J. Pineal Res.

2001; 31:102–108

Interaction of serotonin and melatonin with

sodium, potassium, calcium, lithium and
aluminium
Lack B, Daya S, Nyokong T. Interaction of serotonin and Barbara Lack1, Santy Daya2 and
melatonin with sodium, potassium, calcium, lithium and aluminium. Tebello Nyokong1
J. Pineal Res. 2001; 31:102 – 108. © Munksgaard, 2001 1
Department of Chemistry, Rhodes
University, Grahamstown, 6139, South
Abstract: In the present study, we investigated the ability of serotonin Africa; 2Department of Pharmaceutical
and melatonin to bind metals that occur naturally in the brain. An Sciences, Rhodes University, Grahamstown,
electrochemical technique called adsorptive cathodic stripping 6139, South Africa
voltammetry (AdCSV) was employed to study the metal–serotonin or
metal – melatonin interactions. The results show that both serotonin and
melatonin form stable complexes with lithium and potassium, with Key words: aluminium – calcium – lithium –
serotonin favouring lithium over potassium, and melatonin favouring melatonin – potassium – serotonin –
potassium over lithium. Coordination between either serotonin or sodium
melatonin and calcium was not favoured. The stability of the
Address reprint requests to Tebello
complexes formed between serotonin and the metals decreased with the
Nyokong, Department of Chemistry, Rhodes
metals as follows: Li + \ K + \ Al3 + \ Na + \ Ca2 + . The trend for
University, Grahamstown, 6139, South
melatonin – metal complexes was K + \ Li + \ Na + \ Al3 + \ Ca2 + . Africa, or Santy Daya, Department of
The binding and stable complex formation between both ligands, Pharmaceutical Sciences, Rhodes University,
serotonin and melatonin with lithium, potassium and sodium is of Grahamstown, 6139, South Africa.
biological importance. The binding of serotonin to lithium could E-mail: T.Nyokong@ru.ac.za or
provide an explanation for the therapeutic effects of lithium in S.Daya@ru.ac.za
depression treatment, whereas the binding of aluminium by melatonin
could provide insight into the role of this element in the aetiology of Received April 29, 2000;
Alzheimer’s disease. accepted July 26, 2000.

Introduction reported to be modulated by melatonin [Petit et

The electrochemistry of serotonin, melatonin and
tryptophan in the absence of metals has been
described by Zoulis et al. [1990] and Radi and
Bekhiet [1998] using stripping voltammetry meth-
ods. Recently, Limson et al. [1998] demonstrated
the interaction of melatonin, and its precursors,
with cadmium, copper, iron, lead, zinc and alu-
minium dissolved in acetonitrile using adsorptive
cathodic stripping voltammetry (AdCSV).
Melatonin, and its precursor serotonin, have
been shown to interact with a number of metal
ions either by altering their actions or by altering
their conductance through biological membranes.
Such interactions have been shown to have possi-
ble biological significance, particularly with refer-
ence to melatonin’s ability to protect against tissue
damage [Daya, 1999]. Melatonin is known to stim-
ulate voltage-dependent, sodium-selective currents
and also stimulate tetrodotoxin insensitive
voltage-dependent sodium current by a novel
mechanism [Rich et al., 1999]. Potassium ion
channels and guanylyl cyclases have also been
102
al., 1998]. Melatonin has been shown to play a
role in regulation of potassium, sodium and chlo-
ride in the common dentex, Dentex dentex
[Pavlidis et al., 1999]. The hormone also inhibits
the activity of large conductance calcium-activated
potassium channels [Geary et al., 1998], suggesting
that melatonin inhibits endothelial potassium
channels to decrease flow-induced release of nitric
oxide and blocks smooth muscle potassium chan-
nels to enhance vascular tone. Melatonin has also
been shown to interact with lithium [Lauber and
Vriend, 1989] and aluminium [Limson et al., 1998].
The actions of serotonin have been noted to mod-
ulate both calcium [Hirafuji et al., 1999] and
potassium currents [Herness and Chen, 2000].
In the present study, we describe the interaction
of melatonin and serotonin with sodium, potas-
sium, calcium, lithium and aluminium in aqueous
media and under biological pH conditions. The
study was conducted using a stripping voltamme-
try technique, the AdCSV method [Wang, 1989;
Zoulis et al., 1990; Wang et al., 1993; Zhang et al.,
Printed in Ireland — all rights reser7ed.

