Journal of Inorganic Biochemistry 89 (2002) 54–60

www.elsevier.com / locate / jinorgbio

Redox behavior of melanins: direct electrochemistry of

dihydroxyindole-melanin and its Cu and Zn adducts
Shirley Gidanian, Patrick J. Farmer*
Department of Chemistry, University of California, Irvine, Irvine, CA 92697 -2025, USA
Received 30 July 2001; received in revised form 17 October 2001; accepted 5 November 2001

Abstract

Synthetic melanin films, formed on electrode surfaces by oxidative polymerization of 5,6-dihydroxyindole solution, were used to
directly measure the chromophore’s redox reactivity. Films on optically transparent indium-tin oxide (ITO) electrodes allow correlation of
spectral changes with electrochemical potential. Spectroelectrochemical titrations show an initial reversible transformation that is ascribed
to formation of a unique quinone-imine chromophore. The apparent E1 / 2 for maximum quinone-imine formation is |125 mV (vs.
Ag /AgCl) but at potentials higher than 100 mV, an irreversible bleaching is evident. Correlation of the current with the monomer
concentration implies that only one in six monomers is oxidized to the quinone-imine before the irreversible bleaching occurs. Films
pretreated with CuCl 2 and Zn(CH 3 COO) 2 show elevated quinone-imine absorbances, even under reducing conditions, indicating a
preferential stabilization of this state by coordination to the metals.  2001 Elsevier Science Inc. All rights reserved.

Keywords: Melanin; 5,6-Dihydroxyindole (DHI); Quinone-imine; Spectroelectrochemistry; Redox titration; Cu; Zn

1. Introduction within their cell walls [13]. Fe absorbance by melanin in

Melanins are one of the largest classes of natural
pigments in the animal kingdom, commonly formed by
oxidative polymerization of catecholic species derived
from tyrosine [1–3]. In humans, they are found in the hair
and epidermis, as well as in areas of high nerve activity
such as the middle ear, retina, and substantia nigra of the
brain. They also may occur in a number of other tissues
either as a result of inflammation or as an important
product of malignant tumors (usually termed melanomas)
that arise from the melanin producing cells, the
melanocytes [4]. An assumed physiological function of
melanin is to buffer against photochemical stress, as the
pigment is generated in response to UV damage [5]. But
melanins are also known to mediate oxidative stress by
neutralizing reactive oxygen species (ROS), such as
superoxide and peroxide [6–8], and by absorbing metal
ions and radicals which promote oxidative damage [9–12].
Melanin’s redox activity has been linked to both pro-
and anti-oxidant behavior, with important medical implica-
tions. For example, certain ‘black bacteria’ are protected
against phagocytic mechanisms by a layer of melanin
*Corresponding author. Fax: 11-949-824-2210.
E-mail address: pfarmer@uci.edu (P.J. Farmer).
the brain is thought to promote Fenton-like formation of
hydroxy radicals, and induce Parkinsonism [14]. Likewise,
melanin’s reaction with oxygen and other reactive oxygen
species are likely important in understanding their physio-
logical roles [15,16].
The chemical basis for melanin reactivity is its cate-
cholic subunits, dihydroxyindoles, which can undergo two
electron oxidations to the corresponding quinone form
(Scheme 1). Melanins are subject to irreversible degra-
dation with a loss, or bleaching, of the characteristic
pigment color [16]. Unusual features of both native and
synthetic melanins are persistent radicals, as characterized
by an electron paramagnetic resonance (EPR) signal,
indicative of a sizable population of the semiquinone
within the polymer [9,17,18].
As native melanins are in the form of colloidal particles,
they are difficult to examine using standard homogeneous
solution techniques, and few details are known about the
physicochemical nature of polymeric melanin’s redox
activity [19]. Numerous electrochemical studies on
eumelanin precursors and related catecholamines have
been performed in which a darkening of the sample
solution around the anode has been proposed to indicate
melanin formation, but the colloidal particles thus formed
are typically electrochemically silent [20]. Polymeric

