lable at ScienceDirect

Journal of Psychiatric Research xxx (2015) 1e6

Contents lists avai

Journal of Psychiatric Research

journal homepage: www.elsevier .com/locate/psychires

Lithium increases platelet serine-9 phosphorylated GSK-3b levels indrug-free bipolar disorder during depressive episodes

Rafael T. de Sousa a, Marcus V. Zanetti a, b, c, Leda L. Talib a, Mauricio H. Serpa c,Tiffany M. Chaim c, Andre F. Carvalho e, Andre R. Brunoni a, Geraldo F. Busatto b, c,Wagner F. Gattaz a, b, Rodrigo Machado-Vieira a, b, d, *

a Laboratory of Neuroscience, LIM-27, Institute and Department of Psychiatry, University of Sao Paulo, Brazilb Center for Interdisciplinary Research on Applied Neurosciences (NAPNA), University of Sao Paulo, Brazilc Laboratory of Psychiatric Neuroimaging, LIM-21, Department and Institute of Psychiatry, University of Sao Paulo, Brazild Experimental Therapeutics and Pathophysiology Branch (ETPB), National Institute of Mental Health, NIH, Bethesda, MD, USAe Translational Psychiatry Research Group and Department of Clinical Medicine, Faculty of Medicine, Federal University of Cear�a, Brazil

a r t i c l e i n f o

Article history:Received 15 October 2014Received in revised form27 January 2015Accepted 28 January 2015

Keywords:Bipolar disorderGSK3LithiumTreatmentDepressionNeurobiology

Abbreviations: BD, Bipolar Disorder; GSK3b, GlHAM-D, 21-item Hamilton Depression Scale; PBMC, pcells; phospho-GSK3b, Serine-9 phosphorylated GSKInterview for Axis I DSM-IV-TR Disorders; YMRS, You* Corresponding author. Laboratory of Neuroscien

Institute of Psychiatry, University of Sao Paulo, BrazilE-mail address: machadovieirar@gmail.com (R. M

http://dx.doi.org/10.1016/j.jpsychires.2015.01.0160022-3956/Published by Elsevier Ltd.

Please cite this article in press as: de Sousadisorder during depressive episodes, Journa

a b s t r a c t

Background: Glycogen synthase kinase-3 b (GSK3b) is an intracellular enzyme directly implicated inseveral neural processes relevant to bipolar disorder (BD) pathophysiology. GSK3b is also an importanttarget for lithium and antidepressants. When phosphorylated at serine-9, GSK3b becomes inactive. Fewstudies evaluated serine-9 phosphorylated GSK3b (phospho-GSK3b) levels in BD subjects in vivo and nostudy has assessed it specifically in bipolar depression. Also, the effect of lithium monotherapy on GSK3bhas never been studied in humans.Methods: In 27 patients with bipolar depression, total GSK3b and phospho-GSK3b were assessed inplatelets by enzyme immunometric assay. Subjects were evaluated before and after 6 weeks of lithiumtreatment at therapeutic levels. Healthy subjects (n ¼ 22) were used as a control group.Results: No differences in phospho-GSK3b or total GSK3b were observed when comparing drug-free BDsubjects in depression and healthy controls. Baseline HAM-D scores were not correlated with phospho-GSK3b and total GSK3b levels. From baseline to endpoint, lithium treatment inactivated GSK3b bysignificantly increasing phospho-GSK3b levels (p ¼ 0.010). Clinical improvement (baseline HAM-D d

endpoint HAM-D) negatively correlated with the increase in phospho-GSK3b (p ¼ 0.03).Conclusion: The present results show that lithium inactivates platelet GSK3b in BD during mood epi-sodes. No direct association with pathophysiology of BD was observed. Further studies are needed toclarify the role of GSK3b as a key biomarker in BD and its association with treatment response as well asthe relevance of GSK3b in other neuropsychiatric disorders and as a new therapeutic target per se.

Published by Elsevier Ltd.

1. Introduction

BD is associated with impaired cellular resilience and synapticdysfunction at multiple pathways, which are involved in cell atro-phy and impaired neuroplasticity (Machado-Vieira et al., 2014).

ycogen synthase kinase-3b;eripheral blood mononuclear3b; SCID, Structured Clinicalng Mania Rating Scale.ce (LIM27), Department and.achado-Vieira).

RT, et al., Lithium increasesl of Psychiatric Research (201

Many metabolic and signaling pathways that affect the synthesis ofproteins as well as cell proliferation, adhesion, plasticity andapoptosis are regulated by the enzyme glycogen synthase kinase-3b (GSK3b) (Grimes and Jope, 2001). The activity of GSK3b ismodulated mainly through the phosphorylation of the n-terminalserine-9 of GSK3b, which becomes inactivated in the form ofphosphorylated-GSK3b (phospho-GSK3b) (Chalecka-Franaszek andChuang, 1999). Interestingly, in its active form GSK3b is responsiblefor the activation of several factors related to activation of cellapoptosis and degradation of beta-catenin (Eldar-Finkelman,2002).