1993; Limson and Nyokong, 1997; Limson et al.,
1998].
Materials and methods
All reagents were of analytical grade, and distilled
deionised water was used for the experiments.
Solutions of Na + , K + , Ca2 + , Li + and Al3 + were
prepared from the corresponding metal chloride
salts. Serotonin (as a creatine sulphate complex;
Sigma-Aldrich, St Louis, MO, USA) and mela-
Interactions of serotonin and melatonin
ing solutions of the ligands with an excess of the
metal ions and freeze drying the solution. Nuclear
Magnetic Resonance (1HNMR) data in de-uter-
ated dimethylsulphoxide (DMSO [D6]) was
recorded with the Bruker EMX 400 NMR
spectrometer.
Results
The reduction peak for serotonin in the absence of
metal ions was observed in this study at −1.76 V
tonin (Sigma-Aldrich) solutions were prepared versus AgAgCl. For melatonin, the AdCSV peak
daily; serotonin was dissolved in distilled water, in the absence of metal ions was observed at
while melatonin was dissolved in absolute ethanol
−2.18 V versus AgAgCl. The shapes of the peaks
and subsequently diluted with distilled water. The
shown in Figs. 2 and 3 for metal– serotonin and
final ethanol concentration in the solution was
metal –melatonin complexes, respectively, are sim-
0.2% (v/v). A solution of 0.05 mol dm − 3 tetra-
butylamoniumbromide (TBABr; Sigma-Aldrich) ilar to those of the metal alone, thus suggesting
was used as an electrolyte for the electrochemical that reduction of the metal –serotonin or metal –
experiments (Fig. 1). melatonin complex occurs at the metal. Reduction
Stripping voltammograms were obtained with peaks, as a result of serotonin or melatonin alone,
the BioAnalytical Services (BAS, Lafayette, IN, are not evident in Figs. 2 and 3 at concentrations
USA) CV-50 W voltammetric analyser. A Con- of serotonin and melatonin employed.
trolled Growth Mercury Electrode (CGME; BAS The enhancement in the reduction currents,
Model MF-9058) was employed as the working upon addition of serotonin to the metal containing
electrode. A silver/silver chloride (3 mol dm − 3 solution, is an indication of the in situ formation
KCl) and a platinum wire were employed as the and accumulation of the metal –serotonin or
reference and auxiliary electrodes, respectively. So- metal– melatonin complex onto the Hg electrode.
lutions containing the metal ion and the ligand The shift in the reduction peaks to more negative
(serotonin or melatonin) were de-aerated for a potentials upon addition of serotonin or mela-
minimum period of 5 min, after which a flow of tonin to solutions containing the metal ion,
nitrogen gas was maintained over the solution suggests that the metal – serotonin or metal –
throughout the measurement. The metal –ligand melatonin complexes formed are more difficult to
complex is expected to have formed at this stage. reduce than the free metal. The extent of the shift
A deposition potential ( − 1,200 mV for Al3 + and in reduction peak potential of the metal on addi-
Li + , and −1,400 mV for Na + , Ca2 + , K + ) was tion of serotonin or melatonin is a good measure
applied for a deposition time of 120 s. The adsorp- of the stability of the metal –serotonin or metal –
tion of the metal –ligand complex onto the elec- melatonin complex formed in situ. Table 1 lists the
trode occurs at this stage. The deposition potential
shifts in peak potentials, DEp, for the reduction of
values given above were obtained from optimisa-
the metals on addition of serotonin or melatonin.
tion of the deposition potential for each metal–lig-
and complex at fixed deposition time and at
constant concentrations of the metals and ligands.
The deposition time was similarly optimised using
the optimal deposition potentials and constant
concentrations of the metal and the ligand. The
voltammograms were recorded in a negative direc-
tion from the accumulation potential to at least
0.2 V beyond the reduction peak of the metal
complex. A scan rate of 25 mV s − 1 was applied
during the stripping step. AdCSV experiments
were also performed for the metal ion in the
absence of the ligand or for the ligand in the
absence of the metal ion.
Solid complexes formed between the metal ion
and serotonin or melatonin were prepared by mix- Fig. 1. Molecular structure of serotonin and melatonin.

