Monoamine oxidase and agitation in psychiatric patients Matea Nikolac Perkovic, Dubravka Svob Strac, Gordana Nedic Erjavec, Suzana Uzun, Josip Podobnik, Oliver Kozumplik, Suzana Vlatkovic, Nela Pivac PII: DOI: Reference:
S0278-5846(16)30016-1 doi: 10.1016/j.pnpbp.2016.02.002 PNP 8884
To appear in:
Progress in Neuropsychopharmacology & Biological Psychiatry
Received date: Revised date: Accepted date:
30 November 2015 15 January 2016 2 February 2016
Please cite this article as: Perkovic Matea Nikolac, Strac Dubravka Svob, Erjavec Gordana Nedic, Uzun Suzana, Podobnik Josip, Kozumplik Oliver, Vlatkovic Suzana, Pivac Nela, Monoamine oxidase and agitation in psychiatric patients, Progress in Neuropsychopharmacology & Biological Psychiatry (2016), doi: 10.1016/j.pnpbp.2016.02.002
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Monoamine oxidase and agitation in psychiatric patients
Matea Nikolac Perkovica*, Dubravka Svob Straca*, Gordana Nedic Erjaveca*, Suzana Uzunb, Josip Podobnikc,
CR I
a
PT
Oliver Kozumplikb, Suzana Vlatkovicd, Nela Pivaca
Rudjer Boskovic Institute, Division of Molecular Medicine, Bijenicka cesta 54, 10000 Zagreb, Croatia
b
Clinic for Psychiatry Vrapce, Bolnicka cesta 32, 10000 Zagreb, Croatia
Department of Psychiatry, Psychiatric Hospital for Children and Youth Zagreb, Kukuljeviceva 11, 10000
NU S
c
Zagreb, Croatia d
D
MA
Psychiatric Hospital Dr Ivan Brbot, Popovaca, Croatia
AC
CE P
TE
*Matea Nikolac Perkovic, Dubravka Svob Strac and Gordana Nedic Erjavec equally contributed to this work.
Correspondence to: Prof. Nela Pivac, Ph.D., Senior Scientist Division of Molecular Medicine Rudjer Boskovic Institute Bijenicka cesta 54, 10 000 Zagreb, Croatia Phone: + 385 1 457 1207; Fax: + 385 1 456 1010 E-mail: [email protected]
1
ACCEPTED MANUSCRIPT Abstract Subjects with schizophrenia or conduct disorder display a lifelong pattern of antisocial, aggressive and violent behavior and agitation. Monoamine oxidase (MAO) is an enzyme involved in degradation of
PT
various monoamine neurotransmitters and neuromodulators and therefore has a role in various psychiatric and neurodegenerative disorders and pathological behaviors. Platelet MAO-B activity has
CR I
been associated with psychopathy- and aggression-related personality traits, while variants of the MAOA and MAOB genes have been associated with diverse clinical phenotypes, including aggressiveness,
NU S
antisocial problems and violent delinquency. The aim of the study was to evaluate the association of platelet MAO-B activity, MAOB rs1799836 polymorphism and MAOA uVNTR polymorphism with severe
MA
agitation in 363 subjects with schizophrenia and conduct disorder. The results demonstrated significant association of severe agitation and smoking, but not diagnosis or age, with platelet MAO-B activity. Higher
D
platelet MAO-B activity was found in subjects with severe agitation compared to non-agitated subjects.
TE
Platelet MAO-B activity was not associated with MAOB rs1799836 polymorphism. These results suggested the association between increased platelet MAO-B activity and severe agitation. No significant
CE P
association was found between severe agitation and MAOA uVNTR or MAOB rs1799836 polymorphism, revealing that these individual polymorphisms in MAO genes are not related to severe agitation in subjects
AC
with schizophrenia and conduct disorder. As our study included 363 homogenous Caucasian male subjects, our data showing this negative genetic association will be a useful addition to future metaanalyses.
Keywords: agitation; platelet MAO-B activity; MAOB rs1799836 polymorphism; MAOA uVNTR polymorphism; schizophrenia; conduct disorder
2
ACCEPTED MANUSCRIPT 1. Introduction Different psychiatric disorders, including schizophrenia and conduct disorder, have debilitating effects
PT
on patients and their families, causing them great suffering and carrying enormous mental health and societal burden as well as financial costs to the whole society. These complex and multifactorial mental
CR I
disorders are rarely isolated, and commonly co-occur with other comorbid psychiatric disorders that complicate their clinical features as well as treatment. Patients with these disorders are frequently non-
NU S
responsive or only modestly responsive to current treatment options. Therefore, identification of the risk markers and biomarkers specific for these disorders may improve our understanding of the
MA
etiology/pathophysiology of these diseases, and provide novel drug targets. However, the putative biomarkers for various complex psychiatric diseases, such as schizophrenia and conduct disorder, are still
D
missing (Chan et al., 2014). Although the understanding of biological pathways involved in the development
TE
of these disorders is improving, novel strategies focused to find reliable biomarkers and to improve treatment response are still rare. According to the Biomarkers Definitions Working Group, as part of the
CE P
NIH Director's Initiative on Biomarkers and Surrogate Endpoints, biomarker is defined as “a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic
AC
processes, or pharmacologic responses to a therapeutic intervention or other health care intervention“. Molecular biomarkers (Schmidt et al., 2013) can be assessed on different molecular levels, namely, on the genetic level (DNA biomarkers), the gene expression level (RNA biomarkers), the level of proteins (peptide and protein biomarkers), and epigenetic level, which regulates the gene expression by several mechanisms, namely, DNA methylation, histone modifications, and RNA interference (epigenetic biomarkers). Blood cells represent a good, easily available and non-invasive biological material for biomarker screening (Mohr and Liew, 2007), as they share more than 80% of the transcriptome with other tissues (brain, colon, heart, kidney, liver, lung, prostate, spleen and stomach). Deficits in human brain function can be assessed either “post-mortem” in the specific brain regions, or by the use of neuroimaging and nuclear magnetic resonance (NMR) spectroscopy. Other common approaches include genetics/genomics, transcriptomics, proteomics (Chan et al., 2014) and epigenetics (Shumay et al., 2012). At the metabolite
3
ACCEPTED MANUSCRIPT level, evidence is provided (Chan et al., 2014) that molecular profiling analysis of blood serum/plasma can reveal robust molecular changes in psychiatric patients, suggesting that these changes are detectable besides in the brain, also in other body systems such as the circulating blood. Although peripheral
PT
biomarkers might offer easily accessible tests which could make a significant improvement in diagnosis, treatment, and patient care (Bahn and Chan, 2015), validated and reliable biomarkers of different
CR I
psychiatric disorders are still not identified (Zolad and Diamond, 2013).
The studies investigating the molecular profiles of body fluids (i.e. saliva, plasma or serum) and blood
NU S
cells are promising areas of research aiming to identify peripheral biomarkers for psychiatric disorders (Bahn and Chan, 2015). There is an unmet need to find reliable and validated biomarkers of
MA
neuropsychiatric disorders, especially from the accessible peripheral fluids (e.g. blood), in order to facilitate early diagnoses, detect particular personality traits, symptoms and behaviors, prevent psychopathological
D
behaviors and to advance their treatment, but also to improve prognosis and disease outcome (Chan et al.,
TE
2014). In the research of the biological underpinnings and reliable biomarkers of the specific endophenotypes, pathological behaviors, and different personality traits in neuropsychiatric disorders,
CE P
human platelets have been used as a non-invasive model (Asor and Ben-Shachar, 2012). Platelets and neurons share some similar processes and contain some identical components of the serotonergic
AC
system (Camacho and Dimsdale, 2000; Yubero-Lahoz et al., 2013). In addition, similar pathologies are found in platelets and neurons in many neuropsychiatric diseases (Asor and Ben-Shachar, 2012). Therefore, platelets have been used as an easily obtainable model for neurobiological research in humans. Due to the complexity of the neurobiological changes in neuropsychiatric disorders, not one biomarker, but a combination of multiple biomarkers is expected to be associated with the alterations observed in specific pathological behaviors in neuropsychiatry. The quest for biomarkers is especially important in forensic psychiatry for the prediction of violent and assaultive behaviors, since these markers could improve the accuracy of risk assessments (Gustavson et al., 2010).
1.1. Monoamine oxidase (MAO)
4
ACCEPTED MANUSCRIPT Monoamine oxidase (MAO; E.C. 1.4.3.4) is a flavin-adenosine-dinucleotide (FAD)-containing enzyme which is involved in degradation of various biogenic amines. MAO exists in two forms, MAO-A and MAO-B, responsible for the oxidative deamination of different monoamine neurotransmitters and
PT
neuromodulators (Bortolato and Shih, 2011; Oreland, 2004; Shih et al., 1999). Both MAO subtypes degrade serotonin, melatonin, dopamine, norepinephrine and epinephrine, and different amines
CR I
(tyramine, tryptamine, 2-phenylethylamine, octopamine and 3-iodothyronamine). MAO-A and MAO-B differ in their substrate preferences, immunological properties, molecular weight and anatomical locations,
NU S
and they are inhibited by different inhibitors. MAO-A deaminates primarily serotonin, norepinephrine and epinephrine, while MAO-B degrades β-phenylethylamine and benzylamine. Both isoenzymes degrade
MA
dopamine, tryptamine and tyramine. However, when one subtype is absent, the other isoenzyme can take its role (Bortolato and Shih, 2011).
D
Both forms of MAO are localized on the outer membrane of the mitochondria and widely distributed
TE
throughout the body. Within the central nervous system MAO-B is expressed at highest levels in serotonergic neurons, histaminergic neurons and in glial cells, and MAO-A is localized primarily in
CE P
catecholaminergic (dopaminergic and noradrenergic) neurons (Luque et al., 1995; Saura et al., 1994; Westlund et al., 1988). In the peripheral tissues MAO-A is localized in fibroblasts and placenta (Egashira and
AC
Yamanaka, 1981), liver and gastro-intestinal tract, while MAO-B is found in liver (Billett et al., 2004), blood platelets and lymphocytes (Bond and Cundall, 1977; Bortolato and Shih 2011). MAO inhibitors are used as antidepressants, but due to their multiple drug- and food-interactions and side effects, they are prescribed less frequently (Fiedorowicz and Swartz, 2014). However, these drugs still have a place in the treatment of atypical, treatment-resistant depression or bipolar depression (Shulman et al., 2013). They include phenelzine, tranylcypromine and isocarboxazid, as well as newer, selective and reversible drugs, such as the MAO-A inhibitor, moclobemide, and MAO-B inhibitor, selegiline.
