Neuroscience Letters 421 (2007) 33–36
Oxidative stress parameters in unmedicated and treated bipolar subjects
during initial manic episode: A possible role for lithium antioxidant effects
Rodrigo Machado-Vieiraa,∗, Ana Cristina Andreazzab, Carlos Ivan Vialea, Vanessa Zanattob,
Victor Cereser Jr.b, Rafael da Silva Vargasb, Fl´ avio Kapczinskib, Luiz V. Portelab,
Diogo O. Souzab, Mirian Salvadorc, Valentim Gentild
aMood Disorders Program, HMIPV, Fundacao Faculdade Federal Ciencias Medicas de Porto Alegre and Bipolar Disorder Research Program,
Espirita Hospital of Porto Alegre, Porto Alegre, Brazil
bDepartment of Psychiatry and Biochemistry, Federal University of Rio Grande do Sul, Brazil
cInstitute of Biotechnology and Department of Biomedical Sciences, University of Caxias do Sul, Brazil
dDepartment and Institute of Psychiatry, University of Sao Paulo Medical School, Sao Paulo, Brazil
Received 22 February 2007; received in revised form 7 May 2007; accepted 9 May 2007
Studies have proposed the involvement of oxidative stress and neuronal energy dysfunctions in the pathophysiology of bipolar disorder (BD).
This study evaluates plasma levels of the oxidative/energy metabolism markers, thiobarbituric acid reactive substances (TBARS), superoxide
dismutase (SOD), catalase (CAT), and neuron-specific enolase (NSE) during initial episodes of mania compared to controls in 75 subjects. Two
groups of manic subjects (unmedicated n=30, and lithium-treated n=15) were age/gender matched with healthy controls (n=30). TBARS and
antioxidant enzymes activity (SOD and CAT) were increased in unmedicated manic patients compared to controls. Conversely, plasma NSE levels
were lower during mania than in the controls. In contrast, acute treatment with lithium showed a significant reduction in both SOD/CAT ratio
and TBARS levels. These results suggest that initial manic episodes are associated with both increased oxidative stress parameters and activated
antioxidant defenses, which may be related to dysfunctions on energy metabolism and neuroplasticity pathways. Antioxidant effects using lithium
in mania were shown, and further studies are necessary to evaluate the potential role of these effects in the pathophysiology and therapeutics of
© 2007 Elsevier Ireland Ltd. All rights reserved.
Keywords: Bipolar disorder; Mania; Oxidative stress; Stress; Brain; Mitochondria; Energy metabolism; Psychiatry; Lithium; Neurobiology
Bipolar disorder (BD) is a chronic and severe debilitating men-
tal disorder which affects about 1.2% of the population .
The neurobiological basis of BD may involve dysfunctions on
neurotrophic pathways and energy metabolism [13,15,24,20].
Increased neuronal oxidative stress (OxS) levels generate dele-
terious effects on signal transduction, structural plasticity and
cellular resilience, mostly by inducing lipid peroxidation in
oxidative stress parameters have been reported in BD and
schizophrenia [22,43,28]. Thiobarbituric acid reactive sub-
stances (TBARS) levels is considered a direct index of cell
∗Corresponding author at: Quintino Bocaiuva 1495/1202, 90440-051, Porto
E-mail address: email@example.com (R. Machado-Vieira).
lipid peroxidation and the primary antioxidant system involves
coordinated effects induced by superoxide dismutase (SOD),
catalase (CAT) and glutathione peroxidase (GPx) [29,40]. Ele-
vated SOD/CAT ratio results in increased OxS, which is mostly
. Regarding pharmacological treatment, lithium has shown
to exert antioxidant and neuroprotective effects by increasing
tolerance to OxS [32,44]. Similarly , showed that lithium
treatment prevents excitotoxicity by inhibition of oxidative
The neuronal glycolytic enzyme neuron-specific eno-
lase (NSE) has been shown to present direct effects on
energy metabolism, neuroplastic pathways and cell survival
[14,39,3,12]. Peripherally determined NSE is mainly derived
0304-3940/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved.
R. Machado-Vieira et al. / Neuroscience Letters 421 (2007) 33–36
peripheral levels in diverse neurological disorders [14,37]. Oth-
and these findings are conflicting [6,10].
