Journal of Affective Disorders

Transcription

1 Journal of Affective Disorders 136 (2012) Contents lists available at SciVerse ScienceDirect Journal of Affective Disorders journal homeage: Research reort COMT Met (158) modulates facial emotion recognition in biolar I disorder mood eisodes Márcio Gerhardt Soeiro-de-Souza a,,1, Danielle Soares Bio a,,1, Denise Petresco David a, Domingos Rodrigues dos Santos Jr. a, Daniel Shikanai Kerr b, Wagner Farid Gattaz b, Rodrigo Machado-Vieira b, Ricardo Albeto Moreno a a Mood Disorders Unit (GRUDA), Deartment and Institute of Psychiatry, School of Medicine, University of São Paulo (IPq HC-FMUSP), Brazil b Laboratory of Neuroscience LIM-27, Deartment and Institute of Psychiatry, School of Medicine, University of Sao Paulo (HC-FMUSP), Brazil article info abstract Article history: Received 22 August 2011 Received in revised form 25 October 2011 Acceted 14 November 2011 Available online 4 January 2012 Keywords: Doamine Facial emotions Catechol-O-methyltransferase Biolar disorder Mania Deression Background: One of the many cognitive deficits reorted in biolar disorder (BD) atients is facial emotion recognition (FER), which has recently been associated with doaminergic catabolism. Catechol-O-methyltransferase (COMT) is one of the main enzymes involved in the metabolic degradation of doamine (DA) in the refrontal cortex (PFC). The COMT gene olymorhism rs4680 (Val 158 Met) Met allele is associated with decreased activity of this enzyme in healthy controls. The objective of this study was to evaluate the influence of Val 158 Met on FER during manic and deressive eisodes in BD atients and in healthy controls. Materials and methods: 64 BD tye I atients (39 in manic and 25 in deressive eisodes) and 75 healthy controls were genotyed for COMT rs4680 and assessed for FER using the Ekman 60 Faces (EK60) and Emotion Hexagon (Hx) tests. Results: Biolar manic atients carrying the Met allele recognized fewer surrised faces, while deressed atients with the Met allele recognized fewer angry and hay faces. Healthy homozygous subjects with the Met allele had higher FER scores on the Hx total score, as well as on disgust and angry faces than other genotyes. Conclusion: This is the first study suggesting that COMT rs4680 modulates FER differently during BD eisodes and in healthy controls. This rovides evidence that PFC DA is art of the neurobiological mechanisms of social cognition. Further studies on other COMT olymorhisms that include euthymic BD atients are warranted. ClinicalTrials.gov Identifier: NCT Elsevier B.V. All rights reserved. 1. Introduction There is a growing body of evidence ointing to imaired facial emotion recognition (FER) in biolar disorder (BD) during mood eisodes and euthymia (Harmer et al., 2002; Lembke and Ketter, 2002; Summers et al., 2006; Venn et al., 2004). Recently, the doaminergic system, through Corresonding authors at: Dr. Ovidio Pires de Camos s/n, , São Paulo, Brazil. Tel.: ; fax: address: mgss@us.br (M.G. Soeiro-de-Souza). 1 Co-authorshi. catechol-o-methyltransferase (COMT), has been imlicated in the neurobiology of FER by genetic association studies with functional magnetic resonance imaging (fmri) in healthy subjects and euthymic BD atients (Lelli-Chiesa et al., 2011; Mier et al., 2010). Most data from healthy subjects have rovided system-level evidence suorting a behavioral dissociation by showing an effect of the COMT single nucleotide olymorhism (SNP) rs4680 (Val 158 Met) on amygdala activation during tasks with emotional rocessing comonents. Desite the available data, it is unclear if or how the COMT functional olymorhism modulates FER caacity in healthy oulations and BD atients during eisodes /$ see front matter 2011 Elsevier B.V. All rights reserved. doi: /j.jad

2 M.G. Soeiro-de-Souza et al. / Journal of Affective Disorders 136 (2012) Social cognition refers to the mental oerations underlying social interactions, which can be relatively indeendent from other asects of cognition and is not assessed by traditional neurocognitive tasks (Pinkham et al., 2003). One of the key asects of social cognition is the ability to discriminate accurately between different facially exressed emotions. The ability to rocess and identify facial emotions is an essential comonent of human communication and social interaction. Although social interaction may vary according to cultural norms and customs, cross-cultural studies have reeatedly