1.1.1. Platelet MAO-B activity Platelet MAO-B activity has been proposed to represent a biomarker of different personality traits and psychiatric vulnerability, including sensation-seeking behavior, novelty-seeking personality, extraversion,
5
ACCEPTED MANUSCRIPT poor impulse control and predisposition for taking different risky behaviors (Harro et al., 2004; Oreland, 2004; Oreland and Hallman, 1995). Platelet MAO-B activity, assumed to be a biological marker of psychopathology, was found to be associated with psychopathy- and aggression-related personality
PT
traits (Stalenheim, 2004). It is presumed that MAO-B function in platelets is heritable and its heritability has been estimated around 0.77 (Pedersen et al., 1993). Using the transmission probability model, the
CR I
familial transmission of MAO activity was consistent with either recessive or additive, but not with dominant inheritance (Baron et al., 1985). Platelet MAO-B has similar biochemical and pharmacological
NU S
characteristics as the brain MAO-B (Oreland, 2004), and both brain and platelet MAO-B are encoded by the MAOB gene (Chen et al., 1992). Various MAOB gene variants might influence the protein expression
MA
and the activity of platelet MAO-B (Bortolato and Shih, 2011). In addition, platelet MAO-B activity was shown to be significantly affected by sex, age, and smoking (Oreland, 2004), as well as with other
D
environmental factors, implicating possible epigenetic mechanisms (Shumay et al., 2012).
TE
Many studies have reported reduced platelet MAO-B activity in subjects with disinhibited behaviors such as novelty-seeking behavior, low levels of harm avoidance, high impulsiveness and high levels of sensation
CE P
seeking (Oreland, 2004; Oreland et al., 1999; Ruchkin et al., 2005; Shih et al., 1999). It has been hypothesized that low MAO-B activity may represent a nonspecific marker indicating a predisposition to
AC
distress or psychopathology in general, rather than to specific behavior or specific disorders (Ruchkin et al., 2005). However, alterations in both directions (i.e. increased as well as decreased platelet MAO-B activity) have been proposed to represent biological vulnerability to different behaviors and traits (Paaver et al., 2006), as there is no simple or linear relationship between pathological behaviors, traits and monoaminergic functioning (Paaver et al., 2006; Schalling et al., 1987). Various results suggest that mutations and polymorphisms in MAOA and MAOB genes might differentially affect the turnover rates of dopamine and serotonin. Higher plasma and urine levels of monoamines processed by MAOs (norepinephrine, epinephrine and serotonin), along with their metabolites, such as normetanephrine, have been associated with CT genotype of the MAOA rs1137070 polymorphism (Dorszewska et al., 2013). Moreover, a high activity MAOA uVNTR variant has been also associated with higher levels of monoamine metabolites, homovanillic acid (HVA) and 5-
6
ACCEPTED MANUSCRIPT hydroxyindoleacetic acid (5-HIAA) in lumbar cerebrospinal fluid (CSF) (Jonsson et al., 2000; Williams et al., 2003; Zalsman et al., 2005). It has been found that patients with “Brunner syndrome”, resulting from the mutation in the MAOA gene, have low concentrations of HVA, 5-HIAA and vanillylmandelic acid,
PT
confirming reduced monoamine degradation (Brunner et al., 1993), due to MAO-A deficiency. In addition, the increased MAO-B protein expression, and consequently the increased platelet MAO-B
CR I
activity was shown to be associated with MAOB rs1799836 A allele (Jakubauskiene et al., 2012). However, platelet MAO-B activity was not associated with CSF concentration of different monoamine
NU S
metabolites (HVA, 5-HIAA and 3-methoxy-4-hydroxyphenylglycol) in the study of Gustavson et al. (2010), but it was positively correlated to CSF levels of HVA and 5-HIAA in other studies (Knorring et al., 1986;
D
1.1.2. MAO genetics and epigenetics
MA
Oreland et al., 1981).
TE
Genes coding for MAO-A and MAO-B are located on the X-chromosome (Xp11.23–11.4) and they are both comprised of 15 exons with an identical exon–intron organization (Chen et al., 1992). These genes are
CE P
organized in opposite direction, tail to tail with 24 kb apart. The clinical consequences of the genetic deficits in MAO genes are numerous. After the detection of the recessive X-linked disorder, known as a
AC
“Brunner syndrome”, in male subjects of one Dutch family, characterized by the selective loss of the MAOA activity and diverse pathological symptoms and behaviours such as mild-borderline mental retardation, aggressive and violent behavior, impaired impulse control and disturbed regulation, arson, attempted rape, and exhibitionism (Brunner et al., 1993), there was a quest for other affected subjects. This genotype/phenotype correlation was questioned (Hebebrand and Klug, 1995), as well as confirmed in a few studies including a small number of subjects (Palmer et al., 2015). Affected subjects had disturbances in urine monoamine levels. Other genetic mutations in MAOA and MAOB genes have been associated with diverse clinical phenotypes, as confirmed in animal models as well as in human studies (Bortolato and Shih, 2011). The nonsynonymous variations in the coding sequences of MAO genes might alter the amino acid sequence of the MAO protein or result in a truncated, incomplete or non-functional MAO protein.
7
ACCEPTED MANUSCRIPT However, MAO activity might also be affected by some regulatory polymorphisms located in noncoding regions, including promoter/upstream, downstream and intron regions that may affect transcription factor binding, RNA splicing, RNA stability and degradation, the sequence of non-coding RNA, and
PT
therefore impact both transcription and translation. Such polymorphisms are for instance the MAOA uVNTR polymorphism located in the promoter region, as well as rs1799836 polymorphism found in the
CR I
13th intron of the MAOB gene. Moreover, individual polymorphisms in MAO genes may have minimal functional impact but may be in linkage disequilibrium with a set of polymorphisms that form a
NU S
haplotype influencing MAO gene expression or function, consequently altering MAO activity. In addition, the interaction between various environmental factors and genome might induce gene
MA
expression changes, resulting in MAO activity alterations and consequently in different behavioral phenotypes and various psychiatric disorders. Epigenetic modifications, such as changes in DNA
D
methylation status of the MAOA promoter, have been reported to mediate the normal variability in
TE
brain MAO-A levels (Shumay et al., 2012). In incarcerated offenders with antisocial personality disorder, who show elevated impulsive aggression and criminal behavior, hypermethylation in the MAOA
CE P
promoter region was found compared with the control group (Checknita et al., 2015). This hypermethylation resulted in downregulation of MAOA gene expression, decreased MAO-A production
AC
and activity, as well as increased serum serotonin levels in offenders with antisocial personality disorder, providing insight in the molecular mechanism of aggression (Checknita et al., 2015).
1.1.2.1. MAOB rs1799836 polymorphism The most common MAOB polymorphism is the one found in the 13th intron of the MAOB gene. It is a single nucleotide polymorphism (SNP), an A to G substitution, positioned 36 bp upstream from the intron 13/exon 14 boundary (Ho et al., 1995). It is assumed that, due to its position, MAOB intron 13 A/G polymorphism (rs1799836) affects the stability and/or translation of MAOB mRNA (Balciuniene et al., 2002). An association between G allele of MAOB rs1799836 and increased platelet MAO-B activity was found by Garpenstrand et al. (2000), but our previous studies did not detect any association between this polymorphism and platelet MAO-B activity in large number of healthy subjects (Pivac et al., 2006),
8
ACCEPTED MANUSCRIPT veterans with PTSD (Pivac et al., 2007), or patients with alcohol dependence (Nedic Erjavec et al., 2014). On the other hand, there are studies showing a decreased brain MAO-B activity in G allele carriers, whereas the MAOB mRNA levels are the same regardless of genotype (Balciuniene et al., 2002). Due to the fact that
PT
the mRNA levels remained the same, but MAO-B activity varied depending on the genotype, it has been assumed that there is a cis-regulatory element in linkage disequilibrium with the MAOB rs1799836 which
CR I
alters MAO-B activity in human brain (Balciuniene et al., 2002). This polymorphism can affect the process of MAOB intron 13 removal with an increased efficiency in the case of A allele, which could cause the
NU S
increased protein expression and an enhanced enzymatic activity of MAO-B (Jakubauskiene et al., 2012). In addition, a sex specific association between MAOB rs1799836 and schizophrenia has been detected in
MA
Spanish population, with an increased frequency of G allele in women diagnosed with schizophrenia
D
compared to healthy control women (Gasso et al., 2008).
TE
1.1.2.2. MAOA uVNTR polymorphism
There are several common polymorphisms in the structural and regulatory region of MAOA gene: a
CE P
MAOA (CA)n, a polymorphic dinucleotide repeat polymorphism in intron 2 (Black et al., 1991); a 23 bp variable-number tandem repeat (VNTR) near exon 1 (Hinds et al., 1992); two restriction fragment length
AC
polymorphisms (Fnu4HI and EcoRV) (Lim et al., 1994); a 30 bp VNTR sequence located upstream of transcription initiation site (Sabol et al., 1998); and various SNPs (rs1137070, rs1799835, rs6323, rs1465108, rs979605), (Chester et al., 2015a,b; Peters et al., 2004; Sampaio et al., 2015; Tadic et al., 2003). The most frequently evaluated polymorphism of MAOA is a VNTR in the promoter region. Sabol et al. (1998) first reported a polymorphic region within MAOA gene promotor, located at 1.2 kb upstream of the MAOA coding sequence. This polymorphism is characterized as a 30-bp repeat sequence with 2, 3, 3.5, 4, and 5 copies (Huang et al., 2004) which affects the transcriptional activity of the MAOA gene. Alleles with 3.5 or 4 repeats are transcribed 2-10 times more efficiently compared to the other alleles (Sabol et al., 1998). The possible explanation for the effect of MAOA uVNTR on gene transcription could be that this sequence acts as a binding site for a transcription factor, or it is responsible for altering chromatine structure (Sabol et al., 1998). This polymorphic region could also be involved in the regulation of MAOA
9
ACCEPTED MANUSCRIPT gene expression by DNA methylation mechanism. The most frequent alleles in the population are those with 3 repeats and 4 repeats. The 4 repeat variant has been found to be responsible for the higher (Deckert et al., 1999; Sabol et al., 1998), while the 3 repeat variant is associated with lower MAO-A activity
PT
(Bortolato and Shih, 2011). However, both low activity MAOA variants (i.e. with 3 repeats), but also high activity MAOA variants (i.e. with 4 repeats), were reported to be associated with impulsive aggressiveness,
CR I
antisocial problems and violent delinquency (Antypa et al., 2013; Bortolato and Shih, 2011; Manuck et al.,
NU S
2000; Stetler et al., 2014).