0162-0134 / 01 / $ – see front matter  2001 Elsevier Science Inc. All rights reserved.
PII: S0162-0134( 01 )00405-6
 S. Gidanian, P. J. Farmer / Journal of Inorganic Biochemistry 89 (2002) 54 – 60
Scheme 1. Speciation of monomers within DHI melanin.
melanin films have been shown to display unique electro-
chemical and semiconductor properties [21–23].
In this report, we follow the redox reactivity of polymer-
ized 5,6-dihydroxyindole (DHI) first reported by Horak
and Weeks [24]. As made, the polymerized DHI (poly-
DHI) films have a bluish gray color, with a substantial
absorbance throughout the visible range. Changes in the
absorbance with applied potential are used to follow
reactivity within the melanoid film. By this method, a
direct measure of reversible and irreversible oxidations of
the chromophore is obtained, as well as the effect of
absorbed Cu 21 and Zn 21 within the polymer. Our results
suggest reactivities assignable to specific components of
that structure, as well as to the in-vivo properties of
various forms of melanin.
2. Materials and methods
2.1. General experimental details
5,6-Diacetoxyindole (DAI) was obtained from TCI
America, Portland, OR. Indium-tin oxide plates were
purchased from Quantum Coating, Mt Laurel, NJ. All
other chemicals were reagent grade quality. Mass spectra
were obtained on a VG analytical 707E instrument. Laser
desorption ionization (LDI) measurements were performed
on a Perspective Biosystem DE STR time-of-flight instru-
ment, operating in positive-ion linear mode. Ions were
formed by a pulsed UV laser beam (nitrogen laser, l5337
nm). Absorbance spectra were obtained on a Hewlett-
Packard 8453 UV/ visible spectrometer. Voltammograms
were obtained with commercial gold or hand cut indium-
tin oxide (ITO) plates. A three-electrode voltammetric cell
was used with a platinum counter electrode and Ag /AgCl
as reference electrode. Cyclic voltammetry experiments
were carried out with a BAS-100B-W electrochemical
analyzer. Anaerobic experiments were carried out in a
glovebox, or by purging the analyte solution with nitrogen
for at least 30 min. All reported potentials are versus
Ag /AgCl reference electrode.
AT-cut quartz crystals with a basic resonant frequency
of 5 MHz were purchased from Colorado Crystal Corpora-
tion. The quartz plates (15 mm diameter) with gold
electrodes (8.5 mm diameter) were fixed in the measuring
cell made of aluminum whose temperature was controlled
by a heater (DALE RH-10) attached to a DC power supply
55
(HP E3617A) and the frequency data were recorded by a
quartz crystal microbalance (QCM) measurement system
(HP 8751A 5 Hz–500 MHz network analyzer) with
LabVIEW 6.0 software.
2.2. Poly-DHI film
All steps were performed in a glovebox under a nitrogen
atmosphere. A DHI solution was prepared by hydrolyzing
DAI in situ: DAI (0.030 mmol, 6.0 mg) was dissolved in
12.5 ml of degassed ethanol to which 2 eq. of 0.1 M NaOH
(0.55 ml) needed to hydrolyze the acetoxy groups was
added. This solution was allowed to stand for 1 h and then
diluted up to 25 ml with pH 7.0, 0.05 M phosphate buffer.
Polymerization of DHI on electrode surface was achieved
using cyclic voltammetric deposition for 15 cycles at a
sweep rate of 50 mV s 21 for potentials varying from 20.40
to 0.35 V versus Ag /AgCl. Similar procedures were used
for deposition on gold or ITO substrates.
2.3. Spectroelectrochemical measurements
In a typical procedure, a poly-DHI film deposited on
ITO plate was placed in a three-neck, anaerobic UV-visible
cell containing 0.1 M LiCl / 0.05 M phosphate buffer pH
7.0 solution as electrolyte. After obtaining the absorbance
spectrum of the film without applied potential as a
reference, the ITO electrode was connected to the electro-
chemical analyzer as working electrode, using a platinum
wire as counter electrode and an Ag /AgCl reference
electrode. Variable potentials were applied to the film and
the absorbance spectrum of the film was obtained. Subtrac-
tion of the reference spectrum from these spectra provided
the background subtracted spectra.
2.4. Current efficiency and stoichiometry
In order to quantitate the electron efficiency of the
polymerization process, films were made by bulk electro-
chemical oxidation of DHI solution onto a gold plated
quartz crystal microbalance, and the mass-induced fre-
quency shift determined. In a typical experiment, poly-DHI
film was deposited on one side of the quartz electrode by
bulk electrolysis of the DHI solution at 20 mV, after
recording the original frequency of the crystal. The total
current passed through the electrode was 4.415310 23 C,
which corresponds to 2.287310 28 mol of DHI deposited.
56 S. Gidanian, P. J. Farmer / Journal of Inorganic Biochemistry 89 (2002) 54 – 60

