/mL in methanol)and 1 mL TCA (15% w/v in water) was added to themixture. The tubes were sealed and heated for 10 min in aboiling water bath to release protein-bound MDA. The

samples were then cooled to avoid adsorption of MDAonto insoluble proteins and then centrifuged at 2000 g for15 min. Following centrifugation, 1 mL of this protein-free

supernatant was removed from each tube and a 1-mLaliquot of TBA (0.33% w/v in water) was added to thisfraction. The tubes were sealed and incubated in a boiling

water bath for 60 min. After incubation, the mixture wascooled for 10 min on ice. The TBA-MDA complex wasseparated from other possible interfering thiobarbituricacid-reactive substances (TBARS) using an IsoluteTM C18

solid phase extraction (SPE) column. The SPE column wasprewashed with 2 mL of methanol followed by 2 mL ofdistilled water. One milliliter of the sample was loaded onto

the column, which was subsequently washed with 1 mL ofdistilled water. The TBA–MDA complex was eluted with1 mL methanol. The absorbance was measured at 532 nm

with an UV-visible spectrophotometer (Shimadzu). Meth-anol was used as a blank. An MDA standard curve wasgenerated using 1,1,3,3-tetramethoxypropane as described

above. Standards were dissolved in 1 mL of distilled water,and a series (0–20 nmoles/mL) were prepared. Final resultswere expressed as nmoles/mg protein.

Protein estimation in the liver homogenates was deter-

mined using described methods [17].

Transmission electron microscopy

Male Wistar rats weighing 250–300 g were used in theexperiments. The rats were randomly assigned to one of the

three groups of five rats each, and were housed in acontrolled environment with a 12-hr light:dark cycle; theywere given access to standard laboratory chow and waterad libitum. The control group received the drug vehicle, i.e.

ethanol: 0.9% saline (40:60) for 2 wk. The copper treatedgroup received 2 mg/kg Cu2+ (copper chloride – CuCl2)for 2 wk. The copper/melatonin treated groups received

Cu2+ (2 mg/kg) and melatonin (12 mg/kg) for 2 wk. Therats were injected daily, with either the vehicle, Cu2+ orCu2+/melatonin. Copper/melatonin groups were injected

on different intraperitoneal sites to prevent any interactionof the copper with the melatonin. Protocols for theexperiments were approved by the Rhodes University

Ethics Committee. The rats were killed and tissue samplesof liver were placed in liquid nitrogen and stored in a )70�Cfreezer until used. The tissue pieces were cut using a scalpelinto pieces approximately 2 mm3 and placed in buffered

glutaraldehyde (2.5% glutaraldehyde in 0.1 M sodiumphosphate buffer) for 12–24 hr at 4�C. The glutaraldehydewas replaced with cold buffer and allowed to wash for

10 min. This was repeated twice. The final buffer wash wasfollowed by fixation in a 1% solution of osmium tetroxidein 0.1 M sodium phosphate buffer. After 60–90 min, the

tissue pieces were washed again in two changes of buffer(10 min each). The buffer washes were followed by

Parmar et al.

238

dehydration carried out by immersing the tissue pieces inascending concentrations of ethanol (30, 50, 70, 80, 90 and100% ethanol) for 3–5 min in each solution. A second

change of 100% ethanol was replaced by propylene oxide, atransitional solvent, and allowed to infiltrate for 15 min,after which the propylene oxide was changed. After afurther 15 min, resin infiltration was carried out by

replacing the propylene oxide by solutions containingincreasing concentrations of resin (TAAB 812 and Aralditemixture) propylene oxide. Three mixtures were used (25:75,

50:50, 75:25) and the infiltration carried out for 60–90 minin each mixture. The final mixture was replaced by pureresin and further infiltration allowed to take place over-

night. The tissue pieces were then transferred to mouldscontaining pure resin and allowed to polymerize at 60�C for36 hr. The solidified samples were removed and placed inlabeled tubes until use. The sample blocks were trimmed

and sectioned with an RMC MT-7 ultramicrotome forexamination. The sections were viewed using a Joel JEM1210 transmission electron microscope.