Due to the crucial roles of GSK3b in signaling processes involvedin cellular apoptosis and death in the brain, activation of GSK3b has

platelet serine-9 phosphorylated GSK-3b levels in drug-free bipolar5), http://dx.doi.org/10.1016/j.jpsychires.2015.01.016

Table 1Demographic and clinical characteristics of bipolar depression patients and healthycontrols.

Bipolar (n ¼ 27) Controls (n ¼ 22) p

GenderMale/female, n (%) 7 (25.9)/20 (74.1) 11(50.0)/11 (50.0) 0.08a

Age, years 28.4 (±5.7) 26.5 (±6.8) 0.28b

Bipolar disorder typeType I/type II, n (%) 10 (37.0)/17 (63.0)

Duration of illness, months 37.9 (±18.4)Drug-naïve, n (%) 19 (70.4)Medication-free, n (%) 24 (88.8)History of psychosis, n (%) 4 (14.8)HAM-DBaseline/endpoint 22.2 (±3.0)/7.3 (±6.1)

YMRSBaseline/endpoint 5.6 (±4.9)/4.0 (±8.9)

Response, n (%) 23 (85.2)Remission, n (%) 17 (63.0)Endpoint serum lithium,

mEq/L0.49 (±0.21)

HAM-D e 21-item Hamilton Depression Scale, YMRS e Young Mania Rating Scale.a e Chi-square test.b e Student's t test.

R.T. de Sousa et al. / Journal of Psychiatric Research xxx (2015) 1e62

been proposed to play a role in the pathophysiology of neuropsy-chiatric disorders. GSK3b has a key role in regulating relevantbiological processes in bipolar disorder (BD), such as oxidativestress (Beurel and Jope, 2006; de Sousa et al., 2014), neurogenesis(Berk et al., 2011; Eom and Jope, 2009), and inflammation (Barbosaet al., 2014; Martin et al., 2005).

In BD pathophysiology, evidence supporting GSK3b involve-ment comes from numerous preclinical studies (Jope, 2011) andmolecular genetic studies in humans. In subjects with BD, anincreased copy number variation in GSK3b gene locus has beenfound (Lachman et al., 2007). Also, a GSK3b gene polymorphism hasbeen associated with age of onset (Benedetti et al., 2004a) andantidepressant response to sleep deprivation in bipolar depression(Benedetti et al., 2004b), reinforcing a role for GSK3b in BD.

Studies evaluating GSK3b protein in vivo in BD subjects havebeen controversial; total GSK3b expression in platelets wasdecreased in BD drug-free patients compared to controls (Pandeyet al., 2010), whereas total GSK3b in peripheral blood mono-nuclear cells (PBMC)was shown to be increased in bipolar mania (Liet al., 2010) and unaltered in post mortem brains of BD patients(Beasley et al., 2002). Furthermore, GSK3b activity (as estimated byphospho-GSK3b) showed similar results in PBMC in bipolar maniacompared to healthy controls (Li et al., 2010; Polter et al., 2010) aswell as in PBMC of BD patients not taking lithium (Li et al., 2007). Tothe best of our knowledge, no study has evaluated total or phospho-GSK3b specifically in bipolar depression.

Lithium d the gold standard treatment for BD d has beenassociated with selective inhibition of GSK3b in several preclin-ical studies (Jope, 2011; Machado-Vieira et al., 2009). The inac-tivation of GSK3b is performed by lithium via a directcompetition with magnesium (Ryves and Harwood, 2001) andindirectly through the phosphorylation of its n-terminal Ser-9(Chalecka-Franaszek and Chuang, 1999). Lithium-treated pa-tients with BD showed an increase on phospho-GSK3b in PBMCwhen compared to healthy controls, but in a small sample (n ¼ 9)under multiple treatments (Li et al., 2007). No study to date,however, evaluated the effects of lithium monotherapy on GSK3bprotein levels and activity, and the association of GSK3b withclinical response in bipolar depression.

The present study evaluates phosphorylated and total GSK3blevels in platelets of BD subjects during a depressive episode beforeand after 6 weeks of lithium monotherapy, compared to healthycontrols. Due to the apoptosis-inducing properties of the activatedform of GSK3b, we hypothesized that BD patients would presentlower inactivated phospho-GSK3b but unchanged total GSK3blevels compared to healthy controls, confirming previous findingsof total GSK3b in post mortem brains of BD patients. The knownneuroprotective effects of lithium and the previous preclinicalstudies lead us to hypothesize that lithium treatment in bipolardepression would significantly increase phospho-GSK3b, thusinactivating the enzyme.

