Journal of Neurochemislry Raven Press, Ltd., New York 0 1991 International Society for Neurochernistry
Iron-Melanin Interaction and Lipid Peroxidation: Implications for Parkinson’s Disease Do& Ben-Shachar, *P. Riederer, and M. B. H. Youdim Rappaport Family Research Institute, Department of Pharmacology, Faculty of Medicine. Technion, Ha&, Israel, and *University of Wurzburg, Department of Psychiatry, Laboratory ofClinica1 Neurochemistry, Wurzburg, F.R.G.
Abstract: The vulnerability of substantia nigral (SN) melaninized dopamine neurons to neurodegenerationin Parkinson’s disease and the selective increases of iron and b a d lipid peroxidation in SN indicate that iron-melanin interaction could be crucial to the pathogenesis of this disease. The present study describes, for the first time, the identification and characterization of a high-affinity (KD= 13 IUM) and a lower affinity (KD= 200 nM) binding site for iron on dopamine melanin. The binding of iron to melanin is dependent on pH and the concentration of melanin. Iron chelators, U74500A, desferrioxamine, and to less extent 1,lO-phenanthrolineand chlorpromazine, but not the Parkinson-inducing neurotoxin, 1-methyl4-phenyl-1,2,3,6-tet-
rahydropyridine, can inhibit the binding of iron to melanin and iron-induced lipid peroxidation. Although melanin alone diminishes basal lipid peroxidation in rat cortical homogenates, it can also potentiate that initiated by iron, a reaction inhibited by desferrioxamine. In the absence of an identifiable exogenous or endogenous neurotoxin in idiopathic Parkinson’s disease, iron-melanin interaction in pars compacta of SN may be a strong candidate for the cytotoxic component of oxygen radical-induced neurodegeneration of melaninized dopamine neurons. Key Words: Iron-Melanin-Lipid peroxidation-Parkinson’s disease. Ben-Shachar D. et al. Ironmelanin interaction and lipid peroxidation: Implications for Parkinson’s disease. J. Neurochem. 57, 1609-1 6 14 (1 99 1).
Despite the identification of a neurotoxin, 1-methyl4-phenyl- 1,2,3,6-tetrahydropyridine(MPTP), producing a parkinsonian syndrome after its conversion to 1methyl-4-phenylpyridinium (MPP’) by monoamine oxidase B (MAO-B) (Markey et al., 1986), the etiology of this disease remains obscure. Biochemical (Dexter et al., 1987; Riederer et al., 1989) and nuclear magnetic resonance (Drayer et al., 1986; Rutledge et al., 1987) determinations in parkinsonian brains have shown a selective and highly significant elevation of iron, but not of ferritin, in the substantia nigra (SN), confirming an earlier report by Earle (1968). The ratio of Fez+/ Fe3+ changes from 3: 1 in the control SN to 1: 1 in the parkinsonian SN (Riederer et al., 1989). Such a finding is compatible with unquestioned ability of iron to initiate oxidative stress via generation of maximal rate of cytotoxic free hydroxyl radicals (OH ) (Fenton Reaction) and membrane lipid peroxides (Halliwell and Gutteridge, 1984). These reactions proceed via the well established interaction of Fe2+ and Fe3+ with H202 (Minotti and Aust, 1987) and melanin (Pilas et al.,
1988), respectively, to drive the redox state of iron between its two valences. The selective increase in basal lipid peroxidation in parkinsonian SN (Dexter et al., 1989) has been linked to accumulation of free iron and decreased availability of GSH in this region (Riederer et al., 1989). The well established cytotoxic nature of free tissue iron in depleting tissue GSH, inducing free radical formation, membrane lipid peroxidation, and cellular lesions in systemic organs (Minotti and Aust, 1987; Halliwell, 1989), has added a new perspective to this disorder. The increase of iron in parkinsonian SN and the susceptibility of the melaninized SN dopamine neurons to degeneration in Parkinson’s disease (Hirsch et al., 1988) indicate that iron-melanin interaction could be crucial in the initiation of neurodegeneration resulting from oxidative stress. The catechol polymer resin nature of melanin, formed from autooxidation of dopamine, makes it an avid reservoir for cation binding (Bruenger et al., 1967) and free oxygen radical formation (Pilas et al., 1988). We have proposed that Parkinson’s disease may result from progressive sider-
Received November 13,1990; revised manuscript received March 18, 1991; accepted April 9, 1991. Address correspondenceand reprint requests to Dr. D. Ben-Shachar at Department of Pharmacology, Faculty of Medicine, Technion, POB 9649, Haifa, Israel.
Abbreviations used: MAO-B, monoamine oxidase B; MPP+, 1methyl-4-phenylpyridinium; MPTP, 1-methyl-.l-phenyl-1,2,3,6-tetrahydropyridine; SN, substantia nigra.
-
1609
D. BEN-SHACHAR ET AL.
1610
osis of SN (Youdim et al., 1989), resulting from the accumulation of non-ferritin bound iron. We have investigated, therefore, the binding of 59Fe3+to synthetic dopamine melanin. This approach was considered valid because synthetic melanin has the ability to bind metals and drugs in a manner identical to neuromelanin of the SN (Bruenger et al., 1967; Potts and Au, 1976; Larsson and Tjalve, 1979) and because the role of protein in the binding reactions is negligible (Bruenger et al., 1967). MATERIALS AND METHODS [3H]Spiperone(24-26 Ci/mmol), [3H]flunitrazepam(>60 Ci/mmol), and 59FeC13(14.28 nCi/mg) were purchased from New England Nuclear. Chlorpromazine hydrochloride was a gift from Smith, Kline, and French Laboratories, U.S.A. Desfemoxamine mesylate (desferal) was purchased from Ciba-Geigy. Haloperidol was a gift from Janssen Pharmaceutica, N. V. Beerse, Belgium. MPP+ hydrochloride and MPTP were gifts from S. Markey, N.I.H., Bethesda, MD, U.S.A. The 21-amino acid steroid iron chelator U74500A was a gift from Upjohn, U.S.A. All other materials were purchased from Sigma, U.S.A.
Preparation of melanin Dopamine melanin and noradrenaline melanin were synthesized by the modified procedure of Das et al. (1978), as adapted by DAmato et al. (1986). The final melanin suspensions were lyophilized and stored at 4°C. Just before use, the samples were resuspended in double-distilled water and sonicated for 3 min.
Melanin binding The specific binding of 20 concentrations of 59FeC13(14.28 mCi/mg), ranging from 0.5 to 400 nM, to 3 pg of dopamine melanin was assayed in the presence or absence of 10 or 100 p M FeC13.The total volume was 0.25 ml of 5 mM Tris-HC1 buffer (pH 6.5 at 37°C). The reaction mixture was incubated at 37°C for 2 h. The reaction was terminated by adding 2 ml of cold buffer, and the mixture was filtered through GF/ B filters. The filters were washed three times with 2 ml of cold buffer. The radioactivity on the filters was measured by liquid scintillation spectrometry. Nonspecific binding at 10 p M 59FeC13was about 30% of total binding. The binding of 10 p M 59FeC13to GF/B filters in the absence of dopamine melanin was assessed after 2 h of incubation at 37°C with 2, 50, or 400 nM s9FeC13in 5 mM Tris-HC1 buffer (pH 6.5); the resultant binding to the filters amounted to roughly 7, 9, and 4%, respectively, of total binding in the presence of dopamine melanin (3 pg). To determine the ability of drugs, iron, and non-iron chelators and divalent ions to displace iron from melanin, the binding of 10 nM 59FeC13to 3 pg of melanin was assayed in M of each drug, as described the presence of to above. The specific binding of 10 concentrations of [3H]spiperone,ranging from 0.25 to 20 nM, to 3 p g of dopamine melanin was assayed in the presence or absence of 10 p M haloperidol. The total volume was 0.25 ml of 50 mM Tris-HC1 buffer (pH 7.4 at 37°C). The reaction mixture was incubated at 37°C for 20 min. The reaction was terminated as described above. The specificbinding of 10 concentrations of [3H]flunitrazepam,ranging from 0.125 to 6 nM, to 3 pg of dopamine melanin was assayed in the presence or absence J. Neurochem., Vol. 57. No. 5, 1991
of 1 clonazepam. The total volume was 0.25 ml of 50 mM Tris-HC1 buffer (pH 7.4 at 0°C). The reaction mixture was incubated at 4°C for 90 min. The reaction was terminated as described above for iron binding. The specific binding of 10 nM 59FeC13to six concentrations of dopamine melanin, ranging from 1.0 to 20 pg, was assayed in the presence or absence of 10 pM FeC13, as described above. The effects of six pH levels ranging from 3 to 9 on the specific binding of 10 nM 59FeC13to dopamine-melanin were also examined. pH levels were altered by adding either HCl or NaOH to Tris buffer.