Electrochemistry

Cyclic and stripping voltammograms were obtained withthe Bio Analytical Systems (BAS) CV-50 W voltammetricanalyzer using a BAS C2 cell stand to maintain constant

atmosphere. An undivided cell was employed in all elect-rochemical experiments. A 3-mm diameter glassy carbonelectrode (GCE) was employed as a working electrode forcyclic voltammograms. A hanging mercury drop electrode

(HMDE) was employed as the working electrode in allcathodic adsorptive stripping voltammograms. A silver/silver chloride ([KCl ¼ 3 mol dm)3]) and a platinum wire

were employed as reference and auxiliary electrodes,respectively, in all electrochemical work. Prior to use,GCE was cleaned by polishing with alumina on a Buehler

pad, followed by washing in nitric acid and rinsing in waterand the buffer solution. Between scans, the GCE wascleaned by immersion in a dilute acid solution and rinsedwith water.

For cyclic voltammetric experiments, appropriate con-centrations of Cu1+ and melatonin in pH 7.3 Tris-HClbuffer were introduced into the glass cell, degassed for

5 min with nitrogen before scanning a potential window.For adsorptive stripping experiments, sodium acetate

buffer (pH ¼ 4.4) and appropriate concentrations of Cu1+

and of the ligands (melatonin or glutathione) were intro-duced into an electrochemical cell. The solution was thendeaerated with nitrogen for 5 min, after which a flow of

nitrogen was maintained over the solution throughout themeasurement. The metal complexes of melatonin or gluta-thione are expected to have formed at this stage. Anoptimum deposition potential of 150 mV versus Ag/AgCl

was applied for 60 s to effect the formation and adsorptionof the metal complex onto the HMDE. The voltammogramswere then scanned in the negative direction from 150 mV

versus Ag/AgCl to)300 mV versus Ag/AgCl at the scan rateof 100 mV/s to strip the adsorbed metal–ligand complexfrom the electrode. During the stripping step, current

response caused by the reduction of the metal complexesof melatonin or glutathione, were measured as a function of

potential. All potential values quoted are referenced againstthe silver/silver chloride reference electrode.

Results

As shown in Fig. 1, copper chloride (1 mM) produces asignificant increase in lipid peroxidation in the liver

(P < 0.001), as compared with the control liver group.The copper and copper/melatonin liver groups are signifi-cantly different (P < 0.001), indicating that melatonin

reduces the degree of copper-induced lipid peroxidation.The 2 wk experiment with 2 mg/kg Cu showed extensive

hepatocyte damage. Fig. 2A–C illustrate the control,

copper and copper/melatonin samples, respectively. InFig. 2A the nucleus is spherical, with normal mitochondriaand extensive glycogen. The sample from the copper-treated animals (Fig. 2B), exhibits damaged mitochondria

and a creatinated nucleus. The chromatin in the nucleus iscondensed, and there is less noticeable glycogen. Fig. 2Cshows a copper/melatonin treated sample. Here less dam-

aged mitochondria are noticeable when compared with thecopper sample (Fig. 2B). The nucleus is spherical and issimilar in appearance to the control nucleus (Fig. 2A).

The CV of melatonin in the absence of Cu1+ shows apeak as a result of the oxidation of melatonin at 0.73 Vversus Ag/AgCl (Fig. 3A). Upon several successive scans, a

quasi-reversible couple appears, which may be related tothe adsorption of melatonin at the electrode surface.

The CV of Cu1+ in the absence of melatonin shows anirreversible oxidation peak at 0.24 V versus Ag/AgCl.

Upon several successive scans, this peak does not showany significant increases indicating that it is not beingelectrodeposited at the electrode.

Cyclic voltammetry is an electrochemical method forcharacterizing species in solution. Species in solution tendto have characteristic oxidation/reduction patterns. Chan-

ges in the redox potential of these species indicate that thespecies has been altered, for example, by forming a complexwith another species introduced to the solution. Fig. 3Ashows the CV obtained for a solution of melatonin alone.