2. Methods

2.1. Subjects

From August 2010 to June 2012, 27 outpatients with BD in adepressive episode were studied. Most patients (n ¼ 24, 88.8%)were drug-free for at least 6 weeks before their enrollment in thestudy, with predominance of drug-naïvety (n ¼ 19, 70.4%); thethree remaining patients were using one mood stabilizer (valproicacid or oxcarbazepine). Major characteristics of the sample aredisplayed in Table 1. The diagnosis of BDwas confirmed through theStructured Clinical Interview for Axis I DSM-IV-TR Disorders (SCID)(First et al., 1995). Patients were recruited using media

Please cite this article in press as: de Sousa RT, et al., Lithium increasesdisorder during depressive episodes, Journal of Psychiatric Research (201

advertisement and evaluated and followed-up at the Institute ofPsychiatry, University of Sao Paulo, Brazil.

Patients were required to have a score �18 on the 21-itemHamilton Depression Scale (HAM-D) (Hamilton, 1960) at the firstevaluation to be enrolled. The assessment of symptoms was per-formed by experienced psychiatrists, using the YoungMania RatingScale (YMRS) and the Clinical Global Impression e Severity(Petkova et al., 2000). Exclusion criteria were any medical disorderthat could affect the central nervous system, substance abuse in thelast year or substance dependence, and intellectual disability.

The control group included 22 age-matched (±3 years) healthysubjects (Table 1). Exclusion criteria for controls were a lifetimehistory of any psychiatric disorder (as evaluated with the SCID),including substance use disorders; presence of any disease thatcould affect the central nervous system; or the presence of mood orpsychotic disorders in first-degree relatives. This study was regis-tered at ClinicalTrials.gov (NCT01919892), was approved by theEthics Committee for Analysis of Research Projects (CAPPesq),Clinics Hospital, University of Sao Paulo, and was in accordancewith the Declaration of Helsinki (1964). All participants providedwritten informed consent before entering the study.

2.2. Study design

At baseline, patients received lithium carbonate at 450 mg/day,with subsequent flexible adjustment according to the clinicalimprovement. Plasma lithium levels were controlled to monitorcompliance and avoid toxicity (<1.2 mEq/L); the use of hypnotics(benzodiazepine or zolpidem) as needed was allowed. Psycho-metric assessments were performed at baseline and endpoint(week 6). Response was defined as a decrease of 50% or more in theHAM-D at endpoint, whereas remission criteria were endpointHAM-D<8 and YMRS<8.

2.3. Blood sampling and laboratorial analysis

After 8-h fasting, blood samples of BD patients were collected in0.1 M sodium citrate-coated tubes from 8 to 10 AM at baseline andafter 6 weeks of lithium treatment, whereas healthy controls hadonly one-point blood samples collected in the same conditions. Forthe isolation of platelets, samples were homogenized in 1ml of acidcitrate dextrose solution (ACDNIH-formel-A solution) and

platelet serine-9 phosphorylated GSK-3b levels in drug-free bipolar5), http://dx.doi.org/10.1016/j.jpsychires.2015.01.016

R.T. de Sousa et al. / Journal of Psychiatric Research xxx (2015) 1e6 3

centrifuged for 15 min at 1600 rpm at 20 �C to isolate platelets fromplasma. After pH adjustment to 6.5, plasma rich in platelets wascentrifuged for 10 min at 2400 rpm at 20 �C. Pellets were resus-pended in a wash-solution (sodium citrate 30 mM pH 6.5, potas-sium chloride 5 mM, albumin 500 mg/mL, apyrase 50 mg/mL,calcium chloride 2 mM, magnesium chloride 1 mM, glucose 5 mM)and centrifuged for 8 min at 2400 rpm and 20 �C. Pellets werestored at �70 �C in a solution of Tris sucrose (Tris 50 mM pH 7.4,sucrose 233 mM). Protein levels were determined through amodified Lowry method (Bio-Rad DC Protein Assay); the proteinconcentration was then normalized to 0.1 mg/mL with buffer plusprotease inhibitor provided by manufacturer (Assay Designs, Inc,MI, USA). Platelets were lysed in RIPA cell buffer modified by theaddition of the protease inhibitors PIC (Sigma P8340) and PMSF(Sigma P7626).

The levels of GSK3b protein levels were determined by a specificEnzyme Immunometric Assay (EIA), which assessed total andSerine-9 phosphorylated GSK3b (phospho-GSK3b) (Titer-Zyme EIA& Assay Designs Inc., MI, USA) (Ladeira et al., 2013). The protocolfollowed the instructions of themanufacturer.With serial dilutions,standard curves were obtained: 2000; 1000; 500; 250; 125; 62.5;0.0 pg/mL for phospho-GSK3b, and 5000; 2500; 1250; 625; 312.5;156.3; 78.1; 0.0 pg/mL for total GSK3b. In distinct wells of 96-wellEIA plates, homogenates of platelets corresponding to the sam-ples collected were dispensed; the reactivity to imbedded anti-bodies generated color arrays that were read at 450 nm (Zenyth200 Spectrophotometer, Austria), which provided GSK3b concen-tration in distinct aliquots. The assays were performed by aresearcher who was blind to the source of the samples (patients orcontrols).