Despite recent data describing increased oxidative stress in
neurological and psychiatric disorders, we found no study mea-
suring oxidative stress parameters during manic episodes in
unmedicated or lithium-treated subjects. We now report differ-
ences in the peripheral markers TBARS, SOD, CAT and NSE
during manic episode compared to healthy controls.
Thirty unmedicated adult inpatients (7 men, 23 women;
mean age=26±4 years) and 15 lithium-treated (4 men, 11
women; mean age=26.2±6 years) meeting criteria for a manic
episode (Structured Clinical Interview for Axis I DSM-IV
Disorders-SCID-P)  were included. Subjects included in the
unmedicated group were evaluated in the admission unit of
Espirita Hospital of POA (HEPA). Regarding the past history,
patient sample included 36 never-treated (21 subjects in the
unmedicated group and 14 in the lithium-treated group) and 9
medication-free subjects presenting their second manic episode
(without psychopharmacological treatment for at least 5 weeks
prior to admission; unmedicated group=9, lithium-treated=1).
Young Mania Rating Scale (YMRS) score of ≥25 was con-
sidered inclusion criterion. YMRS scores did not significantly
chiatric comorbidities. Patients presenting rapid cycling course
or mixed episode or using antioxidant vitamins were excluded.
Lithium-treated patients were evaluated from 1 to 3 weeks after
admission (mean=14.5±5days) and before clinical improve-
ment. Except for two patients who were hospitalized for fewer
than 10 days, all lithium-treated patients presented therapeutic
(The control group included 30 healthy subjects (according to
SCID-I non-patient), age and gender-matched (±2 years) with
manic patients. Controls were selected among health profes-
study was approved by the HEPA Ethics Committee, and all
patients or family members provided written informed consent
before study entry.
Samples were collected from 4 to 6pm for unmedicated and
2–4h after lithium intake in the treated group. Blood samples
were obtained using vacutainer tubes and kept on ice. Samples
less than 6 months. All samples and standards were assessed in
rate at 480nm, which was measured using 1mmol/l adrenaline
(pH 2) and 50mmol/l glycine (pH 10.2). The reaction was con-
duced at 30◦C during 3min . CAT assay was carried out
according to . Total protein levels were evaluated using the
Total Proteins Kit from Labtest®. Considering that SOD and
CAT act sequentially in a pathway of ROS elimination, we also
expressed results as SOD/CAT ratios. TBARS levels were mea-
sured on the production of MDA, which, by combining with
thiobarbituric acid, forms a pink chromogen compound .
two monoclonal antibodies identifies the antigen determinants
reacted simultaneously with immobilized monoclonal antibod-
expressed as ?g/l and the coefficient of variation was less than
Statistical analysis was performed using SPSS for Windows
12.0. Results and figure are presented as mean±standard devi-
ation. Significance level was p<0.05. Analysis of variance
was employed for comparison between patients and controls
(one-way ANOVA with Duncan post-hoc test). Possible associ-
ation between clinical variables and biochemical markers was
determined using the Pearson correlation analysis. Possible cor-
relation between age or gender with the peripheral markers was
evaluated using Spearman test.
Increased oxidative lipid peroxidation (TBARS) levels were
observed in unmedicated mania (5.1±1.1) compared to con-
F=62.4) (Fig. 1). The SOD/CAT ratio was significantly higher
in unmedicated mania (ratio=0.83) than in the lithium-treated
group (ratio=0.22, p<0.001). SOD levels were higher in drug-
free mania (7.1±3.5U/g) compared to both lithium-treated
(2.65±1.3) and controls (2.61±1.4, F=27.7, p<0.001). CAT
was significantly increased in the lithium-treated and drug-
na¨ ıve groups (11.8±2.1 and 8.6±4.0?mol/mg, respectively)
in comparison to controls (3.4±1.8, F=42.6, p<0.001). NSE
levels were significantly lower in unmedicated (15.6±6?g/l)
group (22.6±9.4, F=10, p=0.002).
YMRS scores were similar in unmedicated (36.9±5) and
lithium-treated (37.3±4.9) patients (p=0.8). There was no
significant correlation between the biochemical markers and
patients’ YMRS scores, age or gender in the total sample.
These increases in oxidative stress and antioxidant defenses
during manic episodes may derive from a compensatory mech-
anism to a previously increased cellular oxidative stress.
Interestingly, acute treatment with lithium decreased the
mania compared to controls, with enhanced antioxidant defenses in lithium-
treated BD patients compared to unmedicated mania. Decreased NSE was
observed in mania, without reversal effect by lithium. Data are presented as
mean±S.D.; **P<0.001, *P<0.01.