1.2. Schizophrenia
MA
Schizophrenia is a chronic, common disabling psychiatric disorder (APA, 1994) that imposes enormous costs, both on individuals and society at large, affecting approximately 1% of the population. It is a complex
D
diagnostic entity with significant differences in the response to medication, clinical presentation, and
TE
course of the disease (Bray et al., 2010). Various symptoms of schizophrenia usually occur in early adulthood and lead to serious socioeconomic problems. The symptoms of schizophrenia are usually divided
CE P
into positive, negative, cognitive and affective (mood) symptoms. However, its neurobiological basis is poorly understood, which makes the treatment of schizophrenia only partially effective and demands new
AC
treatment strategies (Meyer and Feldon, 2010). Since all effective antipsychotic drugs are dopamine receptor blocking agents (Seeman, 2000), it is postulated that dopaminergic system has an important role in the etiology of schizophrenia. Clinical features of schizophrenia include hallucinations, delusions, thought and speech disorders. Violence, aggression and agitation, as a consequence of delusions, hallucinations and substance abuse, but also as a consequence of the social deterioration, are common in patients with schizophrenia (Soyka, 2011). The findings of the studies that monitored aggressive behavior with the Overt Aggression Scale (OAS) and psychopathology with the Positive and Negative Syndrome Scale (PANSS) in chronic schizophrenia supported the view that positive symptoms may lead to aggression (Nolan et al., 2005). Psychomotor agitation is often associated with aggression, and aggression is frequently associated with violence (Soyka, 2011). The higher risk of committing non-violent and violent crime as well as homicide
10
ACCEPTED MANUSCRIPT have been observed in subjects with schizophrenia (Arseneault et al., 2000; Brennan et al., 2000; Erb et al., 2001; Fazel and Danesh, 2002; Wallace et al., 2004). Hence, it is important to identify aggression and agitation early, and apply therapy quickly, in order to prevent aggressive behavior (Nordstrom and Allen,
PT
2013). The management of aggression in patients with schizophrenia is a complex and challenging clinical
CR I
dilemma (Buckley et al., 2011).
1.3. Conduct disorder
NU S
Abnormal aggressive behavior in children and adolescents is frequent in conduct disorder and oppositional defiant disorder, according to the DSM- IV (APA, 1994). According to the newest 5th edition
MA
of DSM /DSM-5/ (APA, 2013), conduct disorder is characterized by behavior that violates either the rights of others or norms of society, with a cold and unemotional interpersonal style. To be diagnosed with
D
conduct disorder, the symptoms must cause significant impairment in social, academic or occupational
TE
functioning. Individuals with conduct disorder show repetitive and chronic aggressive and antisocial behavior, limited empathy and little concern for the feelings, wishes, and well-being of others, school
CE P
refusal and social communication problems, leading to illicit activities and persistent antisocial, criminal and violent behavior (Frick et al., 2005).
AC
Individuals with conduct disorder often display limited prosocial emotions, lack of empathy and guilt, low sensitivity for punishment and reward, delinquent behavior and they frequently develop substance abuse (Matthys et al., 2013). These subjects are characterized by rule breaking, aggression, destruction, lying, stealing, assaulting and cruelty to animals and people (Spencer et al., 2007). The characteristic symptoms of conduct disorder must be manifested by the presence of three (or more) of the following criteria in the past 12 months, with at least one criterion present in the past six months: aggression to people and animals, destruction of property, deceitfulness or theft, and serious violations of rules. The disorder is typically diagnosed prior to adulthood and might be subdivided into early (childhood) onset and adolescent onset subtypes (Frick and Nigg, 2012). The age at which symptoms of conduct disorder emerge seems to be critical for designating subgroups that differ in severity, course, and etiology (Moffitt et al.,
11
ACCEPTED MANUSCRIPT 2008). In addition, conduct disorder occurs two to three times more likely in boys than in girls (Moffitt and Caspi, 2001).
PT
1.4. The association of schizophrenia and conduct disorder Several studies suggested that children with antisocial behavior and conduct disorder have an
CR I
increased risk for schizophrenia (Robins et al., 1991; Robins, 1993; Robins and Price,1991), and that up to 40% of subjects who developed schizophreniform disorders as adults, also met the criteria for conduct
NU S
disorder in the childhood (Kim-Cohen et al., 2003). Among subjects with schizophrenia, the prevalence of conduct disorder prior to age 15 is 20-45% (Hodgins, 2008; Hodgins et al., 2008). Moreover, prior to the
MA
onset of schizophrenia, a sub-group of patients displayed behavioral problems in childhood and adolescence (Asnarow, 1988; Olin et al., 1997). Among men who develop schizophrenia, conduct problems
D
in childhood and early adolescence are suggested as antecedents of physical aggression towards others
TE
throughout the adulthood (Hodgins et al., 2005). The subjects with both schizophrenia and conduct disorder are characterized by a lifelong pattern of antisocial and aggressive behavior, substance abuse and
CE P
criminality and are suggested to display a distinct subtype of schizophrenia (Schiffer et al., 2013). Accumulating evidence supported the hypothesis that conduct disorder is not a co-morbid disorder,
AC
but rather an early feature of the developing schizophrenia. In line with this hypothesis, Cannon et al. (1990) reported that a pattern of disruptive behavior (overactive, irritable, distractible and aggressive behavior) in the late childhood and early-adolescence was associated with the development of schizophrenia with predominant positive symptoms. Therefore, it has been proposed that the association of conduct disorder with aggressive behavior, following onset of schizophrenia, is due to its impact on factors such as positive symptoms, noncompliance with medication and lack of care and substance misuse (Swartz et al., 1998). On the other hand, some other findings supported the view that among persons with schizophrenia, conduct disorder is a distinct co-morbid disorder. The results of the study by Hodgins et al. (2005) suggested that conduct disorder is directly associated with criminal offending and substance abuse that begins in adolescence and persists across the life-span and that the course of antisocial behavior runs
12
ACCEPTED MANUSCRIPT parallel to schizophrenia. The association between conduct disorder and schizophrenia might be due to the fact that children with genetic predisposition for schizophrenia are more vulnerable than others to etiological factors such as stress or physical abuse, which in interaction with highly prevalent gene variants
PT
(such as polymorphisms in gene encoding MAO-A) result in increased risk for antisocial behavior (Caspi et al., 2002; Foley et al., 2004). Conversely, elevated rates of conduct disorder among subjects with
CR I
schizophrenia could be in part due to higher rates of assortative mating (Parnas, 1988), so that children
NU S
inherit susceptibility genes for both schizophrenia and antisocial behavior (Rhee and Waldman, 2002).
1.5. Agitation
MA
Agitation is represented by a cluster of pathological behavior such as motor restlessness, mental tension, elevated response to various external or internal stimuli, and various symptoms: pacing, hand wringing, fist
D
clenching, pressured speech, yelling and threatening other people, irritability, inappropriate and/or
TE
purposeless verbal or motor activity, decreased sleep and fluctuation of symptoms over time (Lesem et al., 2011; Lindenmayer, 2000). In addition, agitation frequently escalates into violence and is characterized also
CE P
by excessive or inappropriate motor or verbal activity and hostility, which may include irritability, uncooperativeness, threatening gestures, assault and poorly organized and aimless psychomotor activity
AC
(Lesem et al., 2011; Mohr et al., 2005; Montoya et al., 2011; Volavka, 2014). Agitation is frequent in psychotic disorders, especially in acute episodes. It may be presented as verbal or physical aggression, and might be focused to different objects or persons. Verbal aggression, including threats, abuse or incoherent screams, is frequent part of agitation. Since agitation might accompany some mental (schizophrenia, bipolar disorder, post-traumatic stress disorder (PTSD), antisocial personality disorder, conduct disorder, attention deficit hyperactivity disorder (ADHD), alcohol/substance abuse) and neurological (Alzheimer’s disease, autoimmune encephalitis and other dementias) disorders, underlying neurobiology might be associated both to a particular disorder and to agitation (Rosell and Siever, 2015). Moreover, agitation is presumed to be one of the risk factors or indicators for suicide attempt (Bryan et al., 2014). The neurobiological basis of agitation is presumed to involve the disturbances in serotonin, dopamine,
13
ACCEPTED MANUSCRIPT norepinephrine, GABA and glutamate (Lindenmayer, 2000) neurotransmission, but this relationship is far from clear. In addition, agitation is demonstrated as gender specific (Bryan et al., 2014). In psychiatric patients, especially those with schizophrenia, agitation is frequently assessed using the
PT
excited component of PANSS-EC (Citrome et al., 2011; Montoya et al., 2011; Swanson et al., 2004; Volavka, 2014). This PANSS-EC evaluates the scores in particular symptoms: excitement, hostility, tension,
CR I
uncooperativeness and poor impulse control, which can be rated as absent, minimal, mild, moderate, moderate/severe, and severe to extreme (i.e. with scores 1-7). For assessing agitated patients, PANSS-EC is
NU S
easy to use and is one of the simplest and most intuitive scales (Baker et al., 2003; Citrome et al., 2011; Lesem et al., 2011; Lindenmayer et al., 2008; Montoya et al., 2011; Volavka, 2014). Individuals might be
MA
subdivided into those with presence of agitation or severe agitation. Agitation is defined if patients have a total score of ≥ 14 out of 35 on these particular PANSS-EC items (Lesem et al., 2011). On the other hand,
D
severe agitation is present if patients are rated with ≥ 4 (moderate) scores on the each item of the PANSS-
TE
EC and therefore patients are defined as severely agitated if they have a total score of ≥ 20 out of 35 (Baker et al., 2003; Montoya et al., 2011).
CE P
As PANSS is not used in adolescents, agitation in adolescents with conduct disorder is assessed using the particular items of the Psychopathy Checklist Youth Version (PCL:YV) and Overt Aggression Scale-Modified
AC
(OAS-M) (Coccarro et al., 1991; Forth et al., 2003; Rush, 2000). PCL:YV is designed for assessments of psychopathic features in youths separated in 4 factors (arrogance and deceitfulness - factor 1, affective unresponsiveness - factor 2, impulsive and irresponsible life style - factor 3, and criminal behavior - factor 4). OAS-M (Coccarro et al., 1991) is designed to assess various manifestation of aggressive behavior in outpatients. The scale incorporates three overall domains: aggression, irritability and suicidality. The irritability and suicidality items in the OAS-M instrument were adapted from the Schedule for Affective Disorders and Schizophrenia (SADS), suggesting that PANSS-EC and OAS-M could be taken in coincidence. The OAS-M is the only aggression scale that actually assesses concrete aggressive behavior. The use of weighted items in Aggression scale scores takes into account that high severity items are both less frequent and more significant that low severity items. The cut-off scores used in the present study were determined according to clinical experience and data from the earlier published study (Podobnik et al., 2012).
14
ACCEPTED MANUSCRIPT 1.6. Aim of the study Our primary hypothesis was that platelet MAO-B activity and the most frequently studied genetic
PT
polymorphisms in the MAOA and MAOB genes (MAOA uVNTR and MAOB rs1799836) will be significantly related to severe agitation in male patients with schizophrenia and male adolescents with conduct
CR I
disorder. More specifically, we have expected to detect reduced platelet MAO-B activity in agitated subjects and significant association between specific genetic variants of MAOA uVNTR and MAOB
NU S
rs1799836 polymorphisms and severe agitation. The main aim of our study was to define if MAO alternations, studied at functional and genetic levels, contribute to the development of severe agitation.
