Embed Size (px)
Citation preview
The Role of Leptin-Melanocortin System Gene Variants in Antipsychotic-Induced Weight Gain
by
Nabilah Ishrat Chowdhury
A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy
Institute of Medical Science University of Toronto
© Copyright by Nabilah Chowdhury 2015
ii
The Role of Leptin-Melanocortin System Gene Variants in
Antipsychotic-Induced Weight Gain
Nabilah Ishrat Chowdhury
Doctor of Philosophy
The Institute of Medical Sciences
University of Toronto
2015
Abstract
The use of second-generation antipsychotic medications can result in the development of severe
side effects such as weight gain and metabolic syndrome. Previous studies have proposed that a
genetic component contributes between 40% and 80% to the development of this side effect.
However, genetic studies thus far have yielded mixed findings. The genetics of the melanocortin
system has not yet been extensively examined in the context of antipsychotic-induced weight
gain. Objectives: The first objective was to investigate whether melanocortinergic single
nucleotide polymorphisms from genes that were associated with high weight and obesity were
also associated with antipsychotic-induced weight gain. The second objective was to determine
whether gene-gene interactions better predict the onset of antipsychotic-induced weight gain in
psychiatric patients. Hypothesis: It was predicted that the melanocortin-3 receptor,
melanocortin-4 receptor, pro-opiomelanocortin, cocaine-amphetamine regulated transcript, and
agouti related protein would be associated with antipsychotic-induced weight gain, and that a
combination of these genes would provide improved prediction of this phenotype. Results: The
melanocortin-3 and melanocortin-4 receptors were associated with antipsychotic-induced weight
iii
gain. An additive model did not demonstrate that the combined effect in these genes explains
increased variance of the weight gain phenotype. Conclusions: The melanocortin receptor
system may have a more substantial role in predicting antipsychotic-induced weight gain, as
compared to previously studied gene systems. Though further exploration and functional
analyses are required, these findings contribute novel knowledge to the pharmacogenetics of
antipsychotic-induced weight gain, and may help to enhance the development of clinical genetic
tests to guide medication decision-making in the future.
iv
Acknowledgments
I would like to begin by thanking my supervisor, Dr. James Kennedy, for guiding and supporting
me throughout the PhD Degree. He has gone above and beyond my expectations as a mentor,
and I will always be grateful for the excellent supervision he has provided me. I am also very
thankful to Dr. Daniel Müller, who acted as my co-supervisor. Dr. Müller has imparted many
skills and much advice, which have helped me grow as a young scientist. I would also like to
thank my committee members, Drs. Levitan and Remington, for their support, input and for
constantly offering challenging questions that have helped to improve the quality of the research
presented within my thesis.
Dr. Arun Tiwari has played a very influential role during the time of writing this thesis. I am
grateful for his direction in the analyses and methodologies of my work, for the critiques of my
manuscripts and for always taking the time to provide me with support. I could not have
accomplished the achievements during my degree without all of Dr. Tiwari’s help. I would also
like to thank Dr. Renan Souza for teaching me statistical techniques during the early part of my
degree. I would also like to thank Dr. Clement Zai for his support and willingness to teach and
discuss research with trainees, and to let him know that his help was always very much
appreciated. Finally, I would like to thank Dr. Eva Brandl for her expertise and clinical insight,
the many discussions on scientific topics, coffee breaks and chats in her office.
Next, I am very grateful to the Neurogenetics/Pharmacogenetics laboratory staff. Natalie
Freeman was always able to pinpoint difficulties with experiments and provided constructive
criticism, and for that I am very thankful. I would also like to thank Sajid Shaikh for teaching me
how to conduct my very first polymerase chain reaction, and for being a wonderful colleague. I
have also been very lucky to have the help of Olga Likhodi, David Irwin and Maria Tempakaras
in the laboratory. The administrative staff, Ms. Mary Smirniw, Ms. Andrea Smart and Sheraz
Cheema, was very helpful in listening to the inevitable obstacles that arise during a graduate
degree, and for booking me in for those coveted appointments with Dr. Kennedy. I would also
like to thank my lab-mates, past and present, for proofreading my studentship applications,
collaborating to produce scientific work, providing me with many laughs and for their overall
support.
v
Finally, I would very much like to express my gratitude to my parents and younger brother for
their unconditional love and unwavering support. Without them, I would not be where I am
today.
vi
Contributions
The research presented in this thesis consists of three original research studies. Chapter 3 was
published in The Pharmacogenomics Journal, and Chapter 4 was published in The World
Journal of Biological Psychiatry. The research described in Chapter 5 is currently in preparation
for publication in a peer-reviewed journal.
The author conducted all of the experiments, with the guidance and expertise of the individuals
listed below:
Mr. Sajid Shaikh: Supervised and helped Ms. Chowdhury to conduct polymerase chain reaction
for experiments in Chapter 3.
Dr. Arun Tiwari: Conducted the electrophoretic mobility shift assay described in Chapter 3,
and contributed to the experimental design and statistical analyses for genotypic analysis in
Chapters 3, 4 &5.
Dr. Sheng Chen: Provided guidance on quantities and reagents to be used in electrophoretic
mobility shift assay in Chapter 3.
Dr. Fang Liu: Provided laboratory space and reagents for electrophoretic mobility shift assay
experiment in Chapter 3.
Dr. Clement Zai: Conducted statistical analysis of gene-gene interaction in Chapter 5. Guided
haplotype analysis presented in Chapter 5.
vii
Table of Contents
Abstract ............................................................................................................................. ii
Acknowledgements .......................................................................................................... iv
Table of Contents ............................................................................................................ vii
List of Tables ..................................................................................................................... x
List of Figures ................................................................................................................... xi
List of Abbreviations ...................................................................................................... xii
Preface ................................................................................................................................ 1
Chapter 1 Literature Review ........................................................................................... 2
1.0 Introduction ......................................................................................................... 3
1.1 Schizophrenia Background ................................................................................. 3
1.2 Schizophrenia:Epidemiology .............................................................................. 4
1.3 Schizophreia: Etiology ....................................................................................... 5
1.3.1 Environmental Risk Factors ..................................................................... 3
1.3.2 Genetic Risk Factors ................................................................................ 6
1.4 Actions of Antipsychotics at Different Receptors ............................................... 8
1.5 Schizophrenia and Pharmacogenetics of Antipsychotic Induced Weight Gain . 11
1.5.1 Contributing Factors to Antipsychotic-Induced Weight Gain ....................... 11
1.5.2 Gene Systems and Antipsychotic-Induced Weight Gain ............................... 12
1.5.3 Dopamine and Antipsychotic-Induced Weight Gain ..................................... 13
1.5.4 Serotonin and Antipsychotic-Induced Weight Gain ...................................... 16
1.5.5 Histamine System and Antipsychotic-Induced Weight Gain ......................... 19
1.5.6 Adrenergic System and Antipsychotic-Induced Weight Gain ....................... 22
1.6. The Genetics of Obesity .......................................................................................... 23
1.6.1 General Obesity ............................................................................................ 23
1.6.2 Environmental Factors and Obesity .............................................................. 26
1.6.3 Gene and Environment Interaction in Obesity .............................................. 27
1.7. Leptin Genetics in Antipsychotic-Induced Weight Gain .................................... 28
1.7.1 Leptinergic System as a Critical Regulator of Weight Gain ........................ 28
1.7.2 Leptin Genetics in Antipsychotic-Induced Weight Gain ............................... 30
1.7.3 Cocaine and Amphetamine Regulated Transcript ......................................... 31
1.7.4 Endocannabinoid System............................................................................... 32
1.7.5 Neuropeptide Y .............................................................................................. 33
1.7.6 Ghrelin .......................................................................................................... 34
1.7.7 Agouti Related Protein .................................................................................... 34
viii
1.7.8 Brain-Derived Neurotrophic Factor ................................................................ 35
1.7.9 Pro-melanin Concentrating Hormone ............................................................. 36
1.8.0 Melanocortin Receptor Genes .............................................................................. 36
1.9 Methodologies and Statistical Considerations ....................................................... 43
1.9.1. Polymerase Chain Reaction and Genotyping Methodology ........................... 43
1.9.2. Analyses, Sample Size and Power Calculations ............................................. 45
Chapter 2 Experimental Approach & Hypothesis ....................................................... 49
2.0 Overview of Experiments, Hypotheses and Objectives ......................................... 42
2.1.1. General Overview ............................................................................................ 50
2.2 Study One: Genetic Association Study between Antipsychotic-Induced Weight Gain
and the Melanocortin-4 Receptor Gene ........................................................................ 52
2.2.1. Background ..................................................................................................... 52
2.2.2. Hypothesis/Objectives of Study ...................................................................... 51
2.3. Study Two: Investigation of Melancortin-System Gene Variants in Antipsychotic-
Induced Weight Gain ...................................................................................................... 51 2.3.1 Background .................................................................................................. 51
2.3.2 Hypothesis/Objectives of Study ................................................................... 52
2.4. Study Three: An Exploratory Investigation of Melanocortin-3 Receptor Gene Variants
in Antipsychotic-Induced Weight Gain ........................................................................ 52 2.4.1 Background ................................................................................................... 52
2.4.2 Hypotheses/Objectives of Study ................................................................... 53
Chapter 3 Genetic Association Study between Antipsychotic-Induced Weight Gain and the
Melanocortin-4 Receptor Gene ...................................................................................... 54
3.1 Abstract ...................................................................................................................... 55
3.2 Introduction ................................................................................................................ 56
3.3 Materials and Methods ............................................................................................... 59
3.3.1 Subjects ................................................................................................................. 59
3.3.2 Genetic Analysis ................................................................................................... 60
3.3.3 Data Analysis ....................................................................................................... 60
3.3.4 Electrophoretic Mobility Shift Assay ................................................................... 61
3.4 Results ......................................................................................................................... 62
3.4.1. Genetic Analysis .................................................................................................. 62
3.4.2 Electrophoretic Mobility Shift Assay Analysis ................................................... 68
3.5 Discussion ................................................................................................................... 71
Chapter 4 Investigation of Melancortin-System Gene Variants in Antipsychotic-Induced
Weight Gain ..................................................................................................................... 74
ix
4.1 Abstract ....................................................................................................................... 75
4.2 Introduction ................................................................................................................. 76
4.2.1 Proopiomelanocortin ............................................................................................ 77
4.2.2 Cocaine Amphetamine Regulated Transcript ...................................................... 77
4.2.3 Agouti Related Protein ......................................................................................... 78
4.3 Materials & Methods .................................................................................................. 79
4.3.1 Subjects ................................................................................................................ 79
4.3.2 Genotyping ........................................................................................................... 81
4.4 Statistical Analyses ..................................................................................................... 83
4.5 Results ......................................................................................................................... 84
4.6 Discussion ................................................................................................................... 88
Chapter 5 An Exploratory Investigation of Melanocortin-3 Receptor Gene Variants in
Antipsychotic-Induced Weight Gain ............................................................................. 90
5.1 Abstract ...................................................................................................................... 91
5.2 Introduction ................................................................................................................ 92
5.3 Methods ...................................................................................................................... 94
5.3.1 Samples ................................................................................................................. 94
5.3.2 Genotyping Protocol ............................................................................................. 95
5.3.3 Statistical Analysis ................................................................................................ 95
5.4 Results ......................................................................................................................... 96
5.5 Discussion ................................................................................................................ 104
Chapter 6 General Discussion ...................................................................................... 108
6.1 Summary of Findings ............................................................................................ 109
6.2 Obesity and Antipsychotic Induced Weight Gain............................................... 117
6.3 Multi-gene Modeling .............................................................................................. 119
6.4 Postulated Mechanisms of Antipsychotic-Induced Weight Gain ...................... 122
6.5 Limitations and Considerations ............................................................................ 125
6.5.1 Sample Characteristics .................................................................................. 125
6.5.2 Sample Size and Power ................................................................................. 126
6.6 Correction for Multiple Testing ........................................................................... 128
6.7 Replication in Independent Sample Sets.............................................................. 129
6.8 Future Directions ................................................................................................... 131
6.8.1 Measures of Weight Change .......................................................................... 131
6.8.2 Candidate Gene Study vs. Genome Wide Association Study ........................ 132
6.8.3 Remote Regulatory Site Databases .............................................................. 134
6.8.4 Whole Genome Sequencing ........................................................................... 135
6.8.5 Epigenetics ..................................................................................................... 136
6.8.6 Environmental Factors ................................................................................... 137
6.9 Concluding Remarks .............................................................................................. 138
References……………………………………………………………………………...144
x
List of Tables
Table 1-1. Receptor Affinities of Second-Generation Antipsychotic Medications .................... 11
Table 1-2. Receptor Affinities of Second-Generation Antipsychotic Medications .................... 39
Table 3-1. Demographic Characteristics ......................................................................... 63
Table 3-2. MC4R SNPs and Association with Antipsychotic-Induced Weight Gain
at Genotypic Level ........................................................................................... 64
Table 3-3. Oligonucleotides for EMSA .......................................................................... 68
Table 4-1. Demographic Characteristics ......................................................................... 80
Table 4-2. List of SNPs in POMC, CART & AGRP tested for association with AIWG at
the genotypic level ....................................................................................... .85
Table 4-3. Allelic differences in weight change (%) in patients of European ancestry…87
Table 5-1. Demographic Characteristics ........................................................................ .97
Table 5-2. Genoytpic associations between MC3R & antipsychotic induced
weight gain ............................................................................................... 99
Table 5-3. Genotypic associations between MC3R and antipsychotic induced weight gain
in European Patients treated with antipsychotic medications . .................. 101
Table 5-4. Allelic differences in weight change (%) in patients of European ancestry 102
xi
List of Figures
Figure 1-1. Schematic Diagram of the Leptin-Melanocortin System .............................. 38
Figure 1-2. The MC4R rs489693 genotype and antipsychotic induced
weight gain in 4 cohorts of subjects. ............................................................. 41
Figure 1-3. Results from Genome Wide Association Study ............................................ 42
Figure 3-1. Association between MC4R rs8087522 and antipsychotic induced weight
gain. .......................................................................................... 66
Figure 3-2. MC4R (chr18q22) Gene Structure and Linkage Disequilibrium .................. 67
Figure 3-3. Electrophoretic Mobility Shift Assay for SNP rs808752262 ........................ 69
Figure 4-1. Schematic gene diagram of POMC, CART, and AGRP genes. .................... 82
Figure 5-1. Schematic Diagram of the MC3R Gene with the Ten Single Nucleotide
Polymorphism. .............................................................................................. 98
xii
List of Abbreviations
ADRA1A alpha-1 (α1) adrenergic receptor
ADRA2B adrenoceptor alpha 2B
ADRA2A α2A adrenoceptor
ADRB3 beta-3 adrenergic receptor
AGRP agouti related protein
AIM ancestry informative marker
AIWG antipsychotic induced weight gain
AMPK 5' AMP-activated protein kinase
ANOVA analysis of variance
ANCOVA analysis of co-variance
ASTN2 astrotactin 2
ATF7IP2 activating transcription factor 7 interacting protein 2
BDNF brain derived neurotrophic factor
BMI body mass index
CART cocaine and amphetamine regulated transcript
CEPH Centre d'Etude du Polymorphisme Humain
CTCF CCCTC-binding factor
CNR1 cannabinoid receptor 1
CNV copy number variation
DISC 1 disrupted-in-schizophrenia 1
DRD1 dopamine receptor D1
DRD2 dopamine receptor D2
DRD3 dopamine receptor D3
xiii
DRD4 dopamine receptor D4
DRD5 dopamine receptor D5
EMSA electrophoretic mobility shift assay
ENCODE encyclopedia of DNA elements
eQTLs expression quantitative trait loci
FAAH fatty acid amide hydrolase
FHOD3 formin homology 2 domain containing 3
FTO fat mass and obesity
GABRA2 gamma-aminobutyric acid receptor subunit alpha-2
GCG glicentin-related polypeptide -1 encoding gene
GLP-1 glucagon-like peptide
GWAS genome wide association study
GHS-R growth hormone secretagogue receptor
GPR98 G protein-coupled receptor 98
HCRTR2 orexin receptor
HR1 histamine 1 receptors
LD linkage disequilibrium
LEP leptin
LEPR leptin receptor
mbmdr model based multifactor dimensionality reduction
MCH melanin concentrating hormone
MC3R melanocortin-3 receptor
MC4R melanocortin-4 receptor
MIR137 microRNA 137
MSX2 muscle segment homeobox
NDUFS1 NADH-ubiquinone oxidoreductase 75 kDa subunit
xiv
NPY neuropeptide Y
NPY5R neuropeptide Y receptor Y5
NRG1 neuregulin
PCR polymerase chain reaction
PGC psychiatric genetics consortium
PMCH pro-melanin concentrating hormone
POMC pro-opiomelanocortin
PRKAA1 protein kinase, AMP-activated, alpha 1 catalytic subunit
PRKAA2 protein kinase, AMP-activated, alpha 2 catalytic subunit
PRKAB1 protein kinase, AMP-activated, beta 1 non-catalytic subunit
PRKAB2 protein kinase, AMP-activated, beta 2 non-catalytic subuni
PRKAG1 protein kinase, AMP-activated, gamma 1 non-catalytic subunit
PRKAG2 protein kinase, AMP-activated, gamma 2 non-catalytic subunit
PRKAG3 protein kinase, AMP-activated, gamma 3 non-catalytic subunit
PRX2 paired related homeobox protein,
RNF144A ring finger protein 144A
SCZ schizophrenia
SGA second-generation antipsychotic
SH2B1 SH2B adapter protein 1
SNP single nucleotide polymorphism
SOX5 SRY-related HMG-box
5-HT 5-hydroxytryptamine (serotonin)
5-HT1A 5-hydroxytryptamine serotonin receptor 1A
5-HT2A 5-hydroxytryptamine serotonin receptor 2A
5-HT2C 5-hydroxytryptamine serotonin receptor 2C
5-HT6 5-hydroxytryptamine serotonin receptor 6
xv
5-HT7 5-hydroxytryptamine serotonin receptor 7
VGAT vesicular GABA transporter
TMEM18 transmembrane protein 18
NRXN3 neurexin 3
MC4R melanocortin-4 receptor
MEIS2 meis homeobox 2
PRKAR2B protein kinase cAMP-dependent, regulatory, type II, beta.
SEC16B SEC16 homolog B
GNPDA2 glucosamine-6-phosphate deaminase 2
TNNI3K TNNI3 interacting kinase
TSPO translocator protein-18
QPCTL glutaminyl-peptide cyclotransferase-like
xvi
1
Preface
Schizophrenia is a chronic debilitating psychiatric disease, wherein patients exhibit increased
mortality rate compared to the general population (Brown et al., 2000). A subset of patients
experience substantial weight gain when treated with second-generation antipsychotic
medications. Schizophrenia patients may experience stigma due to the weight gain, and the use
of these drugs has been associated with the onset of metabolic side effects such as type 2
diabetes, and cardiovascular diseases (reviewed in Lett et al., 2012). Genetic factors have been
reported to contribute to the development of antipsychotic induced weight gain (AIWG)
(Gebhardt et al., 2010), though no specific biological mechanism has been identified.
Discovering key genetic determinants of AIWG could further understanding of the development
of this serious side-effect, and eventually guide patient treatment. The melanocortinergic gene
pathway has a critical role in weight regulation. However, this biological system has not yet been
extensively investigated in the context of AIWG. We examined whether melanocortinergic gene
variants were associated with increased weight in a psychiatric population.
2
Chapter 1 Literature Review
3
1.0 Introduction
1.1 Schizophrenia: Background
The exact nature of schizophrenia (SCZ) has long been an enigma and remains elusive in
contemporary psychiatry (Brüne et al., 2004). To date, there have been over 40 definitions of
SCZ. In the nineteenth century, descriptions of patients with SCZ were reported across European
hospitals, namely Bethlem hospital in England and the Charite hospital in Berlin (Stone, 2006).
The term SCZ has evolved over time from Dr. Benedict Morel initially describing a sickness
affecting young adults, termed démence precoce. In the late 1800s, Emil Kraeplin, a German
psychiatrist, termed the phrase dementia praecox or premature dementia, and the term SCZ
meaning split mind was set by Eugene Bleuler (Kaplan et al., 2008; McNally et al., 2009).
In the present day, SCZ is diagnosed with criteria set by the “Fifth Edition of the Diagnostic and
Statistical Manual of Mental Disorders – DSM- 5,” or the “Tenth Revision of the International
Classification of Diseases” (ICD-10) (definitions comprehensively reviewed in Hyman, 2007).
SCZ symptoms are typically divided into a set of broad categories, including positive and
negative symptoms (van Os & Kapur, 2009). Positive symptoms include hallucinations,
delusions, thought disorders and disorganized behaviour, while negative symptoms consist of
impaired executive functioning — the ability of higher thought processes, lack of focus or
attention, and impaired working memory.
Currently, no given treatment or medication is efficacious for the entire spectrum of SCZ
symptoms. For example, a medication may reduce the prevalence of positive symptoms, but have
little to no effect on negative or cognitive symptoms. Furthermore, some of the most effective
4
pharmaceutical therapies can result in debilitating side effects, and may even result in the onset
of a co-morbid disorder or disease. For instance, the use of a subset of psychiatric medications
may result in the development of tardive dyskinesia (reviewed in Chowdhury et al., 2011), a
disorder associated with dysregulated motor abilities. The alternative class of psychiatric
medications can result in a substantial increase in weight gain and develop into morbid obesity.
Therefore, further understanding of the etiology of how these side effects develop could benefit
the understanding of drug action at receptors, enhance personalized treatment and improve
outcome of patient quality of life.
1.2 Schizophrenia: Epidemiology
The prevalence and incidence of SCZ has been reported to be variable across geographical
regions. The prevalence, or the proportion of patients diagnosed with SCZ in a given time period,
has been estimated to be between 2.7/1000 and 8.3/1000 persons (Messias et al., 2007). The
approximate lifetime prevalence estimate in the U.S. and Canada is one per one hundred (Eaton
& Chen, 2006). The incidence rate, or the likelihood of developing the disease over a lifetime, is
estimated to be between 0.11/1000/year and 0.70/1000/year individuals (Messias et al., 2007).
SCZ has been reported to affect males and females at a different rate, with a male-to-female ratio
of 1.4:1 (McGrath et al., 2008).
1.3 Schizophrenia: Etiology
The underlying pathophysiology or specific biological cause of SCZ is currently unknown. No
established biological markers or tests exist that can conclusively identify whether a patient is
5
afflicted with the SCZ. Overall, the literature demonstrates that familial factors are important
contributory elements to the development of SCZ, though the exact genetic mechanisms remain
elusive (van Os & Kapur, 2009; Kennedy et al., 2003). As such, both genetic and environmental
factors have been proposed to contribute to the development of SCZ (Sullivan et al., 2003).
1.3.1 Environmental Risk Factors: Schizophrenia
A number of environmental factors have been reported to contribute to the onset of SCZ, and
these may occur at different developmental periods throughout an individual’s lifetime. Factors
that may leave an individual vulnerable to SCZ during the prenatal period include exposure to
influenza (Brown et al., 2006), maternal rubella and respiratory infections (Brown et al., 2006),
low socioeconomic status (Dohrenwend et al., 1992), obstetric complications (Murray & Lewis,
1987), urban residence (Marcelis et al., 1999), prenatal famine (Susser et al., 1996), season of
birth (Torrey et al., 1997) and low folate intake (Marzullo & Fraser, 2005). Post-natal
environmental risk factors for SCZ include living in an urban environment (Krabbendam & van
Os, 2005; McGrath et al. 2008) and cannabis use (van Os, et al., 2002; De Sousa et al., 2013).
1.3.2 Genetic Risk Factors: Schizophrenia
Schizophrenia has a well-established genetic component, though the likelihood of a single major
gene for risk of the disease is highly unlikely. Schizophrenia is a complex polygenic disease, and
almost certainly has several, if not many, contributing genetic factors (Kennedy et al., 2003;
Schizophrenia Psychiatric Genome-Wide Association Study (GWAS) Consortium, 2011).