The crystal was then rinsed with water and ethanol, dried
under nitrogen, and its frequency was measured. A 60-Hz
frequency shift was detected which correspond to 1.2 mg
(0.805 mol) of the material deposited on the surface.
Further deposition experiments were done on both gold
and ITO electrode surface at 20 mV versus Ag /AgCl. The
resulting films were then rinsed several times with deion-
ized water and stored in the glovebox.
Bulk electrolysis of the deposited films allowed quantifi-
cation of the oxidative current passed at various potentials
(after correction for the deposition efficiency). In a typical
set of experiments, a deposition current of 2.7310 23 C
was passed to form a film on Au; whereas, after cleaning,
only 19.7310 25 C were passed during stepwise bulk
electrolysis at 50-mV intervals from 2100 to 50 mV, and
1.1310 23 C total from 2100 to 300 mV. These data were
used to calculate the current passed during the spectrally-
characterized reversible oxidation, and the subsequent
irreversible oxidations.
2.5. Mass spectral characterization
To obtain mass spectral characterization, poly-DHI films
were electrodeposited directly onto stainless steel targets
suitable for use with the MALDI–TOF instrumentation
using cyclic voltammetry technique, by connecting the
sample holder to electrochemical analyzer as the working
electrode.
2.6. Cu- and Zn-treated poly-DHI
Poly-DHI films made by cyclic voltammetry on ITO
electrodes were soaked in a 0.1 M CuCl 2 , or 0.1 M
Zn(CH 3 COO) 2 solution under nitrogen atmosphere over-
night. The electrodes were then rinsed with deionized
water. Spectroelectrochemical experiments on the Cu- or
Zn-treated ITO electrodes were as stated above.
Scheme 2. Oxidative deposition of DHI film.
mechanically stable and insoluble in water or other organic
solvents tested.
The voltammetric response of the poly-DHI films is
slow due to limited charge and ion transport through the
multi-layer film; only broad, low-current waves are seen
for the polymer itself at normal scan rates above 50 mV
s 21 . Cyclic voltammograms of the film at very slow rates
(,1 mV s 21 ) show an oxidation at 100 mV versus Ag /
AgCl (Fig. 2). The reduction peak is broad and small,
indicating a depletion of the redox active component
following oxidation.
3.2. Spectroelectrochemical measurements
Though highly colored, thin films of DHI-melanin are
transparent, and characterizable by their absorption spectra.
Spectroelectrochemical redox titrations of films deposited
on optically transparent film indium-tin oxide (ITO)
electrodes were performed to characterize the change in
absorbance with applied potential.
Fig. 3 illustrates the spectral change of a poly-DHI film
as the applied potential is varied from 250 mV versus
Ag /AgCl to 150 mV. An increase in absorbance is seen at
500 nm, and a decrease at 380 nm, with an isosbestic point
at 417 nm which implies a clean interconversion of
species. Under anaerobic conditions, this absorbance
change is reversible and can be repeated several times. The
difference spectra in Fig. 3 are similar, in both maxima and