R. Machado-Vieira et al. / Neuroscience Letters 421 (2007) 33–36
SOD/CAT ratio and TBARS levels compared to unmedicated
to hydrogen peroxide (H2O2) and oxygen, and CAT converts
H2O2 to water and molecular oxygen. GPx can also act on
could became more important in removing H2O2 than GPx,
mainly when there is a high concentrations of H2O2, since the
KM of catalase for H2O2 is much higher than the GPx one.
An imbalance in SOD/CAT ratio indicates the generation of
reactive species, which can be compensated or not . It is
clear that diverse acute neurological disorders such as stroke
rapidly increase peripheral oxidative stress parameters showing
the potential association with central and peripheral levels. In
the present context, lithium activity on SOD/CAT turnover (by
and in the absence of any clinical illness and other relevant risk
able to suggest that differences in oxidative stress parameters
may be associated, at least in part, with brain metabolism.
Accordingly, a decline in lipid peroxidation and a raise
in CAT levels were described in lithium-treated rats [38,33]
observed lower free radicals and elevated resistance to oxida-
tive stress (by inhibition of the proapoptotic protein GSK-3?)
after lithium treatment. Similarly, risperidone and olanzapine
of psychosis [41,45]. Interestingly , reported a significant
inhibition on the glutamate-induced increase of lipid peroxi-
dation and protein oxidation using lithium at therapeutically
the pathophysiology of BD [30,18]. Increased TBARS levels
were previously observed in BD and schizophrenia, especially
the other hand , described lower plasma levels of SOD in
BD and schizophrenia, suggesting the presence of dysfunctions
in reparative mechanisms on cell membrane . Interestingly,
in mania, SOD activity after 30 days under treatment with anti-
manic agents showed to be negatively correlated to the number
of previous manic episodes .
The discrepancy observed in the literature concerning oxida-
tive stress parameters in BD patients is possibly due to the
disease itself, but is affected, also, by medications, comorbidi-
ties, diet and lifestyle [1,18,28].
This is the first study showing decreased NSE levels in
unmedicated manic subjects and absence of lithium effect on
this parameter. NSE modulates neuronal glycolysis  and BD
has been associated with dysfunctions on energy metabolism
and cell plasticity in neurons and glia [27,26,12] observed the
generation of neurotrophic effects induced by NSE. Similar to
our findings, studies have described lower central NSE lev-
els in patients presenting chronic neurological disorders, such
as Alzheimer and multi-infarct dementia [4,36]. Conversely,
increased central or peripheral NSE levels have been frequently
observed in acute neurological disorders [14,37,35,19,10], in a
postmortem study, found no difference on brain NSE expres-
sion in BD, depression or schizophrenia compared to controls.
Overall, conflicting results regarding NSE in psychiatric disor-
ders may limit the correct interpretation of these findings, but
improved neuronal integrity might be suggested.
Changes in mitochondrial and energy metabolism were
described in BD [15,17]. Mitochondria controls energy and
free radicals production in neurons, through regulatory effects
on cell redox and oxidative metabolism [44,26] proposed the
presence of dysfunctions on neuronal glucose metabolism in
mood disorders. In this context , observed an increase in
gray matter lactate levels in BD, possibly underlying a deficit
in neuronal glycolysis in BD. Thus, it is reasonable to sug-
gest that the altered oxidative stress parameters and decreased
NSE levels observed during manic episodes may play a role in
the energy metabolism dysfunctions associated with BD. Also,
lithium seems to exert antioxidant effects during mania. Limi-
tations of this study include the lack of information regarding
the specificity of the effects for mania and therapeutic effects of
lithium, as well as the extent to which these peripheral markers
and unmedicated subjects were different people, which may
limit the interpretation in terms of direct lithium effects.
Further studies with larger samples may provide additional
data regarding the possible involvement of oxidative stress and
neuroplasticity in the pathophysiology of mania and in the ther-
apeutic effects of lithium.
Patients and Staff from Espirita Hospital of Porto Alegre,
Dr. Luciana Mesko, Luciane Wagner, CAPES, CNPq, Stanley
Medical Research Institute.
 D.S.P. Abdalla, H.P. Manteiro, J.A.C. Oliveira, C.H. Bechara, Activities
of superoxide dismutase and glutathione peroxidase in schizophrenic and
manic depressive patients, Clin. Chem. 32 (1986) 805–807.