Individuals with a family member afflicted with SCZ are more likely to develop the disease
6
(Gottesman & Wolfgramm, 1991), and if both biological parents suffer from SCZ, an individual
has a 44% risk of developing the disease as well. The concordance rate for SCZ development
between siblings is relatively high for dizygotic twins (17%), and for monozygotic twins (48%)
(Gottesman & Wolfgramm, 1991). Based on a number of twin studies, SCZ has also been
reported to have a heritability rate as high as 80% (Gottesman and Shields, 1991; Sullivan et al.,
2003; Owen et al., 2004).
Though it appears that a genetic component contributes to the predisposition of developing SCZ,
concrete genetic mechanisms have not yet been determined. Over the past decades, a large
number of small genetic effects have been associated with SCZ. Taken together, these genetic
variants have been proposed to explain 23% of the variance in SCZ risk (Lee et al., 2012).
Recently, the Psychiatric Genetics Consortium (PGC) reported a five-fold increase for risk for
SCZ based on a polygenic model incorporating several thousand genetic variants from GWAS
studies (Sullivan et al., World Congress of Psychiatric Genetics, Oct. 2012, Hamburg, Germany).
A subset of individuals diagnosed with SCZ has more distinct genetic abnormalities, including
22q11.2 deletion syndrome (Murphy et al., 2002) and 1q21.1 microdeletion syndrome (Stefanson
et al., 2008).
Several candidate gene studies for risk of SCZ have been reported (Ripke et al., 2011). The
serotonin system, particularly the serotonin 2A receptor (5HTR2A), has been implicated as being
associated with SCZ (Inayama et al., 1996; Polesskaya & Sokolov, 2002). The growth and
differentiation factor, neuregulin 1 (NRG1), has been reported to be among the top validated risk
genes for SCZ, psychotic and bipolar disorders (Douet et al., 2014). The neuregulin 1 gene was
first reported to be associated with SCZ in a sample consisting of Icelandic and Scottish patients
(Stefansson et al., 2002), and this finding has been consistently replicated (Stefansson et al.,
7
2003; Williams et al., 2003; Yang et al., 2003; Bakker et al., 2004; Tang et al., 2004; Corvin et
al., 2004; Munafò et al., 2006; Georgieva et al., 2006; Walker et al., 2010). The promoter SNP
NRG1 rs6994992 has also been of high interest, given that this variant has been reported to
predict psychosis (Hall et al., 2006) and can alter promoter activity (Tan et al., 2007).
The disrupted-in-Schizophrenia 1 gene (DISC 1) is also of high interest in the context of studies
examining risk for SCZ. An examination of a large family with high incidence of SCZ, identified
genes and chromosomal translocation that resulted in the truncation of the DISC 1 protein
(Millar et al., 2000; Morris et al., 2003). The DISC1 gene has been implicated to be contributory
to SCZ, as this gene affects neuronal functions dependent on proper cytoskeletal regulation,
neuronal migration, neurite architecture and intracellular transport (Morris et al., 2003).
Recently, the rs1625579 SNP, an intronic putative primary transcript that encodes for the
microRNA 137 (MIR137), was the top hit for a genome-wide association study assessing
common variants for SCZ risk (Ripke et al., 2011). Follow-up work to the association between
the MIR137 and SCZ has been reported in several studies. The MIR137 risk genotype has been
reported to strongly predict earlier age-at-onset of psychosis, and SCZ patients with the MIR137
risk genotype were also reported to have reduced white matter integrity (Lett et al., 2013). A
study by van Erp et al., (2014) reported that the MIR137 rs1625579 TT genotype was associated
with the SCZ risk phenotype dorsolateral of prefrontal cortex hyperactivation, which is
commonly considered a measure of brain inefficiency relevant to SCZ. However, the exact role
of the MIR137 rs1625579 SNP and its effect on white matter microstructure in SCZ patients
requires further study. It is still unknown exactly how the MIR137 rs1625579 SNP mediates risk
for SCZ, though this finding appears to be promising and warrants further investigation.
8
Of interest, one of the classic candidate genes for SCZ, namely the dopamine D2 receptor
(DRD2), has now emerged as significant in the most recent genome-wide association study
(Ripke et al., 2014). The D2 is the only receptor that is a target of all antipsychotic drugs
(Seeman et al., 1976). Dopamine is generally well known to have a role in modulating
antipsychotic response, and is associated with a reduction in positive symptoms. The dopamine
system is an integral component in neural reward circuitry, and thereby may affect appetitive
behaviours. If the dopamine system is altered by D2-blocking medication (i.e., antipsychotics),
this may result in a blunted reward perception, and in turn result in subsequent altered behaviours
such as over-eating. Specifically, patients may not feel satiated after ingesting food content, and
continue to eat in order to feel a sense of reward. Currently, classical first- and second-generation
antipsychotic medications are the main treatment for SCZ, and the detrimental side effects, for
example, weight gain, that occur with the use of these medications remain a troubling issue.
1.4 Actions of Antipsychotics at Receptor Targets
Overall, the majority of antipsychotic medications represent a ‘pharmacological shotgun’
approach to receptor binding, in that they bind to several different targets (Meltzer et al., 1989;
Kapur et al., 2001; Lett et al., 2012). Many antipsychotic medications have a high affinity for the
serotonin 5-HT2A and 5-HT2c receptors, as well as an affinity toward histaminergic,
dopaminergic and many other receptors (see Figure 1). The first antipsychotic medication used to
treat SCZ was chlorpromazine, which was introduced in the 1950s (Meyer & Simpson, 1997).
Chlorpromazine has been proposed to act as an antagonist on dopamine D2 receptors (DRD2),
an action that has been suggested to reduce positive symptoms of SCZ. In past decades, first-
9
generation antipsychotic medications, also called typical antipsychotics, have been used for
SCZ treatment. These medications have been reported to act primarily on DRD2 (Seeman et al.,
1976; Remington et al., 2006). Second-generation antipsychotic medications have a high
affinity for 5-HT2 receptors, though their atypical qualities (e.g., reduced extrapyramidal
symptoms) have been attributed to rapid disassociation from the DRD2 (Kapur et al., 2001).
Though second-generation antipsychotic medications are effective in treating positive symptoms
as well as reducing the risk of developing extra pyramidal side effects, a major health risk
remains when these medications are used for treatment. With the use of second-generation
antipsychotic medications comes the risk of substantial weight gain, or antipsychotic induced
weight gain (AIWG) as well as a number of metabolic abnormalities including impaired glucose
and lipid levels (Müller & Kennedy, 2006). The risk of weight gain is variable across individuals
and among different second-generation antipsychotic medications. The exact biological
processes by which weight gain and metabolic abnormalities occur are elusive not yet well
understood. However, differing receptor-binding profiles have been proposed as a contributing
reason to explain drug-induced weight gain. Those medications that have a relatively high
blockade of 5-hydroxytryptamine (serotonin) receptor 2C (5-HT2C) and histamine 1 receptors
(HR1) have been reported to be associated with higher weight gain, whereas medications that
primarily block DRD2 and dopamine receptor D3 (DRD3) receptors are associated with lower
weight gain. Among the drugs that have the highest propensity for weight gain are clozapine,
where the action at the dopamine receptor D4 (DRD4) receptor was initially proposed to partially
explain its weight-causing properties (Wong & Van Tol, 2003), and olanzapine, where its action
at muscarinic 3 (M3) receptor has been suggested for its diabetes risk characteristic (Weston-
Green et al., 2013; Johnson et al., 2005; Silvestre & Prous, 2005). However, it is unknown which
10
exact specific pharmacological actions are responsible for weight gain and associated metabolic
disturbances. A recent meta-analysis conducted by Lett et al. (2012) maintains that both
olanzapine and clozapine result in more weight gain in comparison to other commonly
prescribed medications, and labeled the two drugs as “high risk.”
The atypical antipsychotic drugs have been characterized by their affinities for serotonergic
receptors and lower affinities for the D2 receptors (Kapur & Seeman, 2001). Both the typical and
atypical antipsychotic drugs interact with a various number of receptors including the
serotonergic 5-hydroxytryptamine serotonin receptor 1A (5-HT1A), 5-hydroxytryptamine
serotonin receptor 2A (5-HT2A), 5-hydroxytryptamine serotonin receptor 2C (5-HT2C), 5-
hydroxytryptamine serotonin receptor 6 (5-HT6), and 5-hydroxytryptamine serotonin receptor 7
(5-HT7) (Roth et al., 1992, 1994, 1998), dopaminergic (DRD2, DRD3, and DRD4) (Van Tol et
al., 1991; Seeman et al., 1997), histaminergic receptors (HR1) (Kroeze et al., 2003), and
adrenergic receptors adrenoceptor alpha 1A (ADRA1A), and adrenoceptor alpha 2B (ADRA2B)
(Nutt, 1994; Hertel et al., 1997, 1999a).
The receptor targets of antipsychotic medications provide a starting point to understand AIWG,
but may not necessarily provide all of the answers. Second-generation antipsychotic medications
and their affinities for different receptor targets are summarized below.
11
Table 1-1. Receptor Affinities of Commonly Prescribed Second-Generation Antipsychotic Medications
Adapted from: Ananth et al., 2004; Onrust et al., 2001; Davis et al., 2002; Leucht, et al., 2009; Komossa et al.,
2010; Brunton et al., 2010; Komossa et al., 2010; Taylor et al., 2012; Leuct et al., 2013; Kishi et al., 2013.
D: dopamine, α: alpha-adrenergic, H: histamine, M: muscarinic 5-HT: serotonin
**** very high affinity, *** high affinity, ** moderate affinity, *low affinity, (--) negligible affinity
Medication D1 D2 D3 D4 5HT2C 5HT2A α α H1 M1 M3
Clozapine ** ** ** *** **** **** **** *** **** **** ***
Olanzapine *** *** *** *** *** **** ** ** **** **** ****
Risperidone * *** ** *** ** **** ** ** ** -- --
Quetiapine * ** ** * * * *** *** **** ** ***
Aripiprazole * **** *** * ** *** ** * ** -- --
Ziprasidone *** *** *** *** *** **** *** ** ** -- --
12
1.5. Schizophrenia and Pharmacogenetics of Antipsychotic
Induced Weight Gain
1.5.1 Contributing Factors to Antipsychotic-Induced Weight Gain
To date, there are no predictive biomarkers with high enough sensitivity and specificity to
determine, with clinical usefulness, whether an individual is susceptible to AIWG. Some known
predictors of AIWG include baseline weight, antipsychotic treatment and duration (Jones et al.,
2001; McIntyre et al., 2001). One of the strongest predictors for AIWG includes family history in
a first-degree relative (Gebhardt et al., 2010). AIWG is without doubt a polygenic phenotype, in
that multiple polymorphisms contribute to this phenotype, and each variant contributes a
relatively small amount to the overall variance. Overall, genetic factors are among the leading
predictors for whether individuals are susceptible to developing this side effect (Gebhardt et al.,
2010; Lett et al., 2012). Further understanding of genetic determinants of AIWG could lead to
greatly improved and more personalized patient treatment regimes.
1.5.2 Gene Systems and Antipsychotic-Induced Weight Gain
Both dopamine and serotonin systems are key regulators of appetite, satiety and food reward.
Second-generation antipsychotic medications have been shown to act upon several receptor types
including serotonergic and dopaminergic receptors. Thus, dysfunction within the serotonergic
and dopaminergic systems has been proposed as potential mechanisms for the onset of AIWG.
13
To date, serotonergic gene variants have been the most extensively explored system in the field
of AIWG. The 5-HT receptors are generally known to regulate weight, appetite and satiety in
healthy populations. The HTR2C promoter polymorphism rs3813929 (−759 C/T) has been
consistently reported to be associated with AIWG, and a meta-analysis showed an association
between this single nucleotide polymorphism (SNP) and AIWG (Sicard et al., 2010). Altered
functioning of dopamine system gene variants has also been proposed as a potential mechanism
for AIWG. Dopamine D2 receptor (DRD2) SNPs have been a primary interest for studies in
AIWG. A recent meta-analysis by Müller et al., (2012) reported that the three DRD2 SNPs,
rs6277 (C957T), rs1079598 and rs1800497 (TaqIA), were associated with AIWG. A functional
promoter region polymorphism in DRD2, the -141C Ins/Del (rs1799732), has been associated
with antipsychotic-induced weight gain SCZ patients treated with either olanzapine or
risperidone (Lencz et al., 2010). This study was conducted in a small sample set (n=58), and
further work in larger samples is required.
Both serotonin and dopamine system genes have been in the spotlight as potential genetic
mechanisms in AIWG, due to their role as primary targets for antipsychotic medications as well
as their critical roles in weight regulation in the general population. Thus far, the 5-HT2C
promoter polymorphism rs3813929 (−759 C/T) has been the most consistently replicated SNP
with AIWG (Lett et al., 2012). However, the associations between serotonergic genes and this
phenotype do not necessarily completely explain the substantial weight gain observed in a
number of psychiatric patients. For instance, the second-generation antipsychotic medication,
aripriprazole, has a high affinity for the 5-HT2C receptor, yet has low weight gain liability
(Nguyen et al., 2012). Thus, it is very likely that serotonergic genes act in tandem with other
genetic factors to regulate the development of AIWG. In addition to dopamine and serotonin,
14
other candidate mechanisms for association with AIWG include the histamine system variants,
as well as the melanocortin system.
1.5.3 Dopamine and Antipsychotic-Induced Weight Gain
The dopamine system, specifically the dopamine D2 receptor (DRD2), has been extensively
studied in its role as a primary target of antipsychotic action. Some antipsychotic medications
have been proposed to alter central reward circuitries, leading to impaired reward responses,
which may be associated with addictive behaviours. The DRD2 Taq1A allele has been
implicated in addictions including cocaine (Noble et al., 1993), alcohol (Blum et al., 1990) and
nicotine use (Munafò et al., 2009), and to a lesser extent, general obesity (Carpenter et al., 2013).
Antipsychotic medication use may result in alteration of binge eating symptomatology, which
may lead to increased food intake (Theisen et al., 2002). It has been postulated that, in a subset of
individuals, atypical antipsychotic medications may alter brain reward circuitry to make food
appear more palatable, which leads to weight gain.
The DRD2 gene encodes the dopamine D2 receptor, and is the most commonly examined gene
in the context of AIWG. In addition, polymorphisms across the other dopamine receptor genes
dopamine receptor D1 (DRD1), dopamine receptor D2 (DRD2), dopamine receptor D3
(DRD3), dopamine receptor D4 (DRD4), dopamine receptor D5 (DRD5), have been investigated
with mixed implications for AIWG. Lane et al. (2006) reported no significant association
between the DRD1, DRD2, and DRD3 SNPs with weight gain in a patient sample treated with
risperidone. A recent analysis conducted by Müller et al. (2012), examined polymorphisms
across the DRD1-DRD5 receptors using tagSNP software. In this study, a nominally significant
association was reported between the DRD2 rs1079598 SNP and the categorical variable of
15
AIWG with a 7% weight gain cut-off (P = 0.03). In a sub-sample stratified for ethnicity, and
medications with the highest propensity for weight gain (clozapine & olanzapine), associations
with weight gain and three DRD2 SNPs, rs6277 (C957T), rs1079598 and rs1800497 (TaqIA),
were reported. The rs6277 SNP is of particular interest given that it has been reported to have a
functional role in vitro and in vivo. The rs6277T allele has been reported to be associated with
decreased mRNA stability and half-life, which in turn was associated with decreased D2 affinity
in the striatum (Duan et al., 2003; Hirvonen et al., 2009). The −141C Ins/Del (rs1799732) DRD2
SNP has been reported to be associated with decreased efficacy of antipsychotic treatment
(Lencz et al., 2010). In addition, this SNP was associated with increased weight in a sample
consisting of first-episode SCZ patients (Lencz et al., 2010). The ‘C’-allele of the rs4436578
SNP has been associated with weight gain in a Chinese SCZ sample treated with clozapine,
olanzapine or risperidone (Hong et al., 2010). The DRD3 receptor gene has not yet been widely
investigated in AIWG, though a recent study reported a non-significant trend between the DRD3
9Gly allele and risperidone-induced weight gain in a population of Spanish SCZ patients
(Almoguera et al., 2013).
The DRD4 has been a receptor of interest in SCZ studies, as it has been reported to have a
stronger affinity for clozapine, compared to DRD2 (Van Tol et al., 1991). However, DRD4 has
been proposed to have a minor role in antipsychotic response and regulation (Wong & Van Tol,
2003). The DRD4 hypofunctional 7-repeat (7R) allele of the exon 3 VNTR has been reported to
confer greater risk for weight gain (p = 0.003) (Popp et al., 2009). However, a study conducted
by Müller et al. (2012) reported no significant association between DRD4 7R allele and AIWG.
Both studies consist of relatively low sample sizes and this variant needs further exploration in
this phenotype.
16
Overall, the role of dopamine in AIWG remains very compelling. A number of postulated
mechanisms of the involvement of dopamine in AIWG have been described in the literature.
Antipsychotic action may mediate hyperphagia via blockade of D2 receptors expressed in the
hypothalamus (Baptista et al., 1987; Baptista, 1990; Parada et al., 1988). The blockade of
dopamine in the anterior pituitary can result in hyperprolactinaemia, which has been correlated to
body mass index in patients being treated with first-generation antipsychotics (Kalinichev et al.,
2005). Finally, arguably the most common proposed DRD2-AIWG mechanism is that the neural
reward circuitry is affected by antipsychotic administration. Striatal D2 receptor availability is
inversely associated with weight gain (Wang et al., 2006), and it has been suggested that
decreased levels of dopamine D2 receptors result in obese individuals to seek reward or
reinforcement in the form of food intake (Roerig et al., 2011).
Recent proposed mechanisms include the peripheral dopamine signalling being altered in
response to antipsychotic use. Melanocortin expression within DRD1 expressing neurons has
recently been explored in terms of body energy homeostasis. The melanocortin-4 receptor
(MC4R) signalling in DRD1 neurons has been reported to suppress food intake and weight gain
in mice (Cui & Lutter et al., 2013). Though preliminary, this finding illustrates a potential
interactive mechanism between the melanocortin and dopamine system that could further
promote understanding of AIWG.
17
1.5.4 Serotonin and Antipsychotic-Induced Weight Gain
Second-generation antipsychotic medications have been reported to have the highest propensity
for weight gain among patients, and are also potent inverse agonists of serotonin (5-HT)
receptors. In terms of weight regulation, the 5-HT receptors expressed within the brain are highly
integrated and influential in hunger and satiety circuitries. Evidence of the role of the 5-HT in
weight regulation includes the finding that d-fenfluramine, a 5-HT agonist, functions as a strong
appetite suppressant (Leibowitz & Alexander, 1998). Among the serotonin receptor genes, the 5-
HT2C rs3813929 promoter SNP (-759C/T) has been the most widely investigated variant that
has been associated with AIWG (Ellingrod et al., 2005; Miller et al., 2005). To a lesser extent,
the 5-HT1RA, 5HTR2A, 5HTR2C and HTR6 have been reported to have a role in AIWG
(reviewed in Mueller & Kennedy, 2006; Lett et al., 2012).
The X-linked 5HT2C rs3813929 (C/T) promoter SNP (-759 C/T) has been reported to be
associated with AIWG across several studies (Reynolds et al., 2002; Templeman et al., 2005;
Ellingrod et al., 2005; Miller et al., 2005; Godlewska et al., 2009; Opgen-Rhein et al., 2010;
McCarthy et al., 2005; Kuzman et al., 2008; De Luca et al., 2007; Sicard et al., 2010; Mulder et
al., 2009). A meta-analysis conducted by De Luca et al. reported no association between AIWG
and the rs3812929 C-allele. However, another meta-analysis conducted by Sicard et al. (2010)
reported the C-759T-G-697C-S ser23Cys haplotype was over-represented in SCZ patients with
weight gain. To date, the 5-HT2C rs3813929 (C/T) SNP remains among the well-studied
candidates in AIWG.
The 5HT2C rs3813929 ‘T’-allele has been associated with significantly less weight gain
compared to the ‘C’-allele in antipsychotic-treated patients (Reynolds et al., 2002). Some
18
subsequent studies determined no association between the rs3813929 SNP and body weight
(Basile et al., 2002; Tsai et al., 2002). Nonetheless, the number of positive associations and
meta-analyses (Sicard et al., 2010) seem to suggest a role for the 5-HT2C rs3813929 in AIWG.
Furthermore, the ‘T’-allele has been reported to reduce promoter activity (Hill et al., 2011),
while the ‘C’-allele has been reported to reduce transcriptional activity (McCarthy et al., 2005;
Buckland et al., 2005; Yuan et al., 2000). The exact action of the 5HT2C rs3813929 SNP is
currently unknown, though the disruption of DNA-protein interaction due to antipsychotic
treatment may be influential in relation to the onset of AIWG. Finally, the rs3812929 SNP has
been reported to be more strongly associated with AIWG in individuals who have been treated
with medications for less than three months, or a relatively short duration (Reynolds et al., 2002;
Ellingrod et al., 2005; Templeman et al., 2005; Opgen-Rhein et al., 2010).
Aside from the HTR2C gene, the 5-hydroxytryptamine serotonin receptor 2A (5-HT2A) gene
has also been reported to be associated with AIWG. The rs6313 (102 T/C) has been associated
with weight gain in a study assessing SCZ patients treated with olanzapine (Gunes et al., 2009).
Furthermore, carriers of the -1438A-102T- 452His haplotype had significantly higher C peptide
levels, a peptide found in insulin, compared to haplotype 3 (-1438A, 102T, and 452Tyr) carriers.
However, no association among 5-HT2A SNPs and serum triglyceride, insulin or cholesterol
levels was reported (Gunes et al., 2009). The 5-HT2A genrs6313 TT genotype has also been
associated with weight gain in a Han Chinese sample of SCZ patients treated with risperidone
(Lane et al., 2006). The 102T allele of 5-HT2A has been associated with olanzapine-induced
weight gain in a Japanese patient population (Ujike et al., 2008). Overall, the 5-HT2A gene is
relatively under-studied in terms of AIWG, but warrants replication and further exploration in
independent sample sets.
19
The overall findings indicate that the 5-HTR2C rs3812929 SNP is a genetic determinant for
AIWG in patients of European descent treated with second-generation antipsychotics for a
relatively short duration. The exact functional role of the 5-HTR2C rs3812929 SNP is unknown,
and the mechanism by which this SNP may contribute to the regulation of AIWG is difficult to
determine. Other genetic mechanisms may act with the serotonin system or independently to
mediate the AIWG effect observed in psychiatric patients.
1.5.5 Histamine System
The histamine system, namely the histamine 1 receptor (H1R), has a well-described role in both
general obesity and AIWG (Reynolds, 2012). Histamine neurotransmitters bind to the
postsynaptic H1Rs , which result in the activation of downstream effectors, resulting in
decreased food intake. Administration of antipsychotics such as olanzapine and clozapine, which
are inverse agonists of the H1 receptors (H1R), result in increased food intake in animal models
(Humbert-Claude et al., 2012). Furthermore, Kroeze et al. (2003) reported that both clozapine
and olanzapine have a high affinity for the H1R, relative to eleven other receptor targets. Thus, it
appears that the effect of these drugs at the H1R is worthy of investigation in order to determine
if antipsychotic action at this histamine receptor has a role in AIWG.