Measurement of lipid peroxidation in brain tissue Brain cortex homogenates (10%wt/vol) from male Wistar rats were prepared in 0.3 M sucrose and incubated in air as described by Rehncrona et al. (1 980). The 0.1-ml aliquots of homogenate were incubated alone at 37°C for 90 min to determine basal lipid peroxidation, or incubated after the addition of low4MFe2(S04)3and in the presence of 3 or 12 p g of dopamine melanin, 10-4 M desferrioxamine, and their combinations. Lipid peroxidation was assayed by measurement of malondialdehyde formation, as described by Dexter et al. (1989).
Statistical analysis The results are expressed as means k SEM. Statistical analysis was performed by two-tailed Student's t tests.
RESULTS In this study, 59Fe3+bound with high affinity to dopamine melanin, and Scatchard analysis of equilibrium saturation data revealed at least two distinct populations of binding sites. The high-affinity site exhibited a dissociation constant (KD)of 13 1.61 nM and a maximal number of binding sites (Bmax) of 1.13 f 0.15 nmol/mg of dopamine melanin. The site with lower affinity had a KDvalue of 200 rt_ 46 nMand B,, value of 17.4 f 1.39 nmol/mg of dopamine melanin (Fig. 1). The 59Fe3+binding sites on dopamine melanin appeared to be relatively specific, as [3H]flunitrazepam did not bind to dopamine melanin and [3H]spiperone demonstrated significantly lower affinity for melanin, exhibiting a single site with a KDvalue of 55 f 4.4 nM
*
KD=ZWnM
\
B,,,=1.13 "K";13;M
0
3.0
-_
nmallmg ,melanin
6.0
9.0
12.0
%FeC13 Specific Bound (nmollmg melanin)
FIG. 1. Scatchard (GraphPAD) plot of specific 59FeC13binding to dopamine melanin. Various concentrationsof "FeCI3 between 0.5 and 400 nM were tested. The computer-fitted curve (solid line) of three separate experiments in triplicate is shown. Dashed lines represent computer-determined apparent high- and low-affinity components of binding.
IRON-MELANIN BINDING AND PARKINSON'S DISEASE
and a B,,, value of 0.3 k 0.1 1 nmol/mg of dopamine melanin. Furthermore, the binding of 59Fe3+to dopamine melanin in the presence of divalent ions, metal chelators, and other substances known to bind to melanin, revealed that only compounds with iron chelating capacity were potent inhibitors (Table 1). The most potent displacers were the iron chelators U74500A and desferrioxamine. Neither MPTP nor MPP' was effective as an inhibitor of s9Fe3fbinding. Chloroquine did not alter the binding of s9Fe3+to dopamine melanin, even at M . By contrast, dopamine increased the binding of s9Fe3fto dopamine melanin (Fig. 2), suggesting the possible participation of dopamine in the accumulation and binding of iron to melanin and formation of more dopamine melanin. The binding of s9Fe3+to dopamine melanin was dependent on the melanin concentration and pH (Figs. 3 and 4, respectively). Maximal binding was observed with 3.0 pg of melanin per 0.25 ml, at pH 6.0. Above and below these values, there was a significant reduction in the amount of 59Fe3+bound per microgram of melanin. The ability of iron to promote lipid peroxidation was substantiated in the present study using rat cerebral cortex homogenate. Although melanin at concentrations of 3.0 and 12 pg/0.25 ml significantly reduced basal lipid peroxidation, it potentiated the lipid peroxidative property of iron (100 p M ) at its higher concentration (Fig. 5). The increases in lipid peroxidation induced by iron over the basal level and over the melanin-treated tissue homogenates were 53 and 2 18%, respectively; these results were significantly different ( p < 0.001) as determined by the calculation of the t values of the ratios. In addition to the ability of desferrioxamine to inhibit s9Fe3+binding to dopamine melanin, this iron chelator was very efficient in inhibTABLE 1. The ability of drugs, iron chelators, and divalent ions to inhibit ''FeCl3 binding to dopamine melanin
U74500A Desferrioxamine Chlorpromazine 1,lO-Phenanthroline FeClz Apomorphine Haloperidol Dopamine NTA MPP' MPTP MnClz Chloroquine Spiperone
10
O
t
b
9
8
7
1611
6
5
4
3
Doparnine [-log MI
FIG. 2. The effect of dopamine on "Fe3+ binding to dopamine melanin. The procedurewas identicalto that described in Table 1, with the inclusion of various concentrations of dopamine. Results are means t SEM of three separate experiments.
iting the iron or iron-melanin-induced lipid peroxidation (Fig. 5). DISCUSSION Iron has been implicated in lipid peroxidation and cell degeneration more often than any other metal (Halliwell and Gutteridge, 1984; Triggs and Willmore, 1984; Dexter et al., 1989; Halliwell, 1989). It is the interaction of increased free iron with H202and the interconversion of iron between its redox states that initiates the liberation of cytotoxic hydroxyl free radicals (OH ) and promotes lipid peroxidation. Human basal ganglia are the most metabolically active parts of the brain, and have the highest activity of MAO-B (Collins et al., 1970; Riederer and Youdim, 1986).The oxidative deamination of dopamine by this enzyme and the autooxidation of dopamine to melanin generate a substantial amount of H202 (Konradi et al., 1986). The diminished levels of endogenous H202 scavengers, ascorbate and GSH, in parkinsonian SN (Riederer et al., 1989) predicts a significant accumulation of H202.In this condition, free Fe2+would be available to interact with H202 and drive the Fenton Reaction, resulting in the liberation of cytotoxic hydroxyl radical (OH. ) and Fe3+(Fig. 6). The optimal condition for participation of iron in OH formation and lipid peroxidation initiation requires a Fe2+/Fe3+
-
60 180
500 1,300 >2,000 22,000 >2,000
>2,000 >2,000 >2,000 >2,000
0
>2,000
>2,000
The binding of 10 nM 59FeClsto 3 pg of melanin was assayed as described in Fig. 1 in the presence of various compounds. Competition curves represent results from 14 concentrations of drugs ranging from to lo-' M (each in triplicate). ICS0values are means of three experiments. NTA, nitrilotriacetic acid.
5
10
15 Melanin [pgl
20
25
FIG. 3. The effect of dopamine melanin concentration on 59Fe3+. The binding procedure consisted of incubation of 10 nM 59FeC13 with various concentrationsof dopamine melanin in 5 mM Tris-HCI buffer, pH 6.5, for 2 h according to the procedure outlined in Materials and Methods for Scatchard analysis. Results are means SEM of five separate experiments.
+
J. Neurachem.. Vat. 57, No. 5, 1991
D. BEN-SHACHAR ET AL.
1612
4
6
8
10
PH
FIG. 4. The effect of pH on "Fe3+ binding to dopamine melanin. The specific binding of 10 nM 59Fe3+to 3 pg of melanin in 5 mM Tris-HCI buffer was assayed as described in Fig. 1. The various pH values were achieved by adding either HCI or NaOH to the Tris buffer. Results are means SEM of four experiments.