 H. Aebi, Catalase in vitro, Method Enzymol. 105 (1984) 121–126.
 E.H. Cooper, Neuron specific enolase, Int. J. Biol. Markers 4 (1994)
 S. Dager, S. Friedman, A. Parow, C. Demapulos, A. Stoll, I. Lyoo, D.
with bipolar disorder, Arch. Gen. Psychiatry 61 (2004) 450–458.
 M.F. Egan, R.S. EL-Mallkh, R.L. Suddath, J.B. Lohr, H.S. Bracha, R.J.
Wyatt, Cerebrospinal fluid and serum levels of neuron-specific enolase in
patients with schizophrenia, Psychiatry Res. 43 (1992) 187–195.
 M.B. First, R.L. Spitzer, M. Gibbon, J.B.W. Williams, L. Benjamin, Struc-
Biometrics Research Department, New York State Psychiatric Institute,
 H.S. Gergerloiglu, H.A. Savas, F. Bulbul, S. Selek, E. Uz, M. Yumru,
Changes in nitric oxide level and superoxide dismutase activity during
antimanic treatment, Prog. Neuropsychopharmacol. Biol. Psychiatry 31
 W. Gsell, R. Conrad, M. Hicketier, E. Sofic, L. Frolich, I. Wichart, K.
Jellinger, G. Moll, G. Ransmayr, H. Beckmann, P. Riederer, Decreased
catalase activity but unchanged superoxide dismutase activity in brains
R. Machado-Vieira et al. / Neuroscience Letters 421 (2007) 33–36
of patients with dementia of Alzheimer type, J. Neurochem. 64 (1995)
 A. Guidotti, J. Auta, J.M. Davis, V. Giorgi, Y. Dwivedi, D.R. Grayson, F.
Impagntirllo, G. Pandey, C. Pesold, R. Sharma, D. Uzunov, E. Costa, V.
Arch. Gen. Psychiatry 57 (2000) 1061–1069.
 H. Halliwel, J.M.C. Gutteridge, Free Radicals in Biology and Medicine,
Oxford University Press, Oxford, UK, 1999, p. 936.
 T. Hattori, N. Takei, Y. Mizuno, K. Kato, S. Kohsaka, Neurotrophic and
embryonic rat brain, Neurosci. Res. 21 (1995) 191–198.
 R.S. Jope, Anti-bipolar therapy: mechanism of action of lithium, Mol.
Psychiatry 4 (1999) 117–128.
 E. Kaiser, R. Kuzmits, P. Pregant, O. Burghuber, W. Worofka, Clinical
biochemistry of neuron specific enolase, Clin. Chim. Acta 183 (1989)
 T. Kato, N. Kato, Mitochondrial dysfunction in bipolar disorder, Bipolar
Disord. 2 (2000) 180–190.
 C. Konradi, M. Eaton, M.L. Macdonald, J. Walsh, F.M. Benes, S. Heck-
ers, Molecular evidence for mitochondrial dysfunction in bipolar disorder,
Arch. Gen. Psychiatry 61 (2004) 300–308.
 M. Kuloglu, B. Ustundag, M. Atmaca, H. Canatan, E. Tezcan, N. Cin-
klinc, Lipid peroxidation and antioxidant enzyme levels in patients with
schizophrenia and bipolar disorder, Cell Biochem. Funct. 20 (2002)
 K.J.B. Lamers, P. Vos, M.M. Verbeek, F. Rosmalen, Vangeelwja, Vanenge-
len, Protein S-100B, neuron-specific enolase (NSE), myelin basic protein
(MBP) and glial fibrillary acidic protein (GFAP) in cerebrospinal fluid
(CSF) and blood of neurological patients, Brain Res. Bull. 61 (2003)
 R. Machado-Vieira, R. Leke, M.O. Dietrich, V. Cereser, V. Zanatto, F.
Kapczisnki, D.O. Souza, L.V. Portela, V. Gentil, Decreased plasma brain
derived neurotrophic factor in unmedicated bipolar patients during manic
episode, Biol. Psychiatry 61 (2007) 142–144.
 S. Mahadik, D. Evans, L. Harbans, Oxidative stress and role of antioxi-
dant and ?-3 essential fatty acid supplementation in schizophrenia, Prog.
Neuropsychopharmacol. Biol. Psychiatry 25 (2001) 463–493.