3. Results and discussion

3.1. DHI-melanin films

To measure the redox behavior of colloidal melanins,

we mimic biological melanogenesis by oxidative poly-
merizing melanin onto electrodes. Voltammetric scans of
an aqueous solution of DHI initiates the polymerization
and deposits a film of poly-DHI on the working electrode
surface (Scheme 2) [24].
The sequential voltammograms obtained during deposi-
tion are shown in Fig. 1. The anodic peak of the first scan
corresponds to the oxidation of DHI on a clean electrode
surface; subsequent cycles display the blocking of the DHI
oxidation and a reductive component due to the growing Fig. 1. Cyclic voltammetric polymerization of 5,6-dihydroxyindole (1.0
poly-DHI film. Film thickness can be controlled by the mM in pH 7.0, 50 mM phosphate buffer) at scan rate of 50 mV s 21 on
scan rate and number of cycles. Once formed, the films are small gold disk electrode for ten cycles.
 S. Gidanian, P. J. Farmer / Journal of Inorganic Biochemistry 89 (2002) 54 – 60
Fig. 2. Cyclic voltammogram of a poly-DHI film in 0.1 M KCl and pH
7.0, 0.1 M phosphate buffer at 0.5 mV s 21 scan rate.
Fig. 3. Open-circuit subtracted absorbance spectra of poly-DHI film on
ITO demonstrating reversibility of oxidation. Sequential spectra obtained
after applying potentials of: (a) 250 mV; (b) 0 mV; (c) 250 mV; and (d) 0
mV. Conditions as described in the text.
isosbestic point, to those obtained during chemical oxida-
tion of L-dopa, attributed to the formation of a quinone-
imine intermediate, dopachrome [26,27]. Thus the revers-
ible spectral changes are due to formation of such quinone-
imine functionalities within the DHI-melanin, formed by
tautomerization from the two-electron oxidized quinone
state (Scheme 3).
At potentials in the range from 150 to 1300 mV, this
characteristic absorbance increases, but the reversibility is
Scheme 3. Tautomerization of quinone to quinone imine.
57
Fig. 4. Open-circuit subtracted spectra of poly-DHI film on ITO demon-
strating irreversible bleaching. Sequential spectra obtained after applying
potentials of: (a) 2300 mV; (b) 2100 mV; (c) 0 mV; (d) 100 mV; (e) 300
mV; (f) 500 mV; and (g) 600 mV. Conditions as described in the text.
lost, indicating a change in the chemical composition (Fig.
4). The irreversible changes are more dramatic at po-
tentials above 1300 mV, where irreversible bleaching is
much increased at higher wavelength, e.g. 700 nm (Fig. 4).
Plotting the absorbance change versus applied potential
gives the apparent E1 / 2 potential for the quinone-imine
formation at |125 mV versus Ag /AgCl (Fig. 5). The loss
of reversibility above 50 mV suggests that the chemical
decomposition is dependent on the concentration of quin-
one-imine monomers.
The absorbance of an as-made film typically shows
substantial quinone-imine absorbance, and can be used to
estimate the resting redox state of the melanin. These
resting potentials vary somewhat from sample to sample,
but are generally in the range of 0 to 50 mV, i.e. reversibly
oxidized. The film’s resting redox state is also affected by
its age and history, e.g. exposure of the dry films to oxygen
Fig. 5. Plot of absorbance versus applied potential for poly-DHI film on
ITO at (a) 500 nm and (b) 700 nm.
58 S. Gidanian, P. J. Farmer / Journal of Inorganic Biochemistry 89 (2002) 54 – 60

slowly increases their observed resting potential. For this

reason, films were made and stored under an inert atmos-
phere. The presence of oxygen has a dramatic effect on the
absorbance changes during spectroelectrochemical elec-
trolysis, and will be the subject of a subsequent report.