rovided evidence in favor of the universality of facial emotions. Six universal emotions have since been established, including hainess, sadness, anger, fear, disgust and surrise, each of which corresonds to a secific arrangement of facial muscles and has artially searable neurocircuitry rocesses (Ekman and Friesen, 1971; Gosselin and Kirouac, 1995). BD atients, even during remission, have sychosocial roblems caused not only by residual symtoms, but also by cognitive deficits and difficulties in social cognition (Burdick et al., 2010; Jabben et al., 2010; Rocca et al., 2009; Solé et al., 2011). Evidence for BD deficits in FER varies in the literature from reorts of no alterations and imroved recognition for disgust (Harmer et al., 2002), isolated fear recognition imairment (Lembke and Ketter, 2002; Venn et al., 2004), to a selective effect of mood state (Rich et al., 2008; Rocca et al., 2009) on surrise recognition. In a recent meta-analysis however, Kohler et al. (2011) concluded that FER imairment in BD reresents a moderate and stable deficit that aears to be moderated by a limited number of demograhic and clinical factors such as self-reorted deression, age at time of testing and years of education. BD atient imairments in FER have been the focus of intense study with fmri disclosing differentiated activation of ventromedial refrontal cortex (PFC), cingulate, hiocamus, amygdala and limbic region (Chen et al., 2006; Dickstein et al., 2007; Foland et al., 2008; Lelli-Chiesa et al., 2011; Malhi et al., 2007). Chen et al. (2006) reorted a significant increase in amygdala activity among BD atients versus control subjects emotion labeling tasks. In this regard, Foland et al. (2008) showed that comared to healthy subjects, manic atients had significantly reduced ventrolateral PFC regulation of amygdala resonse during the emotion labeling task. COMT is an imortant regulator of PFC doaminergic (DA) levels (Lachman et al., 1996). This role renders COMT one of the main enzymes involved in the metabolic degradation of extraneuronal DA in glial cells and ostsynatic neurons (Lachman et al., 1996). Genetic studies have shown that COMT activity levels can vary considerably. The rs4680 SNP in the COMT gene leads to a 3 to 4-fold reduction in COMT enzyme activity in A allele (Met) carriers (Lachman et al., 1996). As a result, they have high levels of PFC DA due to lower enzyme activity while heterozygous subjects have an intermediate level of enzyme activity (Lachman et al., 1996; Weinshilboum et al., 1999). Thus, the COMT olymorhism rs4680 is resonsible for genetically modulating DA levels in PFC. Recently, genetic association fmri studies confirmed that COMT SNP rs4680 influenced emotion stimulus rocessing, showing that the Val allele was associated with greater amygdala activation and that signal change was greater for the Met allele in the ventromedial PFC and ventrolateral PFC (Lelli-Chiesa et al., 2011). Furthermore, studies in healthy carriers of the Val allele reorted imaired erformance (Bosia et al., 2007; Egan et al., 2001; Joober et al., 2002) couled with increased dorsal PFC activation during executive function tasks (Bertolino et al., 2006; Blasi et al., 2005; Mattay et al., 2003; Mier et al., 2010; Schott et al., 2006; Winterer et al., 2006). Nonetheless, healthy subjects with the Met allele are associated with greater reactivity to emotionally negative stimuli, as evidenced by increased activation in the ventral PFC and associated limbic regions (Drabant et al., 2006; Mier et al., 2010; Smolka et al., 2005). Desite all this information, it is unclear if/how the COMT olymorhism imacts FER in healthy controls or in BD atients during manic and deressive eisodes. Based on the otential association revealed by fmri studies between COMT and FER, the objective of this research was to investigate how/if the lower activity of the Met allele of COMT influenced FER scores in BD atients (mania and deression states) and healthy controls. 