MA
This approach was in line with the initiative of the National Institute of Mental Health to evaluate Research Domain Criteria (RDoC) and to assess biomarkers, rather than diagnosis in psychiatry (Insel et
D
al., 2010). Our research was focused on the part of RDoC criteria (i.e. agitation) and genetic and
TE
molecular/cellular factors which might influence clinically relevant functions (Insel et al., 2010). The specific aims of our study were to compare platelet MAO-B activity, the MAOB rs1799836 A and G allele
CE P
frequency as well as the frequency of the low (3-repeat; 3R) and high (4-repeat; 4R) activity variants of
AC
the MAOA uVNTR polymorphism in severely agitated and non-agitated male subjects.
2. Material and methods 2.1. Subjects
This study included 363 male, drug naive subjects, who were subdivided into those with or without severe agitation. There were 187 patients with schizophrenia and 176 adolescents with conduct disorder, diagnosed by consensus of two psychiatrists and according to SCID based or DSM-IV criteria (APA, 1994). Smoking status was determined according to questionnaires and subjects were classified as smokers (smoking ≥10 cigarettes per day) and non-smokers (never-smokers). The “light” smokers, smoking less than 10 cigarettes per day, were excluded. All subjects were subdivided according to the presence or absence of severe agitation, determined according to the a priori chosen cut-offs in the corresponding clinical scores. Out of 187 patients with schizophrenia, there were 26 severely agitated and 161 non-agitated patients, and
15
ACCEPTED MANUSCRIPT 138 smokers and 49 non-smokers. Patients with schizophrenia were evaluated at the Clinics for Psychiatry Vrapce, Zagreb and Psychiatric Hospital Dr Ivan Brbot, Popovaca, Croatia. Adolescents with conduct disorder (N=176) were sampled at the correctional facilities in Zagreb County, Croatia. They were
PT
subdivided into 68 severely agitated and 108 non-agitated subjects, while according to the smoking status they were divided into 144 smokers and 32 non-smokers. Exclusion criteria for all subjects were:
CR I
diagnoses other than schizophrenia or conduct disorder, neurodegenerative disorders, drug and alcohol dependence, any kind of treatment before the study entry, current medication, cognitive behavioral
NU S
therapy or electroconvulsive therapy and smoking between 1-10 cigarettes per day.
MA
2.2. Clinical assessment
Agitation was assessed using the PANSS-EC, the PCL:YV and the OAS-M. For assessing severe agitation,
D
patients with schizophrenia were rated according to the PANSS-EC. This subscale includes particular items:
TE
excitement (P4), hostility (P7), tension (G4), uncooperativeness or lack of cooperation (G8), and poor impulse control (G14). All items are rated on a 7-point scale (1=absent, 2=minimal, 3=mild, 4=moderate,
CE P
5=moderate/severe, 6=severe, 7=extreme). Severe agitation was rated as present when a person had a mean PANSS-EC score of 20 or more out of possible 35 (Baker et al., 2003; Montoya et al., 2011).
AC
In adolescents with conduct disorder, agitation and other forms of aggression are not usually assessed and rated by the PANSS, and these subjects were assessed and rated by the more appropriate OAS-M (aggression restricted to past week) and PCL:YV scales. In the present study adolescents were rated as severely agitated if they had: 1) the subjective irritability in the OAS-M with score ≥ 3 out of 5, corresponding to the tension described in the PANSS-EC; 2) the global irritability in the OAS-M with score ≥ 6 out of 10, corresponding to the poor impulse control described in the PANSS-EC; 3) the finale scale scores for aggression in the OAS-M with ≥ 30 out of 135 scores (including verbal aggression, aggression against property, auto-aggression and physical aggression), corresponding to the excitement described in the PANSS-EC; 4) the total aggression score in the OAS-M with ≥ 40 out of 161 (comprising aggression, irritability and suicidality), corresponding to hostility described in the PANSS-EC; and 5) the total PCL:YV score ≥ 20 out of 40, corresponding to the lack of cooperation described in the PANSS-EC. Therefore,
16
ACCEPTED MANUSCRIPT adolescents with conduct disorder were subdivided into currently severely agitated (indication for hospitalisation or medication) and presently (in time of rating) non-agitated according to PCL:YV and the
PT
OAS-M scores (Forth et al., 2003; Podobnik et al., 2012; Rush, 2000).
2.3. Ethics statement
CR I
All samples were taken after the participants voluntarily signed informed consent. The interview and blood sampling of all subjects were done either at the hospital or in the correctional facilities when they
NU S
were able to understand and agree with the protocol of the study. All studies were conducted with the approval of corresponding Ethics Committees, and were fully compliant with the ethical standards laid
MA
down in the 1975 Declaration of Helsinki.
D
2.4. Sampling and measuring MAO-B activity
TE
Whole blood samples were collected in 8.5 ml yellow-top Vacutainer tubes with 1.5 ml of acid citrate dextrose anticoagulant. Sampling was done in the morning around 8 a.m., after the overnight fasting and
CE P
was part of the routine laboratory visits. Platelets were obtained by series of centrifugation of whole blood and platelet-rich-plasma and then broken by sonication. Platelet MAO-B activity (expressed in nmol 4-
AC
OHQ/mg protein/h) was determined in 363 subjects (patients with schizophrenia and adolescents with conduct disorder) according to the slightly modified method described by Krajl (1965) and using kynuramine as a substrate. Briefly, 100 l of standard (4-hydoxyguinoline, 4-HOQ) and platelet sonicates (both in duplicates) were incubated with 100 l kynuramine (final concentration 73.6 M) at 37°C for 1 hour. The reaction was stopped with 1 N NaOH. Platelet MAO-B activity was measured by fluorimetry of 4-hydroxyquinoline (4-HQ), which is converted by oxidative deamination from kynuramine. The measurement of florescence of 4-HOQ, a product of kynuramine, was performed on Varian Spectrophotofluorimeter Cary Eclipse, with an exciting wavelength of 310 nm and emitted wavelength of 380 nm. Platelet protein levels were measured by the method of Lowry et al. (1951). Platelet MAO-B activity was calculated using the standard curve of 4-HQ and was expressed in nmol of 4-HQ formed per hour per mg of total platelet protein.
17
ACCEPTED MANUSCRIPT 2.5. Genotyping of MAOB rs1799836 polymorphism Genomic DNA was extracted from peripheral blood using the salting out method (Miller et al., 1988).
PT
MAOB intron 13 A/G polymorphisms (rs1799836) was determined in 363 subjects by ABI Prism 7300 Real time PCR System apparatus (Applied Biosystems, Foster city, California, USA), according to the procedures
CR I
described by Applied Biosystems. The primers and probes were purchased from Applied Biosystems as TaqMan® SNP Genotyping Assay (C_8878790_10). All genotyping procedures were done by a researcher
NU S
who was blind to all clinical data. Out of 363, 80 samples (22%) were genotyped again as a quality control
MA
for genotyping analyses.
2.6. Genotyping of MAOA uVNTR polymorphism
D
The MAOA promoter uVNTR polymorphism was genotyped in 354 subjects using PCR-based amplification
TE
with primers MAOA-F: 5’-ACAGCCTCGCCGTGGAGAAG-3’ and MAOA-R’: 5’-GAACGGACGCTCATTCGGA-3’. The final PCR reaction volume was 15 l, which contained ~100 ng of DNA, 1.7 mM MgCl2, 1 x Q solution,
CE P
20 pmol of each primer, 0.25 mM dNTP and 1 U of Taq polymerase (Qiagen). The PCR was performed on 2720 Thermal Cycler (Applied Biosystems) with the cycling conditions: 5 min at 95°C, 35 cycles carried out
AC
at 30 sec at 94°C, 30 sec at 57°C and 1 min at 72°C, and final extension 7 min at 72°C. PCR products were separated on 2.5% agarose gels which were stained with ethidium bromide and visualized using Uvitec BXT20-M transiluminator. All participants were found to harbor alleles carrying 3 or 4 repeats, with the exception of two carriers of the 5-repeat variant. The 3-repeat variants, together with the two 5-repeats variants were defined as low activity variants, while variants with 4-repeat variants were considered as high activity variants. Genotyping for the MAOA uVNTR failed for 9 participants. Out of 363, 40 samples (11%) were genotyped again as a quality control for genotyping analyses.
2.7. Statistical evaluation The results were expressed as numbers and percentages, or median and 25th (Q1) and 75th (Q3) percentiles and evaluated with Sigma Stat 3.5 (Jandel Scientific Corp., San Jose, California, USA). The allele
18
ACCEPTED MANUSCRIPT frequencies were compared using χ2-tests and Yates correction for continuity. In order to verify what category was a major contributor to rejecting the null hypothesis, standardized residual (R) (Field et al., 2012) was calculated. As the study included only male subjects, the deviation from Hardy-Weinberg
PT
equilibrium for both MAOB rs1799836 and MAOA uVNTR could not be tested since the genes encoding MAO-A and MAO-B are located on the short arm of the X chromosome. Normality of distribution was
CR I
assessed with the Kolmogorov-Smirnov test. A multiple linear regression analysis was used to check for the influence of diagnosis, age, smoking status, MAOB rs1799836 and MAOA uVNTR on MAO-B activity.
NU S
Comparison of platelet MAO-B activity and age between different groups of subject was done using Kruskal Wallis ANOVA by ranks or Mann-Whitney test, since normality of the data failed.
MA
To avoid issues with multiple testing (Leucht et al., 2007) and correction for the multiple testing, all statistical tests were applied to cutoffs chosen a priori, and agitation was defined a priori. All tests were
D
two-tailed and α was set at 0.05. G*Power 3 Software (Faul et al., 2007) was used for conducting power
TE
analyses, i.e. to determine a priori sample size and actual power. For multiple regression analysis (with α = 0.05; power (1 − β) = 0.800; a small effect size (ω = 0.30); number of predictors=4), total desired sample
CE P
size was 46, and the actual sample size was 363. For t-test (with α = 0.05; power (1 − β) = 0.800; a small effect size (ω = 0.30)), total desired sample size was 352, and the actual sample size was 363. For ANOVA
AC
(with α = 0.05; power (1 − β) = 0.800; a small effect size (ω = 0.30) and 4 groups), total desired sample size was 128, and the actual sample size was 363. For genetic analyses and χ2-test (with α = 0.05; power (1 − β) = 0.800 and a small effect size (ω = 0.30); d.f. = 1), total desired sample size was 88; and if d.f. = 2; the total desired sample size was 108, while the actual total sample size was 363 for MAOB rs1799836 and 354 for MAOA uVNTR. Therefore the sample size and the statistical power were appropriate to detect significant differences in the studied groups.