2.4. Statistical analyses

The gender distribution of patients and controls wascompared using the Chi-square test. Comparisons betweengroups were performed using student's t test for variables withnormal distribution and with ManneWhitney and WilcoxonSigned Ranks tests when variables had non-normal distribution.Correlations were evaluated with Pearson and Spearman tests,according to variable distribution. All statistical analyses wereperformed using SPSS 21.0 with a significance level set at <0.05(two-tailed).

3. Results

3.1. Clinical data

Depressive and manic symptoms (measured by HAM-D andYMRS scores, respectively) decreased from baseline to endpoint(z ¼ �4.52, p < 0.001, and z ¼ �3.17, p ¼ 0.002, respectively);response was observed in 23 patients and 17 patients achievedremission (Table 1).

3.2. Similar platelet phosphorylated and total GSK3b levels inbipolar depression versus healthy controls

When comparing BD patients with controls at baseline, therewere no significant differences in regards to phospho-GSK3b(patients ¼ 488.5 ± 195.7 pg/mL vs. controls ¼ 554.9 ± 272.0)(t ¼ �0.99, df ¼ 47, p ¼ 0.33) and total GSK3b levels(patients ¼ 3477.4 ± 646.4 pg/mL vs. controls ¼ 3072.8 ± 809.0)(t ¼ 1.95, df ¼ 47, p ¼ 0.06). At baseline, HAM-D scores did notcorrelate with the phospho-GSK3b (p ¼ 0.87) or the total GSK3blevels (p ¼ 0.37).

Please cite this article in press as: de Sousa RT, et al., Lithium increasesdisorder during depressive episodes, Journal of Psychiatric Research (201

3.3. Lithium increased phospho-GSK3b levels but not total GSK3blevels

Lithium treatment increased (inactivated) phospho-GSK3blevels from baseline (488.5 ± 195.7 pg/mL) to endpoint(630.0 ± 263.0) (t ¼ �2.73, df ¼ 26, p ¼ 0.010) (Fig. 1A). No sig-nificant changes were observed in the total GSK3b levels frombaseline (3477.4 ± 646.4 pg/mL) to endpoint (3536.8 ± 845.0)(t ¼ �0.387, df ¼ 26, p ¼ 0.70) (Fig. 1B).

3.4. Clinical improvement was negatively correlated with increasein phospho-GSK3b levels

The overall clinical improvement in HAM-D (baseline HAM-D-endpoint HAM-D) negatively correlated with the increase in(inactivated) phospho-GSK3b levels over time (r ¼ �0.41, p ¼ 0.03)(Fig. 1C) but not with changes in the total GSK3b levels (r ¼ 0.07,p ¼ 0.74) (Fig. 1D). Endpoint HAM-D scores were not correlatedwith endpoint phospho-GSK3b (r ¼ 0.34, p ¼ 0.09) or total GSK3blevels (r ¼ 0.02, p ¼ 0.93).

The comparison of remitters and non-remitters showed nosignificant differences in phospho-GSK3b levels(remitters¼ 568.0 ± 277.5 pg/mL vs. non-remitters¼ 735.4 ± 208.1,p¼ 0.11) or total GSK3b levels (remitters¼ 3445.1 ± 717.9 pg/mL vs.non-remitters¼ 3692.8± 1050.5, p¼ 0.47). Plasma lithium levels atendpoint were not correlated with endpoint phospho-GSK3b levels(r ¼ 0.05, p ¼ 0.82) or total GSK3b levels (r ¼ �0.14, p ¼ 0.48).

4. Discussion

This is the first study evaluating the effects of lithium onplatelets GSK3b levels in BD subjects taking lithium monotherapy.Patients with BD during a depressive episode showed no differencein GSK3b (total and phosphorylated) compared to matched healthycontrols. Lithium selectively increased phospho-GSK3b levels, thusinhibiting GSK3b activity in BD patients. An inverse correlationbetween clinical improvement and increase in phospho-GSK3blevels was observed.

The present study found no difference between BD patients andcontrols regarding phospho-GSK3b expression at baseline, which isin line with studies comparing healthy controls to BD mania (Liet al., 2010) or to patients with BD depression and mania (Polteret al., 2010). However, Ladeira et al. (2013) reported increasedGSK3b phosphorylation in elderly BD patients in comparison tohealthy subjects.