2. Materials and methods 2.1. Samle The atient samle comrised sixty-four medication-free individuals with BD I, aged between 18 and 40 years old (28.16±5.24 years) and currently in manic or deressive eisodes according to DSM-IV TR criteria (DSM-IV, 2000). All atients were articiants in the LICAVAL clinical trial (Camos et al., 2010) and were evaluated immediately after a wash-out eriod of four weeks for antideressants, mood stabilizers and antisychotics, or eight weeks for deot medications. Diagnoses were determined by trained sychiatrists using the Structured Clinical Interview (SCID-I) (First et al., 1997) for DSM-IV TR (DSM-IV, 2000). The Young Mania Rating Scale (YMRS) (Young et al., 1978) and Hamilton Deression Rating Scale (HDRS-21) (Hamilton, 1960) were used to evaluate the intensity of symtoms. The cut-off oint for mania was YMRS 12 and for deression was HDRS 15. The 39 manic atients had a mean YMRS of (±3.44), while the 25 deressive atients had a mean HDRS score of (±7.18). Subjects with neurological disorders, revious head trauma, any illness requiring medical intervention, current substance abuse, or that had undergone electroconvulsive theray in the receding six months, were excluded Control grou Seventy-five healthy volunteers (redominantly medical students) aged between 18 and 35 years old (23.54±3.53) were recruited from the University of São Paulo. All control subjects had no current or ast history of sychiatric disorder according to the evaluation conducted by trained sychiatrists using The Mini International Neurosychiatric Interview (MINI) (Sheehan et al., 1998). Similarly, all subjects had no family history (first degree relatives) of mood or sychotic disorders and had not been in recent use of sychotroic medicine or indulged in substance abuse over the last three months. Only women with a regular menstrual cycle were included. The YMRS (Young et al., 1978) and HDRS-21 (Hamilton, 1960) instruments were used to evaluate

3 372 M.G. Soeiro-de-Souza et al. / Journal of Affective Disorders 136 (2012) subclinical symtoms in controls, yielding a mean YMRS score of 0.67 (±1.05) and mean HDRS score of 0.75 (±1.21) FER tests Facial emotion recognition was tested using the Ekman 60 Faces Test (EK60) emloying a range of hotograhs from the Ekman and Friesen series of Pictures of Facial Affect (Ekman and Friesen, 1976), the most widely used and validated series of hotograhs in facial exression research. From this series, the faces of 10 actors (6 female, 4 male) were chosen, each dislaying six basic emotions ( hainess, sadness, disgust, fear, surrise and anger ). The EK60 can be used to assess recognition of facial exression of basic emotions. The maximum test score (indicating best erformance) is 60 for all six emotions, with 10 oints designated for each basic emotion. The comuter software for the test was available on CD-ROM. Patients were allowed unlimited time to resond. Immediately rior to testing, it was verified that atients and healthy controls semantically understood the words anger, disgust, fear, hainess, sadness and surrise. Patients and healthy controls were asked to rovide an examle for each emotion by answering the questions: Describe a situation when you feel hainess, fear, etc. Any incorrect answer would have led to exclusion from this study but all articiants gave correct answers. The Emotion Hexagon Test (Hx) is a test of facial emotion recognition utilizing ictures of emotional faces derived from Ekman and Friesen's Pictures of Facial Affect (1976). Ekman and Friesen's original ictures were modified using comuter maniulation techniques to generate stimuli of varying levels of difficulty. Each emotional face was merged with a icture deicting another emotion, which it was most likely to be confused with. Three levels of intensity were created for each emotion: 90%, 70% and 50%. Each face was resented for 5 s, after which time, articiants were asked to decide which of the six emotions (hainess, sadness, surrise, disgust, anger and fear) best described the face. Particiants comleted a ractice block followed by 5 test blocks of 30 trials each. Faces were resented in random order. Data from the ractice block and stimuli at the 50% intensity level were not included in the analysis Ethics The research ethics board of the Hosital das Clínicas of the University of São Paulo aroved the study. Written informed consent was obtained from all study articiants Genotying DNA was extracted from eriheral blood according to the salting-out rotocol (Laitinen et al., 1994) and then genotyed for COMT rs4680 using real-time PCR allelic discrimination. PCR amlification for rs4680 was erformed in 5 μl reactions with 5 ng of temlate DNA, 1 TaqMan Universal Master Mix (Alied Biosystems, Foster City, CA), 1 each rimer and robe assay, and H 2 O. Thermal cycling consisted of initial denaturation for10 min at 95 C, followed by 40 cycles of