3.3. Quantification of the reversible and irreversible

oxidations

To determine the number of electrons per monomer

during chromophore formation required a measurement of
both the current and the number of monomers. The current
efficiency for deposition was determined using a quartz
crystal microbalance to measure the mass of DHI-melanin
deposited on the electrode surface. The expected oxidative
current needed for high poly-DHI polymer formation
corresponds to two electrons per monomer. The deposition
current was kept small, to be comparable to the thin films
used in spectroelectrochemical measurements, and was
therefore likely to contain a high component of non-
Faradaic background current. Under these conditions, the
current efficiency for deposition over a set of experiments
was calculated to be 44%, corresponding to a surface
coverage of |8.75 nmol / cm 2 for a close-packed film of
DHI monomers (lower limit |18 layers of poly-DHI).
Using these films, the current observed during the revers-
ible transformation corresponded to only 16% of the total
expected oxidation current. By this measure, only a small
number of monomers are reversibly oxidized before ir-
reversible changes occur.
Laser desorption / ionization mass spectrometry (LDI-
MS) of the poly-DHI film displays repeating peak clusters,
suggesting the sample is composed of oligomers rather
than long polymeric chains. Peak clusters with m /z differ-
ences of 147 mass units are typical, and can be ascribed to
cationic species originating from trimers up to dodecamers
of DHI. Similar clusters have been seen for colloidal
DHI-melanin solutions, and for a number of synthetic and
native melanins [25,28,29]. Quantification of the polydis-
persity of the oligomers is difficult, as desorption selective-
ly favors the smaller chained oligomers. Microscopy
studies have suggested a fundamental repeating unit of
synthetic melanins, melanin protomolecules, modeled as
randomly-coupled pentamers and hexamers of DHI
subunits, as illustrated below. Using this model, the
reversible oxidation corresponds to about two electrons
(one quinone-imine) per protomolecule.
Fig. 6. Plot of absorbance versus applied potential for (a) poly-DHI film,
(b) Cu-treated poly-DHI film, and (c) Zn-treated poly-DHI film at 500
nm.
The total oxidative charge during the quinone-imine
chromophore formation (between 250 and 1300 mV)
was determined to be |1.1 electrons per monomers, about
half that required for a stoichiometric conversion of
quinole to the quinone. Thus the buffering capacity of
melanins in this range is much less than expected.
3.4. Effect of absorbed Cu and Zn
It is known that melanins absorb metal ions in vivo [30]
and these metals may dramatically alter the redox reactivi-
ty of the polymer [16]. The effect of absorbed metal ions
on the melanin films’ spectroelectrochemical behavior was
assayed by pretreatment with metal ion solutions. The
redox titration results for poly-DHI sample as well as
Zn 21 - and Cu 21 -treated samples are shown in Fig. 6,
obtained by first fully reducing the films and varying the
potential positively.
Metal-treated samples show a dramatic increase in
absorbance at 500 nm, which varies by metal, and is
evident before applying potential. This is likely due to the
effect of the metal ions on the equilibrium between
quinone and quinone-imine functionalities in the film
(Scheme 4). The absorbance of Zn-treated poly-DHI film
at open circuit corresponds to an applied potential of |200
mV for metal-free films, and for Cu-treated samples the
absorbance is somewhat above that obtainable before the
irreversible bleaching occurs, implying a high concen-
tration of quinone-imine. Reduction induces the loss of
weakly bound ions, as evident by a loss of absorbance. But

Scheme 4. Effect of metal binding on quinone tautomerization.