+
ratio of 1:1 (Minotti and Aust, 1987). This ratio has been observed in the SN of parkinsonian brains as a result of an increase in the Fe3+form (Riederer et al., 1989). Normally, Fe3+ is bound to endogenous chelators such as ADP or melanin, thus shifting the equilibrium in favor of the chelated form. On the other hand, an increased concentration of Fe3+ also could serve as the catalyst for its conversion to Fez+by melanin, and, in the presence of H202,result in further formation of OH. (Pilas et al., 1988). This reaction has been attributed to the ability of melanin to bind Fe3+as well as to reduce it to Fe2+,depending on environmental conditions. In the absence of significant amounts of fenitin in the SN, iron could be chelated by melanin. The present study demonstrates the high specific capacity of melanin to bind iron. Our experiments with 59Fe3+and the Scatchard analyses of the binding indicate the presence of two (low- and highaffinity) binding sites on dopamine melanin. However, our preliminary studies with synthetic noradrenaline melanin (unpublished data) demonstrated significantly lower capacity for binding 59Fe3+. The binding probably occurs to the free carboxyl groups, phenolic hydroxyl moiety, or quinoid groups present on the melanin polymer (Larsson and Tjalve, 1979). The binding of 59Fe3+ to dopamine melanin appears to be relatively specific because flunitrazepam did not bind to dopamine melanin and spiperone demonstrated low affinity. Iron could be displaced from dopamine melanin only by compounds with iron chelating capacity. This was especially evident in the case of U74500A and desfemoxamine, an iron chelator employed in treatment of systemic organ iron overload (Johnson et al., 1982).Neither MPTP nor MPP', which have high affinity for dopamine melanin and neuromelanin (DAmato et al., 1986), was effective as an inhibitor of iron binding. Evidently, dopamine melanin has a significantly higher number of binding sites for iron (Fig. 1) than for MPP+ (D'Amato et al., 1986). Moreover, chloroquine, which binds to melanin (Larsson and Tjalve, 1979) and displays selectivity for displacing MPP+ from neuromelanin and dopamine melanin, and partially prevents MPTP-induced parkinsonism in monkeys (DAmato et al., 1987), did not J Neurochem.. Vol 57, N o 5. 1991
alter the binding of 59Fe3+to dopamine melanin even M. By contrast, dopamine potentiated the at binding of iron to dopamine melanin, suggesting the possible participation of dopamine in the binding of iron to melanin and the formation of more dopamine melanin as a protective mechanism against iron accumulation. The optimal observed conditions for iron binding to dopamine melanin with 3 p g of melanin/ 0.25 ml at pH 6.0 (Figs. 3 and 4)are compatible with the optimal transport and binding of iron in vivo (Octave et al., 1983) and the induction of oxygen free radicals (Pilas et al., 1988). A decrease in the binding of Fe3+ to dopamine melanin was detected as melanin concentration exceeded 5 pg/0.25 ml. The probability of interaction between the individual melanin chains may increase at higher concentrations of melanin because of their ability to bind iron. Melanin in normal circumstances is considered an effective radical scavenger in vitro and in vivo (Schwabe et al., 1989; Scalia et al., 1990). However, under certain conditions, such as increased availability of Fe3+,melanin can potentiate the formation of oxygen radicals. The production of OH in the presence of melanin is significantly greater when Fe3+is predominant (Pilas et al., 1988) and is further demonstrated by the greater lipid peroxidation of rat cerebral cortex in the presence of Fe3+and higher dopamine melanin concentrations (Fig. 5). This is consistent with our results, demonstrating that high melanin concentrations reduce 59Fe3+ binding to dopamine melanin. According to the reaction pathway in Fig. 6, H202, generated by the deamination and autooxidation of dopamine, participates with free iron and melanin to drive the Fenton 0
n
I
20
0
10
13;
DFO
+ +
DFO
+ +
DFO
+
I Fe3+ Fe3+ F e 3 '
D A - M DA-M 3 12
FIG. 5. Lipid peroxidation properties of iron and dopamine melanin (DA-M). Lipid peroxidation was assayed by measurement of malondialdehyde(MDA) formation in rat cortex homogenate. BL, basal lipid peroxidation; Fe3+,MDA formation in the presence of 10-4 M Fe2(S0&; DA-M, 3 or 12 pg/ml of melanin; DFO, 1 0-4 M desferrioxamine; and their combinations. Results are means SEM of triplicate determinationof five experiments. Values are expressed as nanomoles of MDA production per milligram of protein above basal levels in rat cortical homogenates. *p 0.001 compared with the control basal level (BL); **p i0.001 compared with the 3-pg DA-M response.
*
-=
IRON-MELANIN BINDING AND PARKINSON’S DISEASE
1613
Hz0
Feedback Stimulation
DOPAMINE
MAO-B
+ R-CHO+NH3+H202 FIG. 6. The reaction pathway illustrating the ability of H202and melanin to alter the redox of iron between its two valences. In the event of increased iron and diminished disposition of HZ02by glutathione peroxidase, the net effect would be the formation of cytotoxic hydroxyl (OH - ) and initiation of lipid peroxidation.
Electron
I
Cytotoxic Hydroxyl Radical
j.
Lipid Peroxidation
-
Reaction, thereby liberating cytotoxic OH . The figure demonstrates a system by which H202 and melanin serve as catalysts for the conversion of Fe3+to Fez+. Furthermore, this model may explain the parkinsonism in manganese miners (Donaldson and Barbeau, 1985). The interconversion of manganese, like iron, between its two redox states, resulting in generation of reactive oxygen species (e.g., OH.), has also been proposed as the mechanism of striatal neuronal membrane toxicity (Graham, 1984;Archibald and Tyree, 1987). The ability of melanin to chelate manganese (Potts and Au, 1976; Larsson and Tjalve, 1979) supports this hypothesis. It is also apparent that both iron (Graham et al., 1978) and manganese (Graham, 1984; Liccione and Maines, 1988) deplete tissue levels of GSH, which has been reported to be decreased in parkinsonian brains (Riederer et al., 1989). In the brain, glutathione peroxidase acts as the main cellular defense mechanism in the destruction of H202, by oxidizing GSH to GSSG, in the presence of H202(Liccione and Maines, 1988). It is believed that glutathione peroxidase, and not catalase, in the brain is responsible for removal of monoamine oxidase-generated H202(Oshino and Chance, 1977; Spina and Cohen, 1989) (Fig. 6). In Parkinson’s disease, the heavily pigmented neurons of SN are preferentially lost (Mann and Yates, 1983). Hirsch et al. (1988) clearly demonstrated the greater vulnerability of neuromelanin-containing dopamine neurons to the neurodegenerative process of Parkinson’s disease. We believe that the results of the present study are directly relevant to the pathophysiology of Parkinson’s disease because of the selectivity of iron increase (Dexter et al., 1987; Riederer et al., 1989) and lipid peroxidation (Dexter et al., 1989) in parkinsonian SN. Furthermore, they may explain the
/
selectivity and vulnerability of melaninized dopamine neurons in SN to neurodegeneration. It remains to be seen whether the increased deposition of iron noted in the SN of Parkinsonian brains is within the dopamine neurons and whether it is deposited in these neurons after a neurotoxic insult. Preliminary studies with MPTP-induced dopaminergic neurotoxicity in mice (D. Ben-Shachar, E. Melamed, and M. B. H. Youdim, unpublished data) and monkeys (E. Domino, personal communication) show no alterations in iron content of the striatum. Thus, it seems that neurodegeneration per se does not appear to contribute to elevation of SN iron. The results of the present study predict that future prospects for the treatment or retardation of the process of neurodegeneration in Parkinson’s disease may involve the use of iron chelators, such as U74500A (Braughler et al., 1988),capable of crossing the bloodbrain barrier in a fashion similar to chelators used to treat Wilson’s disease and cases of iron overload in systemic organs (Halliwell and Gutteridge, 1986). Acknowledgment: This work was supported by grants from Golding Research Fund for Parkinson’s Disease (U.S.A.) and Research and Development Grant from the Technion (Haifa) and Israel Center for Psychobiology (Jerusalem).
REFERENCES Archibald F. S. and Tyree C. (1 987) Manganese poisoning and the attack of trivalent manganese upon catecholamines. Arch. Biochem. Biophys. 256,638-650. Braughler J. M., Duncan L. A,, and Chase R. L. (1986) The involvement of iron in lipid peroxidation: importance of fenic to ferrous ratio in initiation. J. Eiol. Chem. 261, 10282-10289.