denaturation at 95 C for 15 s and annealing at 60 C for 1 min. Fluorescence detection occurred in the annealing ste. The amlification and allelic discrimination were erformed in a 7500 Real-Time System (Alied Biosystems, Foster City, CA). Quality control of Real time PCR results was done by direct sequencing on a ABI PRISM 3100 Genetic Analyzer (Alied Biosystems, Foster City, CA) Statistical analyses Subjects were classified into three grous (biolar subjects in mania or deressive eisodes and healthy controls) then stratified according to COMT genotye into Met allele resence [ (Met 158 Val, Met 158 Met)] or absence [ (Val 158 Val)]. Grous were not analyzed by genotye (Met 158- Met, Val 158 Met Val 158 Val) because after stratification by mood eisode and genotye some grous contained less than 10 individuals, which could decrease our statistic ower. Analysis of the three grous was done using the ANOVA test while the Chi-square test was used for categorical data, and the Student's t-test for continuous data. EK60 and Hx scores were comared within each grou using Student's t test. Bonferroni's ost hoc test was erformed for correction when three or more grous were comared. Pearson's correlation test was used to assess the relationshi between symtoms scales and FER tests in each grou. Version 18.0 of the PASW statistics software (SPSS Inc., Chicago, Illinois) was used for all analyses. 3. Results The COMT genotye distribution in the exerimental samles of men and women was in accordance with the Hardy Weinberg equilibrium (χ 2 =0.79) indicating that the samles were reresentative. Allelic frequency was 32.31% for and 67.69% for. No statistically significant differences in sociodemograhics were observed between genotyes in terms of age, gender or years of schooling (Table 1). The COMT genotye did not influence YMRS or HDRS during manic or deressive eisodes Biolar subjects had lower scores on facial emotion recognition than controls ANOVA analysis among the three grous of subjects (mania, deression and controls) for scores on the EK60 and Hx tests revealed significant differences in the following tests: EK60 total score (F=10.68; b0.001), EK60 anger (F=2.98; =0,05), EK60 fear (F=7.63; =0.001), Hx total score (F=10.32; b0.001), Hx fear (F=13.63; =b0.001), Hx hainess (F=3.75; =0.026), and Hx surrise (F=7.55; =0.001) (Table 2). Bonferroni ost hoc analysis for multile variables confirmed the relationshi: controls>mania=deression; in all FER tests with the excetion of Hx hainess and EK60 anger, in which controls=deression>mania (Table 2) Symtoms severity correlated with facial emotion recognition in mania and deression In manic subjects, YMRS score correlated ositively with EK60 fear (r=0.47; =0.002) and Hx fear (r=0.42;

4 M.G. Soeiro-de-Souza et al. / Journal of Affective Disorders 136 (2012) Table 1 Sociodemograhic variables for atients in mania or deression and healthy controls, carrying the COMT / or Val+/ alleles. Mania Deression Controls YMRS 15.5 (7.2) 15.4 (4.3) (6.1) 9.1 (6.9) (1.1) 0.4 (0.7) 0.15 HDRS 18.2 (8.1) 19.1 (7.4) (7.2) 22.7 (6.4) (1.3) 0.5 (0.7) 0.24 Years of study 12.3 (3.3) 12.1 (4.1) (2.8) 12.5 (1.3) (2.2) 14.2 (2.6) 0.99 Age 29.3 (5.1) 27.8 (4.8) (5.0) 28.2 (5.4) (3.4) 23.5 (3.2) 0.86 YMRS = Young Mania Rating Scale; HDRS = Hamilton Deression Rating Scale. =0.006) and negatively with EK60 surrise (r= 0.43; =0.005). Also, HDRS correlated negatively with both Hx disgust (r= 0.35; =0.031) and Hx hainess (r= 0.39; =0.015). In the deression grou, YMRS correlated negatively with EK60 total score (r= 0.41; =0.041) while HDRS correlated negatively with both EK60 hainess (r= 0.42; =0.035) and Hx hainess (r= 0.46; =0.02) In mania or deression subjects recognized less facial emotions than biolar subjects (mania+deression) irresective of mood state recognized fewer surrise faces on the EK60 test than did subjects. individuals in mania recognized fewer surrise faces (t= 2.17; =0.037) on the EK60 test than subjects (Table 3). subjects in deression scored lower than on EK60 hainess (t= 2.32; =0.036), Hx anger (t= 2.25; =0.035) and Hx hainess (t= 2.35; =0.034). Regarding controls, no difference in FER test scores was found between and subjects, which may indicate selective effects of COMT rs4680 for BD FER (Table 3) Scales of deressive and manic symtoms severity interacted with Met allele on FER results Regression analysis detected interaction between Met allele and HDRS only on EK60 hainess (B= 0.06 t= 2.2 =0.02 Partial Eta Squared=0.08). Allele Met and YMRS interaction was observed in EK60 fear (B=0.26 t=2.11 =0.039 Partial Eta Squared=0.07) and EK60 surrise (B= 0.17 t= 2.07 =0.043 Partial Eta Squared=0.07). 