103
Lack et al.

Fig. 2. Adsorptive cathodic stripping voltammograms obtained for 6.7×10 − 8 mol dm − 3 serotonin in the presence of (A) Ca2 + ,
(B) K + , (C) Na + , (D) Li + and (E) Al3 + . (a) Voltammograms for metals alone and (b) voltammograms in the presence of
serotonin. Metal concentration=3.3×10 − 9 mol dm − 3. Scan rate=25 mV s − 1. Electrolyte = 0.05 mol dm − 3 TBABr.

104
 Interactions of serotonin and melatonin

Fig. 3. Adsorptive cathodic stripping voltammograms obtained for 6.7×10 − 8 mol dm − 3 melatonin in the presence of (A) Ca2 + ,
(B) K + , (C) Na + , (D) Li + and (E) Al3 + . (a) Voltammograms for metals alone and (b) voltammograms in the presence of
melatonin. Metal concentration=3.3×10 − 9 mol dm − 3. Scan rate=25 mV s − 1. Electrolyte =0.05 mol dm − 3 TBABr.

105
Lack et al.

Table 1. AdCSV parameters for the metal–serotonin or metal–melatonin complexes

Metal Ep (mV vs. AgAgCl)a Ep (mV) Slopeb (A/mol dm−3)

Serotonin −1760
Sodium −2246 −41 1.97 (0.995)
Potassium −2254 −77 1.04 (0.985)
Calcium −2266 +27 0.88 (0.991)
Lithium −2348 −123 0.59 (0.967)
Aluminium −2360 −71 0.81 (0.955)
Melatonin −2181
Sodium −2290 −85 1.39 (0.996)
Potassium −2318 −141 1.29 (0.991)
Calcium −2309 −16 1.02 (0.986)
Lithium −2323 −97 0.39 (0.982)
Aluminium −2333 −44 0.65 (0.987)

[serotonin]=[melatonin] =6.7×10−8 mol dm−3. Ep = peak potentials of the metal–ligand complexes. Ep is the difference
between the peak potentials of the metal ions and those of the metal–ligand complexes. a Metal concentration=3.3×
10−9 mol dm−3. b Regression coefficients shown in brackets.

For serotonin, the high negative shift observed The current increased with increases in concentra-
(DEp = −123 mV) on addition of serotonin to tion of melatonin and serotonin for concentrations
Li + solutions reflects the stability of the resulting ranging from 1 ×10 − 9 to 1 ×10 − 8 mol dm − 3.
lithium–serotonin complex when compared to the The 1HNMR spectra of melatonin and sero-
other metal –serotonin complexes. The high nega- tonin in the presence and absence of the metal ions
tive DEp value also shows that the resulting com- suggests that an interaction occurs between the
plex is more difficult to reduce when compared to ligands and the metal ions (Tables 2 and 3). For
the reduction of the free metal and the other both serotonin and melatonin, the data presented
metal–serotonin complexes. For melatonin, a high here do not give clear evidence of the actual coor-
negative Ep value was observed for the potassium – dination site of the metals.
serotonin complex showing that a more stable
complex is formed between potassium and mela-
tonin when compared to the other metal –mela-
tonin complexes.
Based on the magnitudes of the shifts in reduc-
tion potentials of the metals on addition of sero-
tonin, the relative stability of the in situ complexes
formed between the metals and serotonin de-
creases with the metal as follows: Li + \ K + \
Al3 + \Na + \Ca2 + . For the complexes between
melatonin and the metals, the stability of the com-
plexes increases with the metal as follows: K + \
Li + \Na + \Al3 + \ Ca2 + . Thus, both Li + and
K + form more stable complexes with serotonin
and melatonin when compared to the other metal
complexes under consideration. Both melatonin
and serotonin do not favour coordination to Ca2+.
Peak currents for the metal–serotonin or
metal– melatonin complexes increased with in-
creases in the concentration of the metal as shown
in Fig. 4 for serotonin and melatonin complexes,
respectively. The plots are linear with regression
coefficients listed in Table 1. The slopes of the
plots shown in Fig. 4 are listed in Table 1. Fig. 4. (A) Variation of currents with metal ion concentra-
tion for 6.7× 10 − 8 mol dm − 3 serotonin and (B) for 6.7×
AdCSV currents increased with increases in the 10 − 8 mol dm − 3 melatonin. Ca2 + ( ), K + (
), Na + (),
concentrations of serotonin and melatonin as Li + () and Al3 + ( ). Scan rate= 25 mV s − 1. Electrolyte =
shown in Fig. 5 at constant concentration of K + . 0.05 mol dm − 3 TBABr.