J. Neurochem., Vol. 57, No. 5 , 1991
1614
D. BEN-SHACHAR E T AL.
Braughler J. M., Burton P. S., Chas R. L., and Pregenzer J. F. (1 988) Novel membrane-localized iron chelators as inhibitors of irondependent lipid peroxidation. Biochem. Pharmacol. 37, 38533860. Bruenger F. W., Stover B. J., and Atherton D. R. (1967) The incorporation of various metal ions into in vivo and in vitro produced melanin. Radiat. Res. 32, 1-12. Collins G. G., Sandler M., Williams E. D., and Youdim M. B. H. (1970) Multiple forms of human brain mitochondria1 monoamine oxidase. Nature 225, 817-820. D’Amato R. J., Lipman Z. P., and Snyder S. H. (1986) Selectivity of the parkinsonian neurotoxin MPTP toxic metabolite MPP’ binds to neuromelanin. Science 231, 987-989. DAmato R. J., Alexander G. M., Schwartzman R. J., Kitt C. A., Price D. L., and Snyder S. H. (1987) Evidence for neuromelanin involvement in MPTP-induced neurotoxicity. Nature 327,324326. Das K. C., Abramson M. B., and Katzman R. (1978) Neuronal pigments: spectroscopic characterization of human brain melanin. J. Neurochem. 30,601-605. Dexter D. T., Well F. R., Agid F., Agid Y . ,Lees A. J., Jenner P., and Marsden C. D. (1987) Increased nigral iron content in postmortem parkinsonian brain. Lancet ii, 12 19- 1220. Dexter D. T., Carter C . J., Wells F. R., Javoy-Agid F., Agid Y ., Lees A., Jenner P., and Marsden C. D. (1989) Basal lipid peroxidation in substantia nigra is increased in Parkinson’s disease. J. Neurochem. 52, 38 1-389. Donaldson J. and Barbeau A. (1985) Manganese neurotoxicity: possible clues to the etiology of human brain disorders, in Metal Ions in Neurology and Psychiatry (Gabay S., Hams J., and Ho B. T., eds), pp. 259-285. Alan R. Liss, New York. Drayer B. P., Olanow W., Burger P., Johnson G. A,, Hertkens R., and Riederer P. ( 1986) Parkinson plus syndrome: diagnosis using high-field MR imaging of brain iron. Radiology 159, 493-498. Earle K. M. (1968) Trace metals in parkinsonian brains. J. Neuropathol. Exp. Neurol. 27, 1-14. Graham D. G. (1984) Catecholamine toxicity: a proposal for the molecular pathogenesis of manganese neurotoxicity and Parkinson’s disease. Neurotoxicology 5, 83-96. Graham D. G., Tiffany S. M., Bell W. R. Jr., and Gutknecht W. F. (1978) Autooxidation versus covalent binding of quinones as the mechanism of toxicity of dopamine, 6-hydroxydopamine, and related compounds towards C- I300 neuroblastoma cells in vitro. Mol. Pharmacol. 14, 644-653. Halliwell B. (1989) Oxidants and the central nervous system, some fundamental questions: is oxidant damage relevant to Parkinson’s disease, Alzheimer’s disease, traumatic injury, or stroke? Acta Neurol. Scand. Suppl. 126,23-33. Halliwell B. and Gutteridge J. M. (1984) Oxygen toxicity, oxygen radicals, transition metals, and disease. Biochem. J. 219, 1-14. Halliwell B. and Gutteridge J. M. C. (1986) Oxygen free radicals and iron in relation to biology and medicine: some problems and concepts. Arch. Biochem. Biophys. 246, 501-5 14. Hirsch E., Graybiel A. M., and Agid Y . (1988) Melanized dopaminergic neurons are differentially susceptible to degeneration in Parkinson’s disease. Nature 334, 345-348. Johnson D. K., Pippard M. J., Murphy T. B., and Rose N. J. ( 1 982) An in vivo evaluation of iron-chelating drugs derived from pyridoxal and its analogs. J. Pharmacol. Exp. Ther. 221, 399-403. Konradi C., Riederer P., and Youdim M. B. H. (1986) Hydrogen peroxide enhances the activity of rnonoamine oxidase type-B but not that of type-A: a pilot study. J. Neural Transm. Suppl. 22,61-74.
J. Neurochem., Vol. 5 7, No. 5 , I991
Larsson B. and Tjalve H. (1979) Studies on the mechanism of drugs binding to melanin. Biochem. Pharmacol. 28, 1181-1 187. Liccione J. J. and Maines M. D. (1988) Selective vulnerability of glutathione metabolism and cellular defence mechanism in rat stnatum to manganese. J. Pharmacol. Exp. Ther. 247,156- 16 I . Mann D. M. and Yates P. 0. (1983) Possible role of neuromelanin in the pathogenesis of Parkinson’s disease. Mech. Ageing Do. 21, 193-203. Markey S. P., Castagnoli N. Jr., Trevor A. J., and Kopin I. J., eds (1986) MPTP: A Neurotoxin Producing a Parkinsonian Syndrome. Academic Press, New York. Minotti G. and Aust S. D. (1987) The requirement for iron 111 in the initiation of lipid peroxidation by iron I1 and hydrogen peroxidase. J. Biol. Chem. 262, 1098-1 104. Octave J. N., Schneider Y. J., Trouet A., and Crichton R. R. (1983) Iron uptake and utilization by mammalian cells. I. Cellular uptake of transfemn and iron. Trends Biochem. Sci. 8, 217-220. Oshino N. and Chance B. (1977) Properties of glutathione release observed during reduction of organic hydroperoxide, demethylation of aminopyrine, and oxidation of some substances in perfused rat liver, and their implications for the physiological function of catalase. Biochem. J. 162, 509-525. Pilas B., Sama T., Kalyanarman B., and Swartz H. M. (1988) The effect of melanin on iron-associated decomposition of hydrogen peroxide. Free Radic. Biol. Med. 4, 285-293. Potts A. M. and Au P. C. (1976) The affinity of melanin for inorganic ions. Exp. Eye Res. 22,487-49 1. Rehncrona S., Smith D. S., Akesson B., Westerberg E., and Siesjo B. H. (1980) Peroxidative changes in brain cortical fatty acids and phospholipids, as characterized during Fe2+-and ascorbic acid-stimulated lipid peroxidation in vitro. J. Neurochem. 34, 1630- 1638. Riederer P. and Youdim M. B. H. (1986) Monoamine oxidase activity and monoamine metabolism in brains of parkinsonian patients treated with L-deprenyl. J. Neurochem. 46, 1359- 1365. Riederer P., Sofic E., Rausch W. D., Schmidt B., Reynolds G. P., Jellinger K., and Youdim M. B. H. (1989) Transition metals, femtin, glutathione, and ascorbic acid in parkinsonian brain. J. Neurochem. 52, 5 15-520. Rutledge J. N., Hilal S. K., Silver A. J., Defendini R., and Fahn S. (1987) Study of movement disorders and brain iron by magnetic resonance. AJR Am. J. Roentgenol. 149, 365-379. Salazar M., Sokoloski T. D., and Patil P. N. (1978) Binding of dopaminergic drugs by the neuromelanin of the substantia nigra, synthetic melanins, and melanin granules. Fed. Proc. 37,24032407. Scalia M., Geremia E. V. C., Santoro C., Baratta D., and Sichel G. (1 990) Lipid peroxidation and unpigmented liver tissues: protective role of melanin. Pigment Cell Res. 3, 1 1 5-1 19. Schwabe K., Lassmann G., Damerau W., and Naundorf H. (1989) Protection of melanoma cells against superoxide radicals by melanin. J. Cancer Res. Clin. Oncol. 115, 597-600. Spina M. B. and Cohen G. J. (1989) Dopamine turnover and glutathione oxidation: implications for Parkinson’s disease. Proc. Natl. Acad. Sci. USA 86, 1398-1400. Tnggs W. J. and Willmore L. J. (1984) In vivo lipid peroxidation in rat brain following intracortical Fe2+injection. J. Neurochem. 42,976-980. Youdim M. B. H., Ben-Shachar D., and Riederer P. (1989) Is Parkinson’s disease a progressive siderosis of substantia nigra resulting in iron- and melanin-induced neurodegeneration? Acta Neurol. Scand. 126, 41-54.