4. Discussion To the best of our knowledge, this is the first research to reort the association between the COMT Met allele and FER scores in biolar I disorder mood eisodes. During mood eisodes, biolar subjects had lower FER scores comared to. subjects in manic eisode recognized fewer surrise faces, while those in a deressive Table 2 FER scores for biolar atients in mania or deression and healthy controls. Mania Deression Controls F Bonferroni Post hoc Ek60 total 45.7 (5.7) 47.4 (6.7) 50.6 (5.1) <0.001 m = d < c ANGER 7.6 (1.8) 8.0 (1.7) 8.3 (1.3) m < d = c DISGUST 7.2 (2.0) 7.8 (1.9) 7.7 (1.9) m = d = c FEAR 5.4 (2.7) 5.5 (3.0) 7.2 (2.3) m = d < c HAPPINESS 9.5 (1.1) 9.6 (0.9) 9.8 (1.0) m = d = c SADNESS 7.3 (2.0) 7.7 (1.9) 8.2 (1.6) m = d = c SURPRISE 8.6 (1.8) 8.6 (2.1) 9.1 (1.4) m = d = c Hx total 99.0 (17.2) 102 (14.6) (9.7) <0.001 m = d < c ANGER 15.6 (5.0) 16.0 (5.7) 17.6 (3.9) m = d = c DISGUST 17.1 (3.7) 17.1 (3.5) 17.8 (3.5) m = d = c FEAR 13.0 (4.6) 14.3 (5.3) 17.3 (3.6) <0.001 m = d < c HAPPINESS 18.4 (4.0) 19.4 (1.3) 19.6 (0.9) m < d = c SADNESS 22.5 (9.5) 18.3 (2.7) (2.35) m = d = c SURPRISE 16.8 (3.79) 16.7 (3.9) 18.7 (1.62) m = d < c FER = Facial emotion recognition; EK60 = Ekman 60 Faces Test; Hx = Emotion Hexagon Test; m = mania; d = deression; c = healthy controls. Text in bold indicates the tests with significant differences according to Bonferroni ost hoc test.

5 374 M.G. Soeiro-de-Souza et al. / Journal of Affective Disorders 136 (2012) Table 3 FER scores for biolar atients in mania or deression and healthy controls, carrying the COMT or alleles. Mania Deression Controls EK60 total 45.4 (5.8) 46.5 (5.3) (7.7) 48.2 (5.0) (4.9) 50.8 (4.5) 0.91 Anger 7.6 (1.8) 7.3 (1.9) (1.6) 7.8 (2.0) (1.3) 8.1 (1.0) 0.48 Disgust 6.9 (2.1) 7.7 (1.7) (2.2) 8.4 (0.9) (1.9) 7.8 (1.9) 0.81 Fear 5.4 (2.6) 5.2 (3.0) (3.1) 5.4 (3.5) (2.4) 6.8 (2.0) 0.64 Hainess 9.6 (0.8) 9.7 (0.4) (1.1) 10.0 (0.0) (1.2) 10.0 (0) 0.23 Sadness 7.6 (2.2) 7.0 (1.7) (2.1) 7.7 (1.7) (1.7) 8.6 (1.3) 0.25 Surrise 8.3 (2.0) 9.3 (0.7) (2.4) 8.8 (1.4) (0.9) 9.3 (0.8) 0.91 Hx total 97.8 (19.3) (14.5) (15.4) (10) (9.5) (10.5) 0.27 Anger 15.6 (5.1) 15.3 (5.1) (6.4) 18.5 (2.9) (4.1) 17.3 (3.4) 0.67 Disgust 16.7 (4.2) 17.5 (3.0) (4.1) 18.0 (1.9) (3.3) 16.9 (4.0) 0.21 Fear 12.5 (4.9) 14.1 (3.9) (5.6) 14.8 (5.9) (3.5) 16.3 (4.2) 0.25 Hainess 18.1 (4.8) 18.7 (2.8) (1.6) 20.0 (0.0) (1.0) 19.8 (0.4) 0.11 Sadness 25.4 (7.5) 17.7 (2.8) (3.2) 19.0 (1.8) (2.4) 19.0 (1.9) 0.94 Surrise 16.9 (4.4) 17.0 (2.6) (4.4) 17.4 (2.9) (1.3) 18.0 (2.0) 0.11 EK60 = Ekman 60 Faces Test; Hx = Emotion Hexagon Test. state recognized fewer anger and hay faces comared to subjects. Notably, the comarison between healthy / controls showed no differences in FER scores. Previous studies on FER suggest that the COMT Val 158 Met olymorhism has a leiotroic effect within the neural networks subserving emotional rocessing (Lelli-Chiesa et al., 2011). Similarly, Mier et al. (2010), in a meta-analysis of all available neuroimaging studies on rs4680 investigating the evidence for a neural substrate of this behavioral leiotroy, found strong and oosing effects for executive cognition aradigms (favoring Met allele carriers) and emotional aradigms (favoring Val), roviding meta-analytical evidence of a neural substrate for the leiotroic behavioral effects of COMT genetic variation. The exlanation for DA influence on FER might be linked to its influence on the amygdala, a key brain structure associated to FER (Adolhs and Tranel, 2003; Adolhs et al., 1994; Morris et al., 1996). A human fmri study reorted that doaminergic drug theray, such as levodoa or DA agonists, artially restored amygdala resonse to an emotional task in Parkinson's disease subjects who showed no significant amygdala resonse during drug-off states (Tessitore et al., 2002). In addition, another fmri study of healthy volunteers has demonstrated that amhetamine otentiated the resonse of the amygdala during an emotional task (Hariri et al., 2002). Moreover, Kienast et al. (2008) reorted that DA storage caacity in human amygdala, measured with 6- [18F] fluoro-ldopa ositron emission tomograhy (PET), was ositively correlated with fmri signal changes in the amygdala. A more recent