Total GSK3b in post mortem brains of BD patients was notaltered in comparison to controls (Beasley et al., 2002; Kozlovskyet al., 2000), whereas studies using only peripheral cells haveshown mixed results. In comparison to healthy controls, in BDpatients across diverse phases of the disease total GSK3b was re-ported to be decreased in platelet (Pandey et al., 2010), andincreased in PBMC (Li et al., 2010). Such differences might beexplained by the heterogeneity of the samples investigated, in re-gard to different treatments, phases, and duration of illness. Pa-tients with shorter duration of illness might present less biologicalalterations than patients chronically exposed to neuronal stress dueto recurrent episodes and effects of medications (Berk et al., 2013;de Sousa et al., 2014).

Lithium is thought to inhibit GSK3b activity directly and indi-rectly. Lithium competes with magnesium to bind to the sameGSK3b site to directly inactivate GSK3b (Ryves et al., 2001). Lithiumindirectly inactivates GSK3b by phosphorylating Ser-9 site, whichcomprises two different mechanisms: lithium activates Akt, leadingto GSK3b phosphorylation (Chalecka-Franaszek and Chuang, 1999),and lithium directly deactivates GSK3b, an action that is followed

platelet serine-9 phosphorylated GSK-3b levels in drug-free bipolar5), http://dx.doi.org/10.1016/j.jpsychires.2015.01.016

Fig. 1. Platelet serine-9 phosphorylated glycogen synthase kinase-3b (phospho-GSK3b) levels (A) and total GSK3b levels (B) in 27 bipolar depression subjects at baseline (blackbars), after 6-week lithium treatment (grey bars) and in 22 controls (white bars). When comparing patients before and after lithium treatment, there was an increase in phospho-GSK3b (p ¼ 0.010) (A), highlighted by significance bars. Vertical lines above the black, grey, and white bars represent standard error means. Changes (baselinedendpoint) indepressive symptoms (measured as HAM-D scores) and platelet GSK3b in bipolar depression treated for 6 weeks with lithium are presented in (C) and (D). (C) Decreases in HAM-Dscores were negatively correlated to increases of phospho-GSK3b levels (r ¼ �0.41, p ¼ 0.03), (D) but not with total GSK3b levels (p ¼ ns). GSK3b levels are presented as pg/mL.(HAM-D ¼ 21-item Hamilton Depression Rating Scale).

R.T. de Sousa et al. / Journal of Psychiatric Research xxx (2015) 1e64

by the inhibition of protein phosphatase-1, thus phosphorylatingGSK3b (Zhang et al., 2003).

The present study reinforces the role of lithium treatment in theinhibition of GSK3b activity through phosphorylation. A huge bodyof preclinical evidence suggests that lithium has effects on GSK3bactivity (Jope, 2011). Lithium has been shown to inhibit GSK3bactivity in cell models (Aubry et al., 2009; Chalecka-Franaszek andChuang, 1999) and to decrease GSK3b activity and gene expressionin neural tissue of rodents (Beaulieu et al., 2004; De Sarno et al.,2002; Mendes et al., 2009). The present results demonstrate, forthe first time, that lithium monotherapy inhibits the activity ofGSK3b in humans.

Also, consistent with the present findings, elderly patients withBD who were under lithium treatment (plus other drugs) showedincreased platelets phospho-GSK3b in platelets compared tohealthy controls (Ladeira et al., 2013). Moreover, BD patients takinglithium with other agents showed an increase on phospho-GSK3brelative to healthy subjects in PBMCs (Li et al., 2007).

Although an increase in GSK3b phosphorylation after lithiumtreatment was here observed, the increase in GSK3b phosphory-lationwas inversely correlated to clinical improvement. The inversecorrelation of GSK3b phosphorylation and clinical improvementfound could be a chance finding, which is in line with the lack ofcorrelation between baseline GSK3b phosphorylation and depres-sive symptoms in our study and consistent with the report of Polteret al. (2010), who also found a lack of correlation between

Please cite this article in press as: de Sousa RT, et al., Lithium increasesdisorder during depressive episodes, Journal of Psychiatric Research (201

depressive symptoms and phospho-GSK3b levels in BD. It is stillunclear if GSK3b modulation by psychotropic drugs is associatedwith clinical improvement (Li and Jope, 2010).

Evidence for a role of GSK3b in mood regulation involve GSK3bmodulating serotonin, dopamine, and glutamatergic systems.GSK3b modulates serotonin 1-B receptors, which was observed inmouse brains (Chen et al., 2011); the antagonism of dopamine D2receptor induces GSK3b phosphorylation as reported in rat brain(Alimohamad et al., 2005) and in human SH-SY5Y cells (Sutton andRushlow, 2012). Also, GSK3b is associated with regulation ofglutamate receptors a-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPAR) and N-methyl-D-aspartate (NMDAR). GSK3bactivation has shown to be required for AMPAR internalization incultured hippocampal neurons (Du et al., 2010) and for NMDAR-dependent long term depression in rat hippocampus (Peineauet al., 2007).