Upon addition of Cu1+, the melatonin peak shifts from0.73 to 0.87 V without much change in current intensity(Fig. 3B). The potential shift from 0.73 to 0.87 V for

Fig. 1. In vitro liver copper (1 mM) and melatonin (5 mM)experiments. *The control group (A) versus copper (B) is statisti-cally significant at P < 0.001. *The copper group (B) versus cop-per-melatonin group (C) is statistically significant at P < 0.001.

Melatonin protects against copper-mediated free radical damage

239

melatonin indicates that melatonin has formed a newspecies with copper. As seen in Fig. 3C, increasing theCu1+ concentration resulted in an increase in currentresponse of the melatonin peak and an enhancement of the

return peak. The weak peak observed for Cu1+ in solutionis absent in the presence of melatonin.

Upon several successive scans of a solution of melatoninand Cu1+ we observe the appearance of a new quasi-

reversible couple at 0.34 V versus Ag/AgCl (Fig. 4). Withincreasing scan number these peaks increase in strength,while the peak caused by melatonin oxidation at 0.87 V

versus Ag/AgCl decreases. This indicates the formation of anew compound electrodeposited at the electrode. Thedecrease in melatonin current response and disappearance

of the peak observed for Cu1+ alone and the increase incurrent strength of the couple at 0.34 V versus Ag/AgCl,indicates that both melatonin and Cu1+ are being used to

produce a new product. This electrode was rinsed in Tris-HCl buffer and placed in a fresh Tris-HCl buffer. No peak

Fig. 2. (A) A control rat hepatocyte. The nucleus is spherical andappears normal. There is extensive glycogen (G) present. Themitochondria (M) appear normal. (B) Hepatocyte of a rat treatedwith copper (2 mg/kg daily for 2 weeks). Extensive damage is seenin the mitochondria (M) and the nucleus is creatinated (�). (C) Ahepatocyte of a copper/melatonin treated rat in the 2 weekexperiment. There are some damaged (fi) pleomorphic mito-chondria (M), with a normal appearing nucleus. Lysosomes (�)are irregularly shaped and dense. Glycogen is present, ·5000.

Fig. 3. Cyclic voltammograms of (A) melatonin (1 · 10)5M) in

buffer in the absence of Cu1+; (B) solution of melatonin(1 · 10)5

M) and Cu1+ (1 · 10)5M) showing a shift in oxidation

potential for melatonin from 0.73 to 0.87 V versus Ag/AgCl in thepresence of Cu1+; (C) further increases in Cu1+ (3 · 10)5

M)results in a decrease in melatonin current.

Fig. 4. Cyclic voltammogram of a solution of Cu1+ (1 · 10)3M)

and melatonin (4 · 10)3M). A new reversible couple is observed at

0.34 V versus Ag/AgCl.

Parmar et al.

240

caused by the oxidation of free melatonin in solution isobserved at 0.87 V versus Ag/AgCl. The redox coupleobserved at 0.071 V versus Ag/AgCl is stable upon multiple

scans of the electrode. The electrode was allowed to dry anda stable blue-purple film was observed at the electrode.

Adsorptive stripping voltammetry has been used withsuccess to examine metal–ligand complex formation at an

electrode. This technique relies on the natural tendency ofanalytes to preconcentrate at the surface of a workingelectrode. Theoretically, the introduction of a successful

ligand to a metal solution will increase preconcentration ofthe metal at the electrode if it forms a metal–ligandcomplex. The increase in preconcentration of the metal as a

metal–ligand complex results in an increase in the currentresponse due to the metal reduction. A shift in thereduction potential further indicates that a new species isbeing reduced. Fig. 5A shows the cathodic adsorptive

stripping voltammogram (CSV) of Cu1+ in the absence ofmelatonin. A peak caused by the reduction of Cu1+ isobserved at 0.02 V versus Ag/AgCl. Fig. 5B shows the CSV

obtained for Cu1+ and melatonin. In the presence ofmelatonin, an increase in the Cu1+ reduction peak and anegative potential shift is observed. Increases in melatonin

concentration increased the current response caused byCu1+ (Figs 5B–E). For 0.04 mM melatonin, a shift to)0.78 V versus Ag/AgCl is observed. The increase in

current strength as well as the potential shift indicates theformation of a melatonin-Cu1+ complex.