study has reorted that DA D1 binding in the amygdala was ositively correlated with amygdala signal change in resonse to fearful faces and concluded that DA D1 recetors might lay a major role in enhancing amygdala resonse when sensory inuts are affective (Takahashi et al., 2010). On the basis of these findings, Mier et al. (2010) roosed that the imaging signals obtained on fmri studies of emotional rocessing tasks in their study relate to this decreased cortical efficiency as a neural correlate of behavioral inflexibility. Additionally, the hysiology of DA modulation of cortical function has been elucidated, where an effect of COMT rs4680 on the interaction of DA synthesis with brain activity, indexed by blood flow, was directly shown by Meyer-Lindenberg et al. (2005). In BD, the COMT Val allele was recently associated with greater amygdala activation during sad facial affect (Lelli-Chiesa et al., 2011). In the same study, the Met allele was associated to greater activation in the ventrolateral PFC in family members with affective morbidity comared against relatives without sychiatric conditions and healthy controls. The imact of DA level alterations on cognition has been extensively investigated. Studies have shown that low doses of D1 agonists imrove working memory and attention regulation (Granon et al., 2000), while high levels of DA release imair PFC function (Mattay et al., 2003). In 1977, Srague and Sleator (1977), reorted that at low sychostimulant doses, hyerkinetic children showed significant imrovement in short-term memory, whereas at higher doses a significant decline in erformance was seen. Mattay et al. (2003) reorted similar findings in healthy volunteers using dextroamhetamine, a drug that otentiates DA activity. These authors observed that subjects carrying the COMT Met/Met genotye (rs4680) had oorer working memory and executive function, while Val carriers showed imroved erformance after a single dose of dextroamhetamine. Healthy (rs4680) subjects have shown decreased hasic, and increased tonic, DA transmission subcortically, and increased DA concentrations cortically (Bilder et al., 2004). This increases the stability of networks mediating sustained working memory reresentations, but makes it more difficult to switch or udate the currently active networks that reresent sustained working memory reresentations or ongoing behavioral rograms (Bilder et al., 2004). However, when the D1/D2 binding otentials are disturbed due to an increase in DA, as occurs during BD eisodes (Gonul et al., 2009), cognitive erformance is theoretically imaired by excessive D1 stimulation. However, although euthymic subjects with BD may show imairments in neurocognitive domains, there is a lack of information on social cognition skills in this oulation. Strengths of the resent study include being the first reort of a ositive association between FER and the COMT genotye in BD and the fact that the samle was comrised

6 M.G. Soeiro-de-Souza et al. / Journal of Affective Disorders 136 (2012) homogeneously of young BD tye I subjects with no current use of medication, allowing a clear observation of the henotyic boundaries of mania and deression in terms of social cognition. In addition, a validated measure of FER was used to assess these biolar subjects. Limitations of the resent study include the grou size, which should ideally have been larger in order to show significant differences more clearly, and also the absence of euthymic subjects. 5. Conclusion This is the first study suggesting that COMT rs4680 modulates facial emotion recognition differently during eisodes of BD among subjects, and in healthy controls. This rovides evidence that PFC DA is art of the neurobiological mechanisms of social cognition. DA recetor stimulation alterations during BD mood eisodes might exlain the contrasting results seen in BD subjects comared to controls. Further studies on other COMT olymorhisms that include euthymic BD subjects are warranted. Role of funding source The Sao Paulo Research Foundation (FAPESP 2010/ / ) financed this study. The Sao Paulo research foundation (FAPESP) is an indeendent ublic foundation with the mission to foster research and the scientific and technological develoment of the State of São Paulo. Conflict of interest The authors do not have any conflict of interest to reort. Acknowledgments We would like to thank the Institute of Psychiatry at the University of Sao Paulo, esecially