 S. Gidanian, P. J. Farmer / Journal of Inorganic Biochemistry 89 (2002) 54 – 60
after the initial reduction, the reversibility of electrochemi-
cal oxidation of both Cu- and Zn-treated films are similar
to that of the metal-free samples, indicating a stable
concentration of bound metal. The maximum concentration
of quinone-imine obtainable remains relatively constant for
all samples, suggesting that the decomposition pathway is
dependent on the quinone-imine concentrations and in-
sensitive to metal content.
Complete reduction (to 2500 mV) only partially de-
creases the absorbance of metal-treated DHI-melanins. The
Zn-treated samples retain 50% of the highest quinone-
imine absorbance, the Cu-treated sample close to 70%,
roughly in line with the affinity of the metal ions for
melanin [31]. This indicates that binding of metals makes a
population of quinone-imine functionalities inert to electro-
chemical reduction, and therefore limits the redox buffer-
ing ability of the melanin.
Metal-binding may also effect the oxidation potential of
the melanins. As derived from redox titrations in Fig. 6,
the apparent E1 / 2 for poly-DHI and for the Cu-treated
sample are close (150 and 125 mV), but that for the
Zn-treated sample is much lower, |0 mV. Repeated
experiments show that Zn-treatment makes the melanin
substantially easier to oxidize. The shift of potential
implies a preference for Zn-coordination to the oxidized
quinone-imine form, which would perturb the redox
equilibrium towards that state under these conditions [32].
The small shift seen in Cu-treated samples would likewise
suggest an equal affinity of both the quinole and quinone
states for Cu-binding in the electroactive DHI monomers.
4. Conclusions
Depositing a melanin film on an electrode surface offers
several advantages in studying its chemical reactivity: (i)
the films are easy to prepare and handle as compared to
colloidal melanin solutions; (ii) they can be formed on
transparent supports which allows coupled electrochemical
and spectroscopic analysis; and (iii) they allow a platform
for following reactions with solution-based species, such
as metal absorption by the melanin.
Using this simple method we have obtained direct
measurements of DHI-melanin’s oxidation, and examined
the effect of absorbed metal ion on this behavior. The
oxidation of melanin proceeds in two distinct steps, the
first of which is reversible, but which corresponds to
formation of one quinone per six monomers. The spectral
changes associated with the reversible oxidation are inter-
preted as formation of quinone-imine functionalities. The
second reactivity is an irreversible bleaching reaction,
beginning after substoichiometric oxidation, with a maxi-
mum of |50% conversion of quinole to the quinone-imine.
Thus natural melanins likely act as disposable buffers of
oxidative stress, in that localized decomposition of mono-
mers occurs upon oxidation, rather than wholescale con-
version of quinoles to quinones.
59
The formation of the quinone-imine tautomers is pro-
moted by absorption of metal ions such as Zn(II) and
Cu(II). The metal-bound quinone-imines are then inert to
reduction, but not to the irreversible bleaching upon
oxidation. The effect of the metal-ion binding on the
oxidation potential of the melanin will depend on the
binding properties of the metal, and may substantially vary
the potential at which the oxidation occurs.
5. Abbreviations
DAI 5,6-diacetoxylindole
DHI 5,6-dihydroxyindole
EPR electron paramagnetic resonance
ITO indium-tin oxide
LDI laser desorption ionization