Other possible explanations for the inverse correlation betweenincrease in phospho-GSK3b and clinical improvement might bethat lithium has a bimodal action on GSK3b phosphorylation, ac-cording to clinical improvement or non-improvement, or even thatthe change on GSK3b phosphorylation corresponds to anepiphenomenon.

Subjects with BD are at higher risk for Alzheimer's Disease(Kessing and Andersen, 2004). One of the neuropathological hall-marks in this disease is the hyper-phosphorylation of tau-proteinby GSK3b (Hooper et al., 2008). In this regard, our present finding

platelet serine-9 phosphorylated GSK-3b levels in drug-free bipolar5), http://dx.doi.org/10.1016/j.jpsychires.2015.01.016

R.T. de Sousa et al. / Journal of Psychiatric Research xxx (2015) 1e6 5

of an inactivation of GSK3b by lithium may provide a rationale forthe reports that lithium treatment reduced the risk for Alzheimer'sDisease in elderly bipolar patients (Kessing et al., 2008; Nunes et al.,2007) and enhanced cognition in patients with high risk for Alz-heimer's Disease (Forlenza et al., 2011).

The strengths of the present study include the homogeneity ofthe sample, which included mostly unmedicated BD patients dur-ing pure depressive episodes, the short duration of illness, and theuse of lithium monotherapy for most patients, thereby reinforcingthe specificity of the findings. A possible limitation of the study isthe sensitivity of the assay used, which might be not enough todetect total GSK3b differences between patients and controls.Although platelet GSK3b might not reflect brain GSK3b, whichcould be a limitation of the study, platelets share several biologicalproperties with neurons and have been widely used in the searchfor biomarkers in neuropsychiatric disorders (Talib et al., 2013;Zainaghi et al., 2007); the regulation of GSK3b platelet activity issubject to similar mechanisms that regulate GSK3b activity in braind phosphoinositide-3-kinase and protein kinase B activation(Barry et al., 2003). Evidence of platelet co-activities in neurons isavailable especially for glutamatergic and serotonergic systems(Plein and Berk, 2001; Stahl, 1977).

The present results support a key role for lithium targeting atGSK3b activity in humans by phosphorylating and thus inactivatingGSK3b in BD. Further studies are necessary to clarify the role ofGSK3b as a biomarker in BD and its association with treatmentresponse as well as the role of GSK3b in other neuropsychiatricdisorders and as a novel therapeutic target.

Role of the funding source

This study was sponsored by Sao Paulo Research Foundation(FAPESP, Brazil; Grant 2009/14891-9, RM-V), but FAPESP had norole in the design of the study, collection, analysis of data, anddecision to publish.

Contributors

Dr. Machado-Vieira designed the study. Dr. Machado-Vieira, Dr.Zanetti, Dr. de Sousa, Dr. Serpa and Dr. Chaim collected clinical data.Dr. Talib performed the laboratorial essays. Dr. Carvalho and Dr.Brunoni contributed for the discussion. Dr. Gattaz and Dr. Busattoprovided invaluable scientific input. All authors contributed to andhave approved the final manuscript.

Conflict of interest

AFC is supported by a research scholarship from ConselhoNacional de Desenvolvimento Científico e Tecnol�ogico (CNPq, LevelII, Brazil) and MVZ receives a research scholarship from Sao PauloResearch Foundation (FAPESP, Brazil). Other authors declare nopotential conflict of interest.

Acknowledgments

Associaç~ao Beneficente Alzira Denise Hertzog da Silva (ABADHS)and JNK Empreendimentos e Incorporaç~oes support Laboratory ofNeurosciences.

References

Alimohamad H, Rajakumar N, Seah YH, RushlowW. Antipsychotics alter the proteinexpression levels of beta-catenin and GSK-3 in the rat medial prefrontal cortexand striatum. Biol Psychiatry 2005;57:533e42.

Please cite this article in press as: de Sousa RT, et al., Lithium increasesdisorder during depressive episodes, Journal of Psychiatric Research (201

Aubry JM, Schwald M, Ballmann E, Karege F. Early effects of mood stabilizers on theAkt/GSK-3beta signaling pathway and on cell survival and proliferation. Psy-chopharmacol Berl 2009;205:419e29.

Barbosa IG, Machado-Vieira R, Soares JC, Teixeira AL. The immunology of bipolardisorder. Neuroimmunomodulation 2014;21:117e22.

Barry FA, Graham GJ, Fry MJ, Gibbins JM. Regulation of glycogen synthase kinase 3in human platelets: a possible role in platelet function? FEBS Lett 2003;553:173e8.

Beasley C, Cotter D, Everall I. An investigation of the Wnt-signalling pathway in theprefrontal cortex in schizophrenia, bipolar disorder and major depressive dis-order. Schizophr Res 2002;58:63e7.