As glutathione is known to bind Cu1+, we ran a similarset of experiments for Cu1+ with glutathione. As before

with melatonin, the presence of glutathione resulted in anincrease in current strength for Cu1+ reduction accompan-ied by a strong negative potential shift to )0.91 V versus

Ag/AgCl. An increase in glutathione concentration alsoincreased the observed current response.

Both cyclic voltammetric and adsorptive stripping vol-tammetric experiments indicate strongly that melatoninforms a complex with Cu1+. From these experiments it

appears that the bond is fairly strong but it is not possibleto conclude whether it is simple ligation or whether it ischelation. Cyclic voltammetric experiments indicate that anew product is being formed.

Discussion

Copper is bound in lysosomes within the cell to limit thedamage caused by free copper ions within the cell. Seques-tration of copper into lysosomes is a cytoplasmic adaptive

process, which protects the cell from free, toxic copper [18].Copper within the nucleus interacts with DNA and

results in cellular damage. Creatinated nuclei are a directresult of copper loading. Chromatin condensation occurs as

early as week 2 of copper loading [19]. Both creatinatednuclei and chromatin condensation evident in Fig. 2B arethus consistent with copper overloading.

Rats are rather tolerant to copper loading. Initially, therats receiving copper exhibit all the signs of acute coppertoxicity, but as time progresses, metallothionein (MT) and

ceruloplasmin (Cp) are induced. This aids in intercellularcopper chelation, and decreases the potential of furthercopper toxicity. Liver, kidney and intestines are able to

form MT when faced with initial copper loading [18, 20].On continuous copper exposure, additional protectivemechanisms are activated. The copper-MT complexes areincorporated into lysosomes, which then release the com-

plex into the biliary ducts for excretion. Other chelatingmolecules such as glutathione can remove the copper andprevent its destructive cellular effects [18]. Melatonin may

provide protection against copper toxicity in a similar wayto glutathione. Melatonin, which has been shown tointeract with Cu2+ in vitro [12], possibly interacts with

copper in vivo as well.The in vitro experiments were performed with the

addition of copper and copper/melatonin to liver homo-genate. When liver homogenate was incubated with copper

(Fig. 1) an increase in lipid peroxidation in the copper andcopper/melatonin groups was seen. Copper alone inducedmore lipid peroxidation (P < 0.001), indicating that mela-

tonin affords a degree of protection against copper-inducedlipid peroxidation, at a copper concentration of 1 mM. It islikely that the free radical induced by copper in these

experiments is the hydroxyl radical (via the Fenton reac-tion), as this is sufficiently reactive to initiate lipid perox-idation. Further studies are required to determine the

optimum concentration of melatonin needed to preventcopper-induced lipid peroxidation.

Previous experiments [12] provide evidence that melato-nin forms a metal–ligand bond with Cu2+. In these studies,

electrochemical studies of Cu1+ indicate strongly thatmelatonin interacts with Cu1+ and that a new productmay be formed from this interaction. In-depth studies are in

progress to determine the nature of this new compoundsuggested. In conclusion, it is suggested that melatonin mayprotect against copper-induced free radical damage by

forming a metal–ligand bond with free copper of bothoxidation states.

Fig. 5. Cathodic adsorptive stripping voltammograms of Cu1+

and melatonin. (A) Cu1+ (4 · 10)5M) in the absence of melatonin;

(B) Cu1+ (4 · 10)5M) in the presence of melatonin (1 · 10)5

M);(C–E), cathodic adsorptive stripping voltammograms of increasingmelatonin concentrations (2 · 10)5 to 4 · 10)5

M, respectively) inthe presence of Cu1+ (4 · 10)5

M).