the members of the Mood Disorders Unit (GRUDA) and Laboratory of Neuroscience (LIM27), for their dedication and hard work, as well as the volunteers for their collaboration in this study. References Adolhs, Ralh, Tranel, D., Amygdala damage imairs emotion recognition from scenes only when they contain facial exressions. Neurosychologia 41 (10), Adolhs, R., et al., Imaired recognition of emotion in facial exressions following bilateral damage to the human amygdala. Nature 372 (6507), Bertolino, A., et al., Prefrontal hiocamal couling during memory rocessing is modulated by COMT Val 158 Met genotye. Biological Psychiatry 60 (11), Bilder, R.M., et al., The catechol-o-methyltransferase olymorhism: relations to the tonic-hasic doamine hyothesis and neurosychiatric henotyes. Neurosychoharmacology: Official Publication of the American College of Neurosychoharmacology 29 (11), Blasi, G., et al., Effect of catechol-o-methyltransferase Val 158 Met genotye on attentional control. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience 25 (20), Bosia, M., et al., Influence of catechol-o-methyltransferase Val 158 Met olymorhism on neurosychological and functional outcomes of classical rehabilitation and cognitive remediation in schizohrenia. Neuroscience Letters 417 (3), Burdick, K.E., Goldberg, J.F., Harrow, M., Neurocognitive dysfunction and sychosocial outcome in atients with biolar I disorder at 15- year follow-u. Acta Psychiatrica Scandinavica 122 (6), Camos, R.N., et al., LICAVAL: combination theray in acute and maintenance treatment of biolar disorder. Trials 11, 72. Chen, C.-H., et al., Exlicit and imlicit facial affect recognition in manic and deressed states of biolar disorder: a functional magnetic resonance imaging study. Biological Psychiatry 59 (1), Dickstein, D.P., et al., Neural activation during encoding of emotional faces in ediatric biolar disorder. Biolar Disorders 9 (7), Drabant, E.M., et al., Catechol O-methyltransferase Val 158 Met genotye and neural mechanisms related to affective arousal and regulation. Archives of General Psychiatry 63 (12), DSM-IV, P.A.T.F.O., Diagnostic and Statistical Manual of Mental Disorders: DSM-IV-TR. American Psychiatric Publishing, Inc. Egan, M.F., et al., Effect of COMT Val108/158 Met genotye on frontal lobe function and risk for schizohrenia. Proceedings of the National Academy of Sciences of the United States of America 98 (12), Ekman, P., Friesen, W.V., Constants across cultures in the face and emotion. Journal of Personality and Social Psychology 17 (2), Ekman, Paul, Friesen, Wallace V., Pictures of facial affect, Consulting Psychologists Press, Palo Alto, CA. First, M.B., Sitzer, R.L., Williams, J.B., Structured Clinical Interview for DSM-IV Axis I Disorders SCID-I. Washington, D.C.: American Psychiatric Press, Inc. Foland, L.C., et al., Evidence for deficient modulation of amygdala resonse by refrontal cortex in biolar mania. Psychiatry Research 162 (1), Gonul, A.S., Coburn, K., Kula, M., Cerebral blood flow, metabolic, recetor, and transorter changes in biolar disorder: the role of PET and SPECT studies. International Review of Psychiatry (Abingdon, England) 21 (4), Gosselin, P., Kirouac, G., Decoding facial emotional rototyes. Canadian Journal of Exerimental Psychology = Revue Canadienne de Psychologie Exérimentale 49 (3), Granon, S., et al., Enhanced and imaired attentional erformance after infusion of D1 doaminergic recetor agents into rat refrontal cortex. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience 20 (3), Hamilton, M., A rating scale for deression. Journal of Neurology, Neurosurgery, and Psychiatry 23, Hariri, A.R., et al., Dextroamhetamine modulates the resonse of the human amygdala. Neurosychoharmacology: Official Publication of the American College of Neurosychoharmacology 27 (6), Harmer, C.J., Grayson, L., Goodwin, G.M., Enhanced recognition of disgust in biolar illness. Biological Psychiatry 51 (4), Jabben, N., et al., Neurocognitive functioning as intermediary henotye and redictor of sychosocial functioning across the sychosis continuum: studies in schizohrenia and biolar disorder. The Journal of Clinical Psychiatry 71 (6), Joober, R., et al., Catechol-O-methyltransferase Val-108/158-Met gene variants associated with erformance on the Wisconsin Card Sorting Test. Archives of General Psychiatry 59 (7), Kienast, T., et al., Doamine in amygdala gates limbic rocessing of aversive stimuli in humans. Nature