3. Results 3.1. Age of participants As expected, age of participants, presented as medians and 95% CI for the median, differed significantly (U=15.5; P=0.0001; Mann Whitney’s test), since patients with schizophrenia were significantly older (42;
19
ACCEPTED MANUSCRIPT 41-46 years) than adolescents with conduct disorder (17; 16-17 years). Severely agitated subjects were significantly (p
S0278-5846(16)30016-1 doi: 10.1016/j.pnpbp.2016.02.002 PNP 8884
To appear in:
Progress in Neuropsychopharmacology & Biological Psychiatry
Received date: Revised date: Accepted date:
30 November 2015 15 January 2016 2 February 2016
Please cite this article as: Perkovic Matea Nikolac, Strac Dubravka Svob, Erjavec Gordana Nedic, Uzun Suzana, Podobnik Josip, Kozumplik Oliver, Vlatkovic Suzana, Pivac Nela, Monoamine oxidase and agitation in psychiatric patients, Progress in Neuropsychopharmacology & Biological Psychiatry (2016), doi: 10.1016/j.pnpbp.2016.02.002
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Monoamine oxidase and agitation in psychiatric patients
Matea Nikolac Perkovica*, Dubravka Svob Straca*, Gordana Nedic Erjaveca*, Suzana Uzunb, Josip Podobnikc,
CR I
a
PT
Oliver Kozumplikb, Suzana Vlatkovicd, Nela Pivaca
Rudjer Boskovic Institute, Division of Molecular Medicine, Bijenicka cesta 54, 10000 Zagreb, Croatia
b
Clinic for Psychiatry Vrapce, Bolnicka cesta 32, 10000 Zagreb, Croatia
Department of Psychiatry, Psychiatric Hospital for Children and Youth Zagreb, Kukuljeviceva 11, 10000
NU S
c
Zagreb, Croatia d
D
MA
Psychiatric Hospital Dr Ivan Brbot, Popovaca, Croatia
AC
CE P
TE
*Matea Nikolac Perkovic, Dubravka Svob Strac and Gordana Nedic Erjavec equally contributed to this work.
Correspondence to: Prof. Nela Pivac, Ph.D., Senior Scientist Division of Molecular Medicine Rudjer Boskovic Institute Bijenicka cesta 54, 10 000 Zagreb, Croatia Phone: + 385 1 457 1207; Fax: + 385 1 456 1010 E-mail: [email protected]
1
ACCEPTED MANUSCRIPT Abstract Subjects with schizophrenia or conduct disorder display a lifelong pattern of antisocial, aggressive and violent behavior and agitation. Monoamine oxidase (MAO) is an enzyme involved in degradation of
PT
various monoamine neurotransmitters and neuromodulators and therefore has a role in various psychiatric and neurodegenerative disorders and pathological behaviors. Platelet MAO-B activity has
CR I
been associated with psychopathy- and aggression-related personality traits, while variants of the MAOA and MAOB genes have been associated with diverse clinical phenotypes, including aggressiveness,
NU S
antisocial problems and violent delinquency. The aim of the study was to evaluate the association of platelet MAO-B activity, MAOB rs1799836 polymorphism and MAOA uVNTR polymorphism with severe
MA
agitation in 363 subjects with schizophrenia and conduct disorder. The results demonstrated significant association of severe agitation and smoking, but not diagnosis or age, with platelet MAO-B activity. Higher
D
platelet MAO-B activity was found in subjects with severe agitation compared to non-agitated subjects.
TE
Platelet MAO-B activity was not associated with MAOB rs1799836 polymorphism. These results suggested the association between increased platelet MAO-B activity and severe agitation. No significant
CE P
association was found between severe agitation and MAOA uVNTR or MAOB rs1799836 polymorphism, revealing that these individual polymorphisms in MAO genes are not related to severe agitation in subjects
AC
with schizophrenia and conduct disorder. As our study included 363 homogenous Caucasian male subjects, our data showing this negative genetic association will be a useful addition to future metaanalyses.
Keywords: agitation; platelet MAO-B activity; MAOB rs1799836 polymorphism; MAOA uVNTR polymorphism; schizophrenia; conduct disorder
2
ACCEPTED MANUSCRIPT 1. Introduction Different psychiatric disorders, including schizophrenia and conduct disorder, have debilitating effects
PT
on patients and their families, causing them great suffering and carrying enormous mental health and societal burden as well as financial costs to the whole society. These complex and multifactorial mental
CR I
disorders are rarely isolated, and commonly co-occur with other comorbid psychiatric disorders that complicate their clinical features as well as treatment. Patients with these disorders are frequently non-
NU S
responsive or only modestly responsive to current treatment options. Therefore, identification of the risk markers and biomarkers specific for these disorders may improve our understanding of the
MA
etiology/pathophysiology of these diseases, and provide novel drug targets. However, the putative biomarkers for various complex psychiatric diseases, such as schizophrenia and conduct disorder, are still
D
missing (Chan et al., 2014). Although the understanding of biological pathways involved in the development
TE
of these disorders is improving, novel strategies focused to find reliable biomarkers and to improve treatment response are still rare. According to the Biomarkers Definitions Working Group, as part of the
CE P
NIH Director's Initiative on Biomarkers and Surrogate Endpoints, biomarker is defined as “a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic
AC
processes, or pharmacologic responses to a therapeutic intervention or other health care intervention“. Molecular biomarkers (Schmidt et al., 2013) can be assessed on different molecular levels, namely, on the genetic level (DNA biomarkers), the gene expression level (RNA biomarkers), the level of proteins (peptide and protein biomarkers), and epigenetic level, which regulates the gene expression by several mechanisms, namely, DNA methylation, histone modifications, and RNA interference (epigenetic biomarkers). Blood cells represent a good, easily available and non-invasive biological material for biomarker screening (Mohr and Liew, 2007), as they share more than 80% of the transcriptome with other tissues (brain, colon, heart, kidney, liver, lung, prostate, spleen and stomach). Deficits in human brain function can be assessed either “post-mortem” in the specific brain regions, or by the use of neuroimaging and nuclear magnetic resonance (NMR) spectroscopy. Other common approaches include genetics/genomics, transcriptomics, proteomics (Chan et al., 2014) and epigenetics (Shumay et al., 2012). At the metabolite
3
ACCEPTED MANUSCRIPT level, evidence is provided (Chan et al., 2014) that molecular profiling analysis of blood serum/plasma can reveal robust molecular changes in psychiatric patients, suggesting that these changes are detectable besides in the brain, also in other body systems such as the circulating blood. Although peripheral
PT
biomarkers might offer easily accessible tests which could make a significant improvement in diagnosis, treatment, and patient care (Bahn and Chan, 2015), validated and reliable biomarkers of different
CR I
psychiatric disorders are still not identified (Zolad and Diamond, 2013).
The studies investigating the molecular profiles of body fluids (i.e. saliva, plasma or serum) and blood
NU S
cells are promising areas of research aiming to identify peripheral biomarkers for psychiatric disorders (Bahn and Chan, 2015). There is an unmet need to find reliable and validated biomarkers of
MA
neuropsychiatric disorders, especially from the accessible peripheral fluids (e.g. blood), in order to facilitate early diagnoses, detect particular personality traits, symptoms and behaviors, prevent psychopathological
D
behaviors and to advance their treatment, but also to improve prognosis and disease outcome (Chan et al.,
TE
2014). In the research of the biological underpinnings and reliable biomarkers of the specific endophenotypes, pathological behaviors, and different personality traits in neuropsychiatric disorders,
CE P
human platelets have been used as a non-invasive model (Asor and Ben-Shachar, 2012). Platelets and neurons share some similar processes and contain some identical components of the serotonergic
AC
system (Camacho and Dimsdale, 2000; Yubero-Lahoz et al., 2013). In addition, similar pathologies are found in platelets and neurons in many neuropsychiatric diseases (Asor and Ben-Shachar, 2012). Therefore, platelets have been used as an easily obtainable model for neurobiological research in humans. Due to the complexity of the neurobiological changes in neuropsychiatric disorders, not one biomarker, but a combination of multiple biomarkers is expected to be associated with the alterations observed in specific pathological behaviors in neuropsychiatry. The quest for biomarkers is especially important in forensic psychiatry for the prediction of violent and assaultive behaviors, since these markers could improve the accuracy of risk assessments (Gustavson et al., 2010).
1.1. Monoamine oxidase (MAO)
4
ACCEPTED MANUSCRIPT Monoamine oxidase (MAO; E.C. 1.4.3.4) is a flavin-adenosine-dinucleotide (FAD)-containing enzyme which is involved in degradation of various biogenic amines. MAO exists in two forms, MAO-A and MAO-B, responsible for the oxidative deamination of different monoamine neurotransmitters and
PT
neuromodulators (Bortolato and Shih, 2011; Oreland, 2004; Shih et al., 1999). Both MAO subtypes degrade serotonin, melatonin, dopamine, norepinephrine and epinephrine, and different amines
CR I
(tyramine, tryptamine, 2-phenylethylamine, octopamine and 3-iodothyronamine). MAO-A and MAO-B differ in their substrate preferences, immunological properties, molecular weight and anatomical locations,
NU S
and they are inhibited by different inhibitors. MAO-A deaminates primarily serotonin, norepinephrine and epinephrine, while MAO-B degrades β-phenylethylamine and benzylamine. Both isoenzymes degrade
MA
dopamine, tryptamine and tyramine. However, when one subtype is absent, the other isoenzyme can take its role (Bortolato and Shih, 2011).
D
Both forms of MAO are localized on the outer membrane of the mitochondria and widely distributed
TE
throughout the body. Within the central nervous system MAO-B is expressed at highest levels in serotonergic neurons, histaminergic neurons and in glial cells, and MAO-A is localized primarily in
CE P
catecholaminergic (dopaminergic and noradrenergic) neurons (Luque et al., 1995; Saura et al., 1994; Westlund et al., 1988). In the peripheral tissues MAO-A is localized in fibroblasts and placenta (Egashira and
AC
Yamanaka, 1981), liver and gastro-intestinal tract, while MAO-B is found in liver (Billett et al., 2004), blood platelets and lymphocytes (Bond and Cundall, 1977; Bortolato and Shih 2011). MAO inhibitors are used as antidepressants, but due to their multiple drug- and food-interactions and side effects, they are prescribed less frequently (Fiedorowicz and Swartz, 2014). However, these drugs still have a place in the treatment of atypical, treatment-resistant depression or bipolar depression (Shulman et al., 2013). They include phenelzine, tranylcypromine and isocarboxazid, as well as newer, selective and reversible drugs, such as the MAO-A inhibitor, moclobemide, and MAO-B inhibitor, selegiline.