Melatonin protects against copper-mediated free radical damage

241

Acknowledgements

Janice Limson thanks the Medical Research Council (South

Africa) for postdoctoral fellowship.

References

1. SAYRE L, PERRY G, SMITH M. Redox metals and neurodegen-

erative disease. Current Opinion Chem Biol, 1999; 3:220–225.

2. WAGGONER D, BARTNIKAS T, GITLIN J. The role of copper in

neurodegenerative disease. Neurobiol Dis 1999; 6:221–230.

3. HORNING M, BLAKEMORE L, TROMBLEY P. Endogenous

mechanisms of neuroprotection: role of zinc, copper, and

carnosine. Brain Res 2000; 852:56–61.

4. GUTTERIDGE J, HALLIWELL B. The measurement and mech-

anism of lipid peroxidation in biological systems. Trends Biol

Sci 1990; 15:129–135.

5. HALLIWELL B, GUTTERIDGE J. Lipid peroxidation, oxygen

radicals, cell damage, and antioxidant therapy. Lancet 1984;

1:1396–1398.

6. HALLIWELL B. Reactive oxygen species and the central nervous

system. J Neurochem 1992; 59:1609–1623.

7. RIKANS L, HORNBROOK K. Lipid peroxidation, anti-oxidant

protection and aging. Biochim Biophys Acta 1997;

1362:116–127.

8. HAYWOOD S, LOUGHRAN M. Copper toxicosis and tolerance in

the rat. II. Tolerance a liver protective adaptation. Liver 1985;

5:567–575.

9. REITER RJ. Antioxidant actions of melatonin. Adv Pharmacol

1997; 38:103–117.

10. TAN DX, CHEN LD, POEGGELER B, MANCHESTER LC,

REITER RJ. Melatonin: a potent endogenous hydroxyl radical

scavenger. Endocr J 1993; 1:57–60.

11. PIERI C, MORRA M, MARONI F, RECCHIONNI R, MARCHESELLI

F. Melatonin: a peroxyl radical scavenger more effective than

vitamin E. Life Sci 1994: 55: PL271–PL276.

12. LIMSON J, NYOKONG T, DAYA S. The interaction of melatonin

and its precursors with aluminium, cadmium, copper, iron,

lead and zinc: An adsorptive voltammetric study. J Pineal Res

1998; 24:15–21.

13. OLIVIERI G, BRACK C, MULLER-SPAHN F, STAHELIN HB,

HERMANN M, RENARD P, BROCKHAUS M, HOCK C. Mercury

induces cell cytotoxicity and oxidative stress and increases

beta-amyloid secretion and tau phosphorylation in SHSY5Y

neuroblastoma cells. J Neurochem 2000; 74:231–236.

14. LACK BA, DAYA S, NYOKONG T. Interaction of serotonin and

melatonin with sodium, potassium, calcium, lithium and alu-

minium. J Pineal Res 2001; 31:102–108.

15. Kubas GJ. Tetrakis (Acetonitrile) Copper (I) Hexafluoro-

phosphate. Inorg Synth 1979; 19:90–92.

16. ANOOPKUMAR-DUKIE S, WALKER R, DAYA S. A sensitive and

reliable method for the detection of lipid peroxidation in bio-

logical tissues. J Pharm Pharmacol 2001; 53:263–266.

17. LOWRY O, ROSENBURG M, FARR A, RANDALL R. Protein

measurement with the folin phenol reagent. J Biol Chem 1951;

193:256.

18. FUENTEALBA I, DAVIS R, ELMES E, JASANI B, HAYWOOD S.

Mechanisms of tolerance in the copper-loaded rat liver. Exp

Mol Pathol 1993; 59:71–84.

19. AGARWAL K, SHARMA A, TALUKDER G. Effects of copper on

mammalian cell components. Chem –Biol Interacts, 1989;

69:1–16.

20. HAYWOOD S. Copper toxicosis and tolerance in the rat. I.

Changes in copper content of the liver and the kidney. J Pathol

1985; 145:149–158.

Parmar et al.

242