In order to demonstrate specifically which receptors were associated with weight gain, Kroeze et
al. (2003) assessed typical (chlorpromazine, perphenazine, trifluoperazine, thioridazine,
thiothixene, fluphenazine, haloperidol, molindone) and atypical antipsychotics (clozapine,
olanzapine, loxapine, sertindole, risperidone, ziprasidone, quietapine, and aripiprazole) for their
affinities to 12 cloned bioegenic amine receptors. It was reported that the high H1R affinities
were associated with antipsychotics that were highly correlated with gain, and the medications
20
associated with low weight gain were found to have low H1R affinities. The H1R affinity was
the single variable that reliably predicted weight gain across all 17 tested antipsychotic
medications. Two histamine 2 receptor polymorphisms (−1018 G/A) and H3R (332Ser/Ser) were
also investigated in an Asian sample for AIWG, and not associated with weight gain (Ujike et
al., 2008). A more recent study examined an association for H1R SNPs with body mass index as
well as glycolated hemoglobin (Hb1Ac) in 430 patients treated with second-generation
antipsychotics (Vehof et al., 2011). Two H1R SNPs (rs346074 and rs346070) were examined,
and a significant interaction was found between the haplotype rs346074-rs346070 and high body
mass index (BMI) (p=0.025) in patients treated with antipsychotics that have a high affinity for
the histamine receptor. The investigation of the histamine system is relatively under-explored in
AIWG, especially in comparison to the dopamine and serotonin system genes. Nonetheless, the
findings described above assessing histamine receptors have been further examined in
combination with the enzyme 5' AMP-activated protein kinase (AMPK) in order to determine
how treatment with antipsychotic medications may result in substantial weight gain.
AMP-kinase mediation through histamine receptors
The 5' AMP-activated protein kinase (AMPK) is a heterotrimer enzyme comprised of catalytic α-
subunits and regulatory β-and γ subunits. The AMPK functions as a gage for monitoring cellular
energy status (Minokishi et al., 2004). The AMPK has a critical role in regulating energy balance
and metabolism, and is mediated by the cellular AMP: ATP ratio (Minokishi et al., 2004).
Clozapine treatment has been shown to result in a four-fold increase in AMPK phosphorylation
in wild-type mice, but has no effect in mice lacking the H1R (Kim et al., 2007). The findings
from the study by Kim et al. (2007) provide evidence that antipsychotic medications such as
clozapine activate hypothalamic AMPK activity, and that this action is depleted when mice lack
21
the H1R. Olanzapine administered to female rats has also been shown to elevate AMPK levels of
phosphorylation (Sejima et al., 2011). Total body weight and food intake was also significantly
elevated after a two-week period (Sejima et al., 2011). Acute administration of olanzapine in rats
has recently been shown to result in hepatic insulin resistance in addition to increased levels of
phosphorylation of AMPK (Girault et al., 2012).
The AMPK and its subunits (subunits that are coded by the protein kinase, AMP-activated,
alpha 1 catalytic subunit (PRKAA1), protein kinase, AMP-activated, alpha 2 catalytic subunit
(PRKAA2), protein kinase, AMP-activated, beta 1 non-catalytic subunit (PRKAB1), protein
kinase, AMP-activated, beta 2 non-catalytic subunit (PRKAB2), protein kinase, AMP-activated,
gamma 1 non-catalytic subunit (PRKAG1), protein kinase, AMP-activated, gamma 2 non-
catalytic subunit (PRKAG2), protein kinase, AMP-activated, gamma 3 non-catalytic subunit
(PRKAG3) genes have been examined in the context of genetic association studies. The intronic
PRKAA1 rs10074991 polymorphism was significantly associated with change in BMI, in a
German SCZ patient sample treated with antipsychotic medications (Jassim et al., 2011). A study
conducted by Souza et al. (2012) investigated schizoaffective patients and evaluated ten AMPK
subunit polymorphisms across the regulatory sub-units. A nominally significant association
between rs3766522 in PRKAB2 (AA vs. AT + TT; p = 0.022) and rs10789038 in PRKAA2 (GG
+ GA vs. AA, p = 0.023) with weight change (%) in patients of European ancestry following
treatment with clozapine or olanzapine was found.
The AMPK gene variants have also been reported to have a role in Type II Diabetes
susceptibility. Variants in the PRKAA2 gene have been associated with increased cholesterol
levels, as well as impaired glucose and lipid metabolism (Horikoshi et al., 2006; Spencer-Jones
et al., 2006) in patients with type 2 diabetes. Though these studies did not investigate AMPK
22
gene variants in a psychiatric population, it appears these SNPs have an influential role not only
in weight gain, but metabolic indices as well.
These exploratory genetic association studies must be interpreted with caution, but provide
evidence in line with animal models that antipsychotic action may result in AMPK
dysregulation. The elevated levels of AMPK phosphorylation due to antipsychotic administration
may result in increased fatty acid synthesis resulting in the rapid and substantial weight gain
observed across humans and rodent models, and this effect may be mediated by antipsychotic
action at the H1R. Most importantly, in the context of this AIWG, the finding that clozapine had
no effect on AMPK phosphorylation in H1R knockout mice suggests that histamine receptors
may have a mediating role on the development of side effects (Kim et al., 2007).
1.5.6 The Adrenergic System in Antipsychotic-Induced Weight
Gain
The adrenergic system may be implicated in the regulation of AIWG, given that a number of
second-generation antipsychotic medications have affinity for the α1- and α2-adrenergic
receptors. Some studies have indicated that the ratio of α2- to β3-adrenergic receptors expressed
in the adipose tissue may regulate the onset of obesity through adipocyte hyperplasia (Lane et al.,
2006; Saiz et al., 2008). The protein coding adrenoceptor alpha 1A (ADRA1A) gene was
assessed in a sample of Chinese SCZ patients treated with SGAs. The tagSNP software was used
to analyze the ADRA1A gene in this sample, and numerous SNPs were associated with AIWG
(Liu et al., 2010). In addition, the α2A adrenoceptor (ADRA2A) ‘C’-allele of the 1291G/C
23
(rs1800544) polymorphism was significantly associated with higher weight in a Korean
population (Park et al., 2006). Our group also reported that the rs1800544 SNP was associated
with SCZ patients treated with SGAs (Sickert et al., 2009). Overall, SGAs exhibit a high affinity
for adrenergic receptors; therefore, it is important to examine this gene system in the context of
AIWG.
1.6 The Genetics of Obesity
1.6.1 General Obesity
General obesity is a complex phenotype that has been more comprehensively studied in the
literature in comparison to AIWG. One standard definition by The World Health Organization
(WHO) for overweight and obesity is abnormal or excessive fat accumulation that may impair
health (Donald & Behan, 2012). A general understanding of factors that may contribute to the
development of substantial weight gain in the general population may provide insight into how
the AIWG phenotype develops.
Similar to AIWG, it has been suggested that the onset of obesity is a complex and dynamic state,
resulting from both environmental and genetic factors. The heritability of obesity has been
estimated to range between 64% and 84% (Stunkard et al., 1986), and has also been examined in
twin studies. The rates of heritability have been reported to be similar between twins reared apart
and twins reared together (Price & Gottesman, 1991). A large study examining 5000 Danish
subjects reported a statistically significant relationship between the body mass index (BMI), a
common measure of weight, of adopted children and their biological parents. No association was
observed between adopted children and adoptive parents (Stunkard et al., 1986b). Interestingly, a
24
close correlation between the BMI of adopted and their biological siblings who were reared
separately was also later reported (Sorensen et al., 1989). Thus, genetic factors have been known
to influence obesity for some decades.
The prevalence of general obesity has been reported to be increasing worldwide, with rates
ranging between 10% and 35% among adults (Hainer et al., 2008). Obesity has been defined as
an excess of body fat, and this state can be measured using dual energy X-ray absorptiometry as
well as isotopic dilution techniques (Goodpaster et al., 2002). However, given costliness, these
techniques are not typically used in general obesity studies. Rather, measures such as body mass
index (BMI), weight in kg/height in m2, are commonly used to assess weight in obesity studies,
and have been correlated to body fat content (Farooqi & O'Rahilly, 2006). Other measures
including waist-to-hip ratio and leptin, which correlates with body fat mass (Comuzzie et al.,
2001), have also been reported as measures of obesity and weight across studies. However, body
mass index remains among the most commonly used assessment of weight in obesity studies.
Clinically, obesity is measured as an excess accumulation of adipose tissue, which results in BMI
that is > 30 kg m−2.
The ideal method approach in measuring a complex phenotype such as obesity is a complex
matter. The measure of BMI follows a normal distribution, with no clear division between
individuals who are ‘clinically obese’ (BMI>30) and the non-obese (O'Rahilly & Farooqi, 2006).
Similarly, in studies that assess AIWG, individuals who exhibit greater than 7% weight gain
during treatment are considered to be affected. This is essentially considered to be dividing
individuals into cases versus controls. Unfortunately, defining the dependent variable in this
capacity may not elucidate the underlying genetic mechanism, or reflect how the biology of
25
weight is regulated. Examining a dichotomous variable in complex phenotypes such as obesity
and AIWG may result in a loss of information, as this method may significantly decrease
statistical power. It has been suggested that statistical information may be lost if transforming
continuous data to a discontinuous scale, and it may be beneficial to measure weight in a
quantitative method (Comuzzie et al. 2001).
Other contributing factors toward the development of obesity include both an individual’s rate of
energy expenditure and food intake. Measuring exact food and nutrient intake is difficult in an
experimental setting. Nonetheless, twin studies have also shown that intake of nutrients, size and
frequency of meals, as well as preference for particular foods is also likely to be governed by
genetic factors (Wade et al., 1981).
Genetic association studies have provided information on the etiology of obesity, though no
single variant has been attributed as casual (Loos et al., 2008; Speliotes et al., 2010). More
recently, genome-wide association studies (GWAS) have revealed novel loci that may be
implicated in obesity. For instance, the SH2B adapter protein 1 (SH2B1), and fat mass and
obesity-associated (FTO) genes have been reported to be significantly associated with changes in
body mass (Thorleifsson et al., 2008; Willer et al., 2009). Despite the strong association between
a gene such as FTO and obesity, this gene does not appear to be associated with AIWG (Shing et
al., 2014), suggesting that obesity and AIWG may be regulated by differing biological
mechanisms.
A large-scale GWAS study examining obesity in 249,796 individuals reported that 32 loci were
associated with this phenotype, while other genome-wide association studies have reported the
melancortin-4 receptor (MC4R) rs17782313 SNP, located 188 kb downstream of the MC4R
26
gene, to be significantly associated with body mass index (Loos et al., 2008). A subsequent
GWAS study conducted by Zhao et al. (2011) reported that polymorphisms across the fat mass
and obesity (FTO), transmembrane protein 18 (TMEM18), neurexin 3 (NRXN3), melanocortin-4
receptor (MC4R), SEC16 homolog B (SEC16B), glucosamine-6-phosphate deaminase 2
(GNPDA2), TNNI3 interacting kinase (TNNI3K), glutaminyl-peptide cyclotransferase-like
(QPCTL), and brain derived neurotrophic factor (BDNF) genes, were associated with common
childhood obesity.
Though these genetic association findings were highly compelling, some limitations should be
considered. The FTO and MC4R genes are considered as top obesity hits across GWAS studies
(Frayling et al., 2007; Loos et al., 2008). However, the additive genetic effect of the FTO and
MC4R genes was reported to be very low, in that the combination of these variants were reported
to account for less than 2% of the variance in adult BMI (Willer et al., 2009; Bogardus et al.,
2009). Thus, the question of the missing heritability for obesity was highlighted, and was
attributed to factors such as the complex, multi-genic nature of obesity, as well as the possibility
of copy number variation (CNV) contributing to obesity. The number of loci reported in the
GWAS studies illustrates the daunting complexity of phenotypes such as obesity and AIWG.
Nonetheless, looking closely at biological markers that influence weight regulation can provide
valuable insight into the mechanisms and processes that govern obesity.
27
1.6.2 Environmental Factors and Obesity
Genetic factors may certainly predispose certain individuals to gain weight, however,
environmental influences should also be considered when studying obesity. Accurate assessment
of gene-environment interaction in obesity and weight phenotypes could provide further
understanding of underlying biological mechanisms of weight gain. However, measuring
environmental factors including diet can be difficult to ascertain, given that exposure to obeso-
genic environments will be different across individuals (O'Rahilly & Farooqi, 2006).
A well-described hypothesis of why modern-day individuals are susceptible to obesity is the idea
of the “thrifty” genotype, which refers to our ancestors and their ability to efficiently store fats
during periods of limited access to food resources. This situation may have been common for our
ancestors, and obesity may have actually been a result of natural selection (Neel et al., 1962).
Individuals who were more adept at storing fat would have been more likely to survive during
long periods of famine (Qi & Cho, 2008). However, in a modern environment, the ‘thrifty’
genotype results in abundance of food resources, increased intake and then obesity, and perhaps
the development of dangerous diseases such as type 2 diabetes. The problem lies in determining
exactly which genes predispose an individual to obesity, given that recent studies have shown
that 32 loci are associated with this phenotype (Speliotes et al., 2010), with each gene
contributing a small effect to the overall variance. Nonetheless, there are some reports of using
knowledge of the genetics of obesity to moderate lifestyle intervention.
28
1.6.3 Gene-Environment Studies in Obesity
A number of studies in the past decade have examined whether specific gene-environment
interactions can influence weight gain. For instance, the Trp64 Arg variant of the beta-3
adrenergic receptor (ADRB3) gene was shown to interact with high energy intake, which led to a
significant increase in obesity (Miyaki et al., 2003). Another approach includes examining the
interaction between genetic variants and physical activity. Of interest, Andreasen et al. (2008)
reported an interaction between physical activity and the FTO rs9939609 SNP in a sample
consisting of Danish participants. Among the physically inactive subjects, significant differences
in BMI were reported between the AA and TT genotypic groups.
In a study consisting of obese patients (BMI > 30), the patients were prescribed a lower-calorie
diet, which was associated with a 25% reduction in spontaneous energy intake for 2.5 months,
and a leptin SNP (−2549C >) was nominally associated with lower BMI loss (Mammes et al.,
1998). In a follow-up study, the leptin receptor SNP Ser (T) 343 Ser(C) ‘C’-allele was
associated with weight loss in women who were prescribed a low-calorie diet (Mammes et al.,
2001). The melanocortin-3 receptor (MC3R) was also reported to have a nominal effect on the
low-calorie diet and exercise on weight loss (Santoro et al., 2007).
Overall, the genetics of obesity encompass a vast number of studies and have been greatly
explored in the general literature (Walley et al., 2006). While genetic factors alone do not appear
to be the sole regulating components of obesity, they do play an important role in understanding
the etiology. The approaches taken to understand genetic factors in obesity, and the subsequent
findings, provide an effective springboard to begin to understand how AIWG may be regulated.
29
1.7 Leptin Genetics in Antipsychotic-Induced Weight Gain
1.7.1 Leptinergic System as a Critical Regulator of Weight Gain
Several studies over the past decades have reported dozens of peptides, hormones and other
molecules that are all critical for the regulation of food and nutrient intake, energy homeostasis,
satiety and appetite. These molecules all interact in complex ways to determine weight and
metabolic measures. The hypothalamus has been identified as a key regulator of energy and
weight regulation. The leptin hormone, a critical regulator of weight, has downstream effects
upon receptors that are heavily expressed within the hypothalamus. Dysregulation of leptin has
been reported to play a role in the pathogenesis of obesity. In the late 1990s, it was discovered
that the ob gene encodes the hormone leptin, which is primarily expressed in adipose tissue. The
ob/ob knockout mice have been reported to exhibit hyperphagia, decreased energy expenditure
and decreased immune function, traits that are also exhibit by starved animals (Friedman &
Halas et al., 1998). When leptin is administered to ob/ob knockout mice, these effects are
reversed. Lean mice have been shown to decrease in adiposity when treated with leptin infusions.
Thus, these initial findings determined a key physiological for leptin in informing an organism’s
body during food deprivation (Friedman & Halas et al., 1998).
Leptin receptors are expressed in various hypothalamic nuclei: the arcuate nucleus, lateral
hypothalamus, dorsomedial hypothalamus and paraventricular nucleus (reviewed in Cone et al.,
2005). Leptin acts in a feedback loop in order to maintain stores of body fat. If an organism is
starved and their body fat decreases, this leads to decreased levels of circulating leptin. As a
30
result, food intake will exceed energy expenditure. If an organism’s adiposity increases, leptin
levels also increase. Leptin acts as part of a feedback loop to maintain constant stores of fat.
During periods of starvation or decreased adiposity, leptin levels decrease, which in turn leads to
a state of positive energy balance wherein food intake exceeds energy expenditure. Conversely,
an increase in adiposity leads to an increase in the levels of leptin and a state of negative energy
balance (Cone et al., 2005).
Within the hypothalamus, several different neuropeptides and neurotransmitters interact with
leptin to regulate biological processes such as food intake and energy expenditure. The cocaine
amphetamine regulated transcript, melanocortin-4 receptor, promelanin concentrating hormone,
agouti related protein, neuropeptide-Y and ghrelin are examples of the most critical known
molecules to regulate leptin signalling. Though the exact mechanism of how the leptin-
melanocortin system is affected in humans when treated with antipsychotic medications is not
well understood, olanzapine administration in rats has been shown to result in an increase in
weight, along with levels of neuropeptide Y and reduced proopiomelanocortin expression
(Weston-Green et al., 2012).
1.7.2 Leptin Genetics in Antipsychotic-Induced Weight Gain
In addition to dopamine and serotonin (described above), the leptin gene has also been
extensively studied in AIWG. The gene-encoding leptin and leptin receptor have been of great
interest, given that leptin has an established role in hunger and appetite signalling. To date, two
SNPs, the functional promoter −2548A/G polymorphism of the leptin gene, as well as
223Gln/Arg polymorphisms in the leptin receptor, have been reported to be associated with
AIWG. The most prominent and consisting finding with leptin and AIWG is the association with
31
the rs7799039 (−2548A/G) SNP (De Luca et al., 2007; Sicard et al., 2010). In 2008, Zhang et al.
(2008) reported the −2548A/G G-allele was associated with AIWG in a male subset of patients.
In addition, Gregoor et al. (2011) was able to demonstrate that the ‘G’-allele was associated with
AIWG in a sample comprised of males of European ancestry. These findings are suggestive of a
sex-specific effect for the (−2548A/G) SNP. An additional study by Calarge et al. (2009)
reported the opposite allele, ‘A’-allele, as being associated with AIWG in an adolescent sample.
It has been suggested the −2548A/G SNP may influence weight gain differently between age
groups (e.g., adolescents vs. adults) (Reynolds et al., 2006; Lett et al., 2012). However, a recent
study by Nurmi et al. (2013) reported that the G-allele of the rs7799039 SNP resulted in
increased weight gain compared to A-allele homozygotes (P=1.4 × 10−4) in a dominant model.
Thus, the overall findings regarding the leptin promoter SNP rs7799039 (−2548A/G) indicate the
GG genotypic group is the risk genotype for AIWG across samples.
The leptin receptor, primarily expressed in the hypothalamus (Leinninger et al., 2009), has also
been examined within pharmacogenetic and AIWG studies. The LEPR 223Arg allele has been
reported to be associated with lower risk for developing obesity in a patient sample consisting of
adolescents treated with risperidone. The LEPR rs817983 (656N/K) SNP has been associated
with weight gain resulting from risperidone treatment (Ruanŏ et al., 2007).
Leptin and leptin receptors remain compelling candidates to explore, and the involvement of this
system is relevant to study in the development of AIWG. In addition to the leptin receptor, the
hormone leptin has multiple targets that have not yet been extensively investigated in AIWG.
32
1.7.3 Cocaine and Amphetamine Regulated Transcript
The cocaine amphetamine regulated transcript (CART) is a neuropeptide highly expressed in the
nucleus accumbens, hypothalamus and ventral tegmental area. The CART is reported to be a
modulator of drug reward and drug re-enforcement (Jaworski & Jones, 2006; Jean et al., 2007).
Though the role of CART in neural reward circuitry is its most extensively studied role, this
neuropeptide is also an important mediator of energy homeostasis. It was initially suggested that
CART had a role in food intake (Koylu et al., 1997), which was later supported by the finding
that CART immunoreactivity was observed in brain areas that mediate feeding behaviour
(Elmquist et al., 1999). More recently, it has been demonstrated that CART is co-expressed with
another critical peptide involved in weight regulation, the neuropeptide Y, in the dorsomedial
hypothalamus. Leptin has been shown to stimulate neurons within the dorsomedial
hypothalamus, though its direct influence on CART remains a topic of further investigation (Lee
et al., 2013). Interestingly, clozapine has been shown to reduce CART expression in the nucleus
accumbens, and it has been suggested that the dysregulated food intake is induced by
introduction of antipsychotic medications in rats (Beaudry et al., 2004).
1.7.4 Endocannabinoid System
The endocannabinoid system has recently been brought into the spotlight as a compelling
contributing mechanism to weight and obesity. Stimulation of cannabinoid 1 receptors (CNR1)
results in an increase of food intake and in body weight. The CNR1 receptors are stimulated by
ghrelin and inhibited by leptin. The CNR1 antagonist, rimonabant, in combination with a healthy
diet, has shown reductions in weight, waist circumference and cardiometabolic risk factors, and
animal models also provide evidence for the role of the CNR1 in weight regulation (Jamshidi &
33
Taylor, 2001; Ravinet Trillou et al., 2003). Mice without CNR1 expression show reduced eating
or hypophagia as well as resistance to obesity. In human studies, the CNR1 and acid amide
hydrolase (FAAH) genes have been examined in terms of obesity and related phenotypes. In a
study by Monteleone et al. (2010), the cDNA 385C/A (rs324420) SNP of the FAAH gene was
reported to be associated with greater than 7% weight change in a sample of European ancestry
patients diagnosed with psychosis. No association was found between the CNR1 359 G/A
(rs1049353) SNP and weight gain. In 2010, Tiwari et al. conducted a study examining tag SNPs
that provided improved coverage of the CNR1 gene. The CNR1 rs806378 T-allele was
associated with second-generation AIWG in SCZ patients of European ancestry (P = 0.008).
They then determined that the T-allele creates a binding site for the transcription factor
arylhydrocarbon receptor translocator (ARNT), suggesting a functional effect. Finally, Nurmi et
al. (2013) replicated the CNR1 SNP rs806378 T-allele association with body weight gain in a
pediatric sample treated with risperidone for eight weeks. Overall, the role for the CNR1 gene in
AIWG is among the most compelling findings in this field.
1.7.5 Neuropeptide Y
The neuropeptide Y (NPY) is an integral component of the feeding and appetitive neural
circuitry governed by the hypothalamus, and has been described as a potent stimulator of food
intake (Stanley et al., 1986). The NPY is expressed through the central nervous system, including
the hypothalamic nuclei, and has been associated with altered feeding behaviour. Rats with over-
expression of NPY resulted in increased body weight and lower metabolic efficiency (Zheng et
al., 2013). In terms of AIWG, the neuropeptide Y receptor Y5 (NPY5R) rs6837793 SNP has been
associated with weight gain in patients treated with risperidone (Ruaño et al., 2007). In the same
34
study, no association was reported between this SNP and in patients treated with olanzapine. The
functional NPY promoter SNP rs16147 is associated with higher waist-hip ratio in women
(Mutschler et al., 2013). In a recent study conducted by Tiwari et al. (2013), five polymorphisms
(rs10551063, rs16147, rs5573, rs5574 & rs16475) in the NPY gene were investigated in SCZ
patients treated with olanzapine or clozapine. Three NPY SNPs (rs16147, rs5573 & rs5574) were
significantly associated with weight change, suggesting that NPY may have an influence on the
onset of AIWG.
1.7.6 Ghrelin
The hormone ghrelin mediates perception of hunger by acting upon receptors expressed within
the hypothalamus. Ghrelin acts upon ghrelin receptors, or growth hormone secretagogue receptor
(GHS-R), located in the arcuate and paraventricular nuclei of the hypothalamus, as well as NPY
receptors (Heiman & Witcher, 2006). In terms of antipsychotic action, it is currently unclear
whether ghrelin has a major role in terms of regulating AIWG. Atypical antipsychotic
medications have been associated with decreased morning fasting ghrelin levels in the first two
weeks of treatment, and increased ghrelin levels for longer treatment durations (Sentissi et al.,
2008; Jin et al., 2008). Overall, it has been suggested that ghrelin is up-regulated by second-
generation antipsychotic medications, and that altered ghrelin signalling may contribute to the
onset of AIWG (Yan et al., 2013).