1.1.1. Platelet MAO-B activity Platelet MAO-B activity has been proposed to represent a biomarker of different personality traits and psychiatric vulnerability, including sensation-seeking behavior, novelty-seeking personality, extraversion,
5
ACCEPTED MANUSCRIPT poor impulse control and predisposition for taking different risky behaviors (Harro et al., 2004; Oreland, 2004; Oreland and Hallman, 1995). Platelet MAO-B activity, assumed to be a biological marker of psychopathology, was found to be associated with psychopathy- and aggression-related personality
PT
traits (Stalenheim, 2004). It is presumed that MAO-B function in platelets is heritable and its heritability has been estimated around 0.77 (Pedersen et al., 1993). Using the transmission probability model, the
CR I
familial transmission of MAO activity was consistent with either recessive or additive, but not with dominant inheritance (Baron et al., 1985). Platelet MAO-B has similar biochemical and pharmacological
NU S
characteristics as the brain MAO-B (Oreland, 2004), and both brain and platelet MAO-B are encoded by the MAOB gene (Chen et al., 1992). Various MAOB gene variants might influence the protein expression
MA
and the activity of platelet MAO-B (Bortolato and Shih, 2011). In addition, platelet MAO-B activity was shown to be significantly affected by sex, age, and smoking (Oreland, 2004), as well as with other
D
environmental factors, implicating possible epigenetic mechanisms (Shumay et al., 2012).
TE
Many studies have reported reduced platelet MAO-B activity in subjects with disinhibited behaviors such as novelty-seeking behavior, low levels of harm avoidance, high impulsiveness and high levels of sensation
CE P
seeking (Oreland, 2004; Oreland et al., 1999; Ruchkin et al., 2005; Shih et al., 1999). It has been hypothesized that low MAO-B activity may represent a nonspecific marker indicating a predisposition to
AC
distress or psychopathology in general, rather than to specific behavior or specific disorders (Ruchkin et al., 2005). However, alterations in both directions (i.e. increased as well as decreased platelet MAO-B activity) have been proposed to represent biological vulnerability to different behaviors and traits (Paaver et al., 2006), as there is no simple or linear relationship between pathological behaviors, traits and monoaminergic functioning (Paaver et al., 2006; Schalling et al., 1987). Various results suggest that mutations and polymorphisms in MAOA and MAOB genes might differentially affect the turnover rates of dopamine and serotonin. Higher plasma and urine levels of monoamines processed by MAOs (norepinephrine, epinephrine and serotonin), along with their metabolites, such as normetanephrine, have been associated with CT genotype of the MAOA rs1137070 polymorphism (Dorszewska et al., 2013). Moreover, a high activity MAOA uVNTR variant has been also associated with higher levels of monoamine metabolites, homovanillic acid (HVA) and 5-
6
ACCEPTED MANUSCRIPT hydroxyindoleacetic acid (5-HIAA) in lumbar cerebrospinal fluid (CSF) (Jonsson et al., 2000; Williams et al., 2003; Zalsman et al., 2005). It has been found that patients with “Brunner syndrome”, resulting from the mutation in the MAOA gene, have low concentrations of HVA, 5-HIAA and vanillylmandelic acid,
PT
confirming reduced monoamine degradation (Brunner et al., 1993), due to MAO-A deficiency. In addition, the increased MAO-B protein expression, and consequently the increased platelet MAO-B
CR I
activity was shown to be associated with MAOB rs1799836 A allele (Jakubauskiene et al., 2012). However, platelet MAO-B activity was not associated with CSF concentration of different monoamine
NU S
metabolites (HVA, 5-HIAA and 3-methoxy-4-hydroxyphenylglycol) in the study of Gustavson et al. (2010), but it was positively correlated to CSF levels of HVA and 5-HIAA in other studies (Knorring et al., 1986;
D
1.1.2. MAO genetics and epigenetics
MA
Oreland et al., 1981).
TE
Genes coding for MAO-A and MAO-B are located on the X-chromosome (Xp11.23–11.4) and they are both comprised of 15 exons with an identical exon–intron organization (Chen et al., 1992). These genes are
CE P
organized in opposite direction, tail to tail with 24 kb apart. The clinical consequences of the genetic deficits in MAO genes are numerous. After the detection of the recessive X-linked disorder, known as a
AC
“Brunner syndrome”, in male subjects of one Dutch family, characterized by the selective loss of the MAOA activity and diverse pathological symptoms and behaviours such as mild-borderline mental retardation, aggressive and violent behavior, impaired impulse control and disturbed regulation, arson, attempted rape, and exhibitionism (Brunner et al., 1993), there was a quest for other affected subjects. This genotype/phenotype correlation was questioned (Hebebrand and Klug, 1995), as well as confirmed in a few studies including a small number of subjects (Palmer et al., 2015). Affected subjects had disturbances in urine monoamine levels. Other genetic mutations in MAOA and MAOB genes have been associated with diverse clinical phenotypes, as confirmed in animal models as well as in human studies (Bortolato and Shih, 2011). The nonsynonymous variations in the coding sequences of MAO genes might alter the amino acid sequence of the MAO protein or result in a truncated, incomplete or non-functional MAO protein.
7
ACCEPTED MANUSCRIPT However, MAO activity might also be affected by some regulatory polymorphisms located in noncoding regions, including promoter/upstream, downstream and intron regions that may affect transcription factor binding, RNA splicing, RNA stability and degradation, the sequence of non-coding RNA, and
PT
therefore impact both transcription and translation. Such polymorphisms are for instance the MAOA uVNTR polymorphism located in the promoter region, as well as rs1799836 polymorphism found in the
CR I
13th intron of the MAOB gene. Moreover, individual polymorphisms in MAO genes may have minimal functional impact but may be in linkage disequilibrium with a set of polymorphisms that form a
NU S
haplotype influencing MAO gene expression or function, consequently altering MAO activity. In addition, the interaction between various environmental factors and genome might induce gene
MA
expression changes, resulting in MAO activity alterations and consequently in different behavioral phenotypes and various psychiatric disorders. Epigenetic modifications, such as changes in DNA
D
methylation status of the MAOA promoter, have been reported to mediate the normal variability in
TE
brain MAO-A levels (Shumay et al., 2012). In incarcerated offenders with antisocial personality disorder, who show elevated impulsive aggression and criminal behavior, hypermethylation in the MAOA
CE P
promoter region was found compared with the control group (Checknita et al., 2015). This hypermethylation resulted in downregulation of MAOA gene expression, decreased MAO-A production
AC
and activity, as well as increased serum serotonin levels in offenders with antisocial personality disorder, providing insight in the molecular mechanism of aggression (Checknita et al., 2015).
1.1.2.1. MAOB rs1799836 polymorphism The most common MAOB polymorphism is the one found in the 13th intron of the MAOB gene. It is a single nucleotide polymorphism (SNP), an A to G substitution, positioned 36 bp upstream from the intron 13/exon 14 boundary (Ho et al., 1995). It is assumed that, due to its position, MAOB intron 13 A/G polymorphism (rs1799836) affects the stability and/or translation of MAOB mRNA (Balciuniene et al., 2002). An association between G allele of MAOB rs1799836 and increased platelet MAO-B activity was found by Garpenstrand et al. (2000), but our previous studies did not detect any association between this polymorphism and platelet MAO-B activity in large number of healthy subjects (Pivac et al., 2006),
8
ACCEPTED MANUSCRIPT veterans with PTSD (Pivac et al., 2007), or patients with alcohol dependence (Nedic Erjavec et al., 2014). On the other hand, there are studies showing a decreased brain MAO-B activity in G allele carriers, whereas the MAOB mRNA levels are the same regardless of genotype (Balciuniene et al., 2002). Due to the fact that
PT
the mRNA levels remained the same, but MAO-B activity varied depending on the genotype, it has been assumed that there is a cis-regulatory element in linkage disequilibrium with the MAOB rs1799836 which
CR I
alters MAO-B activity in human brain (Balciuniene et al., 2002). This polymorphism can affect the process of MAOB intron 13 removal with an increased efficiency in the case of A allele, which could cause the
NU S
increased protein expression and an enhanced enzymatic activity of MAO-B (Jakubauskiene et al., 2012). In addition, a sex specific association between MAOB rs1799836 and schizophrenia has been detected in
MA
Spanish population, with an increased frequency of G allele in women diagnosed with schizophrenia
D
compared to healthy control women (Gasso et al., 2008).
TE
1.1.2.2. MAOA uVNTR polymorphism
There are several common polymorphisms in the structural and regulatory region of MAOA gene: a
CE P
MAOA (CA)n, a polymorphic dinucleotide repeat polymorphism in intron 2 (Black et al., 1991); a 23 bp variable-number tandem repeat (VNTR) near exon 1 (Hinds et al., 1992); two restriction fragment length
AC
polymorphisms (Fnu4HI and EcoRV) (Lim et al., 1994); a 30 bp VNTR sequence located upstream of transcription initiation site (Sabol et al., 1998); and various SNPs (rs1137070, rs1799835, rs6323, rs1465108, rs979605), (Chester et al., 2015a,b; Peters et al., 2004; Sampaio et al., 2015; Tadic et al., 2003). The most frequently evaluated polymorphism of MAOA is a VNTR in the promoter region. Sabol et al. (1998) first reported a polymorphic region within MAOA gene promotor, located at 1.2 kb upstream of the MAOA coding sequence. This polymorphism is characterized as a 30-bp repeat sequence with 2, 3, 3.5, 4, and 5 copies (Huang et al., 2004) which affects the transcriptional activity of the MAOA gene. Alleles with 3.5 or 4 repeats are transcribed 2-10 times more efficiently compared to the other alleles (Sabol et al., 1998). The possible explanation for the effect of MAOA uVNTR on gene transcription could be that this sequence acts as a binding site for a transcription factor, or it is responsible for altering chromatine structure (Sabol et al., 1998). This polymorphic region could also be involved in the regulation of MAOA
9
ACCEPTED MANUSCRIPT gene expression by DNA methylation mechanism. The most frequent alleles in the population are those with 3 repeats and 4 repeats. The 4 repeat variant has been found to be responsible for the higher (Deckert et al., 1999; Sabol et al., 1998), while the 3 repeat variant is associated with lower MAO-A activity
PT
(Bortolato and Shih, 2011). However, both low activity MAOA variants (i.e. with 3 repeats), but also high activity MAOA variants (i.e. with 4 repeats), were reported to be associated with impulsive aggressiveness,
CR I
antisocial problems and violent delinquency (Antypa et al., 2013; Bortolato and Shih, 2011; Manuck et al.,
NU S
2000; Stetler et al., 2014).