Beaulieu JM, Sotnikova TD, Yao WD, Kockeritz L, Woodgett JR, Gainetdinov RR, et al.Lithium antagonizes dopamine-dependent behaviors mediated by an AKT/glycogen synthase kinase 3 signaling cascade. Proc Natl Acad Sci U S A2004;101:5099e104.

Benedetti F, Bernasconi A, Lorenzi C, Pontiggia A, Serretti A, Colombo C, et al.A single nucleotide polymorphism in glycogen synthase kinase 3-beta promotergene influences onset of illness in patients affected by bipolar disorder. Neu-rosci Lett 2004a;355:37e40.

Benedetti F, Serretti A, Colombo C, Lorenzi C, Tubazio V, Smeraldi E. A glycogensynthase kinase 3-beta promoter gene single nucleotide polymorphism isassociated with age at onset and response to total sleep deprivation in bipolardepression. Neurosci Lett 2004b;368:123e6.

Berk M, Berk L, Dodd S, Cotton S, Macneil C, Daglas R, et al. Stage managing bipolardisorder. Bipolar Disord 2013;16:471e7.

Berk M, Kapczinski F, Andreazza AC, Dean OM, Giorlando F, Maes M, et al. Pathwaysunderlying neuroprogression in bipolar disorder: focus on inflammation,oxidative stress and neurotrophic factors. Neurosci Biobehav Rev 2011;35:804e17.

Beurel E, Jope RS. The paradoxical pro- and anti-apoptotic actions of GSK3 in theintrinsic and extrinsic apoptosis signaling pathways. Prog Neurobiol 2006;79:173e89.

Chalecka-Franaszek E, Chuang DM. Lithium activates the serine/threonine kinaseAkt-1 and suppresses glutamate-induced inhibition of Akt-1 activity in neu-rons. Proc Natl Acad Sci U S A 1999;96:8745e50.

Chen L, Zhou W, Chen PC, Gaisina I, Yang S, Li X. Glycogen synthase kinase-3b is afunctional modulator of serotonin-1B receptors. Mol Pharmacol 2011;79:974e86.

De Sarno P, Li X, Jope RS. Regulation of Akt and glycogen synthase kinase-3 betaphosphorylation by sodium valproate and lithium. Neuropharmacology2002;43:1158e64.

de Sousa RT, Zarate CA, Zanetti MV, Costa AC, Talib LL, Gattaz WF, et al. Oxidativestress in early stage Bipolar Disorder and the association with response tolithium. J Psychiatr Res 2014;50:36e41.

Du J, Wei Y, Liu L, Wang Y, Khairova R, Blumenthal R, et al. A kinesin signalingcomplex mediates the ability of GSK-3beta to affect mood-associated behaviors.Proc Natl Acad Sci U S A 2010;107:11573e8.

Eldar-Finkelman H. Glycogen synthase kinase 3: an emerging therapeutic target.Trends Mol Med 2002;8:126e32.

Eom TY, Jope RS. Blocked inhibitory serine-phosphorylation of glycogen synthasekinase-3alpha/beta impairs in vivo neural precursor cell proliferation. BiolPsychiatry 2009;66:494e502.

First MB, Spitzer RL, Gibbon M, Williams JB. Structured clinical interview for DSM-IV Axis I disorders. Patient Ed. New York: Biometrics Research, New York Psy-chiatric Institute; 1995

Forlenza OV, Diniz BS, Radanovic M, Santos FS, Talib LL, Gattaz WF. Disease-modifying properties of long-term lithium treatment for amnestic mildcognitive impairment: randomised controlled trial. Br J Psychiatry 2011;198:351e6.

Grimes CA, Jope RS. The multifaceted roles of glycogen synthase kinase 3beta incellular signaling. Prog Neurobiol 2001;65:391e426.

Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960;23:56e62.

Hooper C, Killick R, Lovestone S. The GSK3 hypothesis of Alzheimer's disease.J Neurochem 2008;104:1433e9.

Jope RS. Glycogen synthase kinase-3 in the etiology and treatment of mood dis-orders. Front Mol Neurosci 2011;4:16.

Kessing LV, Andersen PK. Does the risk of developing dementia increase with thenumber of episodes in patients with depressive disorder and in patients withbipolar disorder? J Neurol Neurosurg Psychiatry 2004;75:1662e6.

Kessing LV, Søndergård L, Forman JL, Andersen PK. Lithium treatment and risk ofdementia. Arch Gen Psychiatry 2008;65:1331e5.

Kozlovsky N, Belmaker RH, Agam G. Low GSK-3beta immunoreactivity in post-mortem frontal cortex of schizophrenic patients. Am J Psychiatry 2000;157:831e3.

Lachman HM, Pedrosa E, Petruolo OA, Cockerham M, Papolos A, Novak T, et al.Increase in GSK3beta gene copy number variation in bipolar disorder. Am J MedGenet B Neuropsychiatr Genet 2007;144B:259e65.