Thus far, two published reports have shown that SNPs within the ghrelin gene were not
associated with AIWG (Ujike et al., 2008; Ruano et al., 2007). However, low coverage of the
35
gene and sample sizes are limitations for these studies. Our group recently analyzed 18
polymorphisms across the ghrelin and ghrelin receptor genes, and also report no significant
association with AIWG (Chowdhury & Shams et al., in prep).
1.7.7 Agouti Related Protein
The AGRP has a critical role within feeding behaviour, as the primary endogenous antagonist of
melanocortin receptors, and its expression is down-regulated by leptin (Haskell-Luevano et al.,
2001; Nijenhuis et al., 2001). Animal models provide evidence that the AGRP has a role in
obesity, as AGRP over-expression in mice results in obesity, increased body length, hyperplasia,
as well as hyperglycemia and hyperinsulinemia (Graham et al., 1997; Michaud et al., 1997)
(Argyropoulos et al., 2002; Bonilla et al., 2006). In the context of AIWG, its exact role is
unclear. Plasma levels of AGRP were reported to be no different between controls and SCZ
patients treated with either ziprasidone or olanzapine, despite increases in both body mass and
leptin levels (Ehrlich et al., 2012). We recently conducted a study investigating the role of AGRP
in AIWG and reported no association in our local sample set (Chowdhury et al., 2014).
Nonetheless, the AGRP is a critical regulator of weight and requires consideration in the context
of AIWG.
36
1.7.8 Brain Derived Neurotrophic Factor
The brain derived neurotrophic factor (BDNF) has a widespread role in the central nervous
system of regulating the development of neurons, as well as hippocampal neurogenesis and
plasticity. Circulating levels of BDNF have been reported to be altered by the use of
antipsychotics (Lipska et al., 2003). The BDNF 66 Val/Met polymorphism (rs6265) has an
integral role in mental illness and cognitive function (Green et al., 2011). BDNF rs6265 has been
associated with body mass index and eating behaviour in a GWAS study (Thorleifsson et al.,
2009). Taken together, these findings provide a strong rationale study BDNF in the context of
AIWG. The BDNF Val66Met (rs6265) polymorphism is the most well-studied SNP in the
literature, and has been reported to be associated with antipsychotic-induced weight gain in
males (p = 0.01) (Zhang et al., 2008). In this study, it was reported that the BDNF gene was a
risk factor for weight gain in male SCZ patients on long-term antipsychotic treatment. A recent
study conducted by Zai et al. (2012) found no association between the BDNF rs6265 and AIWG.
However, a two-marker haplotype including the rs6265 and rs1519480 SNPs were reported to be
significantly associated with weight change (Zai et al., 2012). The BDNF is a downstream
effector of leptin and appears to have several regulatory roles aside from weight gain, elucidating
the complexity of the AIWG phenotype.
1.7.9 Pro-Melanin Concentrating Hormone
The pro-melanin concentrating hormone (PMCH) is the pre-cursor of melanin concentrating
hormone (MCH), and MCH levels are higher in leptin-deficient mice compared to controls (Qu
et al., 1996). Knock-out PMCH mice are typically lean, hyperphagic and exhibit an increased
metabolic rate (Shimada et al., 1998). The PMCH rs7973796 ‘G’-allele has been shown to be
37
associated with increased odds of obesity in the SCZ patient sample treated with olanzapine
(Chagnon et al., 2007). The PMCH and MCH also have a regulatory role in skin pigment
expression, but further understanding of its role in AIWG is warranted.
1.8 The Melanocortin System Genes
The melanocortin system is a neuronal pathway that is a critical regulator of energy homeostasis
(reviewed in Cone et al., 2005). The unique circuitry projects and receives input from a vast
number of peptides, regulators and hormones, and is largely expressed within the hypothalamus.
The melanocortin system is among the forefront of investigated biological factors for obesity
drug development, yet relatively understudied in the context of AIWG.
Peripheral hormone signals including leptin, insulin and ghrelin act upon a number of targets,
including neuronal bodies expressed in the hypothalamus. Both POMC and CART cell bodies
are expressed within the arcuate nucleus of the hypothalamus. Neurons expressing POMC send
descending projections to the brainstem, innervating the nucleus of the solitary tract, and the
medulla, nucleus ambiguous and spinal cord (reviewed in Cone et al., 2005). The POMC peptide
produces the α-melanocyte stimulating hormone, a primary activator of melanocortin receptors.
The POMC neurons are also targets of another critical regulator of weight, the serotonin 5-HT2C
receptor (5HT2CR). Serotonergic compounds, including fenfluramine, activate POMC neurons,
and administration of 5HT2CR agonists result in increased POMC expression levels in rodent
models (Xu et al., 2011). Another critical upstream regulator for the melanocortin peptides are
the NPY/AGRP-expressing neurons, and AGRP is an endogenous antagonist for both the
38
melanocortin-3 and melanocortin-4 receptors (see Figure 1). In mice with NPY/AGRP ablation,
rapid decreases in food intake and body weight have been reported (Xu et al., 2005). The
NPY/AGRP expressing neurons also produce γ-aminobutyric acid (GABA), which inhibit
POMC neuron action, thus suppressing the anorexigenic effects of POMC. In a mouse model,
mice lacking vesicular GABA transporter (VGAT) in NPY/AGRP-related protein neurons
exhibit reduced body weight and GABA release (Vong et al., 2011).
39
Figure 1-1. Schematic Diagram of the Leptin-Melanocortin System
Reprinted with permission from Molecular Psychiatry (Lett et al., 2012).
The melanocortin system has been reported to have an integral role in terms of studies on obesity
and metabolic abnormalities (Cone et al., 2005). The five melanocortin receptors (MC1R-
MC5R) have differential physiological roles. The melanocortin-1 receptor (MC1R) is a regulator
40
of pigmentation and anti-inflammation. The adrenocorticotropic hormone (ACTH) is the ligand
for the melanocortin-2 receptor (MC2R), which regulates steroidogenesis, and the melanocortin-
5 receptor (MC5R) plays a role in sebaceous gland section (further described in Yang et al.,
2003; Cone et al., 2005). In terms of weight regulation, the melanocortin-3 receptor (MC3R) and
the melanocortin-4 receptor (MC4R) have been the focus of several obesity studies over the past
decade (reviewed in Santini et al., 2009) (see Table 1-2).
Adapted from Yang et al., (2003)
Table 1-2. Melanocortin Receptor Expression and Function
Receptor Primary Areas of Expression Function
Melanocortin-1 Receptor (MC1R) melanocyte anti-inflammation
Melanocortin-2 Receptor (MC2R) adrenal Cortex steroidogenesis, pigmentation
Melanocortin-3 Receptor (MC3R) brain, intestinal tract, placenta Weight homestasis, reproduction,
cardiovascular function
Melanocortin-4 Receptor (MC4R) hypothalamus, spinal cord weight homestasis, sexual function
Melanocortin-5 Receptor (MC5R) exocrine glands skeletal muscle,
adipose tissue
exocrine gland secretion
Overall, the associations of both MC3R & MC4R and obesity provide a basis to study the
melanocortin system in AIWG. If common variants of the MC3R and MC4R genes have a strong
41
regulating component within obesity, then common variants of these genes may also influence
susceptibility and/or onset to AIWG.
Both MC3R and MC4R interact with leptin to regulate appetite, satiety and food intake
signalling. Leptin is secreted from adipose tissue, and activates specific hypothalamic regions
including pro-opiomelanocortin (POMC) expressing neurons. The pre-cursor polypeptide
hormone (POMC) is cleaved into several peptides, including the α, β and γ melancortin peptides,
which stimulate receptors such as the MC4R and MC3R expressed in the paravantricular nucleus
of the hypothalamus. Overall, the activation of the melanocortin receptors result in increased
weight, as demonstrated in animal models and human genetic studies (Santini et al., 2009).
Mice with inactivated MC4R exhibit obesity and hyperphagia as well as altered metabolic
indices including hyperglycemia. In humans, MC4R mutations have been shown to result in
obesity, though the association varies between different ethnic populations. A frameshift
mutation was associated with obesity in a study that examined 63 severely obese children (Yeo
et al., 1998; Vaisse et al., 1998). MC4R mutations have been reported to be the most common
single cause of monogenetic obesity and may be responsible for a range 0.5–4% across different
ethnic populations (Miraglia et al., 2002). Recently, Lee et al. (2012) reported that MC4R
mutations appeared to be more prevalent in patients of Northern European descent. A study by
Lubrano-Berthelier et al., (2003) quantitatively analyzed the effect of a mutation on MC4R cell
surface expression, and reported that 81.3% of pediatric obesity-associated heterozygous MC4R
mutations led to intracellular retention of the receptor.
Two common MC4R variants have been reported to be protective against obesity (Val103Ile and
Ile25Leu). The V103I MC4R polymorphism was negatively associated with obesity and lower
42
body mass index in a sample consisting of 7937 participants (Heid et al., 2004). More recently,
the rs17782313 SNP was highlighted in a recent genome-wide association study investigating
large obese populations (Loos et al., 2008). In the context of AIWG, it is currently unknown
whether second-generation antipsychotic medication treatment has a direct impact upon the
MC4R. However, second-generation antipsychotic medications have been shown to have an
effect upon circulating leptin levels (Hägg et al., 2002; Atmaca et al., 2003). Patients treated with
clozapine and olanzapine have been reported to have increases in weight, body mass index and
leptin levels (Kraus et al., 1999). Given that leptin is stimulatory of MC4R, the exploration of its
role in AIWG is warranted. Antipsychotic action could have substantial downstream effects on
the two MC3R and MC4R receptors, resulting in the extreme weight gain observed in psychiatric
patients treated with second-generation antipsychotic medications. Our group collaborated with a
research group at Zucker Hillside Hospital, New York on a genome wide association study
(GWAS). The results of the GWAS illustrated twenty SNPs at chromosome 18, exceeding a
statistical threshold of P < 10-5 (see Figure 1-2).
Reprinted with permission from the Archives of General Psychiatry (Malhotra et al., 2012).
Figure 1-2. The MC4R
rs489693 genotype and
antipsychotic induced weight
gain in four samples of
subjects. The A/A genotype
was associated with
antipsychotic induced gain
across four separate samples.
Discovery represents initial
finding, Replic1 indicates the
first replication cohort;
Replic2, the second replication
cohort; Recplic3, the third
replication cohort.
43
The association between one SNP, rs489693, located near the melanocortin 4 receptor (MC4R)
gene, and AIWG was found in the discovery sample, as well as three additional independent
samples with rs489693 demonstrating consistent recessive effects. A subsequent meta-analysis
resulted in a genome-wide significant effect P =5.59 X 10-12 (see Figure 1-3). Thus, the MC4R
gene appears to have an important association with AIWG, and highly warrants investigation in
psychiatric pharmacogenetic studies.
Reprinted with permission from the Archives of General Psychiatry (Malhotra et al., 2012).
Figure 1-3. Results from Genome Wide Association Study. A) Manhattan plot displaying chromosal
position for single-nucleotide polymorphisms on x-axis, and −log10P values on y-axis. Peak values are
located on chromosome 18. B) Quantile-quantile plot displays statistical significance levels (−log10P
values) of correlation and trend tests for change in BMI in the discovery cohort, plotted against expected
values under the null hypothesis. With the exception of the most strongly associated SNPs on
chromosome 18, there is no deviation from the diagonal (λgenomic control = 1.00).
44
These data implicate MC4R in extreme SGA-induced weight gain and related metabolic
disturbances. A priori identification of high-risk subjects could lead to alternative treatment
strategies in this population.
The MC3R has a relatively less well-defined role in the context of obesity. Nonetheless, the
MC3R has been reported to influence weight and feeding measures. Recent reports have shown
that rare loss-of-function mutations of MC3R may partially indicate a role for this gene having a
regulatory role in obesity (Mencarelli et al., 2008; Tao et al., 2004). Among the most frequently
reported findings in human genetic studies are the reports of two common non-synonymous
SNPs (Thr6Lys and Val81Ile) in high-linkage disequilibrium being associated with childhood
adiposity (Feng et al., 2005). Interestingly, in vitro studies have shown that the two SNPs have
been associated with reduced function and protein expression in comparison to the wildtype
genotype (Feng et al., 2005). However, MC3R polymorphisms within the coding region and 5’
flanking sequences were reported to have no association with obesity in a sample assessing obese
women (Li et al., 2000). A case-control study investigation of MC3R and MC4R mutations
reported that the prevalence of rare MC4R variants was higher in obese participants compared to
controls, while no difference was reported for MC3R mutations between the two groups (Calton
et al., 2009). The MC3R rs3746619 SNP was examined in a study of SCZ patients treated with
second-generation antipsychotic medications, and no association was reported (Moons et al.,
2011). Altogether, the MC3R and MC4R appear to have non-redundant roles in regulating
weight and appetite, given that mice with abolished MC3R and MC4R gain more weight than
either MC3R -/- or MC4R -/-knock-out mice (Chen et al., 2000).
Given the prominent role of the melanocortin system in general obesity, namely the MC3R and
MC4R, it is surprising that these factors have not yet been extensively studied as possible
45
underlying contributors to the onset of AIWG. This research delves into the exploration of
identifying novel associations between melanocortin system gene variants and AIWG.
1.9 Methodologies and Statistical Considerations
1.9.1 Polymerase Chain Reaction and Genotyping
The primary method used for analyzing genetic associations presented in this work was
conducted by utilizing genotyping methodology. Genotyping with polymerase chain reaction
(PCR) allows for identification of a specific allele. DNA, which refers to any cell in an organism
that contains a nucleus, from individuals is required for the genotyping process. For the purposes
of our studies, DNA was extracted from blood cells, though it can be taken from saliva or cheeks
as well. The polymerase chain reaction (PCR) is a common molecular biology technique used to
amplify DNA in order to generate thousands to millions of copies of a pre-determined DNA
sequence. (Holland et al., 1991). Further details are provided in Chapters 3, 4 and 5, as PCRs
were utilized for all three experimental studies.
There are numerous ways to conduct a PCR within a laboratory setting, as well as various
alterations. For the purposes of this thesis, genotyping was conducting use distinct methods: 1)
TaqMan allele-specific assays, 2) Restriction Fragment Length Polymorphism and 3) Illumina
GoldenGate genotyping assays. Full descriptions and protocols are available within each chapter,
and general methodologies for the techniques are outlined below.
46
TaqMan allele-specific assay
The TaqMan assay used for our studies utilizes a general PCR guideline with some small
alterations with the procedure/mechanism (Bartlett et al., 2003). The TaqMan assays provide
optimized genotyping of SNPs. The assay uses a 5’ nuclease assay for detecting and amplifying
specific SNPS in purified DNA sequences. The forward and reverse primers amplify the
sequence of interest using two TaqMan probes. The TaqMan MGB probes provide a fluorescent
signal for the amplification of each allele. One probe labelled with the VIC dye detects Allele 1,
and the FAM probe detects allele 2 (Holland et al., 1991).
Restriction Fragment Length Polymorphism
Restriction Fragment Length Polymorphism (RFLP) is a relatively older technology, which we
utilized in Chapter 4. At times, it is not feasible to design or obtain primers via TaqMan
technology, due to non-binding of primers to a given fragment of a DNA sequence. This was the
case for our analysis of the POMC rs3754860 polymorphism. Thus, we used RFLP methodology
in order to genotype this SNP and determine an association with our phenotype. RFLP
distinguishes the difference in homologous DNA sequences, which can be detected by the
presence of fragments of different lengths after digestion of the DNA samples. The digestion is
conducted with the use of specific restriction endonucleases. Once the digestion is complete, gel
electrophoresis is conducted and ethidium bromide staining is used to visualize unique DNA
patterns at a pre-specified genotype.
(http://www.ncbi.nlm.nih.gov/projects/genome/probe/doc/TechRFLP.shtml).
47
Electrophoretic Mobility Shift Assay
The electrophoretic mobility shift assay is a methodology that can be used to determine the
existence of a potential interaction between DNA and protein. A control lane (DNA without a
probe) is used in one lane, and the visualization of the band on the gel will exhibit unbound
DNA. In the second lane, if a DNA fragment is able to bind to a protein, the DNA will bind to
the protein complex and a band will appear in the gel. (Fried et al., 1989; Smith et al., 2009).
1.9.2 Analyses, Sample Size and Power Calculations
Genetic Analyses
The primary statistical analyses used to assess differences between genotypic groups were
performed using SPSS 15.0. Categorical variables were tested using a χ² test, and continuous
variables were determined using both analysis of variance (ANOVA), and analysis of co-
variance (ANCOVA). Demographic and clinical characteristics of the samples assessed in our
studies were determined using ANOVA. Though BMI has been reported to be a consistent
measure across obesity and AIWG literature (O'Rahilly & Farooqi, 2006), patient height data
was unavailable across all of our local samples. Thus, we used the comparison of mean weight
change (%) across genotypic categories as our primary genetic association test. This method
incorporates baseline weight of patients into our calculations.
48
Haplotype Analyses
Linkage disequilibrium (LD) refers to the correlation across alleles located near each other,
which may be transmitted across generations, or descend from an ancestral chromosome (Reich
et al., 2001). This grouping of alleles at adjacent locations which may combine together, may
also be referred to as a haplotype. The level of LD between two or more SNPs is expressed as
either D’ or r2, with values closer to 1.0 representing relatively higher correlations (Li et al.,
2003).
In our studies, both a 2-marker and 3-marker sliding window haplotype analyses across genes of
interest (MC4R, MC3R) were analyzed. We used the sliding window approach so that we could
systematically explore additional haplotypes in areas of lower correlation or linkage
disequilibrium.
Gene-gene Interaction Analysis
The AIWG phenotype is multifactorial, with a number of genetic and environmental components
that may influence the onset of AIWG in a given individual. However, both gene-gene and gene-
environment interactions require analyses by specific statistical methods, aside from the
ANCOVA described above. When multiple genes are assessed for an interaction effect, a
number of cells may have very few or no data points, and this is referred to as the ‘curse of
dimensionality’ (Bellman et al., 1961). If there are too few data points, then large standard errors
may result, which can lead to an increase in the number of Type I errors (Concato et al., 1996;
Peduzzi et al., 1996; Hosmer and Lemeshow, 2000). The multifactor dimensionality reduction
49
(MDR) uses a data-reduction approach that can address the limitations of using parametric tests
in gene-gene interaction analyses (Hahn et al., 2003). More recently, the Model-Based
Multifactor Dimensionality Reduction (mbmdr) was proposed as a statistical tool, given that this
program can analyze quantitative traits and has increased power compared to the MDR program
(Cattaert et al., 2011). Thus, we utilized mbmdr for our gene-gene interaction analyses.
Power Calculations, Sample Size and Correction for Multiple Testing
Our current samples include 266 patients assessed for drug-induced weight gain. This sample
size will continue to increase over the next few years as more subjects are recruited. The power
of a genetic association depends on multiple variables. Factors that contribute toward statistical
power in genetic association studies include the degree of linkage disequilibrium in the
population, minor allele frequency, effect size, sample size, standard deviation or variance of the
population mean. Greater than 80% power is generally accepted in order to determine genetic
effects.
In genetic association studies, it is typical to divide empirical p-values by the number of tests
performed (Bonferroni correction) (Abdi et al., 2007). The Bonferroni method has been
suggested to be too conservative and applies to a global null hypothesis, as opposed to the
specific individual hypotheses. An additional consideration is that type I errors cannot decrease
without an increase in type II errors (the probability of accepting the null hypothesis when the
alternative is true), and type II errors are no less false than type I (Perneger et al., 1998). In the
case of our MC3R study (Chapter 5), the variants are highly correlated or linked, thus the
Bonferroni method was not an appropriate test to implement. Some tests that include the
50
calculation for the LD between SNPs include SNP Spectral Decomposition (SNPSpD) (Nyholt
2004) and False Discovery Rate (FDR) (Benjamini et al., 2001). The FDR test is more suitable
and appropriate for correcting for a large number of tests (e.g., assessing over 10,000 genes
across individuals in a study) (Benjamini et al., 2001). For the purposes of our studies consisting
of 266 patients, the SNPSpD was the appropriate test to apply for correction for multiple testing
in our study assessing the melanocortin-3 receptor (Chapter 5), where SNPs were highly linked.
51
Chapter 2 Hypotheses and Experimental Approach
52
2.0 Overview of Experiments, Hypotheses and Objectives
2.1.1 General Overview
The general aims of our research include the exploration of the effects of melanocortinergic
system gene variants in antipsychotic induced weight gain (AIWG), as well as any additive
effects of these genes. The genetic effects of leptin, serotonin and dopamine gene variants have
been extensively investigated in the literature, and several variants from these systems have been
examined on our samples. The present work will focus on downstream targets of leptin and
constituents of the leptin-melanocortin system that have been determined to have a role in weight
regulation in healthy populations. Our data are presented in three separate papers, two of which
have been published in peer-reviewed journals; the third is in preparation for publication.
2.2 Study One: Genetic Association Study between Antipsychotic-
Induced Weight Gain and the Melanocortin-4 Receptor Gene.
2.2.1 Background
The first paper (Chapter 3) focuses on the melancortin-4 receptor (MC4R). The purpose of this
study was to explore the potential effect of MC4R variation in the AIWG phenotype. Several
genome-wide association studies detected a signal on chromosome 18, and the MC4R
rs17782313 has been associated with obesity phenotypes in healthy populations (Loos et al.,
2008). However, the MC4R gene had not yet been investigated in psychiatric pharmacogenetic
53
studies. We hypothesized that MC4R variation influences AIWG, and that a gene association
will be found between MC4R SNPs and this phenotype.
An additional objective in this study included the investigation of a potential functional role of
the MC4R rs8087522 SNP, located in the promoter region of this gene. We hypothesized that
this SNP alters protein expression.
2.2.2 Hypothesis and Objectives of Study
Hypothesis: the MC4R gene variants are associated with AIWG. Melanocortin-4 receptor
(MC4R) mutations have been reported to be a cause of monogenic obesity, where one single
mutation is associated with disease, and common MC4R variants have been identified to be
associated with obesity (Loos et al, 2008).
2.3 Study Two: Investigation of Melancortin-System Gene Variants
in Antipsychotic-Induced Weight Gain
2.3.1 Background
The second paper (Chapter 4) examines the role between pro-opiomelanocortin, cocaine
amphetamine regulated transcript and agouti related protein in AIWG. The POMC, CART and
AGRP are integral to the leptin-melancortin system, and have been associated with general
obesity and weight gain (reviewed in Santini et al., 2009). The CART is an especially intriguing
target in the context of AIWG. It has been suggested that the CART peptide has a potential role
54
in dysregulated reward and food intake as a result of antipsychotic treatment (Beaudry et al.,
2004).
2.3.2 Hypothesis and Objectives of Study
Hypothesis: one or more variants across the POMC, CART and AGRP genes will be
associated with AIWG in our sample. The pro-opiomelanocortin, cocaine amphetamine
regulated transcript, and agouti related proteins have significant roles in general weight
regulation. The CART peptide has been reported to be influenced by clozapine administration in
a rodent model (Beaudry et al., 2004). However, the association between any of these three
genes and AIWG has remained relatively unexplored.
2.4 Study Three: An Exploratory Investigation of Melanocortin-3
Receptor Gene Variants in Antipsychotic-Induced Weight Gain
2.4.1 Background
The third paper (Chapter 5) examines the potential relationship between the melanocortin-3
receptor and AIWG. In addition, this paper explores the potential gene-gene interaction between
MC4R and MC3R with AIWG. This is based upon the finding that in animal models, MC3R and
MC4R have non-redundant roles in weight gain. Mice that lack both MC3R and MC4R receptors
gain more weight in comparison to MC3R -/- or MC4R -/- mice (Chen et al., 2000). Thus, we
will conduct pair-wise interactions between MC3R and MC4R SNPs. The melanocortin-3
receptor has two functional variants (rs3746619 & rs3827103), which have been associated with
55
obesity (Feng et al., 2005) and reduced in vitro protein expression. Given the biological roles of
MC3R and MC4R described above, gene-gene epistatic effects will occur between
melacortinergic genes and AIWG.