QCM quartz crystal microbalance
ROS reactive oxygen species
Acknowledgements
We thank Dr Frank Meyskens for his interest and
inspiration, Dr M. Bayachou and Michelle Tran for their
initial efforts on this project, and Dr Bruno Szpoganicz for
thoughtful discussions and interpretations. Dr Peter
Taborek is thanked for the QCM analysis. This work was
supported by grants from the Chao Family Cancer Center.
References
[1] G. Prota, in: G. Prota (Ed.), Melanins and Melanogenesis, Academic
Press, San Diego, 1992, pp. 153–184.
[2] R. Crippa, V. Horak, G. Prota, P. Svoronos, L. Wolfram, in: A. Brossi
(Ed.), Alkaloids, Academic Press, New York, 1989, pp. 253–323.
[3] P.A. Riley, Int. J. Biochem. Cell Biol. 29 (19) (1997) 1235–1239.
[4] G. Prota, M. d’Ischia, D. Mascagnena, Melanoma Res. 4 (1994)
351–358.
[5] N. Kollias, R.M. Sayre, L. Zeise, R. Chedekel, J. Phototchem.
Photobiol. B Biol. 9 (1991) 135–160.
[6] F.L. Meyskens, P.J. Farmer, J. Fruehauf, Pigment Cell Res. 14
(2001) 148–154.
[7] W. Koritowski, P. Hintz, R.C. Sealy, B. Kalayanaraman, Biochem.
Biophys. Res. Commun. 131 (2) (1985) 659–665.
[8] M. Rozanowska, T. Sarna, E.J. Land, T.G. Truscott, Free Radic.
Biol. Med. 26 (1999) 518–525.
[9] R.C. Sealy, C.C. Felix, J.S. Hyde, H.M. Swarz, W.A. Pryor, in: Free
Radicals in Biology, Vol. IV, Academic Press, New York, 1980, pp.
210–260.
[10] B. Larsson, H. Tjalve, Acta Physiol. Scand. 104 (1978) 479–484.
[11] T. Sarna, W. Froncisz, J.-S. Hyde, Arch. Biochem. Biophys. 202 (1)
(1980) 289–303.
[12] L. Novellino, A. Napolitano, G. Prota, Chem. Res. Toxicol. 12
(1999) 985–992.
[13] E.S. Jacobson, N.D. Jenkins, J.M. Todd, Infect. Immunol. 62 (1994)
4085–4086.
[14] D. Ben-Shachar, M.B.H. Youdim, Prog. Neuropsychopharmacol.
Biol. Psychiatry 17 (1993) 139–150.
60 S. Gidanian, P. J. Farmer / Journal of Inorganic Biochemistry 89 (2002) 54 – 60
[15] T. Sarna, H.M. Swarz, in: G. Scott (Ed.), Atmospheric Oxidation
and Antioxidants, Vol. III, Elsevier, Amsterdam, 1993, pp. 129–165.
[16] W. Kortawski, T. Sarna, J. Biol. Chem. 265 (21) (1990) 12410–
12416.
[17] T. Sarna, J.S. Hyde, H.M. Swartz, Science 192 (1976) 1132–1134.
[18] R.C. Sealy, J.S. Hyde, C.C. Felix, I.A. Menon, G. Prota, Science
217 (1982) 545–547.
[19] T. Sarna, H.M. Swartz, in: J.J. Nordlund, R.E. Boissy, V.J. Hearing,
R.A. King, J.P. Ortone (Eds.), The Pigmentary System: Physiology
and Pathophysiology, Oxford University Press, New York, 1998, pp.
333–358.
[20] G. Dryhurst, Chem. Rev. 90 (1990) 795–811.
[21] G.M. Robinson, E.I. Iwuoha, M.R. Smyth, Electrochim. Acta 34
(1998) 3489–3496.
[22] J. McGinness, P. Corry, P. Proctor, Science 183 (1974) 853–855.
[23] M.M. Jastrzebska, S. Jussila, H. Isotalo, J. Mater. Sci. 33 (1998)
4023–4028.
[24] V. Horak, G. Weeks, Bioorg. Chem. 21 (1993) 24–33.
[25] A. Napolitano, A. Pezzella, G. Prota, R. Seraglia, P. Traldi, Rapid
Commun. Mass Spectrom. 10 (1996) 468–472.
[26] J.N. Rodrigues-Lopez, J. Tudela, R. Varon, F. Garcia-Canovas,
Biochim. Biophys. Acta 1076 (1991) 379–386.
[27] J. Cabanes, F. Garcia-Canovas, J.A. Lazano, F. Garcia-Carmona,
Biochim. Biophys. Acta 923 (1987) 187–195.
[28] A. Bertazzo, C.V.L. Costa, G. Allegri, D. Favretto, P. Traldi, J. Mass
Spectrom. 34 (1999) 922–929.
[29] A. Pezzella, M. d’Ischia, A. Napolitano, A. Palumbo, G. Prota,
Tetrahedron 33 (24) (1997) 8281–8286.
[30] C. Sarzanini, E. Mentasti, O. Abollino, M. Fasano, Mar. Chem. 39
(1992) 243–250.
[31] B. Szpoganicz, S. Gidanian, P. Kong, P.J. Farmer, J. Inorg. Biochem.
89 (2002) 45–53.
[32] R.C. Kapoor, B.S. Aggarwal, in: Principles of Polarography, Wiley,
New York, 1991, pp. 54–65, Chapter 5.