 S. Mahadik, S. Mukherjee, Free radical pathology and antioxidant defense
in schizophrenia: a review, Schizophr. Res. 19 (1996) 11–17.
 S. Mahadik, S. Mukherjee, R. Scheffer, E.E. Correnti, J.S. Mahadik, Ele-
vated plasma lipid peroxides at the onset of nonaffective psychosis, Biol.
Psychiatry 43 (1998) 674–679.
 H.K. Manji, G.J. Moore, G. Rajkowska, G. Chen, Neuroplasticity and
cellular resilience in mood disorders, Mol. Psychiatry 5 (2000) 578–593.
 H.P. Mirsa, I. Fridovich, The role of superoxide anion in the autoxidation
217 (1972) 3170–3175.
 A. Moretti, A. Gorini, R.F. Villa, Affective disorders, antidepressant drugs
and brain metabolism, Mol. Psychiatry 8 (2003) 773–785.
 P.K. Ranjekar, A. Hinge, M.V. Hegde, M. Ghate, A. Kale, S. Sitasawad,
and membrane essential polyunsaturated fatty acids in schizophrenic
and bipolar mood disorder patients, Psychiatry Res. 121 (2003) 109–
 R. Reddy, M.P. Sahebarao, S. Mukhergee, J.N. Murthy, Enzymes of the
antioxidant defense system in chronic schizophrenic patients, Biol. Psy-
chiatry 30 (1991) 409–412.
 R. Reddy, J. Yao, Free radical pathology in schizophrenia: a review,
Prostaglandins Leukot. Essent. Fatty Acids 55 (1996) 33–43.
 A.J. Rush, Toward an understanding of bipolar disorder and its origin, J.
Clin. Psychiatry 64 (2003) 4–8.
synthase kinase 3 beta is involved in the resistance to oxidative stress in
neuronal HT22 cells, Brain Res. 1005 (2004) 84–89.
 L Shao, L.T. Young, J.F. Wang, Chronic treatment with mood stabilizers
lithium and valproate prevents excitotoxicity by inhibiting oxidative stress
in rat cerebral cortical cells, Biol. Psychiatry 58 (2005) 879–884.
 H. Stevens, C. Jakobs, A.E.J. Jager, R.T. Cunningham, J. Korf, Neuron-
specific enolase and N-acetyl-aspartate as a potential peripheral markers of
ischaemic stroke, Eur. J. Clin. Invest. 29 (1999) 6–11.
specific enolase is decreased in multi-infarct dementia, but unchanged
in Alzheimer’s disease, J. Neurol. Neurosurg. Psychiatry 51 (1988)
 G. Sulter, J.W. Elting, J. Keyser, Increased serum neuron specific eno-
lase concentrations in patients with hyperglycemic cortical ichemic stroke,
Neurosci. Lett. 253 (1991) 71–73.
 A. Tandon, D.K. Dhawan, J.P. Nagpaul, Effect of lithium on hepatic lipid
peroxidation and antioxidative enzymes under different dietary protein
regimens, J. Appl. Toxicol. 18 (1998) 187–190.
57 (1991) 1178–1184.
 J.F. Wang, J.E. Azzam, L.T. Young, Valproate inhibits oxidative damage to
116 (2003) 485–489.
 Z. Wei, O. Bai, J.S. Richardson, D.D. Mousseay, X.M. Li, Olanzapine
protects PC12 cells from oxidative stress induced by hydrogen peroxide,
J. Neurosci. Res. 73 (2003) 364–368.
 E.D. Wills, Mechanisms of lipid peroxide formation in animal tissues,
Biochem. J. 99 (1966) 667–676.
 J.K. Yao, R.D. Reddy, D.P. Van Kammen, Oxidative damage and
CNS Drugs 15 (2001) 287–310.
 G. Yuan, G. Adhikary, A.A. Mccormick, J.J. Holcroft, G.K. Kumar, N.R.
Prabhakar, Role of oxidative stress in intermittent hypoxia-induced imme-
diate early gene activation in rat PC12 cells, J. Physiol. 557 (2004)
 X.Y. Zhang, D.F. Zhou, L.Y. Cao, P.Y. Zhang, G.Y. Wu, Y.C. Shen, The
effect of risperidone treatment on superoxide dismutase in schizophrenia,
J. Clin. Psychopharmacol. 23 (2003) 128–131.