106

Fig. 5. Variation of currents with ligand concentration for
serotonin () and melatonin (
). [K + ]=3.3×10 − 9 mol solubility improved. Solvent, DMSO (D6).
dm − 3. Scan rate =25 mV s − 1. Electrolyte=water containing
TBABr.
Discussion
The binding and stable complex formation be-
tween both ligands, serotonin and melatonin, with
lithium, potassium and sodium is of biological
importance. It is known that melatonin inhibits
the activity of large conductance calcium-activated
potassium channels [Geary et al., 1998] and also
reduces the KCl-induced insulin response to glu-
cose [Peschke et al., 1997). The reason for this
could be as a result of direct chelation of potas-
sium by melatonin. Although this study shows
that calcium forms a very weak and unstable com-
plex with melatonin, in light of the findings that
melatonin inhibits neuronal calcium overload in
mouse brain cells [Zhang and Zhang, 1999], chela-
tion of potassium by melatonin could prevent
opening of the voltage gated calcium channels. In
this respect, melatonin is known to enhance acti-
vation and inactivation kinetics in lens epithelial
cells and trabecular meshwork cells and to also
stimulate the voltage-dependant selective currents
in these tissues [Rich et al., 1999]. Whether this is
as a result of chelation of metals is not known.
Table 2. 1HNMR shifts observed for the serotonin or melatonin
NH peaks following addition on the metal ions
d Shift (ppm)
Melatonin
Metal Ring Chain Serotonin
Lithium +0.241 +0.278 +0.169
Potassium +0.144 +0.020 +0.022
Sodium +0.084 +0.095 +0.029
Calcium +0.030 +0.040 +0.004
Solvent, DMSO (D6). Positive values represent downfield
shifts, negative values upfield shifts.
Interactions of serotonin and melatonin
1
Table 3. HNMR peaks observed for NH2 and OH resonances
for serotonin in the presence of the metal ions
d (ppm)*
Metal OH NH2
Lithium 8.804 8.430
Potassium 8.638 8.261
Sodium 8.651 8.244
Calcium 8.578 –
* The low solubility of serotonin in de-uterated DMSO did
not allow for clear observation of the OH and NH2 peaks;
however, in the presence of the above metal ions, the
The finding that serotonin forms a strong and
stable complex with Li + is also important in terms
of the possible mode of action in the psychophar-
macology of this metal. The mode of action of
Li + in the treatment of manic-depressive disorder
has not yet been characterised. One possible expla-
nation offered by the present study is that the
formation of a serotonin Li complex could intro-
duce an increase in the half-life of serotonin by
preventing its rapid destruction by enzymes such
as monoamine-oxidase. This would allow for a
greater pool of serotonin in the neurons. The
finding that melatonin also binds this metal with
tenacity, but to a lesser degree than serotonin,
could explain why this metal reduces plasma,
pineal and ocular melatonin levels [Lauber and
Vriend, 1989]. It is possible that the melatonin –Li
complex is not detected as melatonin per se, or
that the serotonin– Li complex prevents the con-
version of serotonin to melatonin.
The binding of Al3 + by both serotonin and
melatonin is of significance in Alzheimer’s disease
where this metal in known to accumulate in the
brain [Limson et al., 1998]. Whilst it is uncertain
whether Al3 + does indeed play a role in the aetiol-
ogy of Alzheimer’s disease, its introduction into
the brain can mimic many of the pathological
features of Alzheimer’s disease, except for the cor-
tical somatostatin deficit [Beal et al., 1989]. It thus
appears that chelation of metal ions by these in-
doleamines, as shown in the present study as well
as the stability of the complexes formed, indicate
that such complexes can have important biological
and pharmacological consequences.
In conclusion, all the metals investigated formed