Iron-Melanin Interaction and Lipid Peroxidation: Implications for Parkinson’s Disease Do& Ben-Shachar, *P. Riederer, and M. B. H. Youdim Rappaport Family Research Institute, Department of Pharmacology, Faculty of Medicine. Technion, Ha&, Israel, and *University of Wurzburg, Department of Psychiatry, Laboratory ofClinica1 Neurochemistry, Wurzburg, F.R.G.
Abstract: The vulnerability of substantia nigral (SN) melaninized dopamine neurons to neurodegenerationin Parkinson’s disease and the selective increases of iron and b a d lipid peroxidation in SN indicate that iron-melanin interaction could be crucial to the pathogenesis of this disease. The present study describes, for the first time, the identification and characterization of a high-affinity (KD= 13 IUM) and a lower affinity (KD= 200 nM) binding site for iron on dopamine melanin. The binding of iron to melanin is dependent on pH and the concentration of melanin. Iron chelators, U74500A, desferrioxamine, and to less extent 1,lO-phenanthrolineand chlorpromazine, but not the Parkinson-inducing neurotoxin, 1-methyl4-phenyl-1,2,3,6-tet-
rahydropyridine, can inhibit the binding of iron to melanin and iron-induced lipid peroxidation. Although melanin alone diminishes basal lipid peroxidation in rat cortical homogenates, it can also potentiate that initiated by iron, a reaction inhibited by desferrioxamine. In the absence of an identifiable exogenous or endogenous neurotoxin in idiopathic Parkinson’s disease, iron-melanin interaction in pars compacta of SN may be a strong candidate for the cytotoxic component of oxygen radical-induced neurodegeneration of melaninized dopamine neurons. Key Words: Iron-Melanin-Lipid peroxidation-Parkinson’s disease. Ben-Shachar D. et al. Ironmelanin interaction and lipid peroxidation: Implications for Parkinson’s disease. J. Neurochem. 57, 1609-1 6 14 (1 99 1).
Despite the identification of a neurotoxin, 1-methyl4-phenyl- 1,2,3,6-tetrahydropyridine(MPTP), producing a parkinsonian syndrome after its conversion to 1methyl-4-phenylpyridinium (MPP’) by monoamine oxidase B (MAO-B) (Markey et al., 1986), the etiology of this disease remains obscure. Biochemical (Dexter et al., 1987; Riederer et al., 1989) and nuclear magnetic resonance (Drayer et al., 1986; Rutledge et al., 1987) determinations in parkinsonian brains have shown a selective and highly significant elevation of iron, but not of ferritin, in the substantia nigra (SN), confirming an earlier report by Earle (1968). The ratio of Fez+/ Fe3+ changes from 3: 1 in the control SN to 1: 1 in the parkinsonian SN (Riederer et al., 1989). Such a finding is compatible with unquestioned ability of iron to initiate oxidative stress via generation of maximal rate of cytotoxic free hydroxyl radicals (OH ) (Fenton Reaction) and membrane lipid peroxides (Halliwell and Gutteridge, 1984). These reactions proceed via the well established interaction of Fe2+ and Fe3+ with H202 (Minotti and Aust, 1987) and melanin (Pilas et al.,
1988), respectively, to drive the redox state of iron between its two valences. The selective increase in basal lipid peroxidation in parkinsonian SN (Dexter et al., 1989) has been linked to accumulation of free iron and decreased availability of GSH in this region (Riederer et al., 1989). The well established cytotoxic nature of free tissue iron in depleting tissue GSH, inducing free radical formation, membrane lipid peroxidation, and cellular lesions in systemic organs (Minotti and Aust, 1987; Halliwell, 1989), has added a new perspective to this disorder. The increase of iron in parkinsonian SN and the susceptibility of the melaninized SN dopamine neurons to degeneration in Parkinson’s disease (Hirsch et al., 1988) indicate that iron-melanin interaction could be crucial in the initiation of neurodegeneration resulting from oxidative stress. The catechol polymer resin nature of melanin, formed from autooxidation of dopamine, makes it an avid reservoir for cation binding (Bruenger et al., 1967) and free oxygen radical formation (Pilas et al., 1988). We have proposed that Parkinson’s disease may result from progressive sider-
Received November 13,1990; revised manuscript received March 18, 1991; accepted April 9, 1991. Address correspondenceand reprint requests to Dr. D. Ben-Shachar at Department of Pharmacology, Faculty of Medicine, Technion, POB 9649, Haifa, Israel.
Abbreviations used: MAO-B, monoamine oxidase B; MPP+, 1methyl-4-phenylpyridinium; MPTP, 1-methyl-.l-phenyl-1,2,3,6-tetrahydropyridine; SN, substantia nigra.
-
1609
D. BEN-SHACHAR ET AL.
1610
osis of SN (Youdim et al., 1989), resulting from the accumulation of non-ferritin bound iron. We have investigated, therefore, the binding of 59Fe3+to synthetic dopamine melanin. This approach was considered valid because synthetic melanin has the ability to bind metals and drugs in a manner identical to neuromelanin of the SN (Bruenger et al., 1967; Potts and Au, 1976; Larsson and Tjalve, 1979) and because the role of protein in the binding reactions is negligible (Bruenger et al., 1967). MATERIALS AND METHODS [3H]Spiperone(24-26 Ci/mmol), [3H]flunitrazepam(>60 Ci/mmol), and 59FeC13(14.28 nCi/mg) were purchased from New England Nuclear. Chlorpromazine hydrochloride was a gift from Smith, Kline, and French Laboratories, U.S.A. Desfemoxamine mesylate (desferal) was purchased from Ciba-Geigy. Haloperidol was a gift from Janssen Pharmaceutica, N. V. Beerse, Belgium. MPP+ hydrochloride and MPTP were gifts from S. Markey, N.I.H., Bethesda, MD, U.S.A. The 21-amino acid steroid iron chelator U74500A was a gift from Upjohn, U.S.A. All other materials were purchased from Sigma, U.S.A.
Preparation of melanin Dopamine melanin and noradrenaline melanin were synthesized by the modified procedure of Das et al. (1978), as adapted by DAmato et al. (1986). The final melanin suspensions were lyophilized and stored at 4°C. Just before use, the samples were resuspended in double-distilled water and sonicated for 3 min.
Melanin binding The specific binding of 20 concentrations of 59FeC13(14.28 mCi/mg), ranging from 0.5 to 400 nM, to 3 pg of dopamine melanin was assayed in the presence or absence of 10 or 100 p M FeC13.The total volume was 0.25 ml of 5 mM Tris-HC1 buffer (pH 6.5 at 37°C). The reaction mixture was incubated at 37°C for 2 h. The reaction was terminated by adding 2 ml of cold buffer, and the mixture was filtered through GF/ B filters. The filters were washed three times with 2 ml of cold buffer. The radioactivity on the filters was measured by liquid scintillation spectrometry. Nonspecific binding at 10 p M 59FeC13was about 30% of total binding. The binding of 10 p M 59FeC13to GF/B filters in the absence of dopamine melanin was assessed after 2 h of incubation at 37°C with 2, 50, or 400 nM s9FeC13in 5 mM Tris-HC1 buffer (pH 6.5); the resultant binding to the filters amounted to roughly 7, 9, and 4%, respectively, of total binding in the presence of dopamine melanin (3 pg). To determine the ability of drugs, iron, and non-iron chelators and divalent ions to displace iron from melanin, the binding of 10 nM 59FeC13to 3 pg of melanin was assayed in M of each drug, as described the presence of to above. The specific binding of 10 concentrations of [3H]spiperone,ranging from 0.25 to 20 nM, to 3 p g of dopamine melanin was assayed in the presence or absence of 10 p M haloperidol. The total volume was 0.25 ml of 50 mM Tris-HC1 buffer (pH 7.4 at 37°C). The reaction mixture was incubated at 37°C for 20 min. The reaction was terminated as described above. The specificbinding of 10 concentrations of [3H]flunitrazepam,ranging from 0.125 to 6 nM, to 3 pg of dopamine melanin was assayed in the presence or absence J. Neurochem., Vol. 57. No. 5, 1991
of 1 clonazepam. The total volume was 0.25 ml of 50 mM Tris-HC1 buffer (pH 7.4 at 0°C). The reaction mixture was incubated at 4°C for 90 min. The reaction was terminated as described above for iron binding. The specific binding of 10 nM 59FeC13to six concentrations of dopamine melanin, ranging from 1.0 to 20 pg, was assayed in the presence or absence of 10 pM FeC13, as described above. The effects of six pH levels ranging from 3 to 9 on the specific binding of 10 nM 59FeC13to dopamine-melanin were also examined. pH levels were altered by adding either HCl or NaOH to Tris buffer.