2.4.2 Hypothesis/Objectives of Study
We hypothesize that MC3R variants are associated with antipsychotic induced weight gain.
Furthermore, we hypothesize that an interaction between MC3R and MC4R SNPs in AIWG
will be detected. Ultimately, a gene-gene interaction analysis within the wider leptin-
melanocortin system may provide improved modelling of AIWG risk.
56
Chapter 3
Genetic association study between antipsychotic-induced
weight gain and the melanocortin-4 receptor gene
Contents of this chapter have been reprinted from:
Nature Publishing Group
Chowdhury NI1, Tiwari AK, Souza RP, Zai CC, Shaikh SA, Chen S, Liu F, Lieberman JA,
Meltzer HY, Malhotra AK, Kennedy JL, Müller DJ.The Pharmacogenomics Journal (2013) 13,
272–279; doi:10.1038/tpj.2011.66;
A link to the published paper can be found at
http://www-ncbi-nlm-nih-gov.myaccess.library.utoronto.ca/pubmed/22310352
57
3.1 Abstract
Antipychotic induced weight gain (AIWG) may result in the metabolic syndrome in
schizophrenia (SCZ) patients. Downstream variants of the melanocortin-4 receptor (MC4R) gene
have been associated with obesity in various populations. Thus, we examined single-nucleotide
polymorphisms (SNPs) in the MC4R region for association with AIWG in SCZ patients. Four
SNPs (rs2229616, rs17782313, rs11872992, and rs8087522) were genotyped in 224 patients who
underwent treatment for SCZ and were evaluated for AIWG for up to 14 weeks. We compared
weight change (%) across genotypic groups using analysis of covariance for three SNPs (r2=0.8).
European-ancestry patients who were rs8087522 A-allele carriers (AG+AA) on olanzapine
gained significantly more weight than non-carriers (P = 0.027, n = 69). These observations were
marginal after correction for multiple testing. We performed in vitro electrophoretic mobility
shift assay that suggested that the presence of the A-allele may create a transcription factor-
binding site. Further investigation is warranted for both these exploratory findings.
58
3.2 Introduction
Schizophrenia (SCZ) is a debilitating psychiatric disease that affects 1% of the general
population. The widely prescribed atypical antipsychotic medications are effective in treating
core positive symptoms of SCZ. Despite the success of atypical antipsychotic drugs in symptom
reduction, their clinical use remains burdensome. This is due to their association with substantial
weight gain and the metabolic syndrome. The metabolic syndrome is composed of several
factors, including insulinemia, abdominal obesity and hypertension (Meyer et al., 2005). Both
severe weight gain and the metabolic syndrome can have potentially fatal consequences,
including development of diabetes mellitus and coronary heart disease (Müller et al., 2006)
Studies have confirmed that the rates of the metabolic syndrome in SCZ samples are between 40
and 50% (McEvoy et al., 2005; Meyer et al., 2005; Callaghan., 2005). The underlying
mechanisms of onset of weight gain and the metabolic syndrome in SCZ patients have yet to be
elucidated. Approximately 30–40% of patients experience substantial weight gain and the
metabolic syndrome after initial treatment with second-generation antipsychotic drugs.
Discovery of genetic factors associated with metabolic side effects can lead to improved
treatment regimes and novel targets for drug development.
Genetic mechanisms are among the strongest known factors in regulating second-generation
antipsychotic (SGA)-induced weight gain. A high inter-individual variability exists between
patients who are treated with SGAs in terms of the side effect profile. This is demonstrated by
the finding that up to 30% of patients develop weight gain and the metabolic syndrome, whereas
some patients do not (Müller et al., 2006). One of the strongest reports supporting genetic
regulation of this side effect is from the recent twin and sibling study that reported heritability of
59
SGA-induced weight gain is between(h2 = 0.6–0.8) (Gebhardt et al., 2010). Thus,
pharmacogenetic-based algorithms have a high potential to identify individuals who are at risk
for development of antipsychotic-induced weight gain (AIWG). These, in turn, can greatly aid in
the decision-making process for antipsychotic treatment.
In general, obesity has become an epidemic in North America (Wang et al., 2006). The
pathophysiology of appetite and weight regulation is an area that is being extensively
investigated in order to develop counteracting strategies and treatment. The leptinergic system
has been reported to be heavily involved in physiological food intake and weight regulation. In
an intact homeostatic feedback system, leptin activates pro-opiomelanocortin neurons in the
arcuate nucleus of the hypothalamus. The pro-opiomelanocortin peptide is the precursor for
neuropeptides, including α-melanocyte-stimulating hormone (a-MSH) (Santini et al., 2009). The
α -MSH activates the melanocortin receptor-4 (MC4R), and the signals from MC4R-expressing
neurons are hypothesized to modulatevarious pathways through downstream effectors that result
in energy expenditure (Arch et al., 2005). Animal model studies utilizing gene targeting found
that mice lacking the MC4R receptors showed a mature-onset obesity syndrome, associated with
metabolic abnormalities such as hyper-insulinemia and hyperglycemia (Huszar et al., 1997; He et
al., 2010; Srisai et al., 2011). Furthermore, mutations of MC4R can cause monogenic obesity in
humans.
Recently, a more common genetic variant located 188 kb downstream from the MC4R gene
(rs17782313) has been associated with increased fat mass, weight and waist circumference (Loos
et al., 2008; Speliotes et al., 2010). The MC4R gene was recently identified as a gene for obesity
susceptibility in a genome-wide association study (GWAS) of 249 796 individuals (Speliotes et
al., 2010). The minor C-allele of this MC4R variant has previously been associated with higher
60
intake of total energy and dietary fat, resulting in greater long-term weight change in a healthy
population (Loos et al., 2008) Another well-studied polymorphism, the MC4R 103I (rs2229616)
marker, has been associated with decreased waist circumference along with other measures of
the metabolic syndrome (decreased glycosylated hemoglobin) (Brönner et al., 2006; Heid et al.,
2005).
Interestingly, the same rs17782313 marker was not significantly associated with AIWG in a
sample of pediatric and adolescent patients who were naıve to antipsychotic medication using a
GWAS approach (Malhotra et al., 2012) However, our group, in collaboration with Malhotra et
al., (2012) discovered and replicated an association between another marker at MC4R rs489693
and AIWG. The discovery cohort in this study consisted of antipsychotic-naıve adolescents who
were administered one of several antipsychotics (olanzapine, risperidone, quetiapine or
aripiprazole) and weight gain was measured for up to 12 weeks. The rs489693 marker was also
associated with weight gain in a sample comprising SCZ patients who had been given clozapine
with no prior second-generation antipsychotic treatment. The association of rs489693 was again
replicated in a third sample that consisted of SCZ patients who were prescribed antipsychotic
treatments in a naturalistic setting. Taking into consideration the strong positive associations of
MC4R polymorphisms with obesity and AIWG, we evaluated whether additional MC4R gene
polymorphisms are associated with AIWG. In this study, we investigated the possible association
between rs17782313 and rs2229616, which have been consistently implicated with obesity in the
literature (Loos et al., 2008; Speliotes et al., 2010; Brönner et al., 2006; Heid et al., 2008). In
addition, we also investigated rs8087522and rs11872992, which are both located in the promoter
region of the MC4R gene.
61
3.3 Materials and Methods
3.3.1 Subjects
A total of 224 patients with chronic SCZ or schizoaffective disorder were included in this study,
and sample characteristics have been described previously (Tiwari et al., 2010). Approval from
the institutional ethics committee and informed consent were obtained for all patients. Patients
were primarily recruited from three different sites. Sample-A: Charite University Medicine
(Berlin, Germany) (DJ Muller, n= 93). Patients were diagnosed with either SCZ or
schizoaffective disorder according to the DSM-IV and ICD-10 criteria. They were given
antipsychotic medication and assessed for up to 6 weeks (Table 1). Sample-B: Case Western
Reserve University in Cleveland (Ohio, USA) (HY Meltzer, n=76). DSM III-R-diagnosed SCZ
patients were longitudinally studied and were either treatment-refractory or intolerant to
treatment with typical antipsychotics. Prior to treatment with clozapine, patients underwent a
washout period of 2–4 weeks during which no medications were given unless deemed clinically
essential by the physician. The dosage of clozapine was not fixed and serum levels were
monitored to ascertain compliance. Patients were treated with clozapine for a minimum of 6
weeks and weight change (%) was calculated from baseline weight (for details see Masellis et
al., 1998 and Basile et al., 2002). Sample-C: Hillside Hospital in Glen Oaks (New York, USA)
(JA Lieberman, n=55). DSM-IV-diagnosed patients with chronic SCZ or schizoaffective disorder
that had shown suboptimal response to previous atypical anti-psychotic drug treatment were
included in this study. Despite previous treatment, these patients had continued to express
positive symptoms and poor cognitive functioning over a period of 2 years. In a 14-week double-
blind study, patients were randomly assigned to receive clozapine (500 mg day-1), olanzapine
62
(20 mg day-1) risperidone (8 mg day-1) or haloperidol (20 mg day-1) (for details see Volavka et
al., 2002 and Müller et al., 2005). Patient ethnicity was ascertained using self-reported ancestry
information over the last three generations.
3.3.2 Genetic Analysis
Venous blood (10–20 ml) was obtained from study participants and DNA was extracted at the
Centre for Addiction and Mental Health using the high-salt method (Lahiri & Nurnberger et al.,
1991). All single-nucleotide polymorphisms SNPs; rs17782313, rs8087522, rs11872992 and
rs2229616; Figure 1) were genotyped using TaqMan SNP genotyping assays (Applied
Biosystems, Foster City, CA, USA). PCR was performed in a final volume of10ml, using 20 ng
genomic DNA and 1 x TaqMan Universal PCR master mix. The following conditions were
applied for amplification: 95 °C for 10 min, followed by 50 cycles of 92 °C for 15 s and 60 °C)
for 1 min. Alleles were called using the Allelic Discrimination module of the Sequence
Detection software supplied with the ABI model 7500 machine. Genotyping was performed
blind to the subject’s weight gain status. Ten percent of the sample was randomly re-genotyped
for quality control and 100% concordant genotype calling was observed.
3.3.3 Data analysis
Statistical analyses were performed for individual SNP using SPSS 15.0. Categorical variables
were tested by χ² test. Continuous variables were tested by analysis of variance or covariance.
Weight change (%) from baseline was compared across the genotypic categories by analysis of
covariance using duration of treatment as a covariate. Distribution of age and sex across
genotypic groups was assessed. If either age or sex, or both, showed a significant effect, they
63
were included as covariates in the analysis of covariance. Haplotype blocks were examined and
weight change (%) with UNPHASED 3.1 (Dudbridge et al., 2003). Only haplotypes with
frequencies greater than 5% were included in the analyses. Linkage disequilibrium and Hardy–
Weinberg equilibrium were determined using Haploview 4.1 (Barret et al., 2006). Power
calculations were performed using Quanto 1.2.4. Assuming a minor allele frequency of 0.20 in a
sample size of n=224, in an additive model we had more than 80% power to detect a mean
weight difference of 1.5% between carriers and non-carriers of the risk genotype (Gauderman et
al., 2007).
3.3.4 Electrophoretic Mobility-Shift Assay
We performed an electrophoretic mobility-shift assay (EMSA) as described by Tiwari et al
(2010). We initially analyzed in silico for presence of binding sites using the Genomatix
MatInspector program. Biotin-labeled and unlabeled double-stranded oligonucleotides for each
allele of rs8087522 (A and G), and unlabeled oligonucleotides with the specific consensus-
binding site for the transcription factor for Paired Related Homeobox 2 (PRX2) and MSH
homeobox 2 (MSX2) were synthesized. Nuclear extract was prepared from SH-SY5Y
neuroblastoma cells (ATCC:CRL2266, Manassas, VA, USA) using the NE-PER Nuclear and
cytoplasmic extraction kit (Pierce, Rockford, IL, USA). The complementary pairs of nucleotides
were suspended in STE Buffer (10 mM Tris (pH 8.0), 50 mM NaCl, 1 mMEDTA) and annealed
using a thermocycler (Step-1: 95 °C for 5 min; Step-2: 95 °C (-1°Ccycle, 70 times); Step-3:
HOLD at 4 °C 1. The re-annealed double-stranded oligonucleotides (20 fmol) were incubated
with 2ml of the nuclear extract, 1 x binding buffer, 2.5% glycerol, 5 mM MgCl2, 50 ng ml-1poly-
(dIdC)and 0.05% NP-40 for 20 min at room temperature. For the competition experiments,
64
unlabeled oligonucleotides at 200-fold molar excess (4 pmol) were pre-incubated with the
nuclear extract before adding the biotin-labeled oligo-nucleotides. DNA–protein complexes were
fractioned on5% polyacrylamide gel in 0.5 x TBE for 35 min. Complexes were electrotransferred
to a nylon membrane at 380 mA for 45 min, and cross-linked at 120 mJ cm-2 for 45 s. The
biotinylated oligonucleotides were detected by chemiluminescence using the chemiluminescent
nucleic acid detection module of the Light Shift Chemiluminescent EMSA kit (Pierce). The
processed nylon membrane was exposed to an X-ray film for 45 s.
3.4 Results
3.4.1 Genetic analysis
The demographic characteristics of the study sample are presented in Table 1. Significant
differences in clinical variables were observed among clinical sites; however, similar rates of
change in weight were observed (Table 3-1). The three SNPs, rs17782313, rs8087522 and
rs11872992, analyzed in this study were in Hardy–Weinberg equilibrium (P >0.05; Table 3-2).
65
Reprinted with permission from the The Pharmacogenomics Journal (Chowdhury et al., 2012)
Table 3-1. Demographic characteristic from each of the clinical sites.
Characteristics Sample A
(n=93)a
Sample B
(n=76)a
Sample C
(n=55)a
Total
sample
(n=224)
p-value
Gender
Male
55
(59.1 %)
49
(64.4%)
46
(83.6%)
150
(66.9%)
0.008
Female 38
(40.9%)
27
(35.5%)
9
(16.4%)
74
(33.0%)
Age (years) 35.01±12.0 32.84±8.0 40.5±9.3 35.63±10.50 p<0.001b
Initial body weight
(kg) 80.7 ± 15.6 74.3 ± 13.6 84.82 ± 17.6 79.5 ±15.9 p<0.002b
Weight change (kg) 3.32 ±3.9 3.56± 4.5 4.53±6.5 3.69±4.8 0.570b
Weight change (%) 4.06 ± 4.7 5.09±6.4 5.85±8.4 4.85±6.4 0.555b
Study duration
(weeks) 5.09±1.6 6.00 ± 0 11.83 ± 3.7 7.03±3.4 p<0.001b
Ethnicity
European-
Americans
91
(97.8%)
55
(72.4%)
11
(20.0%)
157
p<0.001
African-Americans 2
(2.2%)
21
(27.6%)
33
(60.0%)
56
Others 0
(0%)
0
(0%)
11
(4.91%)
11
Treatment
Clozapine 12 76 11 99 p<0.001
Haloperidol 5 0 11 16
Olanzapine 15 0 21 36
Risperidone 28 0 12 40
Other 31 0 0 33 a The samples were included from three different clinical sites
b Kruskal-Wallis Test
66
Reprinted with permission from the The Pharmacogenomics Journal (Chowdhury et al., 2012)
Table 3-2. List of SNPs in MC4R tested for association with antipsychotic Induced Weight Gain at
Genotypic Level
Total sample
All Drugs
European Americans Clozapine/
Olanzapine
European Americans
Clozapine Only
SNP Genotype
Weight Change
(%)a, n=224
P-
value
b
Weight Change
(%)a
n=96
P-
valueb
WeightChange
%a
n=76
P-value
rs11872992 GG 5.03±6.6(176) 0.460 4.37±5.7(68) 0.560 4.37±5.8(57) 0.688
GA 3.49±4.7(34) 3.776±4.9(16) 3.09±4.1(12)
AA 7.80±10.2(3) 7.80±10.2(3) 2.01 ± 2.8 (2)
rs8087522 GG 3.99±5.8 (91) 0.231 3.22±4.8(42) 0.110 2.43±4.3(32) 0.080b
GA 4.73±6.3(89) 5.27±5.8(33) 5.55±6.01(27)
AA 6.73±6.4(27) 4.92±5.2(10) 4.92±5.2(10)
rs17782313 TT 4.97± 6.4(118) 0.306 4.50±5.9(51) 0.214 3.81±5.6(39) 0.582
TC 4.75± 6.5(73) 5.15±5.6(27) 5.24±5.5(24)
CC 4.62±6.0(10) 3.38±4.3(7) 3.41±5.2(5)
aMean ± SD (number of individuals).
bSNPs with low numbers in a genotypic group were merged with the heterozygous genotypic group.
cFor GG vs AA+AG. cAll the p-values were calculated using ANCOVA. rs8087522, European Americans on Clozapine: GG vs AA+AG,
2.43±4.3 vs 5.38±5.7, (p=0.027); Bonferroni corrected p= 0.081.
67
The rs2229616 SNP was monomorphic in our sample and thus was excluded from our statistical
analyses. No variation of the SNP was detected, and thus there was only one genotypic group to
assess. Duration of treatment varied across clinical sites and thus was entered as a covariate in all
association analysis. In the total sample, we observed no significant association between
genotypes and weight change (Table 2). However, our sample consisted of patients of different
ethnicities (mostly of European and African ancestry) and receiving drugs with different
propensities to cause weight gain. Clozapine and olanzapine have been reported to be associated
with the highest risk of clinically significant weight gain and show overlapping side-effect
profiles (Nasrallah et al., 2008).Thus, we performed a sub-analysis stratifying patients by their
self-reported ethnicities and whether they received clozapine or olanzapine. Patients classified in
the African American group were too low in number to allow analysis with sufficient power. In
the patients of European ancestry who received either clozapine or olanzapine, no significant
genotypic association was observed for rs11872992 and rs17782313, whereas a non-significant
trend was found for the rs8087522 marker, where the GA or AA genotypes gained more weight
than GG homozygotes (P=0.11) (Table 3-2).Both clozapine and olanzapine have a significantly
higher propensity to induce weight gain compared with other antipsychotic drugs (Nasrallah et
al., 2008).
We stratified the European-ancestry sample by individual drug type to uncover associations with
weight gain in relation to these two individual drugs. In patients of European ancestry who were
treated with clozapine (n=69), we found that carriers of the MC4R rs8087522 ‘A’-allele (AA
+AG) gained significantly more weight (5.38% ± 5.7) than the GG homozygotes (2.43% ±4.3;
P= 0.027) (Table 3). However, these results were not significant after taking into account the
multiple SNPs that we have analyzed (3 SNPs, Bonferroni-corrected, P=0.081) (see figure 3-2).
68
Figure 3-1. Association between
MC4R rs8087522
and antipsychotic
induced weight gain.
The MC4R
rs8087522 ‘A’-allele
reached a trend for
association with
weight gain in
patients of European
ancestry who were
treated with
clozapine: GG vs
AA+AG, 2.43±4.3 vs
5.38±5.7, (p=0.027);
Bonferroni corrected:
p= 0.081.
Allelic analyses also suggested that the ‘A’-allele carriers gained more weight (5.60% ± 5.4) than
carriers of the ‘G’-allele (3.3% ± 5.4; P = 0.04) (Table 3-2). There were no significant allelic or
genotypic associations observed with weight gain in patients of African American ancestry,
owing to the very small sample size. In the allelic analyses in the European sub-sample, the
minor C-allele of the rs17782313 marker, a trend for association (P=0.09) with weight gain was
observed in the combined sample of clozapine- and olanzapine-treated patients. A non-
significant trend for allelic association was also observed with rs11872992, whereas no allelic
association was observed for the rs8087522 marker (Table 3) despite its modest association in
the genotypic analyses. Linkage disequilibrium was calculated using Haploview 4.1. D’-values
P = 0.081
69
were high for all pairwise comparisons among the SNPs and r2was low (Figure 1). Two marker
haplotype analyses were conducted and no associations were found for the three SNPs and
weight gain ( P > 0.05) (Figure 3-1).
Reprinted with permission from the The Pharmacogenomics Journal (Chowdhury et al., 2012).
Figure 3-2. MC4R
(chr18q22) gene structure
and linkage disequilibrium
in European-ancestry
patients. Location of
polymorphisms with
respect to the exon. The
black filled box represents
the coding region and the
gray box represents the
non-coding region.
Linkage disequilibrium
observed in patients with
European ancestry.
Standard color scheme has
been implemented
(Haploview 4.1). The two
dark-shaded diamonds
represent D=1 and
LOD<2; the white
diamond, D’<1 and
LOD<2. The values in the
box represent r2.
3.4.2. Analysis of Electrophoretic-mobility shift assay
The SNP rs8087522, associated with clozapine-induced weight gain, is present in the promoter
region of MC4R and is in the vicinity of a genomic region that has been reported to code for
transcription factor-binding sites, according to the University of California Santa Cruz (UCSC)
genome browser. EMSA is an in vitro method to assess DNA–protein interactions that reflect
transcription factor-binding sites, which may be relevant to gene expression. An in silico
70
prediction of transcription factor binding using a 31-bp oligonucleotide surrounding the SNP
rs8087522 was performed using Mat Inspector (Genomatix, 2010). Presence of the A-allele was
predicted to bind to 20 transcription factors. The transcription factors PRX2-binding protein-
1(PRX2, matrix similarity of 1.00) and MSX2 protein (matrix similarity of 0.995) had the
highest similarity with the region encompassing the A-allele (Table 3-2). Therefore, these two
transcription factors were chosen for further analysis. An EMSA was performed and putative
binding of a nuclear protein to the A-allele was observed (Figure 3-2).
Reprinted with permission from the The Pharmacogenomics Journal (Chowdhury et al., 2012).
Table 3-3. Oligonucleotides for EMSA
Name Oligonucleotides Reference
rs8087522-A 5’-GCCAGTAGTGGTTCAATTAAAATACCTGAAA-3 This Study
Rev-5’-TTTCAGGTATTTTAATTGAACCACTACTGGC-3’
rs8087522-G 5’-GCCAGTAGTGGTTCAGTTAAAATACCTGAAA-3’ This Study
Rev-5’-TTTCAGGTATTTTAACTGAACCACTACTGGC-3’
PRX2 (S8) 5’- TAACTAATTAAC-3’ (Norris & Kern, 2001)
R5’-GTTAATTAGTTA-3’
MSX-2 F5’-GGCGAATTAGGAATTAGG-3’ (Towler et al., 1994)
R5’-CCTAATTCCTAATTCGCC-3’
71
Figure 3-3. EMSA for SNP rs8087522. Identification of nuclear proteins binding to the A-allele (lanes 1–
5) and G-allele (lanes 6–8). In the competition experiment, using 200-fold excess of unlabeled A-allele
containing an oligonucleotide, binding is significantly reduced, suggesting that binding of this nuclear
protein is specific to the A-allele. We tested two known transcription factors (PRX and MSX2) for
binding to the A-allele. Oligonucleotides containing specific binding sites for these transcription factors
(lanes 4 and 5) were unable to complete out theA-allele. EMSA, electrophoretic mobility-shift assay;
MSX, MSH homeobox; PRX, Paired Related Homeobox.
Reprinted with permission from the The Pharmacogenomics Journal (Chowdhury et al., 2012).
Oligonucleotides containing specific binding sites for transcription factors PRX2 and MSX2
were unable to compete out the labeled oligonucleotide with the A-allele, indicating that these
may not be the transcription factors binding at this site (Figure 3-2). Binding of nuclear protein
to the G-allele was likely to be result of non-specific binding as the unlabeled G-allele containing
the oligonucleotide could not disrupt this binding.
72
3.5 Discussion
In this study, based on the strong biological rationale for MC4R in weight regulation, we
investigated MC4R polymorphisms for their potential roles in AIWG. We investigated
rs17782313 and rs2229616, which have been associated with obesity previously (Loos et al.,
2008; Masellis et al., 1998), as well as two SNPs in the MC4R promoter region (rs8087522 and
rs11872992) that could have potential roles in gene expression.