Neuroscience 11 (12), Kohler, C.G., Hoffman, L.J., Eastman, L.B., Healey, K., Moberg, P.J., Facial emotion ercetion in deression and biolar disorder: a quantitative review. Psychiatry Research 188 (3), Lachman, H.M., et al., Human catechol-o-methyltransferase harmacogenetics: descrition of a functional olymorhism and its otential alication to neurosychiatric disorders. Pharmacogenetics 6 (3), Laitinen, J., Samarut, J., Hölttä, E., A nontoxic and versatile rotein salting-out method for isolation of DNA. Biotechniques 17 (2), 316, 318, Lelli-Chiesa, G., et al., The imact of the Val 158 Met catechol-omethyltransferase genotye on neural correlates of sad facial affect rocessing in atients with biolar disorder and their relatives. Psychological Medicine 41 (4), Lembke, A., Ketter, T.A., Imaired recognition of facial emotion in mania. The American Journal of Psychiatry 159 (2), Malhi, G.S., et al., Is a lack of disgust something to fear? A functional magnetic resonance imaging facial emotion recognition study in euthymic biolar disorder atients. Biolar Disorders 9 (4), Mattay, V.S., et al., Catechol O-methyltransferase val158-met genotye and individual variation in the brain resonse to amhetamine. Proceedings of the National Academy of Sciences of the United States of America 100 (10), Meyer-Lindenberg, A., et al., Midbrain doamine and refrontal function in humans: interaction and modulation by COMT genotye. Nature Neuroscience 8 (5), Mier, D., Kirsch, P., Meyer-Lindenberg, A., Neural substrates of leiotroic action of genetic variation in COMT: a meta-analysis. Molecular Psychiatry 15 (9), Morris, J.S., et al., A differential neural resonse in the human amygdala to fearful and hay facial exressions. Nature 383 (6603), Pinkham, A.E., et al., Imlications for the neural basis of social cognition for the study of schizohrenia. The American Journal of Psychiatry 160 (5),

7 376 M.G. Soeiro-de-Souza et al. / Journal of Affective Disorders 136 (2012) Rich, B.A., et al., Face emotion labeling deficits in children with biolar disorder and severe mood dysregulation. Develoment and Psychoathology 20 (2), Rocca, C.C., Heuvel, E., Caetano, S.C., Lafer, B., Facial emotion recognition in biolar disorder: a critical review. Revista Brasileira de Psiquiatria (São Paulo, Brazil: 1999) 31 (2), Schott, B.H., et al., The doaminergic midbrain articiates in human eisodic memory formation: evidence from genetic imaging. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience 26 (5), Sheehan, D.V., et al., The Mini-International Neurosychiatric Interview (M.I.N.I.): the develoment and validation of a structured diagnostic sychiatric interview for DSM-IV and ICD-10. The Journal of Clinical Psychiatry 59 (Sul 20), quiz Smolka, M.N., et al., Catechol-O-methyltransferase Val 158 Met genotye affects rocessing of emotional stimuli in the amygdala and refrontal cortex. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience 25 (4), Solé, B., Bonnin, C.M., Torrent, C., Balanzá-Martínez, V., Tabarés-Seisdedos, R., Poovic, D., Martínez-Arán, A., Vieta, E., 2011 Aug 17. Neurocognitive imairment and sychosocial functioning in biolar II disorder. Acta Psychiatr Scand. doi: /j x. [Eub ahead of rint]. Srague, R.L., Sleator, E.K., Methylhenidate in hyerkinetic children: differences in dose effects on learning and social behavior. Science 198 (4323), Summers, M., et al., Biolar I and biolar II disorder: cognition and emotion rocessing. Psychological Medicine 36 (12), Takahashi, H., et al., Contribution of doamine D1 and D2 recetors to amygdala activity in human. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience 30 (8), Tessitore, A., et al., Doamine modulates the resonse of the human amygdala: a study in Parkinson's disease. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience 22 (20), Venn, H.R., et al., Percetion of facial exressions of emotion in biolar disorder. Biolar Disorders 6 (4), Weinshilboum, R.M., Otterness, D.M., Szumlanski, C.L., Methylation harmacogenetics: catechol O-methyltransferase, thiourine methyltransferase, and histamine N-methyltransferase. Annual Review of Pharmacology and Toxicology 39, Winterer, G., et al., Prefrontal electrohysiologic noise and catechol- O-methyltransferase genotye in schizohrenia. Biological Psychiatry 60 (6), Young, R.C., et al., A rating scale for mania: reliability, validity and sensitivity. The British Journal of Psychiatry: The Journal of Mental Science 133,