Measurement of lipid peroxidation in brain tissue Brain cortex homogenates (10%wt/vol) from male Wistar rats were prepared in 0.3 M sucrose and incubated in air as described by Rehncrona et al. (1 980). The 0.1-ml aliquots of homogenate were incubated alone at 37°C for 90 min to determine basal lipid peroxidation, or incubated after the addition of low4MFe2(S04)3and in the presence of 3 or 12 p g of dopamine melanin, 10-4 M desferrioxamine, and their combinations. Lipid peroxidation was assayed by measurement of malondialdehyde formation, as described by Dexter et al. (1989).
Statistical analysis The results are expressed as means k SEM. Statistical analysis was performed by two-tailed Student's t tests.
RESULTS In this study, 59Fe3+bound with high affinity to dopamine melanin, and Scatchard analysis of equilibrium saturation data revealed at least two distinct populations of binding sites. The high-affinity site exhibited a dissociation constant (KD)of 13 1.61 nM and a maximal number of binding sites (Bmax) of 1.13 f 0.15 nmol/mg of dopamine melanin. The site with lower affinity had a KDvalue of 200 rt_ 46 nMand B,, value of 17.4 f 1.39 nmol/mg of dopamine melanin (Fig. 1). The 59Fe3+binding sites on dopamine melanin appeared to be relatively specific, as [3H]flunitrazepam did not bind to dopamine melanin and [3H]spiperone demonstrated significantly lower affinity for melanin, exhibiting a single site with a KDvalue of 55 f 4.4 nM
*
KD=ZWnM
\
B,,,=1.13 "K";13;M
0
3.0
-_
nmallmg ,melanin
6.0
9.0
12.0
%FeC13 Specific Bound (nmollmg melanin)
FIG. 1. Scatchard (GraphPAD) plot of specific 59FeC13binding to dopamine melanin. Various concentrationsof "FeCI3 between 0.5 and 400 nM were tested. The computer-fitted curve (solid line) of three separate experiments in triplicate is shown. Dashed lines represent computer-determined apparent high- and low-affinity components of binding.
IRON-MELANIN BINDING AND PARKINSON'S DISEASE
and a B,,, value of 0.3 k 0.1 1 nmol/mg of dopamine melanin. Furthermore, the binding of 59Fe3+to dopamine melanin in the presence of divalent ions, metal chelators, and other substances known to bind to melanin, revealed that only compounds with iron chelating capacity were potent inhibitors (Table 1). The most potent displacers were the iron chelators U74500A and desferrioxamine. Neither MPTP nor MPP' was effective as an inhibitor of s9Fe3fbinding. Chloroquine did not alter the binding of s9Fe3+to dopamine melanin, even at M . By contrast, dopamine increased the binding of s9Fe3fto dopamine melanin (Fig. 2), suggesting the possible participation of dopamine in the accumulation and binding of iron to melanin and formation of more dopamine melanin. The binding of s9Fe3+to dopamine melanin was dependent on the melanin concentration and pH (Figs. 3 and 4, respectively). Maximal binding was observed with 3.0 pg of melanin per 0.25 ml, at pH 6.0. Above and below these values, there was a significant reduction in the amount of 59Fe3+bound per microgram of melanin. The ability of iron to promote lipid peroxidation was substantiated in the present study using rat cerebral cortex homogenate. Although melanin at concentrations of 3.0 and 12 pg/0.25 ml significantly reduced basal lipid peroxidation, it potentiated the lipid peroxidative property of iron (100 p M ) at its higher concentration (Fig. 5). The increases in lipid peroxidation induced by iron over the basal level and over the melanin-treated tissue homogenates were 53 and 2 18%, respectively; these results were significantly different ( p < 0.001) as determined by the calculation of the t values of the ratios. In addition to the ability of desferrioxamine to inhibit s9Fe3+binding to dopamine melanin, this iron chelator was very efficient in inhibTABLE 1. The ability of drugs, iron chelators, and divalent ions to inhibit ''FeCl3 binding to dopamine melanin
U74500A Desferrioxamine Chlorpromazine 1,lO-Phenanthroline FeClz Apomorphine Haloperidol Dopamine NTA MPP' MPTP MnClz Chloroquine Spiperone
10
O
t
b
9
8
7
1611
6
5
4
3
Doparnine [-log MI
FIG. 2. The effect of dopamine on "Fe3+ binding to dopamine melanin. The procedurewas identicalto that described in Table 1, with the inclusion of various concentrations of dopamine. Results are means t SEM of three separate experiments.
iting the iron or iron-melanin-induced lipid peroxidation (Fig. 5). DISCUSSION Iron has been implicated in lipid peroxidation and cell degeneration more often than any other metal (Halliwell and Gutteridge, 1984; Triggs and Willmore, 1984; Dexter et al., 1989; Halliwell, 1989). It is the interaction of increased free iron with H202and the interconversion of iron between its redox states that initiates the liberation of cytotoxic hydroxyl free radicals (OH ) and promotes lipid peroxidation. Human basal ganglia are the most metabolically active parts of the brain, and have the highest activity of MAO-B (Collins et al., 1970; Riederer and Youdim, 1986).The oxidative deamination of dopamine by this enzyme and the autooxidation of dopamine to melanin generate a substantial amount of H202 (Konradi et al., 1986). The diminished levels of endogenous H202 scavengers, ascorbate and GSH, in parkinsonian SN (Riederer et al., 1989) predicts a significant accumulation of H202.In this condition, free Fe2+would be available to interact with H202 and drive the Fenton Reaction, resulting in the liberation of cytotoxic hydroxyl radical (OH. ) and Fe3+(Fig. 6). The optimal condition for participation of iron in OH formation and lipid peroxidation initiation requires a Fe2+/Fe3+
-
60 180
500 1,300 >2,000 22,000 >2,000
>2,000 >2,000 >2,000 >2,000
0
>2,000
>2,000
The binding of 10 nM 59FeClsto 3 pg of melanin was assayed as described in Fig. 1 in the presence of various compounds. Competition curves represent results from 14 concentrations of drugs ranging from to lo-' M (each in triplicate). ICS0values are means of three experiments. NTA, nitrilotriacetic acid.
5
10
15 Melanin [pgl
20
25
FIG. 3. The effect of dopamine melanin concentration on 59Fe3+. The binding procedure consisted of incubation of 10 nM 59FeC13 with various concentrationsof dopamine melanin in 5 mM Tris-HCI buffer, pH 6.5, for 2 h according to the procedure outlined in Materials and Methods for Scatchard analysis. Results are means SEM of five separate experiments.
+
J. Neurachem.. Vat. 57, No. 5, 1991
D. BEN-SHACHAR ET AL.
1612
4
6
8
10
PH
FIG. 4. The effect of pH on "Fe3+ binding to dopamine melanin. The specific binding of 10 nM 59Fe3+to 3 pg of melanin in 5 mM Tris-HCI buffer was assayed as described in Fig. 1. The various pH values were achieved by adding either HCI or NaOH to the Tris buffer. Results are means SEM of four experiments.