Ladeira RB, Joaquim HP, Talib LL, Nunes PV, Forlenza OV. Higher proportion ofinactive Gsk3b in platelets of elderly patients with bipolar disorder: an effect oftreatment? Rev Bras Psiquiatr 2013;35:274e8.

Li X, Friedman AB, Zhu W, Wang L, Boswell S, May RS, et al. Lithium regulatesglycogen synthase kinase-3beta in human peripheral blood mononuclear cells:implication in the treatment of bipolar disorder. Biol Psychiatry 2007;61:216e22.

platelet serine-9 phosphorylated GSK-3b levels in drug-free bipolar5), http://dx.doi.org/10.1016/j.jpsychires.2015.01.016

R.T. de Sousa et al. / Journal of Psychiatric Research xxx (2015) 1e66

Li X, Jope RS. Is glycogen synthase kinase-3 a central modulator in mood regula-tion? Neuropsychopharmacology 2010;35:2143e54.

Li X, Liu M, Cai Z, Wang G. Regulation of glycogen synthase kinase-3 during bipolarmania treatment. Bipolar Disord 2010;12:741e52.

Machado-Vieira R, Manji HK, Zarate CA. The role of lithium in the treatment ofbipolar disorder: convergent evidence for neurotrophic effects as a unifyinghypothesis. Bipolar Disord 2009;11(Suppl. 2):92e109.

Machado-Vieira R, Soeiro-De-Souza MG, Richards EM, Teixeira AL, Zarate CA.Multiple levels of impaired neural plasticity and cellular resilience in bipolardisorder: developing treatments using an integrated translational approach.World J Biol Psychiatry 2014;15:84e95.

Martin M, Rehani K, Jope RS, Michalek SM. Toll-like receptor-mediated cytokineproduction is differentially regulated by glycogen synthase kinase 3. NatImmunol 2005;6:777e84.

Mendes CT, Mury FB, de S�a Moreira E, Alberto FL, Forlenza OV, Dias-Neto E, et al.Lithium reduces Gsk3b mRNA levels: implications for Alzheimer Disease. EurArch Psychiatry Clin Neurosci 2009;259:16e22.

Nunes PV, Forlenza OV, Gattaz WF. Lithium and risk for Alzheimer's disease inelderly patients with bipolar disorder. Br J Psychiatry 2007;190:359e60.

Pandey GN, Ren X, Rizavi HS, Dwivedi Y. Glycogen synthase kinase-3beta in theplatelets of patients with mood disorders: effect of treatment. J Psychiatr Res2010;44:143e8.

Peineau S, Taghibiglou C, Bradley C, Wong TP, Liu L, Lu J, et al. LTP inhibits LTD in thehippocampus via regulation of GSK3beta. Neuron 2007;53:703e17.

Please cite this article in press as: de Sousa RT, et al., Lithium increasesdisorder during depressive episodes, Journal of Psychiatric Research (201

Petkova E, Quitkin FM, McGrath PJ, Stewart JW, Klein DF. A method to quantify raterbias in antidepressant trials. Neuropsychopharmacology 2000;22:559e65.

Plein H, Berk M. The platelet as a peripheral marker in psychiatric illness. HumPsychopharmacol 2001;16:229e36.

Polter A, Beurel E, Yang S, Garner R, Song L, Miller CA, et al. Deficiency in theinhibitory serine-phosphorylation of glycogen synthase kinase-3 increasessensitivity to mood disturbances. Neuropsychopharmacology 2010;35:1761e74.

Ryves WJ, Harwood AJ. Lithium inhibits glycogen synthase kinase-3 by competitionfor magnesium. Biochem Biophys Res Commun 2001;280:720e5.

Stahl SM. The human platelet. A diagnostic and research tool for the study ofbiogenic amines in psychiatric and neurologic disorders. Arch Gen Psychiatry1977;34:509e16.

Sutton LP, RushlowWJ. The dopamine D2 receptor regulates Akt and GSK-3 via Dvl-3. Int J Neuropsychopharmacol 2012;15:965e79.

Talib LL, Valente KD, Vincentiis S, Gattaz WF. Correlation between platelet and brainPLA(2) activity. Prostaglandins Leukot Essent Fatty Acids 2013;89:265e8.

Zainaghi IA, Forlenza OV, Gattaz WF. Abnormal APP processing in platelets of pa-tients with Alzheimer's disease: correlations with membrane fluidity andcognitive decline. Psychopharmacol Berl 2007;192:547e53.

Zhang F, Phiel CJ, Spece L, Gurvich N, Klein PS. Inhibitory phosphorylation ofglycogen synthase kinase-3 (GSK-3) in response to lithium. Evidence forautoregulation of GSK-3. J Biol Chem 2003;278:33067e77.

platelet serine-9 phosphorylated GSK-3b levels in drug-free bipolar5), http://dx.doi.org/10.1016/j.jpsychires.2015.01.016