complexes with both serotonin and melatonin.
However, the stability and affinity for the ligands
to the various metals varied greatly. Both Li + and
K + formed more stable complexes with serotonin
and melatonin when compared to the other metal
complexes under consideration. Melatonin and
serotonin do not favour coordination to Ca2 + .
107
Lack et al.
Acknowledgments
This work was supported by Rhodes University, the South
African Medical Research Council and the National Re-
search Foundation in South Africa.
Literature cited
BEAL, M.F., M.F. MAZUREK, D.W. ELLISON, N.W.
KOWALL, P.R. SOLOMON, W.W. PENDLEBURY (1989) Neu-
rochemical characteristics of aluminum-induced neurofibril-
lary degradation in rabbits. Neuroscience 29:339–346.
DAYA, S. (1999) The role of melatonin as a neuroprotectant in
Alzheimer’s disease. Spec. Med. 21:528–533.
GEARY, G.G., S.P. DUCKLES, D.N. KRAUSE (1998) Effect of
melatonin in the rat tail artery: Role of potassium channels
and endothelial factors. Br. J. Pharmacol. 123:1533–1540.
HERNESS, M.S., Y. CHEN (2000) Serotonergic agonists inhibit
calcium-activated potassium and voltage-dependent sodium
currents in rat taste receptor cells. J. Membr. Biol.
15(173):127–138.
HIRAFUJI, M., F. KAWAHARA, T. EBIHARA, A. NEZU, A.
TANIMURA, M. MINAMANI (1999) 5-Hydroxytryptamine in-
duces transient Ca2 + influx through Ni2 + -sensitive Ca2 +
channels in rat vascular smooth muscle cells. Eur. J. Pharma-
col. 380:163–170.
LAUBER, J.K., J. VRIEND (1989) Melatonin reduction by
lithium and albinism in quail and hamsters. Gen. Comp.
Endocrinol. 76:414–420.
LIMSON, J., T. NYOKONG, S. DAYA (1998) The interaction of
melatonin and its precursors with aluminium, cadmium,
copper, iron, lead and zinc. J. Pineal Res. 24:15– 31.
LIMSON, J., T. NYOKONG (1997) Substituted catechols as
complexing agents for the determination of bismuth, lead,
copper and cadmium by adsorptive stripping voltammetry.
Anal. Chim. Acta 344:87–95.
108
PAVLIDIS, M., L. GREENWOOD, M. PAALAVUO, H. MOLSA,
J.T. LAITINEN (1999) The effect of photoperiod on diel
rhythms in serum melatonin, cortisol, glucose and elec-
trolytes in the common dentex, Dentex dentex. Gen. Comp.
Endocrinol. 113:240– 250.
PESCHKE, E., D. PESCHKE, T. HAMMER, V. CSERNUS (1997)
Influence of melatonin and serotonin on glucose-stimulated
insulin release from perifused rat pancreatic islets in vitro. J.
Pineal Res. 23:156– 163.
PETIT, L., B. GUARDIOLA, P. DELAGRANGE, R. JOCKERS,
A.D. STROSBERG (1998) Signalling be melatonin receptors.
Therapie 53:421–428.
RADI, A., G.E. BEKHIET (1998) Voltammetry of melatonin at
carbon electrodes and determination in capsules. Bioelec-
trochem. Bioenerg. 45:275– 279.
RICH, A., G. FARRUGIA, J.L. RAE (1999) Effects of melatonin
on ionic currents in cultured ocular tissues. Am. J. Physiol.
276:C923– C929.
WANG, J. (1989) Voltammetry after non-electrolytic precon-
centration. In: Electroanalytical Chemistry, A.J. Bard, ed.
Wiley, New York, pp. 1–84.
WANG, J., J. LU, C. YARNITZKY (1993) Highly sensitive and
selective measurements of lead by stripping voltammetry/po-
tentiometry following adsorptive accumulation of the lead
o-cresolphthalexon complex. Anal. Chim. Acta 280:61–67.
ZHANG, Q.Z., J.T. ZHANG (1999) Inhibitory effects of mela-
tonin on free intracellular calcium in mouse brain cells.
Chung Kuo Li Hsueh Pao 20:206–210.
ZHANG, Z.Q., S.-Z. CHEN, H.-M. LIN, H. ZHANG (1993)
Simultaneous determination of copper, nickel, lead, cobalt
and cadmium by adsorptive stripping voltammetry. Anal.
Chim. Acta 272:227–232.
ZOULIS, N.E., D.P. NIKOLELIS, C.E. EFSTATHIOU (1990) Pre-
concentration of indolic compounds at a carbon paste elec-
trode and the indirect determination of L-tryptophan in
serum by adsorptive stripping voltammetry. Analyst
115:291– 295.