+
ratio of 1:1 (Minotti and Aust, 1987). This ratio has been observed in the SN of parkinsonian brains as a result of an increase in the Fe3+form (Riederer et al., 1989). Normally, Fe3+ is bound to endogenous chelators such as ADP or melanin, thus shifting the equilibrium in favor of the chelated form. On the other hand, an increased concentration of Fe3+ also could serve as the catalyst for its conversion to Fez+by melanin, and, in the presence of H202,result in further formation of OH. (Pilas et al., 1988). This reaction has been attributed to the ability of melanin to bind Fe3+as well as to reduce it to Fe2+,depending on environmental conditions. In the absence of significant amounts of fenitin in the SN, iron could be chelated by melanin. The present study demonstrates the high specific capacity of melanin to bind iron. Our experiments with 59Fe3+and the Scatchard analyses of the binding indicate the presence of two (low- and highaffinity) binding sites on dopamine melanin. However, our preliminary studies with synthetic noradrenaline melanin (unpublished data) demonstrated significantly lower capacity for binding 59Fe3+. The binding probably occurs to the free carboxyl groups, phenolic hydroxyl moiety, or quinoid groups present on the melanin polymer (Larsson and Tjalve, 1979). The binding of 59Fe3+ to dopamine melanin appears to be relatively specific because flunitrazepam did not bind to dopamine melanin and spiperone demonstrated low affinity. Iron could be displaced from dopamine melanin only by compounds with iron chelating capacity. This was especially evident in the case of U74500A and desfemoxamine, an iron chelator employed in treatment of systemic organ iron overload (Johnson et al., 1982).Neither MPTP nor MPP', which have high affinity for dopamine melanin and neuromelanin (DAmato et al., 1986), was effective as an inhibitor of iron binding. Evidently, dopamine melanin has a significantly higher number of binding sites for iron (Fig. 1) than for MPP+ (D'Amato et al., 1986). Moreover, chloroquine, which binds to melanin (Larsson and Tjalve, 1979) and displays selectivity for displacing MPP+ from neuromelanin and dopamine melanin, and partially prevents MPTP-induced parkinsonism in monkeys (DAmato et al., 1987), did not J Neurochem.. Vol 57, N o 5. 1991
alter the binding of 59Fe3+to dopamine melanin even M. By contrast, dopamine potentiated the at binding of iron to dopamine melanin, suggesting the possible participation of dopamine in the binding of iron to melanin and the formation of more dopamine melanin as a protective mechanism against iron accumulation. The optimal observed conditions for iron binding to dopamine melanin with 3 p g of melanin/ 0.25 ml at pH 6.0 (Figs. 3 and 4)are compatible with the optimal transport and binding of iron in vivo (Octave et al., 1983) and the induction of oxygen free radicals (Pilas et al., 1988). A decrease in the binding of Fe3+ to dopamine melanin was detected as melanin concentration exceeded 5 pg/0.25 ml. The probability of interaction between the individual melanin chains may increase at higher concentrations of melanin because of their ability to bind iron. Melanin in normal circumstances is considered an effective radical scavenger in vitro and in vivo (Schwabe et al., 1989; Scalia et al., 1990). However, under certain conditions, such as increased availability of Fe3+,melanin can potentiate the formation of oxygen radicals. The production of OH in the presence of melanin is significantly greater when Fe3+is predominant (Pilas et al., 1988) and is further demonstrated by the greater lipid peroxidation of rat cerebral cortex in the presence of Fe3+and higher dopamine melanin concentrations (Fig. 5). This is consistent with our results, demonstrating that high melanin concentrations reduce 59Fe3+ binding to dopamine melanin. According to the reaction pathway in Fig. 6, H202, generated by the deamination and autooxidation of dopamine, participates with free iron and melanin to drive the Fenton 0
n
I
20
0
10
13;
DFO
+ +
DFO
+ +
DFO
+
I Fe3+ Fe3+ F e 3 '
D A - M DA-M 3 12
FIG. 5. Lipid peroxidation properties of iron and dopamine melanin (DA-M). Lipid peroxidation was assayed by measurement of malondialdehyde(MDA) formation in rat cortex homogenate. BL, basal lipid peroxidation; Fe3+,MDA formation in the presence of 10-4 M Fe2(S0&; DA-M, 3 or 12 pg/ml of melanin; DFO, 1 0-4 M desferrioxamine; and their combinations. Results are means SEM of triplicate determinationof five experiments. Values are expressed as nanomoles of MDA production per milligram of protein above basal levels in rat cortical homogenates. *p 0.001 compared with the control basal level (BL); **p i0.001 compared with the 3-pg DA-M response.
*
-=
IRON-MELANIN BINDING AND PARKINSON’S DISEASE
1613
Hz0
Feedback Stimulation
DOPAMINE
MAO-B
+ R-CHO+NH3+H202 FIG. 6. The reaction pathway illustrating the ability of H202and melanin to alter the redox of iron between its two valences. In the event of increased iron and diminished disposition of HZ02by glutathione peroxidase, the net effect would be the formation of cytotoxic hydroxyl (OH - ) and initiation of lipid peroxidation.
Electron
I
Cytotoxic Hydroxyl Radical
j.
Lipid Peroxidation
-
Reaction, thereby liberating cytotoxic OH . The figure demonstrates a system by which H202 and melanin serve as catalysts for the conversion of Fe3+to Fez+. Furthermore, this model may explain the parkinsonism in manganese miners (Donaldson and Barbeau, 1985). The interconversion of manganese, like iron, between its two redox states, resulting in generation of reactive oxygen species (e.g., OH.), has also been proposed as the mechanism of striatal neuronal membrane toxicity (Graham, 1984;Archibald and Tyree, 1987). The ability of melanin to chelate manganese (Potts and Au, 1976; Larsson and Tjalve, 1979) supports this hypothesis. It is also apparent that both iron (Graham et al., 1978) and manganese (Graham, 1984; Liccione and Maines, 1988) deplete tissue levels of GSH, which has been reported to be decreased in parkinsonian brains (Riederer et al., 1989). In the brain, glutathione peroxidase acts as the main cellular defense mechanism in the destruction of H202, by oxidizing GSH to GSSG, in the presence of H202(Liccione and Maines, 1988). It is believed that glutathione peroxidase, and not catalase, in the brain is responsible for removal of monoamine oxidase-generated H202(Oshino and Chance, 1977; Spina and Cohen, 1989) (Fig. 6). In Parkinson’s disease, the heavily pigmented neurons of SN are preferentially lost (Mann and Yates, 1983). Hirsch et al. (1988) clearly demonstrated the greater vulnerability of neuromelanin-containing dopamine neurons to the neurodegenerative process of Parkinson’s disease. We believe that the results of the present study are directly relevant to the pathophysiology of Parkinson’s disease because of the selectivity of iron increase (Dexter et al., 1987; Riederer et al., 1989) and lipid peroxidation (Dexter et al., 1989) in parkinsonian SN. Furthermore, they may explain the
/
selectivity and vulnerability of melaninized dopamine neurons in SN to neurodegeneration. It remains to be seen whether the increased deposition of iron noted in the SN of Parkinsonian brains is within the dopamine neurons and whether it is deposited in these neurons after a neurotoxic insult. Preliminary studies with MPTP-induced dopaminergic neurotoxicity in mice (D. Ben-Shachar, E. Melamed, and M. B. H. Youdim, unpublished data) and monkeys (E. Domino, personal communication) show no alterations in iron content of the striatum. Thus, it seems that neurodegeneration per se does not appear to contribute to elevation of SN iron. The results of the present study predict that future prospects for the treatment or retardation of the process of neurodegeneration in Parkinson’s disease may involve the use of iron chelators, such as U74500A (Braughler et al., 1988),capable of crossing the bloodbrain barrier in a fashion similar to chelators used to treat Wilson’s disease and cases of iron overload in systemic organs (Halliwell and Gutteridge, 1986). Acknowledgment: This work was supported by grants from Golding Research Fund for Parkinson’s Disease (U.S.A.) and Research and Development Grant from the Technion (Haifa) and Israel Center for Psychobiology (Jerusalem).
REFERENCES Archibald F. S. and Tyree C. (1 987) Manganese poisoning and the attack of trivalent manganese upon catecholamines. Arch. Biochem. Biophys. 256,638-650. Braughler J. M., Duncan L. A,, and Chase R. L. (1986) The involvement of iron in lipid peroxidation: importance of fenic to ferrous ratio in initiation. J. Eiol. Chem. 261, 10282-10289.