The rs2229616 marker has been reported to be mildly protective against weight gain in the
general population (Brönner et al., 2006; Heid et al., 2008). However, it was monomorphic in
our sample and was therefore excluded from statistical analyses. We did not observe any
significant associations with the rs17782313 and rs11872992 polymorphisms and AIWG in the
total sample set. There was no significant association between rs17782313 or rs11872992 and
weight gain in a subset of European-ancestry patients who received clozapine or olanzapine
treatment. Notably, the rs17782313 minor C-allele had a non-significant trend for association
with weight gain in our allelic analyses in our European-ancestry patients (P=0.09). This same
allele has been previously associated with increased body mass index in a GWAS of 16 876
individuals of European descent (Renström et al., 2009) , However, the rs17782313 marker was
not significantly associated with weight gain or metabolic parameters in the recent GWAS
assessing a pediatric drug-naıve population exposed to antipsychotic medication (Malhotra et al.,
2012). Given that our hypothesis-driven study design and sample were different from those of
the pediatric GWAS investigation by Malhotra et al (2012) direct comparisons are not possible.
Overall, more work is needed to elucidate the role of this SNP in AIWG.
73
By contrast, we observed a modest association with the rs8087522 variant and AIWG in the
subset of European patients who were exposed to clozapine. Our genetic results indicated that
patients of European American ancestry who were carriers of the rs8087522 AA genotype were
more likely to gain weight when given clozapine for treatment. In the allelic analysis, a modest
trend of association with weight gain was found for carriers of the A-allele. This polymorphism
is in the promoter region of MC4R (Figure 3-1), suggesting a potential regulatory role for
transcription factor binding. In order to discern the functional relevance of this marker, we
performed in silico predictions and observed that several transcription factors bind to the A-
allele. Using EMSA we provide suggestive evidence that the rs8087522 A-allele, but not the G-
allele, is bound by a yet unknown nuclear protein, and thus may influence MC4R gene
expression
To our knowledge, this is the first study investigating the role of the rs8087522 marker in AIWG.
The limitations of the study include a relatively small sample size, compared with other studies,
but it is important to note that we have sufficient power. One caveat with our sample is that
patients in some sub-samples were not drug-naıve when they entered our study. Thus, they may
have already gained weight owing to prior antipsychotic exposure. Despite this, clozapine, the
antipsychotic with the highest risk for weight gain, was administered for the first time to a
substantial number of patients (every patient in Sample-B), and our analyses showed that these
patients did gain a considerable amount of weight. In this study, we report a nominally
significant association between clozapine-induced weight gain and the rs8087522 variant in our
European American-ancestry sample. Furthermore, antipsychotic-induced side-effect profiles
vary across ethnicities, and we did not have sufficient power to detect an effect for AIWG in our
74
African American-ancestry sample. Further investigation of MC4R gene variants in African
Americans is warranted.
Our collaborative group reported that the rs489693 marker is associated with AIWG in three
separate samples (Malhotra et al., 2012). The region where this marker is located has been
highlighted in several GWASs regarding obesity (Loos et al., 2008; Speliotes et al., 2010).The
rs489693 is located 190 kb downstream from the MC4R gene and has no known functional
relevance. There is no gene located in the region, but the variant has been associated with weight
gain in the general population. It may be controlled by one or more remote regulatory sites
located at considerable distance from rs489693 (for example, chromosomal folding) (Espinoza et
al., 2011). The rs489693 marker may also be linked to another ‘true’ marker that is the actual
biological regulator of changes in weight gain. We also examined possible combinations of
markers across the MC4R that were significant either in the study by Malhotra et al. (2012) or
within our Toronto-based AIWG samples. There were no significant findings for these marker
combinations.
To summarize, we did not replicate the reported association of rs17782313 with obesity (Loos et
al., 2008; Speliotes et al., 2010; Renström et al., 2009) in AIWG. Similarly, we found no
association with the rs11872992 SNP, which is located within the promoter region of MC4R, and
AIWG.A strong biological rationale remains for the study of MC4R and metabolic parameters,
including weight gain. The MC4R has a significant regulatory role in food intake and energy
homeostasis. Furthermore, the MC4R genetic findings may be used in an attempt to define new
targets for discovery of novel medications that do not have severe weight-gain side effects. A
recent study reported that an MC4R agonist has potent anti-obesity efficacy in a rodent model
(He et al., 201). Furthermore, recent studies have shown that in some cases the MC4R protein, if
75
altered by mutation, is unable to reach the plasma membrane as the receptor may misfold in the
endoplasmic reticulum, and these defective MC4R proteins are therefore targeted for degradation
(Granell et al., 2010). The authors suggest that drugs, which improve MC4R folding or decrease
the endoplasmic reticulum-associated degradation of the receptor, may treat some forms of
hereditary obesity. Understanding the common variants of the MC4R gene may also have
implications in drug development for general obesity as well as AIWG.
In conclusion, we report a nominally significant association between rs8087522 and AIWG, and
we found that this SNP affects potential transcription factor-binding sites, which may result in
functional effects. A recent case-control study reported no association between the rs8087522
and high weight (Beckers et al., 2011). Our overall findings still suggest that the rs8087522
marker in the MC4R gene may be of interest for further investigation in AIWG and weight
regulation in general.
76
Chapter 4
Investigation of melanocortin-system gene variants in
antipsychotic induced weight gain
Contents of this chapter have been published in the World Journal of
Biological Psychiatry: Chowdhury NI, Souza RP, Tiwari AK, Brandl EJ, Sicard M,
Meltzer HY, Lieberman JA, Kennedy JL, Müller DJ. The Pharmacogenomics Journal (2013)
13, 272–279; doi:10.1038/tpj.2011.66;
A link to the published paper can be found at
http://www-ncbi-nlm-nih-gov.myaccess.library.utoronto.ca/pubmed/24564533
77
4.1 Abstract
The common use of second-generation antipsychotic medications may result in substantial
weight gain and metabolic syndrome in a subset of schizophrenia patients. Distinct populations
of neurons expressed in the hypothalamus have regulatory roles in weight control and energy
homeostasis involving the cocaine amphetamine and regulated-transcript (CART), the
polypeptide pro-opiomelanocortin (POMC) and the agouti related protein (AGRP). Thus, we
investigated the potential role of CART, POMC and AGRP genetic variants in antipsychotic
induced weight gain. Five CART single nucleotide polymorphisms (SNPs) (rs10515115,
rs3763153, rs3857384, rs11575893, rs16871471), three POMC SNPs (rs6713532, rs1047521,
rs3754860), and one AGRP SNP were genotyped in 218 patients treated with different
antipsychotics for chronic schizophrenia and evaluated for AIWG. We compared weight change
(%) across genotypic groups using analysis of covariance. In our overall sample, the CART and
AGRP variants were not significantly associated with AIWG. The POMC SNP rs3754860 was
significantly associated with AIWG (pcorrected = 0.036). In our sub-sample accounting for
ethnicity (European descent) and drug-type (olanzapine and clozapine), the POMC rs3754860
and rs1042571 SNPs reached a trend for association (p = 0.097 and p = 0.089, respectively) with
AIWG. In this exploratory study, we found that POMC gene variants were nominally associated
with antipsychotic induced weight gain.
78
4.2 Introduction
Antipsychotic drugs are an essential component in the drug treatment of schizophrenia.
However, the variability of treatment outcome and side effect manifestation remains a critical
problem in terms of effective treatment regime and compliance (reviewed in Lett et al. 2012).
One devastating and potentially fatal effect of antipsychotic drug use is substantial weight gain
and metabolic disturbances. The use of atypical antipsychotics such as clozapine and olanzapine
are associated with dramatic secondary effects in the basal metabolism ultimately increasing risk
for obesity, diabetes and heart disease (Ray et al. 2009). Despite the serious impact on overall
physical and psychological health, mechanisms and predictors for antipsychotic induced weight
gain (AIWG) remain unclear. Pharmacogenetic approaches may provide distinct advantages in
the search for the factors associated with AIWG and its predictors. Weight gain may be
considered a robust phenotype for pharmacogenetic studies because of evidence on the
heritability of weight regulation shown by twin, adoption and family studies (Gebhardt et al.
2010, Ternouth et al. 2011, Wehmeier et al. 2005). Recent studies suggest that genes expressed
in the hypothalamus regulating energy homeostasis may have a role in AIWG (Brandl et al.,
2012, Tiwari et al., 2012; Malhotra et al., 2012).
In the arcuate nucleus of the hypothalamus, distinct populations of neurons influence feeding
behaviour (Cone et al., 2005). Adipocytes release leptin, which, in turn, stimulate neurons
expressing cocaine amphetamine and regulated transcript (CART) and polypeptide pro-
opiomelanocortin (POMC) inhibit feeding behaviour (anorexigenic effects). Additionally, leptin
acts upon the agouti related protein (AGRP), reducing the AGRP expression, which is an
79
endogenous antagonist for melanocortin receptors. Thus, an intact leptin-melanocortin system
results in an anorexigenic effect (Santini et al. 2009).
4.2.1 Proopiomelanocortin
Sub-chronic olanzapine treatment has been shown to result in down regulation of pro-
opiomelanocortin expression in the rat hypothalamus (Ferno et al. 2011). However, Davoodi et
al., (2009), reported that female rats treated with olanzapine gained body weight and exhibited
increased food intake, but showed no difference in POMC mRNA expression compared to
controls (Davoodi et al. 2009). Genetic studies assessing healthy human populations for weight-
related measures have provided evidence that certain POMC gene variants may predispose some
individuals to become obese (Baker et al., 2005; Ternouth et al., 2011).
4.2.2 Cocaine Amphetamine Regulated Transcript
CART is a hypothalamic neuropeptide, and hypothalamic expression levels of both POMC and
CART are decreased during food deprivation in rodents (Brady et al. 1990, Savontaus et al.
2002, Ziotopoulou et al. 2000). CART is also directly stimulated by leptin (Cheung and Mao
2012). Acute and chronic treatment with clozapine has been shown to reduce the expression of
CART mRNA in the shell of the nucleus accumbens (Beaudry et al. 2004). Two studies reported
different CART variants (Phe34Leu and Delta A1457) associated with obesity in Italian children
and adolescents (del Giudice et al. 2001, Rigoli et al. 2010) but other CART variants were
reported not to be associated with obesity in Danish Caucasians (Echwald et al. 1999), Pima
Indians (Walder et al. 2000) or severely obese Caucasian children (Challis et al. 2000).
80
4.2.3 Agouti Related Protein
AGRP regulates feeding behaviour by activating melanocortin receptors, and its expression is
down-regulated by leptin (Lett et al. 2012, Santini et al. 2009). Mice that are either leptin
deficient (ob/ob), or leptin receptor deficient (db/db) exhibit increased levels of AGRP mRNA in
the hypothalamus (Ollmann et al. 1997). The AGRP over-expression in mice results in obesity,
increased body length, hyperplasia, as well as hyperglycemia and hyperinsulinemia (Graham et
al. 1997; Michaud et al. 1997). Notably, administration of olanzapine in rats resulted in an up
regulation of AGRP in the arcuate nucleus (Ferno et al. 2011). Polymorphisms of the AGRP
gene have been implicated in genetic association studies across populations. The GG genotypic
group for the c.199G-->A) polymorphism were significantly associated with fatness and
abdominal adiposity in a healthy human population (Argyropoulos et al. 2002). Additionally, the
functional AGRP promoter SNP -38C/T has been reported to be related in body measures in
African American populations (Argyropoulos et al. 2002, Bonilla et al. 2006), indicating that
ethnicity can play a major role in this context.
Given the roles of POMC, CART and AGRP in weight regulation, we evaluated whether these
variants are associated with AIWG in schizophrenia subjects treated mostly with clozapine. We
assessed SNPs that were associated with weight measures in the literature as well as utilized
tagSNPs covering the common variations across POMC, CART and AGRP genes.
81
4.3 Materials and Methods
4.3.1 Subjects
A total of 218 patients with chronic schizophrenia (SCZ) or schizoaffective disorder according to
DSM-III or DSM-IV criteria were included in this study. Approval from the institutional ethics
committee and informed consent were obtained for all patients. The patient ethnicity was
ascertained using self-reported ancestry information over the last three generations. The sample
characteristics have been described in more detailed previously (Tiwari et al. 2010). Sample-A:
Charité University Medicine (Berlin, Germany) (DJ Müller, n=88). Patients were treated with
antipsychotic medication and assessed for up to 6 weeks in this naturalistic study. The SCZ
patients were treated with medications deemed most suitable, including fluphenazine,
aripiprazole, quetiapine, ziprasidone, amisulpride, clozapine, olanzapine, haloperidol, and
risperidone [for details see Tiwari et al., 2010].Sample-B: Case Western Reserve University in
Cleveland (Ohio, USA) (HY Meltzer, n=73). Schizophrenia patients were treated with clozapine.
They then underwent a washout period of 2–4 weeks during which no medications were given
unless deemed clinically essential by the physician. Serum levels were monitored to ascertain
compliance [for details see Masellis et al, 1998 and Basile et al., 2002]. Sample-C: Hillside
Hospital in Glen Oaks (New York, USA) (JA Lieberman, n=48). Patients diagnosed with
schizophrenia had shown suboptimal response to previous atypical antipsychotic drug treatment.
These patients were included in this 14-week double-blind study and were randomly assigned to
receive clozapine, olanzapine, risperidone or haloperidol [for details Volavka et al., 2002 and
Müller et al., 2005]. Briefly, patients were primarily recruited from three separate recruitment
82
sites, and patients were mostly treated with antipsychotic drugs that carry high risk for weight
gain (see Table 1) varying across the sample sites for up to 14 weeks.
Table 4-1. Demographic characteristics from each of the clinical site
CHARACTERISTICS SAMPLE A
(N=88)A
SAMPLE B
(N=73)A
SAMPLE C
(N=57)A
TOTAL
SAMPLE
(N=218)
P-VALUE
Gender
Male 50 47 48 145 0.003
Female 38 26 9 73
Age (years) 35.78 ± 12.2 33.38 ± 8.6 40.47 ± 9.1 36.21 ± 10.6 0.001
Initial body weight
(kg)
79.5 ± 15.9 74.8 ± 13.7 84.9 ± 17.4 79.26 ± 16.0 0.002
Weight change (kg) 3.26 ± 3.9 3.81 ± 4.4 4.43 ± 6.4 3.75 ± 4.8 0.374
Weight change (%) 4.00 ± 4.8 5.42 ± 6.3 5.73 ± 8.4 4.93 ± 6.4 0.21
Study duration (weeks) 5.09 ± 1.6 6.00 ± 0.00 11.57 ± 3.9 7.07 ± 3.5 P > 0.0001
Ethnicity P >0.0001
European Americans 87 52 13 152
African Americans 1 21 33 55
Others 0 0 11 11
Treatment
Clozapine 12 73 12 97 P >0.0001
Haloperidol 5 0 11 16
Olanzapine 15 0 22 37
Risperidone 27 0 12 39
Othersc 28 0 0 28
a The samples were included from three different clinical sites
b Includes antipsychotic drugs: fluphenazine, aripiprazole, quetiapine, ziprasidone and amisulpride
83
4.3.2 Genotyping
Blood samples were collected from the clinical sites and sent to the Centre for Addiction and
Mental Health (CAMH) in Toronto, Canada. Genomic DNA was extracted from white blood
cells using the high-salt method (Lahiri and Nurnberger, 1991). The POMC (rs3754860,
rs6713532, rs1047521) (see Figure 4-1a) and CART (rs10515115, rs3763153, rs3857384,
rs11575893, rs16871471) (see Figure 4-2b) variants were genotyped using TaqMan® SNP
Genotyping Assays (Applied Biosystems, Foster City, CA). The AGRP SNP rs1338993 (see
Figure 4-3c) was genotyped Illumina GoldenGate Assay (Illumina, San Diego, California, USA).
The POMC rs3754860 was genotyped using PCR restriction enzyme digestion and agarose gel
electrophoresis. This SNP is present 1798bp 5’ to exon 1 (a C/T substitution), was genotyped
using primers F: 5’-GGC AAC ATA GTG AAA CCC TGT C 3’ and R: 5’-TCC AAA TGG
ACC CAA CTT 3’, annealing temperature 72 °C and 1.5 mmol MgCL2 and 2.5 mmol 10X
Buffer w/ KCl. The PCR product (298 bp) was digested with the restriction enzyme RsaI (New
England Biolabs). The PCR products with the C-allele were digested into two fragments of 190
and 108 bp, whereas the T-allele did not carry the RsaI restriction site and was not digested
(298bp fragment). The digested products were resolved in a 2.5% ethidium bromide stained
agarose gel and reference DNA samples of known genotypes were included in each run.
84
Figure 4-1. Schematic gene diagram of POMC, CART, and AGRP genes. The black filled box represents the coding
region and the gray box represents the non-coding region. Standard color scheme has been implemented (Haploview
4.1). The two dark-shaded diamonds represent D=1 and LOD<2; the white diamond, D’<1 and LOD<2. The values
in the box represent r2. a) Proopiomelanocortin (POMC) (chr2p23) gene structure and linkage disequilibrium in
European-ancestry patients. Location of polymorphisms with respect to the exons. The black filled box represents
the coding region and the gray box represents the non-coding region. Standard color scheme has been implemented
(Haploview 4.1). The two dark-shaded diamonds represent D=1 and LOD<2; the white diamond, D’<1 and LOD<2.
The values in the box represent r2. b) CART (5q13) gene structure and linkage disequilibrium in Euripean ancestry
patients. C) AGRP (16q22) gene structure.
85
Laboratory staff was blind to the clinical data. Ten percent of the total sample was randomly re-
genotyped to confirm genotyping accuracy. Haploview 4.1 (Barrett et al. 2005) was used to
select tagSNPs. All polymorphisms that had a minimum allele frequency more (MAF) than 5%
and Hardy-Weinberg equilibrium (HWE) p > 0.001 in the CEPH population (Utah residents with
ancestry from northern and western Europe) of the HapMap database (The International HapMap
Consortium, 2007) were included for tag SNP selection.
4.4 Statistical Analysis
The statistical analyses were conducted using SPSS 15.0. Continuous variables were tested using
analysis of variance (ANOVA) or co-variance (ANCOVA), and categorical variables were tested
using χ² tests. Our primary dependent measure was percentage (%) weight change from baseline,
compared across genotypic groups using ANCOVA with duration of treatment as a covariate.
Haplotypes were constructed and weight change (%) was compared using UNPHASED 3.1
(Dudbridge 2003), and only haplotypes with frequencies greater than 5% were included in the
analyses. Linkage disequilibrium (LD) and Hardy-Weinberg equilibrium were determined using
Haploview 4.1 (Barrett et al. 2005). Power calculations were carried out using Quanto 1.2.4.
(Gaudermann et al., 2006). Assuming an additive model, a minor allele frequency of 0.15 and a
sample size of n=218, we had more than 80% power to detect a mean difference of 2.5%
86
between carriers and non-carriers of the risk genotype. The demographic characteristics of the
study sample are presented in Table 1.
4.5 Results
Significant differences in clinical variables were observed among clinical sites; nonetheless,
comparable rates of change in weight were observed (Table 1). All SNPs analyzed in this study
were in Hardy–Weinberg equilibrium (p > 0.05; Table 3). Linkage disequilibrium was calculated
using Haploview 4.1. (Barrett et al., 2005).
The duration of drug treatment significantly differed between clinical sites and was therefore
entered a covariate in the association analyses. In the overall sample, the POMC rs3754860 SNP
was associated with weight gain in patients of all races treated with all drugs (p = 0.004,
Bonferroni corrected: p = 0.036). Subjects that were G-allele carriers, GG homozygotes (5.35%
± 6.8) and GA (5.04% ± 4.8) showed increased weight gain compared to the AA homozygotes (-
2.00 ± 7.7) (Table 2). No significant observations were observed for CART and AGRP SNPs
(Table 2).
Table 4-2. List of SNPs in POMC, CART & AGRP tested for association with AIWG at the genotypic
level
87
Gene SNP Genotype
Total Sample – All
Patients on All
Medications
(%)a, n=218
p-valueb
Weight Change
European
ancestry-clozapine
or olanzapine
(%)a, n=87
p-valueb
POMC
rs3754860 (RsaI)
CC 5.35 ± 6.8 (118)
0.004
3.92 ± 5.9 (41)
0.139c CT 5.04 ± 4.8 (66) 5.65 ± 5.0 (31)
TT -2.00 ± 7.7 (9) 0.04 ± 1.28 (4)
rs6713532
TT 4.01 ± 5.9 (87)
0.224
4.08 ± 5.4 (46)
0.637 TC 4.77 ± 6.0 (83) 4.88 ± 6.0(35)
CC 6.56 ± 8.2 (32) 2.61 ± 2.3 (3)
rs1042571 (C8246T)
GG 4.80 ± 6.4 (151)
0.413
4.50 ± 5.0 (58)
0.089 GA 5.21 ± 6.2 (44) 5.10 ± 6.3 (21)
AA 2.29 ± 7.2 (9) -0.35 ± 6.3(6)
CART
AGRP
rs10515115
(-2815T>A)
TT 3.88 ± 4.9 (42)
0.526
3.27± 4.9(19)
0.659 TA 5.08 ± 7.1(98) 4.66± 58(40)
AA 4.99 ± 6.01(68) 4.46± 5.4(27)
rs3763153 (-1474T>C)
AA 4.92 ± 6.3 (100)
0.264
3.01 ± 4.8 (32)
0.174 AG 5.10 ± 6.1 (92) 5.42± 6.0(43)
GG 1.92 ± 9.0 (11) 4.24 ± 5.6(8)
rs3857384
(-1157C>T)
AA 5.14 ± 7.3 (8)
0.294
-
0.718 AG 6.02± 6.8 (68) 4.04 ± 5.5 (22)
GG 4.16 ± 6.0 (132) 4.45 ± 5.5 (64)
rs11575893 (IVS1+172C>T)
TT 5.41 ± 4.8(6)
0.828
5.87 ± 9.2 (2)
0.787 TC 5.05 ± 7.5 (43) 4.94 ± 3.8 (15)
CC 4.68 ± 6.1(164) 4.17 ± 5.7 (69)
rs16871471
TT 5.32 ± 7.0(9)
0.928
-
0.718 TA 5.03 ± 6.4 (59) 4.50 ± 4.2 (25)
AA 4.82 ± 6.5(131) 4.21 ± 6.0 (58)
rs1338993
AA 4.13 ± 5.9 (146)
0.152
4.46 ± 5.8 (70)
0.716 GA 5.98 ± 6.7 (49) 5.06 ± 5.1 (13)
GG 7.52 ± 8.1 (22) - a Mean ± s.d. (number of individuals) b SNPs with low numbers in a genotypic group were merged with heterozygous genotypic group c For AA versus GA+GG: All P-values were calculated by ANCOVA. rs3754860, European Americans
on olanzapine or clozapine: AA versus GA+GG, 0.03 ± 1.1 versus 4.54 ± 5.6 (P = 0.097).
Our overall study sample consisted of patients of two primary and distinct ethnicities (European
and African ancestry). The patients were all also treated with a variety of different
antipsychotics. Due to these differences in the overall sample consisting of 218 subjects, further
sub-analyses were carried out.
88
Significant differences in the amount of AIWG between Africans and Europeans were observed
in our sample (Table 1). In addition, the patients of African American ancestry (n = 55) exhibited
different minor allele frequencies (data not shown) than the Europeans, and were therefore
excluded from any further analyses to avoid risk for spurious findings.
Refined Sub-Sample Analysis
Both clozapine and olanzapine have been established as antipsychotics being associated with the
higher propensity for weight gain and have similar pharmacological profiles (Nasrallah 2008).
Thus we conducted a refined sub-analysis that incorporated patients by their self-reported
ethnicities (European-ancestry vs. African-American ancestry) and whether these patients
received either clozapine or olanzapine.
In our refined sample of European-ancestry patients treated with either olanzapine or clozapine
(n=87), the POMC rs3754860 SNP reached a trend for association with weight gain for AA
homozygotes (0.03% ± 1.1) vs. G-allele carriers, GA+GG (4.54% ± 5.6 ) (p = 0.097); (Table 2).