1.2. Schizophrenia
MA
Schizophrenia is a chronic, common disabling psychiatric disorder (APA, 1994) that imposes enormous costs, both on individuals and society at large, affecting approximately 1% of the population. It is a complex
D
diagnostic entity with significant differences in the response to medication, clinical presentation, and
TE
course of the disease (Bray et al., 2010). Various symptoms of schizophrenia usually occur in early adulthood and lead to serious socioeconomic problems. The symptoms of schizophrenia are usually divided
CE P
into positive, negative, cognitive and affective (mood) symptoms. However, its neurobiological basis is poorly understood, which makes the treatment of schizophrenia only partially effective and demands new
AC
treatment strategies (Meyer and Feldon, 2010). Since all effective antipsychotic drugs are dopamine receptor blocking agents (Seeman, 2000), it is postulated that dopaminergic system has an important role in the etiology of schizophrenia. Clinical features of schizophrenia include hallucinations, delusions, thought and speech disorders. Violence, aggression and agitation, as a consequence of delusions, hallucinations and substance abuse, but also as a consequence of the social deterioration, are common in patients with schizophrenia (Soyka, 2011). The findings of the studies that monitored aggressive behavior with the Overt Aggression Scale (OAS) and psychopathology with the Positive and Negative Syndrome Scale (PANSS) in chronic schizophrenia supported the view that positive symptoms may lead to aggression (Nolan et al., 2005). Psychomotor agitation is often associated with aggression, and aggression is frequently associated with violence (Soyka, 2011). The higher risk of committing non-violent and violent crime as well as homicide
10
ACCEPTED MANUSCRIPT have been observed in subjects with schizophrenia (Arseneault et al., 2000; Brennan et al., 2000; Erb et al., 2001; Fazel and Danesh, 2002; Wallace et al., 2004). Hence, it is important to identify aggression and agitation early, and apply therapy quickly, in order to prevent aggressive behavior (Nordstrom and Allen,
PT
2013). The management of aggression in patients with schizophrenia is a complex and challenging clinical
CR I
dilemma (Buckley et al., 2011).
1.3. Conduct disorder
NU S
Abnormal aggressive behavior in children and adolescents is frequent in conduct disorder and oppositional defiant disorder, according to the DSM- IV (APA, 1994). According to the newest 5th edition
MA
of DSM /DSM-5/ (APA, 2013), conduct disorder is characterized by behavior that violates either the rights of others or norms of society, with a cold and unemotional interpersonal style. To be diagnosed with
D
conduct disorder, the symptoms must cause significant impairment in social, academic or occupational
TE
functioning. Individuals with conduct disorder show repetitive and chronic aggressive and antisocial behavior, limited empathy and little concern for the feelings, wishes, and well-being of others, school
CE P
refusal and social communication problems, leading to illicit activities and persistent antisocial, criminal and violent behavior (Frick et al., 2005).
AC
Individuals with conduct disorder often display limited prosocial emotions, lack of empathy and guilt, low sensitivity for punishment and reward, delinquent behavior and they frequently develop substance abuse (Matthys et al., 2013). These subjects are characterized by rule breaking, aggression, destruction, lying, stealing, assaulting and cruelty to animals and people (Spencer et al., 2007). The characteristic symptoms of conduct disorder must be manifested by the presence of three (or more) of the following criteria in the past 12 months, with at least one criterion present in the past six months: aggression to people and animals, destruction of property, deceitfulness or theft, and serious violations of rules. The disorder is typically diagnosed prior to adulthood and might be subdivided into early (childhood) onset and adolescent onset subtypes (Frick and Nigg, 2012). The age at which symptoms of conduct disorder emerge seems to be critical for designating subgroups that differ in severity, course, and etiology (Moffitt et al.,
11
ACCEPTED MANUSCRIPT 2008). In addition, conduct disorder occurs two to three times more likely in boys than in girls (Moffitt and Caspi, 2001).
PT
1.4. The association of schizophrenia and conduct disorder Several studies suggested that children with antisocial behavior and conduct disorder have an
CR I
increased risk for schizophrenia (Robins et al., 1991; Robins, 1993; Robins and Price,1991), and that up to 40% of subjects who developed schizophreniform disorders as adults, also met the criteria for conduct
NU S
disorder in the childhood (Kim-Cohen et al., 2003). Among subjects with schizophrenia, the prevalence of conduct disorder prior to age 15 is 20-45% (Hodgins, 2008; Hodgins et al., 2008). Moreover, prior to the
MA
onset of schizophrenia, a sub-group of patients displayed behavioral problems in childhood and adolescence (Asnarow, 1988; Olin et al., 1997). Among men who develop schizophrenia, conduct problems
D
in childhood and early adolescence are suggested as antecedents of physical aggression towards others
TE
throughout the adulthood (Hodgins et al., 2005). The subjects with both schizophrenia and conduct disorder are characterized by a lifelong pattern of antisocial and aggressive behavior, substance abuse and
CE P
criminality and are suggested to display a distinct subtype of schizophrenia (Schiffer et al., 2013). Accumulating evidence supported the hypothesis that conduct disorder is not a co-morbid disorder,
AC
but rather an early feature of the developing schizophrenia. In line with this hypothesis, Cannon et al. (1990) reported that a pattern of disruptive behavior (overactive, irritable, distractible and aggressive behavior) in the late childhood and early-adolescence was associated with the development of schizophrenia with predominant positive symptoms. Therefore, it has been proposed that the association of conduct disorder with aggressive behavior, following onset of schizophrenia, is due to its impact on factors such as positive symptoms, noncompliance with medication and lack of care and substance misuse (Swartz et al., 1998). On the other hand, some other findings supported the view that among persons with schizophrenia, conduct disorder is a distinct co-morbid disorder. The results of the study by Hodgins et al. (2005) suggested that conduct disorder is directly associated with criminal offending and substance abuse that begins in adolescence and persists across the life-span and that the course of antisocial behavior runs
12
ACCEPTED MANUSCRIPT parallel to schizophrenia. The association between conduct disorder and schizophrenia might be due to the fact that children with genetic predisposition for schizophrenia are more vulnerable than others to etiological factors such as stress or physical abuse, which in interaction with highly prevalent gene variants
PT
(such as polymorphisms in gene encoding MAO-A) result in increased risk for antisocial behavior (Caspi et al., 2002; Foley et al., 2004). Conversely, elevated rates of conduct disorder among subjects with
CR I
schizophrenia could be in part due to higher rates of assortative mating (Parnas, 1988), so that children
NU S
inherit susceptibility genes for both schizophrenia and antisocial behavior (Rhee and Waldman, 2002).
1.5. Agitation
MA
Agitation is represented by a cluster of pathological behavior such as motor restlessness, mental tension, elevated response to various external or internal stimuli, and various symptoms: pacing, hand wringing, fist
D
clenching, pressured speech, yelling and threatening other people, irritability, inappropriate and/or
TE
purposeless verbal or motor activity, decreased sleep and fluctuation of symptoms over time (Lesem et al., 2011; Lindenmayer, 2000). In addition, agitation frequently escalates into violence and is characterized also
CE P
by excessive or inappropriate motor or verbal activity and hostility, which may include irritability, uncooperativeness, threatening gestures, assault and poorly organized and aimless psychomotor activity
AC
(Lesem et al., 2011; Mohr et al., 2005; Montoya et al., 2011; Volavka, 2014). Agitation is frequent in psychotic disorders, especially in acute episodes. It may be presented as verbal or physical aggression, and might be focused to different objects or persons. Verbal aggression, including threats, abuse or incoherent screams, is frequent part of agitation. Since agitation might accompany some mental (schizophrenia, bipolar disorder, post-traumatic stress disorder (PTSD), antisocial personality disorder, conduct disorder, attention deficit hyperactivity disorder (ADHD), alcohol/substance abuse) and neurological (Alzheimer’s disease, autoimmune encephalitis and other dementias) disorders, underlying neurobiology might be associated both to a particular disorder and to agitation (Rosell and Siever, 2015). Moreover, agitation is presumed to be one of the risk factors or indicators for suicide attempt (Bryan et al., 2014). The neurobiological basis of agitation is presumed to involve the disturbances in serotonin, dopamine,
13
ACCEPTED MANUSCRIPT norepinephrine, GABA and glutamate (Lindenmayer, 2000) neurotransmission, but this relationship is far from clear. In addition, agitation is demonstrated as gender specific (Bryan et al., 2014). In psychiatric patients, especially those with schizophrenia, agitation is frequently assessed using the
PT
excited component of PANSS-EC (Citrome et al., 2011; Montoya et al., 2011; Swanson et al., 2004; Volavka, 2014). This PANSS-EC evaluates the scores in particular symptoms: excitement, hostility, tension,
CR I
uncooperativeness and poor impulse control, which can be rated as absent, minimal, mild, moderate, moderate/severe, and severe to extreme (i.e. with scores 1-7). For assessing agitated patients, PANSS-EC is
NU S
easy to use and is one of the simplest and most intuitive scales (Baker et al., 2003; Citrome et al., 2011; Lesem et al., 2011; Lindenmayer et al., 2008; Montoya et al., 2011; Volavka, 2014). Individuals might be
MA
subdivided into those with presence of agitation or severe agitation. Agitation is defined if patients have a total score of ≥ 14 out of 35 on these particular PANSS-EC items (Lesem et al., 2011). On the other hand,
D
severe agitation is present if patients are rated with ≥ 4 (moderate) scores on the each item of the PANSS-
TE
EC and therefore patients are defined as severely agitated if they have a total score of ≥ 20 out of 35 (Baker et al., 2003; Montoya et al., 2011).
CE P
As PANSS is not used in adolescents, agitation in adolescents with conduct disorder is assessed using the particular items of the Psychopathy Checklist Youth Version (PCL:YV) and Overt Aggression Scale-Modified
AC
(OAS-M) (Coccarro et al., 1991; Forth et al., 2003; Rush, 2000). PCL:YV is designed for assessments of psychopathic features in youths separated in 4 factors (arrogance and deceitfulness - factor 1, affective unresponsiveness - factor 2, impulsive and irresponsible life style - factor 3, and criminal behavior - factor 4). OAS-M (Coccarro et al., 1991) is designed to assess various manifestation of aggressive behavior in outpatients. The scale incorporates three overall domains: aggression, irritability and suicidality. The irritability and suicidality items in the OAS-M instrument were adapted from the Schedule for Affective Disorders and Schizophrenia (SADS), suggesting that PANSS-EC and OAS-M could be taken in coincidence. The OAS-M is the only aggression scale that actually assesses concrete aggressive behavior. The use of weighted items in Aggression scale scores takes into account that high severity items are both less frequent and more significant that low severity items. The cut-off scores used in the present study were determined according to clinical experience and data from the earlier published study (Podobnik et al., 2012).
14
ACCEPTED MANUSCRIPT 1.6. Aim of the study Our primary hypothesis was that platelet MAO-B activity and the most frequently studied genetic
PT
polymorphisms in the MAOA and MAOB genes (MAOA uVNTR and MAOB rs1799836) will be significantly related to severe agitation in male patients with schizophrenia and male adolescents with conduct
CR I
disorder. More specifically, we have expected to detect reduced platelet MAO-B activity in agitated subjects and significant association between specific genetic variants of MAOA uVNTR and MAOB
NU S
rs1799836 polymorphisms and severe agitation. The main aim of our study was to define if MAO alternations, studied at functional and genetic levels, contribute to the development of severe agitation.