J. Neurochem., Vol. 57, No. 5 , 1991
1614
D. BEN-SHACHAR E T AL.
Braughler J. M., Burton P. S., Chas R. L., and Pregenzer J. F. (1 988) Novel membrane-localized iron chelators as inhibitors of irondependent lipid peroxidation. Biochem. Pharmacol. 37, 38533860. Bruenger F. W., Stover B. J., and Atherton D. R. (1967) The incorporation of various metal ions into in vivo and in vitro produced melanin. Radiat. Res. 32, 1-12. Collins G. G., Sandler M., Williams E. D., and Youdim M. B. H. (1970) Multiple forms of human brain mitochondria1 monoamine oxidase. Nature 225, 817-820. D’Amato R. J., Lipman Z. P., and Snyder S. H. (1986) Selectivity of the parkinsonian neurotoxin MPTP toxic metabolite MPP’ binds to neuromelanin. Science 231, 987-989. DAmato R. J., Alexander G. M., Schwartzman R. J., Kitt C. A., Price D. L., and Snyder S. H. (1987) Evidence for neuromelanin involvement in MPTP-induced neurotoxicity. Nature 327,324326. Das K. C., Abramson M. B., and Katzman R. (1978) Neuronal pigments: spectroscopic characterization of human brain melanin. J. Neurochem. 30,601-605. Dexter D. T., Well F. R., Agid F., Agid Y . ,Lees A. J., Jenner P., and Marsden C. D. (1987) Increased nigral iron content in postmortem parkinsonian brain. Lancet ii, 12 19- 1220. Dexter D. T., Carter C . J., Wells F. R., Javoy-Agid F., Agid Y ., Lees A., Jenner P., and Marsden C. D. (1989) Basal lipid peroxidation in substantia nigra is increased in Parkinson’s disease. J. Neurochem. 52, 38 1-389. Donaldson J. and Barbeau A. (1985) Manganese neurotoxicity: possible clues to the etiology of human brain disorders, in Metal Ions in Neurology and Psychiatry (Gabay S., Hams J., and Ho B. T., eds), pp. 259-285. Alan R. Liss, New York. Drayer B. P., Olanow W., Burger P., Johnson G. A,, Hertkens R., and Riederer P. ( 1986) Parkinson plus syndrome: diagnosis using high-field MR imaging of brain iron. Radiology 159, 493-498. Earle K. M. (1968) Trace metals in parkinsonian brains. J. Neuropathol. Exp. Neurol. 27, 1-14. Graham D. G. (1984) Catecholamine toxicity: a proposal for the molecular pathogenesis of manganese neurotoxicity and Parkinson’s disease. Neurotoxicology 5, 83-96. Graham D. G., Tiffany S. M., Bell W. R. Jr., and Gutknecht W. F. (1978) Autooxidation versus covalent binding of quinones as the mechanism of toxicity of dopamine, 6-hydroxydopamine, and related compounds towards C- I300 neuroblastoma cells in vitro. Mol. Pharmacol. 14, 644-653. Halliwell B. (1989) Oxidants and the central nervous system, some fundamental questions: is oxidant damage relevant to Parkinson’s disease, Alzheimer’s disease, traumatic injury, or stroke? Acta Neurol. Scand. Suppl. 126,23-33. Halliwell B. and Gutteridge J. M. (1984) Oxygen toxicity, oxygen radicals, transition metals, and disease. Biochem. J. 219, 1-14. Halliwell B. and Gutteridge J. M. C. (1986) Oxygen free radicals and iron in relation to biology and medicine: some problems and concepts. Arch. Biochem. Biophys. 246, 501-5 14. Hirsch E., Graybiel A. M., and Agid Y . (1988) Melanized dopaminergic neurons are differentially susceptible to degeneration in Parkinson’s disease. Nature 334, 345-348. Johnson D. K., Pippard M. J., Murphy T. B., and Rose N. J. ( 1 982) An in vivo evaluation of iron-chelating drugs derived from pyridoxal and its analogs. J. Pharmacol. Exp. Ther. 221, 399-403. Konradi C., Riederer P., and Youdim M. B. H. (1986) Hydrogen peroxide enhances the activity of rnonoamine oxidase type-B but not that of type-A: a pilot study. J. Neural Transm. Suppl. 22,61-74.
J. Neurochem., Vol. 5 7, No. 5 , I991
Larsson B. and Tjalve H. (1979) Studies on the mechanism of drugs binding to melanin. Biochem. Pharmacol. 28, 1181-1 187. Liccione J. J. and Maines M. D. (1988) Selective vulnerability of glutathione metabolism and cellular defence mechanism in rat stnatum to manganese. J. Pharmacol. Exp. Ther. 247,156- 16 I . Mann D. M. and Yates P. 0. (1983) Possible role of neuromelanin in the pathogenesis of Parkinson’s disease. Mech. Ageing Do. 21, 193-203. Markey S. P., Castagnoli N. Jr., Trevor A. J., and Kopin I. J., eds (1986) MPTP: A Neurotoxin Producing a Parkinsonian Syndrome. Academic Press, New York. Minotti G. and Aust S. D. (1987) The requirement for iron 111 in the initiation of lipid peroxidation by iron I1 and hydrogen peroxidase. J. Biol. Chem. 262, 1098-1 104. Octave J. N., Schneider Y. J., Trouet A., and Crichton R. R. (1983) Iron uptake and utilization by mammalian cells. I. Cellular uptake of transfemn and iron. Trends Biochem. Sci. 8, 217-220. Oshino N. and Chance B. (1977) Properties of glutathione release observed during reduction of organic hydroperoxide, demethylation of aminopyrine, and oxidation of some substances in perfused rat liver, and their implications for the physiological function of catalase. Biochem. J. 162, 509-525. Pilas B., Sama T., Kalyanarman B., and Swartz H. M. (1988) The effect of melanin on iron-associated decomposition of hydrogen peroxide. Free Radic. Biol. Med. 4, 285-293. Potts A. M. and Au P. C. (1976) The affinity of melanin for inorganic ions. Exp. Eye Res. 22,487-49 1. Rehncrona S., Smith D. S., Akesson B., Westerberg E., and Siesjo B. H. (1980) Peroxidative changes in brain cortical fatty acids and phospholipids, as characterized during Fe2+-and ascorbic acid-stimulated lipid peroxidation in vitro. J. Neurochem. 34, 1630- 1638. Riederer P. and Youdim M. B. H. (1986) Monoamine oxidase activity and monoamine metabolism in brains of parkinsonian patients treated with L-deprenyl. J. Neurochem. 46, 1359- 1365. Riederer P., Sofic E., Rausch W. D., Schmidt B., Reynolds G. P., Jellinger K., and Youdim M. B. H. (1989) Transition metals, femtin, glutathione, and ascorbic acid in parkinsonian brain. J. Neurochem. 52, 5 15-520. Rutledge J. N., Hilal S. K., Silver A. J., Defendini R., and Fahn S. (1987) Study of movement disorders and brain iron by magnetic resonance. AJR Am. J. Roentgenol. 149, 365-379. Salazar M., Sokoloski T. D., and Patil P. N. (1978) Binding of dopaminergic drugs by the neuromelanin of the substantia nigra, synthetic melanins, and melanin granules. Fed. Proc. 37,24032407. Scalia M., Geremia E. V. C., Santoro C., Baratta D., and Sichel G. (1 990) Lipid peroxidation and unpigmented liver tissues: protective role of melanin. Pigment Cell Res. 3, 1 1 5-1 19. Schwabe K., Lassmann G., Damerau W., and Naundorf H. (1989) Protection of melanoma cells against superoxide radicals by melanin. J. Cancer Res. Clin. Oncol. 115, 597-600. Spina M. B. and Cohen G. J. (1989) Dopamine turnover and glutathione oxidation: implications for Parkinson’s disease. Proc. Natl. Acad. Sci. USA 86, 1398-1400. Tnggs W. J. and Willmore L. J. (1984) In vivo lipid peroxidation in rat brain following intracortical Fe2+injection. J. Neurochem. 42,976-980. Youdim M. B. H., Ben-Shachar D., and Riederer P. (1989) Is Parkinson’s disease a progressive siderosis of substantia nigra resulting in iron- and melanin-induced neurodegeneration? Acta Neurol. Scand. 126, 41-54.