The POMC SNP rs1042571 reached a trend for association with change in weight gain
percentage in patients of European ancestry, treated with either olanzapine or clozapine (p =
0.089). The GG (4.50% ± 5.0) and GA (5.10% ± 6.3) genotypic groups gained more weight than
AA homozygotes (-0.35% ± 6.3). No allelic associations were observed in our sample of patients
of European ancestry, treated with either clozapine or olanzapine (Table 4-3).
89
Table 4-3. Allelic differences in weight change (%) in patients of European ancestry
Gene SNP HapMAP
CEU
European-ancestry-all
drugs
European
ancestry-clozapine
or olanzapine
Allele
Frequency
p-
Value
HWE
p-
value
Frequency p-
value
POMC rs3754860 C 0.63 0.71 (193) 0.566 0.53 0.72 (114) 0.980
T 0.33 0.29 (75) 0.28 (40)
rs6713532 T 0.75 0.72 (210) 0.695 0.88 0.75 (130) 0.953
C 0.23 0.28 (82) 0.25 (44)
rs1042571 G 0.76 0.82 (245) 0.638 0.05 0.81 (143) 0.159
A 0.24 0.17 (51) 0.19 (33)
CART rs10515115 A 0.59 0.50 (158) 0.481 1.00 0.54 (96) 0.322
T 0.41 0.50 (140) 0.46 (80)
rs3763153 A 0.66 0.68 (198) 0.578 0.98 0.64 (110) 0.360
G 0.34 0.32 (92) 0.36 (62)
rs3857384 G 0.93 0.86 (258) 0.556 0.30 0.87 (156) 0.904
A 0.07 0.14 (42) 0.13 (22)
rs11575893 C 0.89 0.89 (268) 0.878 0.07 0.89 (158) 0.281
T 0.11 0.11 (32) 0.11 (20)
rs16871471 A 0.76 0.92 (231) 0.784 0.08 0.84 (145) 0.644
T 0.24 0.18 (49) 0.16 (29)
AGRP rs1338993 A N/A* 0.93(211) 0.479 0.741 0.94 (153)
0.485 G N/A* 0.07 (7) 0.06 (13)
*Data not available in hapmap
4.6 Discussion
In this study, we investigated the role of three candidate genes, pro-opiomelanocortin (POMC),
cocaine amphetamine regulated transcript (CART), and agouti related protein (AGRP) in AIWG.
90
We included SNPs that were previously associated with weight measures across various ethnic
populations, and utilized tagSNP to capture meaningful coverage of these genes.
We found a significant association between the POMC SNP rs3754860 and percentage change in
weight in our overall sample (pcorrected = 0.036). This signal became non-significant in our refined
sample of European-Ancestry patients treated with clozapine and olanzapine. A previous report
on this gene variant showed no association between this SNP in BMI and diabetes mellitus in
patients of Caucasian origin with chronic heart failure (Bienertova-Vasku et al. 2009). A study
conducted by (Baker et al. 2005) also reported no association between the POMC rs3754860
SNP and waist-to-hip ratio in a family study comprised of over 1000 individuals.
The POMC rs1042571 SNP reached a trend for association with AIWG in patients of European
ancestry treated with clozapine or olanzapine. No allelic association was detected in this sub-
sample. The POMC rs1042571 has also been associated with measures including BMI, waist,
visceral adipose tissue, and subcutaneous adipose tissue in obese patients of Hispanic descent
(Sutton et al. 2005) and more recently, was associated with an obesity phenotype: the ratio
between fat and protein intake in Dutch males (Ternouth et al. 2011).
Despite the biological evidence supporting a role for both the CART and AGRP genes in weight
regulation (Cheung & Mao et al., 2012), we did not observe any significant associations with
polymorphisms of these genes in the total sample set. To the best of our knowledge, these genes
and their impact on AIWG have not yet been thoroughly investigated in pharmacogenetic
studies. Plasma values of total cholesterol, LDL-cholesterol, and HDL-cholesterol were
91
associated with Delta A1457 CART variant in Chinese diabetic subjects (Fu et al. 2002) but not
in Caucasians (Challis et al. 2000); (Vasseur et al. 2007).
Some limitations of this study should be considered. These include a relatively small sample
size, heterogeneity of the sample to differing ethnicities, and different medication treatments
between sub-samples. We did not observe any significant association in sub-groups of patients of
European ancestry who received clozapine or olanzapine. Our findings with the SNP rs3754860
suggest that the heterogeneity and population stratification in the overall sample may be
contributing to our initial positive association in the total sample.
In summary, we demonstrated a potential influence of a POMC genetic variation in AIWG.
However, we could not detect a positive association between AGRP and CART SNPs in AIWG.
Nonetheless, our finding, in combination with recent reports of the melanocortin system
(Chowdhury et al., 2012; Tiwari et al., 2012), underline the importance of the role of this system
in antipsychotic induced weight gain and deserves further investigation in larger sample sets.
92
Chapter 5
An Exploratory Investigation of Melanocortin-3 Receptor
Gene Variants in Antipsychotic-Induced Weight Gain
This manuscript is in preparation
Nabilah I. Chowdhury1, Arun K. Tiwari1, Eva J. Brandl1,2, Clement C. Zai1,2, Jeffrey A.
Lieberman3, Herbert Y. Meltzer4, James L. Kennedy1,2, Daniel J. Müller1,2
5.1 Abstract
Second generation antipsychotics treat core symptoms of schizophrenia (SCZ) and other
disorders. However, the use of this class of medications may result in the development of severe
side effects such as substantial weight gain and metabolic syndrome. Early diagnosis of
metabolic disturbances through genetic testing could reduce the morbidity and mortality of
patients. Based on the melanocortin-3 receptor (MC3R) gene’s role in the regulation of weight
and food intake, we investigated the potential role of MC3R single nucleotide polymorphisms
93
(SNPs) in antipsychotic induced weight gain (AIWG).Ten MC3R SNPs (rs6127698, rs6024731,
rs1543570, rs6024730, rs6014649, rs1926064, rs6014646, rs11697509, rs3827103) were
selected for coverage of common variation in the gene using tagSNP software. The SNPs were
genotyped with the Illumina GoldenGate Genotyping Assays in 266 chronic SCZ patients who
underwent treatment with various antipsychotics and were evaluated for weight gain for up to 14
weeks. We compared weight change (%) across genotype groups using analysis of covariance. In
this study, we observed that the MC3R gene was associated with AIWG. Specifically, genotypic
associations were found between the correlated MC3R polymorphisms in our total sample set.
Genotypic associations were also found between MC3R SNPs(rs6014649, rs11697509,
rs3746619, rs3827103, rs1543570) and weight gain (p<0.05) in a refined sub-sample consisting
of European ancestry patients treated with either olanzapine or clozapine (n=104). This finding
requires validation in larger sample sets. However, the potential role of MC3R in AIWG may
further provide understanding toward the genetic etiology of AIWG, which could in turn lead to
greatly novel drug target development for obesity or psychiatric medications.
94
5.2 Introduction
Between 40-80% of schizophrenia (SCZ) patients treated with second-generation antipsychotic
medications experience substantial increase in body weight, or antipsychotic induced weight gain
(AIWG) (Umbricht et al, 2001; Green et al., 2000). Second-generation antipsychotic medications
– namely clozapine and olanzapine - are associated with the most dramatic weight increases in
patients (Allison et al., 1999; Lett et al., 2012). Despite the effectiveness of second-generation
antipsychotics, AIWG and associated metabolic side effects greatly impair patient quality of life
and may lead to conditions such as type 2 diabetes and cardiovascular diseases (Rummel-Kluge
et al., 2010; Correll et al., 2011).Various neural systems regulating appetite, satiety and lipid
metabolism have been implicated in the etiology of AIWG, but a clear mechanism by which this
side effect develops remains largely unknown (Roerig et al., 2011). Further understanding of
AIWG and predicting this side effect could lead to improved treatment in psychiatric patients.
The onset of AIWG is highly variable between patients, and a genetic component has been cited
as among one of the leading factors of susceptibility to develop this side effect (Gebhardt et al.,
2010). The antipsychotic action at the serotonergic (Reynolds et al., 2002; Reynolds et al., 2006),
and histaminergic (Kroeze et al., 2003) systems has been reported to be associated with AIWG,
though no causal mechanism has been verified. Adding to the complexity of AIWG is that the
pharmacological action of antipsychotic drugs is complex, as various antipsychotics have
differing receptor-binding profiles, which may have downstream effects on weight and
metabolism (Nasrallah et al., 2008).
95
The Melanocortin System
The melanocortin receptors are encoded by a gene family (MC1R-MC5R) and have preferential
binding to separate ligands, resulting in different functions. The biological roles of the
melanocortin receptors have been well described in their capacity to regulate melanogenesis
(MC1R), (Sulem et al., 2007) mediate cortisol levels via interaction with the adrenocorticotropic
hormone (MC2R) (Schiöth el al., 2005), and have been implicated in obesity (MC3R-MC5R).
The MC3R gene has been reported to have mixed associations with high body mass index and
weight measures in the literature, while MC4R mutations have been reported to cause monogenic
obesity (Huszar et al., 1997; Farooqi et al., 2003). The MC5R gene has been implicated in lipid
metabolism (Sánchez et al., 2009; Rodrigues et al., 2013) but its function in weight regulation is
not known.
In the context of AIWG, a recent genome wide association study yielded a highly significant
association between the MC4R rs489693 in AIWG, and this association was subsequently
replicated three distinct samples (Malhotra et al., 2012) Additionally, an MC4R SNP
(rs8087522) located in the promoter region of the gene, was associated with clozapine induced
weight gain (Chowdhury et al., 2013). A study conducted by Moons et al., (2011) examined the
MC3R Thr6Lys (rs3746619) in olanzapine induced weight gain and reported no significant
association. Nonetheless, the MC3R gene is an important candidate to examine in terms of a
contributory role to the development of AIWG. The double-variant combination of MC3R
Thr6Lys (rs3746619) and Val81Ile (rs3827103), located in the 5’UTR, have been reported to be
associated with higher body weight measures such as body mass index (BMI), and body fat mass
in obese pediatric patients. (Feng et al., 2005). The two mis-sense MC3R variants have been
associated with decreased cyclic AMP generation in vitro (Feng et al., 2005).
96
One shortcoming of previous MC3R gene association studies is the limited coverage of the
MC3R gene. In this study, we captured 100% of the common variation of the MC3R gene in
order to comprehensively investigate the role of this gene in AIWG.
5.3 Methods
5.3.1 Samples
Our samples have been described in detail in other studies (Masellis et al., 1998; Volavka et al.,
2002; Tiwari et al., 2010). Briefly, patients with schizophrenia or schizoaffective disorder
according to DSM-III-R or DSM-IV criteria were recruited at three different sites: Patients in
Sample A: (N=88; DJM; Charité University Medicine, Berlin, Germany) received various
antipsychotics in a naturalistic study design (Müller et al., 2012) Sample B: (N=74; HYM; Case
Western Reserve University, Cleveland, Ohio, USA) consisted of patients who were treated with
clozapine (first exposure to a second generation antipsychotic) for six weeks (Kane et al., 1988;
Masellis et al., 1998). In Sample C: (N=54; JAL; Hillside Hospital in Glen Oaks, New York,
USA), patients were treated with clozapine, risperidone, olanzapine or haloperidol in a double-
blind study design (Volavka et al., 2002). For Sample D: (N=48, DJM2, Centre for Addiction
and Mental Health, Toronto,ON), patients were prescribed antipsychotic medication and
followed up for a minimum of 6 weeks (Goncalves et al., 2013).All ancestry information across
the four samples was self-reported. All subjects gave written informed consent prior to study
entry in accordance with institutional ethics guidelines and the Declaration of Helsinki.
97
Due to the heterogeneity among the samples in terms of ethnicity and medication, we selected
patients of European ancestry treated with high-risk medication for AIWG (clozapine,
olanzapine) to create a more homogeneous subset (N=104),. The baseline weight and duration of
treatment significantly influenced the amount of weight gain during the study (see Table 5-1),
and therefore we used these as covariates in our analyses. Details on all samples are provided in
Table 5-1.
5.3.2 Genotyping
DNA extraction was performed with a standard high-salt method (Lahiri & Nurnberger et al.,
1991). The MC3R SNP genotyping was conducted using customized GoldenGate Genotyping
Assays (Illumina Inc, San Diego, CA, USA). The MC3R SNPs were selected from information
available in the literature (Feng et al., 2005; Moons et al., 2011; Santos et al., 2011), and tag
SNPs from the European ancestry population: CEU, in the HapMap database (r2 ≥ 0.8, minor
allele frequency ≥ 0.05; Haploview 4.2, Barrett et al., 2005). Ten MC3R SNPs within and near
the MC3R gene (rs6014646, rs6024730, rs6024731, rs6014649, rs11697509, rs6127698,
rs3746619, rs3827103, rs1926064, rs1543570) were genotyped in the DNA samples. Samples
with less than 95% successful genotyping were excluded from our study. To ensure genotyping
accuracy, 10% of samples were randomly re-genotyped, and the concordance rate was 100%.
5.3.3. Statistical Analysis
Categorical variables were analyzed with Pearson’s χ2, and continuous variables with analysis of
covariance (ANCOVA) using percentage (%) weight change as outcome and baseline weight
98
(kg) and treatment duration (weeks) as covariates. The Statistical Package for the Social
Sciences (SPSS), version 15.0, was used for our genetic association tests. Hardy-Weinberg-
equilibrium, linkage disequilibrium and haplotype associations were calculated using Haploview
Version 4.2. The change in weight change percentage (%) was compared across haplotypes using
in UNPHASED version 3.1.4. Correction for multiple testing was carried out by estimating the
number of independent tests using single nucleotide polymorphism spectral decomposition
(SNPSpD) (Nyholt, 2004; Li & Ji, 2005).
5.4 Results
Demographic and clinical characteristics of the samples are shown in Table 5-1. Weight gain
was significantly correlated with treatment duration (p< 0.0001), and with baseline weight
(p=0.005). We observed significant differences across different variables, including baseline
weight, age, ethnicity, sex and study medication across the samples. Weight change (%) from
baseline weight did not differ significantly among samples (Table 5-2). Joint statistical analysis
of weight change was conducted for the combined samples, and confounding factors were
factored into the analyses. In our total sample (N = 266), eight correlated SNPS near the MC3R
gene were associated with AIWG (Figure 5-1).
Table 5-1. Demographic Characteristics
99
Characteristics
Sample A
(n=88) a
Sample
B(n=73)a
Sample
C(n=57)a
Sample
D(n=48)a
Total
Sample
(n=266)
p-value
Gender 0.007
Male 50 (56.8%) 47 (64.3%) 48 (84.2%) 32 (66.7%)
177
(66.5%)
Female 38 (43.1%) 26 (35.6%) 9 (15.8%) 17 (35.4%)
90
(33.8%)
Age (years) 35.78 ±
12.2 33.38 ± 8.6 40.47 ± 9.1 37.9 ± 12.4
36.52 ±
11.0 0.002
Initial body
weight (kg) 79.5 ± 15.9 74.8 ± 13.7 84.9 ± 17.4 81.8 ± 19.4 79.7± 16.7 0.005
Weight change
(kg) 3.26 ± 3.9 3.81 ± 4.4 4.43 ± 6.4 3.89 ± 7.1 3.77± 5.3 0.261
Weight change
(%) 4.00 ± 4.8 5.42 ± 6.3 5.73 ± 8.4 3.89 ± 7.1 4.74 ± 6.5 0.649
Study
Duration
(weeks) 5.09 ± 1.6 6.00 ± 0.0 11.57 ± 3.9 11.20 ± 5.2 7.79 ± 4.1 p <0.001
Ethnicity p <0.001
European
Americans 87 (98.9%) 52 (71.2%) 13 (22.8%) 26 (54.1%)
178
(66.9%)
African
Americans 1 (0.01%) 21 (28.8%) 33 (57.9%) 2 (0.04%)
57
(21.4%)
Others 0 0 11 (19.2%) 1 (0.02%)
29
(0.11%)
Treatment p <0.001
Clozapine 12 (13.6%) 73 (100%) 12 (21.1%) 27 (56.3%)
124
(46.6%)
Haloperidol 5 (0.056%) 0 11 (19.3%) 0 16 (0.06%)
Olanzapine 15 (17.4%) 0 22 (38.6%) 8 (16.7%) 45 (16.9%)
Risperidone 27 (30.6%) 0 12 (21.1%) 5 (10.4%) 44 (16.5%)
Others 28 (31.8%) 0 0 9 (18.8%) 37 (13.9%)
aSamples were included from four different clinical sites.
b Includes antipsychotic drugs: fluphenazine, aripiprazole, quetiapine, ziprasidone and amisulpride.
100
a)
b)
Figure 5-1.a) Schematic diagram of the MC3R gene with the nine single nucleotide polymorphisms we
examined for weight percentage change in psychiatric patients treated with either clozapine or olanzapine.
b) The linkage disequilibrium structure of the MC3R SNPs is displayed, with the numbers representing
rsquared values, and the intensity of red relating to the r-squared values between SNP pairs.
101
The allele frequency of rs1926064 SNP was low (p = 0.034) in our samples and was not
analyzed further. Of the remaining nine MC3R SNPs, eight (rs6014646, rs6024730, rs6024731,
rs6014649, rs11697509, rs3746619, rs3827103 rs1543570) were associated with AIWG in our
overall sample (see Table 5-2). To account for differing ethnic backgrounds in the patient
population, we ran analyses on a sub-sample consisting of patients of European ancestry treated
with olanzapine or clozapine (see Table 5-3).
Table 5-2. Genotypic associations between MC3R and Antipsychotic Induced Weight
in Overall Sample
SNP Genotype N = 260 Weight-gain%a p-value
Overall Sample
rs6014646 AA 92 3.51 ± 6.8 0.007
AT 123 4.50 ± 5.4
TT 57 7.57 ± 8.2
rs6024730 GG 141 3.76± 6.1 0.001
AG 103 5.09± 6.4
AA 17 10.18± 8.5
rs6024731 AA 114 3.77 ± 6.4 0.001
AG 115 4.87 ± 5.7
GG 26 9.10 ± 9.1
rs6014649 GG 189 3.64± 5.8 8.66E-06
AG 64 6.77± 6.9
AA 9 12.86 ± 9.4
rs11697509 CC 167 3.58 ±5.9 3.79E-05
CG 83 5.63 ± 6.5
GG 12 12.64 ± 8.7
rs6127698 TT 66 3.78 ± 6.9 0.175
TG 116 4.51 ± 6.1
GG 81 5.75± 6.9
rs3746619 CC 173 3.66 ±5.9 6.75E-05
AC 77 6.08± 6.6
AA 11 12.03 ± 8.8
rs3827103 GG 178 3.77 ± 5.9 1.37E-04
AG 73 5.93 ± 6.8
AA 11 12.03 ± 8.8
rs1543570 CC 153 3.53± 6.0 3.00E-03
AC 82 5.85 ± 6.4
AA 27 7.97 ± 8.2
a. Mean ± S.D. (number of individuals
102
The MC3R SNPs rs6014646, rs6024730, rs6024731, rs6014649, rs11697509, rs3746619,
rs3827103, & rs1543570 were associated with AIWG in our sub-sample consisting of patients of
European ancestry. The rs6014649, rs3746619, rs11697509, rs3827103, and rs1543570 SNPs
consisted of genotypic groups with n = 3 or lower (see Table 5-3). Thus, the group sizes for the
minor allele homozygotes consisted of low numbers, and the homozygote groups were merged
with heterozygote groups to examine dominant models. Of interest, the two highly linked SNPs
in the literature, rs3746619 & rs3827103, were also associated with AIWG in our samples,
among additional SNPs (see Table 5-3).
The total number of independent tests according to the SNPSpD method was four. After
correction for multiple testing (α = 0.0125), five MC3R SNPs remained significant (rs6014649,
rs3746619, rs11697509, rs3827103, rs1543570). In our allelic analyses in the European sub-
sample treated with clozapine and olanzapine, eight SNPs (rs6014646, rs6024730, rs6024731,
rs6014649, rs11697509, rs6127698, rs3746619, rs3827103, rs1543570) had significant allelic
associations with AIWG. The allelic associations are listed within Table 5-4. The rs6127698
SNP had no significant allelic association with AIWG.
103
Table 5-3. Genotypic associations between MC3R and Antipsychotic Induced Weight in
European Patients treated with antipsychotic medications
SNP Genotype N = 104 Weight-gain% p-value
rs6014646 AA 45 2.65±6.0 0.019
AT 46 4.40±4.5
TT 12 7.7±7.5
rs6024730 GG 61 3.06±5.6 0.066
AG 37 4.97±5.6
AA 5 8.28±6.5
rs6024731 AA 55 2.84±5.8 0.038
AG 39 5.00±4.9
GG 8 7.62±8.3
rs6014649 GG 84 3.14±5.4 0.003b
AG 18 7.21±5.6
AA 2 12.03±6.1
rs11697509 CC 78 3.06±5.4 0.000464b
CG 22 5.91±5.3
GG 2 14.5±6.0
rs6127698 TT 28 3.07±5.1 0.429
TG 49 4.13±5.7
GG 21 5.18±6.6
rs3746619 CC 81 3.07±5.5 0.006b
AC 21 6.47±5.5
AA 2 12.03±6.1
rs3827103 GG 81 3.18±5.5 0.007b
AG 21 6.47±6.1
AA 2 12.03±6.1
rs1543570 CC 73 3.11±5.6 0.007b
AC 28 5.50±5.2
AA 3 12.15±4.3 a. Mean ± S.D. (number of individuals).
b. SNPs with low numbers in a genotypic group were merged with the heterozygous genotypic group.All p-values
were calculated using ANCOVA. All p-values corrected with Nyholt testing. rs6014649: AA+AG (7.68% ± 5.7) vs.
GG (3.01% ± 5.5), p = 0.004; rs3746619: AA + AC (6.96% ± 5.7) vs. CC (304% ± 5.5), p = 0.016; rs11697509:
GG+CG (6.94% ± 6.0) vs. CC (2.94% ± 5.4), p = 0.008; rs3827103, AA + AG (6.96% ± 5.7) vs. GG (3.04% ± 5.5),
p = 0.016; rs1543570, the 6.14% ± 5.4) vs. GG (2.96% ± 5.7), p = 0.044
104
Table 5-4. Allelic differences in weight change (%) in patients of European ancestry
MC3R
Polymorphism
Allele
European
Ancestry
All
Medications
P-value
European Ancestry
Olanzapine/clozapine
P-value
rs6014646
A
0.667
0.0293 0.668 0.009
T 0.333 0.332
rs6024730
G 0.776 0.0649 0.778
A 0.224 0.222
rs6024731
G 0.725 0.0469 0.737 0.012
A 0.275 0.263
rs6014649
G 0.905 0.00373 0.898 0.00049
A 0.095 0.102
rs11697509
C 0.87 0.00341 0.866 0.0003
G 0.130 0.134
rs6127698
T 0.548 0.201 0.566 0.16
G 0.452 0.434
rs3746619
C 0.895 0.01 0.883 0.0023
A 0.105 0.117
rs3827103
G 0.895 0.01 0.883 0.0023
A 0.105 0.117
rs1543570
C 0.861 0.032 0.84 0.003
A 0.139 0.16
European panel; CEU: Utah (USA) residents with northern and western European ancestry by the Centre
d'Etude du PolymorphismeHumain (CEPH) (http://hapmap.ncbi.nlm.nih.gov/).
Power Analysis
The power analysis for this study was conducted using Quanto 1.2.4. Our study had more than
80% power to detect a difference of 3.6% between risk-genotype carriers and non-carriers in our
sample of patients in an additive model, assuming a minor allele frequency of 0.05. The variance
explained for this phenotype was 2.91%.