MA
This approach was in line with the initiative of the National Institute of Mental Health to evaluate Research Domain Criteria (RDoC) and to assess biomarkers, rather than diagnosis in psychiatry (Insel et
D
al., 2010). Our research was focused on the part of RDoC criteria (i.e. agitation) and genetic and
TE
molecular/cellular factors which might influence clinically relevant functions (Insel et al., 2010). The specific aims of our study were to compare platelet MAO-B activity, the MAOB rs1799836 A and G allele
CE P
frequency as well as the frequency of the low (3-repeat; 3R) and high (4-repeat; 4R) activity variants of
AC
the MAOA uVNTR polymorphism in severely agitated and non-agitated male subjects.
2. Material and methods 2.1. Subjects
This study included 363 male, drug naive subjects, who were subdivided into those with or without severe agitation. There were 187 patients with schizophrenia and 176 adolescents with conduct disorder, diagnosed by consensus of two psychiatrists and according to SCID based or DSM-IV criteria (APA, 1994). Smoking status was determined according to questionnaires and subjects were classified as smokers (smoking ≥10 cigarettes per day) and non-smokers (never-smokers). The “light” smokers, smoking less than 10 cigarettes per day, were excluded. All subjects were subdivided according to the presence or absence of severe agitation, determined according to the a priori chosen cut-offs in the corresponding clinical scores. Out of 187 patients with schizophrenia, there were 26 severely agitated and 161 non-agitated patients, and
15
ACCEPTED MANUSCRIPT 138 smokers and 49 non-smokers. Patients with schizophrenia were evaluated at the Clinics for Psychiatry Vrapce, Zagreb and Psychiatric Hospital Dr Ivan Brbot, Popovaca, Croatia. Adolescents with conduct disorder (N=176) were sampled at the correctional facilities in Zagreb County, Croatia. They were
PT
subdivided into 68 severely agitated and 108 non-agitated subjects, while according to the smoking status they were divided into 144 smokers and 32 non-smokers. Exclusion criteria for all subjects were:
CR I
diagnoses other than schizophrenia or conduct disorder, neurodegenerative disorders, drug and alcohol dependence, any kind of treatment before the study entry, current medication, cognitive behavioral
NU S
therapy or electroconvulsive therapy and smoking between 1-10 cigarettes per day.
MA
2.2. Clinical assessment
Agitation was assessed using the PANSS-EC, the PCL:YV and the OAS-M. For assessing severe agitation,
D
patients with schizophrenia were rated according to the PANSS-EC. This subscale includes particular items:
TE
excitement (P4), hostility (P7), tension (G4), uncooperativeness or lack of cooperation (G8), and poor impulse control (G14). All items are rated on a 7-point scale (1=absent, 2=minimal, 3=mild, 4=moderate,
CE P
5=moderate/severe, 6=severe, 7=extreme). Severe agitation was rated as present when a person had a mean PANSS-EC score of 20 or more out of possible 35 (Baker et al., 2003; Montoya et al., 2011).
AC
In adolescents with conduct disorder, agitation and other forms of aggression are not usually assessed and rated by the PANSS, and these subjects were assessed and rated by the more appropriate OAS-M (aggression restricted to past week) and PCL:YV scales. In the present study adolescents were rated as severely agitated if they had: 1) the subjective irritability in the OAS-M with score ≥ 3 out of 5, corresponding to the tension described in the PANSS-EC; 2) the global irritability in the OAS-M with score ≥ 6 out of 10, corresponding to the poor impulse control described in the PANSS-EC; 3) the finale scale scores for aggression in the OAS-M with ≥ 30 out of 135 scores (including verbal aggression, aggression against property, auto-aggression and physical aggression), corresponding to the excitement described in the PANSS-EC; 4) the total aggression score in the OAS-M with ≥ 40 out of 161 (comprising aggression, irritability and suicidality), corresponding to hostility described in the PANSS-EC; and 5) the total PCL:YV score ≥ 20 out of 40, corresponding to the lack of cooperation described in the PANSS-EC. Therefore,
16
ACCEPTED MANUSCRIPT adolescents with conduct disorder were subdivided into currently severely agitated (indication for hospitalisation or medication) and presently (in time of rating) non-agitated according to PCL:YV and the
PT
OAS-M scores (Forth et al., 2003; Podobnik et al., 2012; Rush, 2000).
2.3. Ethics statement
CR I
All samples were taken after the participants voluntarily signed informed consent. The interview and blood sampling of all subjects were done either at the hospital or in the correctional facilities when they
NU S
were able to understand and agree with the protocol of the study. All studies were conducted with the approval of corresponding Ethics Committees, and were fully compliant with the ethical standards laid
MA
down in the 1975 Declaration of Helsinki.
D
2.4. Sampling and measuring MAO-B activity
TE
Whole blood samples were collected in 8.5 ml yellow-top Vacutainer tubes with 1.5 ml of acid citrate dextrose anticoagulant. Sampling was done in the morning around 8 a.m., after the overnight fasting and
CE P
was part of the routine laboratory visits. Platelets were obtained by series of centrifugation of whole blood and platelet-rich-plasma and then broken by sonication. Platelet MAO-B activity (expressed in nmol 4-
AC
OHQ/mg protein/h) was determined in 363 subjects (patients with schizophrenia and adolescents with conduct disorder) according to the slightly modified method described by Krajl (1965) and using kynuramine as a substrate. Briefly, 100 l of standard (4-hydoxyguinoline, 4-HOQ) and platelet sonicates (both in duplicates) were incubated with 100 l kynuramine (final concentration 73.6 M) at 37°C for 1 hour. The reaction was stopped with 1 N NaOH. Platelet MAO-B activity was measured by fluorimetry of 4-hydroxyquinoline (4-HQ), which is converted by oxidative deamination from kynuramine. The measurement of florescence of 4-HOQ, a product of kynuramine, was performed on Varian Spectrophotofluorimeter Cary Eclipse, with an exciting wavelength of 310 nm and emitted wavelength of 380 nm. Platelet protein levels were measured by the method of Lowry et al. (1951). Platelet MAO-B activity was calculated using the standard curve of 4-HQ and was expressed in nmol of 4-HQ formed per hour per mg of total platelet protein.
17
ACCEPTED MANUSCRIPT 2.5. Genotyping of MAOB rs1799836 polymorphism Genomic DNA was extracted from peripheral blood using the salting out method (Miller et al., 1988).
PT
MAOB intron 13 A/G polymorphisms (rs1799836) was determined in 363 subjects by ABI Prism 7300 Real time PCR System apparatus (Applied Biosystems, Foster city, California, USA), according to the procedures
CR I
described by Applied Biosystems. The primers and probes were purchased from Applied Biosystems as TaqMan® SNP Genotyping Assay (C_8878790_10). All genotyping procedures were done by a researcher
NU S
who was blind to all clinical data. Out of 363, 80 samples (22%) were genotyped again as a quality control
MA
for genotyping analyses.
2.6. Genotyping of MAOA uVNTR polymorphism
D
The MAOA promoter uVNTR polymorphism was genotyped in 354 subjects using PCR-based amplification
TE
with primers MAOA-F: 5’-ACAGCCTCGCCGTGGAGAAG-3’ and MAOA-R’: 5’-GAACGGACGCTCATTCGGA-3’. The final PCR reaction volume was 15 l, which contained ~100 ng of DNA, 1.7 mM MgCl2, 1 x Q solution,
CE P
20 pmol of each primer, 0.25 mM dNTP and 1 U of Taq polymerase (Qiagen). The PCR was performed on 2720 Thermal Cycler (Applied Biosystems) with the cycling conditions: 5 min at 95°C, 35 cycles carried out
AC
at 30 sec at 94°C, 30 sec at 57°C and 1 min at 72°C, and final extension 7 min at 72°C. PCR products were separated on 2.5% agarose gels which were stained with ethidium bromide and visualized using Uvitec BXT20-M transiluminator. All participants were found to harbor alleles carrying 3 or 4 repeats, with the exception of two carriers of the 5-repeat variant. The 3-repeat variants, together with the two 5-repeats variants were defined as low activity variants, while variants with 4-repeat variants were considered as high activity variants. Genotyping for the MAOA uVNTR failed for 9 participants. Out of 363, 40 samples (11%) were genotyped again as a quality control for genotyping analyses.
2.7. Statistical evaluation The results were expressed as numbers and percentages, or median and 25th (Q1) and 75th (Q3) percentiles and evaluated with Sigma Stat 3.5 (Jandel Scientific Corp., San Jose, California, USA). The allele
18
ACCEPTED MANUSCRIPT frequencies were compared using χ2-tests and Yates correction for continuity. In order to verify what category was a major contributor to rejecting the null hypothesis, standardized residual (R) (Field et al., 2012) was calculated. As the study included only male subjects, the deviation from Hardy-Weinberg
PT
equilibrium for both MAOB rs1799836 and MAOA uVNTR could not be tested since the genes encoding MAO-A and MAO-B are located on the short arm of the X chromosome. Normality of distribution was
CR I
assessed with the Kolmogorov-Smirnov test. A multiple linear regression analysis was used to check for the influence of diagnosis, age, smoking status, MAOB rs1799836 and MAOA uVNTR on MAO-B activity.
NU S
Comparison of platelet MAO-B activity and age between different groups of subject was done using Kruskal Wallis ANOVA by ranks or Mann-Whitney test, since normality of the data failed.
MA
To avoid issues with multiple testing (Leucht et al., 2007) and correction for the multiple testing, all statistical tests were applied to cutoffs chosen a priori, and agitation was defined a priori. All tests were
D
two-tailed and α was set at 0.05. G*Power 3 Software (Faul et al., 2007) was used for conducting power
TE
analyses, i.e. to determine a priori sample size and actual power. For multiple regression analysis (with α = 0.05; power (1 − β) = 0.800; a small effect size (ω = 0.30); number of predictors=4), total desired sample
CE P
size was 46, and the actual sample size was 363. For t-test (with α = 0.05; power (1 − β) = 0.800; a small effect size (ω = 0.30)), total desired sample size was 352, and the actual sample size was 363. For ANOVA
AC
(with α = 0.05; power (1 − β) = 0.800; a small effect size (ω = 0.30) and 4 groups), total desired sample size was 128, and the actual sample size was 363. For genetic analyses and χ2-test (with α = 0.05; power (1 − β) = 0.800 and a small effect size (ω = 0.30); d.f. = 1), total desired sample size was 88; and if d.f. = 2; the total desired sample size was 108, while the actual total sample size was 363 for MAOB rs1799836 and 354 for MAOA uVNTR. Therefore the sample size and the statistical power were appropriate to detect significant differences in the studied groups.
3. Results 3.1. Age of participants As expected, age of participants, presented as medians and 95% CI for the median, differed significantly (U=15.5; P=0.0001; Mann Whitney’s test), since patients with schizophrenia were significantly older (42;
19
ACCEPTED MANUSCRIPT 41-46 years) than adolescents with conduct disorder (17; 16-17 years). Severely agitated subjects were significantly (p