105
Haplotype Analysis
Haplotype analyses were conducted using UNPHASED 3.1.4. Our haplotype analyses were
conducted by using the marker distribution in our haplotype block (see Figure 5-1). We
determined the estimated additive values (EAV) for the haplotypes. The EAV indicates the
means of quantitative variable (weight change %) for a haplotype relative to other haplotypes
within a haplotype window. All of the statistical analyses in this study were based on alpha <
0.05 as nominally significant. The first haplotype assessed in AIWG was the rs6014646-
rs6024730-rs6024731-rs6014649 T-A-A-A was associated with AIWG (p = 0.004893, EAV T-A-
A-A = 0.1328, C.I.: 0.036 - 0.230). The second tested haplotype was rs3746619-rs3827103-
rs1543570 A-A-A, and was significantly associated with AIWG (p = 0.01777, EAV A-A-A =
0.1059 , C.I.: 0.0158 - 0.196). Given the exploratory nature of our analysis, we did not correct
the haplotypes for multiple testing.
Gene-Gene Interaction Analysis
Gene-gene interaction analyses were performed between MC3R and MC4R SNPs utilizing
stepwise linear regression from the mbmdr software. For the purposes of comprehensiveness, all
MC3R SNPs from this study (rs6014646, rs6024730, rs6024731, rs6014649, rs11697509,
rs6127698, rs3746619, rs3827103, rs1926064, rs1543570) were assessed with top MC4R SNPs
associated with AIWG in the study conducted by Malhotra et al., (2012), (rs12967878,
106
rs12970134, rs2045439, rs489693, rs646749 rs6567160), and a recent study conducted by our
own group (rs8087522, rs11872992, rs17782313) (Chowdhury et al., 2012).
The significance was determined using a permutation test with 10,000 simulated replicates (Calle
et al., 2010). The gene-gene interaction analyses resulted in no significant interaction between
any MC3R and MC4R SNPs.
5.5 Discussion
We investigated the association between common MC3R polymorphisms and antipsychotic
induced weight gain (AIWG). The melanocortin receptors system has biological functions in
energy homeostasis, pigmentation, and glucorticoid mediation. Several common MC4R gene
variants as well as mutations have been reported to be associated with weight gain and
development of obesity in non-psychiatric populations. However, MC3R gene variation had not
yet been broadly examined in AIWG.
Our current study found that eight of the nine tested MC3R SNPs are associated with AIWG.
The rs6024730, rs6024731, rs6014649, rs11697509. rs6127698, rs3746619, rs3827103 SNPs
were associated with AIWG in the overall sample. However, our total sample (N=266) is
comprised of various ethnicities. Therefore, we examined a sub-sample comprising of patients of
European-ancestry, treated with either clozapine or olanzapine (N=104).Five SNPs remained
associated with AIWG under a dominant model.
Two functional MC3R SNPs, 6Thr>Val (rs3746619) and 81Val>Ile (rs3827103), have been
reported to be associated with obesity across different populations (Feng et al., 2005) and are
107
associated with a reduction in cAMP signaling in vitro (Feng et al., 2005). However, a recent
study by Cieslak et al. (2013), reported that MC3R 81Ile allele was not associated with obesity in
a Polish population. Four non-synonymous coding variants (82S, N128S, L249F, and R257S)
were identified in a sample of Obese Belgian sample (Zegers et al., 2013). These mutations have
been reported to be associated with obesity in previous studies as well (Zegers et al., 2011;
Calton et al., 2009; Mencarelli et al., 2011). The MC3R 6Lys–81Ile haplotype has also been
associated with substrate oxidation in response to moderate exercise in obese children (Obregón
et al., 2011). In a study conducted by Moons et al., (2011), the MC3R 6Thr>Val SNP was not
associated with olanzapine induced weight gain in SCZ patients. In our study, we found that the
two non- synonymous polymorphisms, 6Thr>Val (rs3746619) and 81Val>Ile (rs3827103), were
associated with AIWG in our sample. The sample described in Moons et al., (2011) consisted of
male schizophrenia patients. Our sample was comprised of both male and female patients, and
the sample was assessed for weight gain for up to six weeks. Thus, there are notable differences
between the patients described in Moons et al., (2011), and the population described in the
current study.
A recent study conducted by Santos et al., (2011) reported that the MC3R rs6014646 ‘T’-allele
was nominally associated with weight loss in a clinical trial consisting of an obese adult
population. The association between the MC3R rs6014646 SNP and AIWG did not survive
correction for multiple testing. In addition, the MC3R rs6014646-rs6024731, rs6024731-
rs6014649, rs6014649-rs11697509, rs11697509-rs6127698, rs6127698-rs1543570 haplotypes
achieved statistically or near-statistically significant results in relation to weight loss (p > 0.05).
108
However, given the high correlation between the MC3R SNPs in this study (Figure 5-1), the
haplotype results must be interpreted with caution and require replication in independent samples
for validity.
The melanocortin-4 receptor (MC4R) has a highly established role in general obesity (reviewed
in Friedman et al., 2000; Blakemore 2014). Recently, two separate studies highlighting the
genetic association between MC4R variants and AIWG (Chowdhury et al., 2012; Malhotra et al.,
2012) have been reported. We reported that the MC4R rs8087522 promoter SNP may have a
nominal association with olanzapine induced weight gain in European ancestry patients, and that
this SNP may have a functional role in regulating transcription factor binding.
In addition, a genome wide association study found several SNPs located near MC4R to be
associated with AIWG (Malhotra et al., 2012). However, our gene-gene interaction analysis
showed no significant interaction between the MC4R SNPs recently reported in the literature and
the MC3R SNPs in this study. The results from this analysis suggest that MC3R and MC4R may
act independently in their roles in regulating AIWG. Though the results found in our study are
compelling, some limitations should be considered. Our sub-sample of patients of European
ancestry was low in sample size, and the numbers within genotypic groups became low in our
sub-analyses. Nonetheless, there may still be a small effect of MC3R on the onset of AIWG. We
had sufficient power to detect small effects of the MC3R on AIWG. These findings require
replication and further study in independent sample sets. Another consideration is that the SNPs
were highly correlated in our sample set (see Figure 5-1). Thus, one MC3R SNP may be the
critical regulator of AIWG, while the remaining SNPs may serve as a proxy. Investigation of
MC3R SNPs in an independent sample set may further elucidate the specific polymorphisms
which play a role in AIWG.
109
Overall, our results are indicative of an influential role for the MC3R gene in AIWG. Future
studies could assess MC3R SNPs in larger sample sets in order to validate these findings, and to
further determine whether the MC3R gene interacts with other leptin-melanocortin system genes
which may result in increased risk to gain weight.
110
Chapter 6 General Discussion
111
6.0 General Discussion, Conclusions and Future directions
6.1 Summary of Findings
The three main studies (Chapters 3-5) presented in this thesis focused upon demonstrating a role
for associations between melanocortin system genes and antipsychotic induced weight gain
(AIWG). The main experiments in this study demonstrated the following results: (a) the MC4R
rs8087522 SNP was nominally associated with AIWG, and that the rs8087522 ‘A’-allele may
have a functional effect, (b) each of the POMC, AGRP and CART genes were not significantly
associated with AIWG in our samples, and the roles of these genes may require further
exploration in larger sample sizes, (c) the correlated MC3R SNPs rs6014649, rs3746619,
rs11697509, rs3827103, rs1543570 SNPs are significantly associated with AIWG.
In the first study, we investigated whether the MC4R gene is associated with AIWG in SCZ
samples. MC4R is located in on 18q22, and the gene encodes a protein that functions as g-
protein coupled receptor. The protein interacts with peptides including the adrenocorticotropic
and melanocyte stimulating hormones. The MC4R gene is comprised of one exon, with no
intron, and mutations in this gene have been a reported cause of autosomal dominant obesity
(Yeo et al., 1998; Vaisse et al., 1998). To the best of our knowledge, our group was the first to
examine the MC4R gene and its possible association with AIWG in a psychiatric sample. The
MC4R SNP rs17782313 was initially identified in a genome wide association study examining
risk variants for obesity in an adult and pediatric population (Loos et al., 2008). Interestingly, the
MC4R rs17782313 and MC4R rs11872992 SNPs were reported to be associated with high BMI
in this study (Zegers et al., 2011). This rs17782313 SNP is located approximately 188 kb away
112
from the MC4R gene. Since the publication of this study, dozens of articles have been published
indicating that the MC4R rs17782313 is associated with several weight measures including high
BMI, as well as dysregulated eating behaviours (Zobel et al., 2009; Stutzmann et al., 2009;
Renström et al., 2009; Cauchi et al., 2009; Grant et al., 2009; Timpson et al., 2009; Li et al.,
2010; Kring et al., 2010; Hardy et al., 2010; Cheung et al., 2010; Liu et al., 2010; Petry et al.,
2010; Valladares et al., 2010; Orkunoglu-Suer et al., 2011; Croteau-Chonka et al., 2011; Tao et
al., 2012; Thomsen et al., 2012; Lombard et al., 2012; Xi et al., 2012; Yang et al., 2013; Ortega-
Azorín et al., 2012; Dwivedi et al., 2013; Zlatohlavek et al., 2013; Mejía-Benítez et al., 2012; Xi
et al., 2013; Sull et al., 2013; Marcadenti et al., 2013; Dušátková et al., 2013; Jääskeläinen et al.,
2013; Hortsmann et al., 2013; Ho-Urriola et al., 2014; Mutombo et al., 2013; Srivastava et al.,
2014; Yilmaz et al., 2014). Overall, the MC4R rs17782313 ‘C’-allele is associated with risk for
obesity (Loos et al., 2008). Despite some reports that the MC4R rs17782313 is not significantly
associated with obesity (Vasan et al. 2012; Valette et al., 2012; Arrizabalaga et al., 2014), the
majority of studies have reported that rs17782313 SNP has an association with weight measures.
A recent meta-analysis examined 61 studies for rs17782313 SNP in obesity, and reported a
significant risk for obesity (obesity risk (OR=1.18, 95% CI=1.15-1.21, p<0.001) (Xi et al.,
2012).
In our first study, we investigated whether the MC4R SNPs rs8087522, rs17782313 &
rs11872992 were associated with AIWG. We found that the MC4R rs8087522 SNP ‘A’-allele
was associated with clozapine induced weight gain, and that the ‘A’-allele may regulate or create
a transcription factor binding site. We interpreted these findings with caution, considering that
the sample size of patients treated with clozapine was relatively low (n = 69). The MC4R
113
rs8087522 SNP has been previously reported not to be associated with obesity in a case control
study comprised of obese and healthy individuals (Beckers et al., 2011). In a recent study
(Czerwensky et al., 2013), MC4R rs17782313 ‘C’-allele carriers were reported to be at risk for
AIWG in a sample of patients treated with atypical antipsychotics (clozapine, olanzapine,
risperidone, paliperidone, quetiapine, or amisulpride), and were assessed in a naturalistic study
design. Limitations of this study include that patients were assessed for antipsychotic treatment
for only four weeks.
The MC4R rs489693 SNP has been examined within obesity, though this SNP has not been
reported as a primary hit in the context of obesity GWAS studies. Our research group conducted
a collaborative research project with a laboratory affiliated with Zucker Hillside Hospital (Long
Island, NY) and the GWAS revealed that the rs489693 was significantly associated with AIWG
in a pediatric sample treated with second generation antipsychotic medications, along with more
than ten other SNPs located near the MC4R gene (Malhotra et al., 2012). Furthermore, the
association between rs489693 was associated with AIWG in three distinct samples, a rarity in
psychiatric genetics. Replication of the association provides strong evidence that the rs489693
SNP has a role in AIWG. However, further study of the exact influence of the MC4R rs489693
upon weight regulation and AIWG are required. Previously, the MC4R rs489693 SNP was
associated with obesity in a large sample size of over 30 000 participants collected from various
consortiums (Heard-Costa et al., 2009). However, the specific investigation of this SNP remains
relatively unexplored. More recently, the MC4R rs489693 SNP 'A'-allele was significantly
associated with AIWG in a sample of patients treated with olanzapine, clozapine, risperidone,
paliperidone, quetiapine or amisulpride (Czerwensky et al., 2013). However, the 'C'-allele was
114
associated with high BMI in a pediatric sample treated with risperidone (Nurmi et al., 2013).
Nonetheless, the rs489693 ‘A’-allele association across distinct samples remains a strong finding
in the pharmacogenetics of AIWG (Malhotra et al., 2012; Czerwensky et al, 2013).
The potential functional role of SNPs such as rs17782313 and rs489693 are currently unknown.
The two SNPs are located downstream of MC4R gene, approximately ~200,000 base pairs away
from the gene. Examinations of the Encyclopedia of DNA Elements (ENCODE) functional
information using the UCSC genome browser and HaploReg did not indicate a potential
functional effect. Given the evidence described above, it appears that certain MC4R SNPs are
implicated within the realm of general obesity (rs17782313), and that the rs489693 specifically
appears to be more strongly associated with AIWG. A recent study by Yilmaz et al., (2014),
reported that the rs17782313 was more associated with depressed mood and over-eating
behaviours, while no association was detected between either rs489693 or rs8087522 and over-
eating measures. Nonetheless, further study of potential functional effect would help to
determine how exactly MC4R SNPs may act in order to alter the onset of AIWG.
In the second manuscript, Chapter 3, we investigated whether hypothalamic peptide genes,
including POMC, CART and AGRP were associated with AIWG. These critical key endogenous
ligands that bind to MC4R were examined in Chapter 3. As discussed above, leptin has been
reported to account for a small portion of the variance of AIWG. However, the role of leptin’s
biological targets located in the hypothalamus has not yet been extensively studied in terms of
AIWG. Leptin stimulates POMC and CART expressing neurons within the arcuate nucleus of
the hypothalamus, eventually resulting in decreased food intake and body weight (Santini et al.,
115
2009). The POMC gene encodes for different neuropeptides, including the adrenocorticotropic
hormone, and α-melanocortin stimulating hormone. The α-melanocortin stimulating hormone
exerts its anorexigenic effects largely through melanocortin-3 and melanocortin-4 receptor. To
the best of our knowledge, after review of the literature the genetics of the proopiomelanocortin
system and AIWG had not yet been thoroughly examined. Adult male Sprague Dawley rats
treated with olanzapine exhibit increased weight gain, and decreased proopiomelanocortin
expression in the arcuate nucleus (Weston-Green et al., 2011). This finding supports the
biological rationale to study POMC gene variants in AIWG and the hypothesis that variation
within the POMC gene would be associated with AIWG. The CART is a neuropeptide which is
highly expressed within the hypothalamus and the nucleus accumbens. In 2004, Beaudry et al.,
demonstrated that CART mRNA levels are altered in the nucleus accumbens of rats treated with
clozapine. Recently, leptin has been shown to have a stimulatory effect on CART production in
mice that were not exposed to acute chronic stress (Xu et al., 2014). Within human genetics
obesity studies, the findings with CART remained mixed. The CART variants Phe34Leu and
Delta A1457 were associated with obesity in Italian children and adolescents (del Giudice et al.
2001; Rigoli et al. 2010). Though relatively little is known in regard to the role of CART in
AIWG, the CART has been cited as a critical regulator of weight, energy homeostasis and is
regulated by leptin (Hunter et al., 2004). Thus, the investigation of CART was relevant in the
context of AIWG.
The agouti-related protein (AGRP) is an endogenous antagonist of the MC4R. Olanzapine
administration in rats has been reported to result in an up-regulation of AGRP in the arcuate
nucleus in rats (Ferno et al., 2011). However, AGRP plasma levels do not differ between patients
116
treated with olanzapine or ziprasidone (Ehrlich et al., 2012). In terms of AGRP and AIWG, our
study was unable to demonstrate an association between AGRP SNPs and this phenotype. We
utilized tagSNPs for full coverage of the AGRP gene, however, no associations were detected.
Thus, it appears that AGRP may not have a critical role in mediating AIWG. However, studies in
larger sample sets are required to rule out minor contribution of AGRP to AIWG.
Overall, Chapter 3 demonstrated that genes that have a high profile in terms of weight
regulation in general obesity do not necessarily translate to being implicated in AIWG. The
POMC, CART and AGRP were not associated with AIWG in our samples. However, our
findings do not definitively negate the potential roles of these genes in AIWG, especially if the
effect sizes are small. Investigation of these SNPs in additional sample sets is warranted. The
role of the NPY is further discussed in published work by our group (Tiwari et al., 2013), and
also appears to have significant role in the development of AIWG. The NPY rs16147 ‘C’-allele
was associated with weight change, as well as two additional NPY polymorphisms (rs5573 and
rs5574). A significant gene-gene interaction between the NPY rs16147 and cannabinoid receptor
1 rs806378 was also detected, illustrating complex epistatic interactions that may be contributory
toward AIWG.
In Chapter 4, we examined the role of the melanocortin-3 receptor in AIWG. The melanocortin-
3 receptor has been reported to have a role in weight regulation according to animal studies –
MC3R knockout mice exhibit increased weight despite hypophagia. Furthermore, mice which
are lacking both MC3R and MC4R genes have been reported to be more obese than either MC3R
-/- or MC4R -/- mice (Chen et al., 2000). Thus, it appears that MC3R has role distinct from
117
MC4R in weight regulation, and that these two receptors have their own, non-redundant
functions. The MC3R has not been extensively studied in AIWG, either in human or animal
studies, and it is unknown whether antipsychotic treatment alters MC3R expression, or MC3R
plasma levels. Two non-synonymous SNPS (Thr6Lys & Val81Ile) have been associated with
higher weight in pediatric samples, and reduced MC3R expression in an in vitro study (Feng et
al., 2005). The MC3R variants associated with weight gain were more prevalent among obese
children of African American ancestry compared to children of European ancestry (Feng et al.,
2005). Thus, the exact impact of MC3R in obesity is not well understood; however, it remains a
compelling candidate gene for AIWG.
The MC3R rs3746619 was not significantly associated with AIWG in a sample of schizophrenia
patients treated with second generation antipsychotic medications (Moons et al., 2011).
However, that study had a limited sample size (n = 261), and patients were treated with
medications in a naturalistic manner. Furthermore, only the rs3746619 SNP was assessed in this
sample, while previous obesity studies had shown that the combination of this SNP and an
additional SNP in high LD were associated with higher weight in children (Thr6Lys and
Val81Ile). Our study investigated SNPs that covered the common genetic variation in the MC3R
gene, in order to comprehensively understand the impact of MC3R variations in AIWG. Our
sample size is also limited, however, our power calculation determined that at a given reasonable
minor allele frequency, we could detect a genetic effect within our sample. We were able to
determine that five MC3R variants were associated with AIWG in our samples, under a
dominant model, and corrected for multiple testing. However, these findings have not yet been
replicated in an independent sample with patients assessed for AIWG, and should therefore be
118
considered as preliminary. Future studies would include larger sample sets to validate whether
the MC3R has an influence on the development of AIWG.
Given that the MC3R and MC4R have been postulated to have non-redundant roles in terms of
weight gain and energy homeostasis in animal models (Chen et al., 2000). Thus, it was
imperative to assess potential interactions between the two genes. We used the R-based MB-
MDR package to assess pairwise MC3R x MC4R interactions with percentage change in weight
as the phenotype. No significant interactions were detected between MC3R SNPs in our study, or
the MC4R SNPs (Chowdhury et al., 2013) or the findings from our collaborative study (Malhotra
et al., 2012). Overall, our findings with these analyses indicate that the MC3R and MC4R do not
act together to contribute to the development of AIWG. To the best of our knowledge, this study
was the first to attempt to examine the potential additive effects of MC3R and MC4R in AIWG.
However, the potential of epistatic effects between biological factors that have an association
with AIWG warrant further investigation in larger sample sets.
The evidence provided in this thesis provides support for the melanocortin system as an
important regulator of AIWG, and genes encoding peptides within the system may provide
insight as to which patients may go on to develop AIWG. However, further molecular genetic
studies, and recently available databases such as the ENCODE may help in investigating the
exact mechanism of AIWG, and more precisely which genetic variants are actually contributing
to the development of AIWG phenotype.
119
6.2 Obesity and Antipsychotic Induced Weight Gain
Obesity has become a serious worldwide health concern, and a risk factor for many severe health
problems including type 2 diabetes and heart disease. The interaction between environmental and
genetic factors likely contributes to the onset of obesity. Adding to the complexity of
understanding the biological processes contributing to obesity is the many genetic risk factors
that have been identified. Genome wide association studies have highlighted both the fat mass
and obesity-associated (FTO) gene and and the melanocortin-4 receptor as being highly
associated with obesity (Loos et al., 2008; Speloites et al., 2010). The MC4R is a critical
regulator of weight and is expressed within hypothalamic nuclei, a key component of the brain in
regulating appetite and satiety. Ligands such as α-melanocyte stimulating hormone and agouti
related protein interact with the melanocortin receptor in order to regulate hunger signaling.
MC4R rare mutations are associated with severe obesity (Vaisse et al, 1998; Hebebrand et al.,
2010). The MC4R non-synonymous SNPs (rs2229616 / Val103Ile and rs52820871 / Ile251Leu)
have been reported to be associated with obesity (Geller et al., 2004; Heid et al., 2005;
Stutzmann et al., 2009; Young et al., 2007; Wang et al., 2010; Meyre et al., 2009; Xiang et al.,
2006). The MC4R 103Ile variant has been reported to increase MC4R function, resulting in
lower body weight. Thus, the MC4R 103Ile variant has been reported to have a protective effect
(Xiang et al., 2006). In addition, the MC4R SNP rs17782313 has been reported to be associated
with obesity in GWAS (Loos et al., 2008; Speliotes et al., 2010). However, in our study, the
MC4R rs17782313 SNP was not associated with AIWG (Chowdhury et al., 2012). In our
collaborative study (Malhotra et al., 2012), the MC4R rs17782313 was among the top SNPS
associated with AIWG, though this signal was not subsequently replicated among independent
samples. To the best of our knowledge, the top MC4R SNP associated with AIWG (rs489693:
120
Malhotra et al., 2012; Czwerensky et al., 2014) has not been reported to be associated with
general obesity. We also investigated common variations of three independent genes (POMC,
CART & AGRP) and found no association with AIWG (Chowdhury et al., 2014). Variants of
these genes have not been associated with general obesity in genome wide association studies,
though the roles of POMC, CART and AGRP have been well established in studies investigating
the physiology of weight gain in animal models (Biebermann et al., 2012). Interestingly, the
melanocortin-3 receptor SNPs investigated in our study (rs3746619 and rs3827103) were both
significantly associated with AIWG as well as general obesity (Feng et al., 2005). Nonetheless,
the MC3R variants have not been implicated with general obesity in large scale GWAS
investigations. Thus, taken together, these findings indicate that distinct genetic mechanisms may
govern general obesity vs. AIWG. In addition, a distinct epistatic effect of melanocortinergic
system SNPs may be important in the development of AIWG.
6.3 Multi-Gene Modeling
In Chapter 5, we utilized the model based multifactor dimensionality reduction (mbmdr)
software to determine whether an interaction between the MC3R and MC4R genes contributed to
AIWG in an additive manner. In our study, we found no evidence of an interaction effect across
the variants between the two genes. Nonetheless, the effect of the combination of SNPs across
genes may provide crucial insight toward the genetic regulation of AIWG. Currently, there are
no reliable biomarkers or tests that can predict the onset of AIWG.
121
Our group recently used the most promising findings from across our multiple previous and
ongoing studies to attempt to develop a predictive model which could help determine which
patients were at risk for gain weight. The SNPs used in the model include variations from the
gamma-aminobutyric acid receptor subunit alpha-2 (GABRA2), translocator protein-18 (TSPO),
NADH-ubiquinone oxidoreductase 75 kDa subunit (NDUFS1), orexin receptor (HCRTR2),
glicentin-related polypeptide -1 encoding gene (GCG), melanocortin-4 receptor (MC4R),
cannabinoid receptor 1, (CNR1) and neuropeptide Y (NPY) genes.
The rationale of investigating the CNR1 and the NPY in AIWG is more extensively described in
Chapter 1 of this thesis (see sections 1.7.4 and 1.7.5, respectively). Tiwari et al., (2010)
investigated the potential role of CNR1 in AIWG. Twenty SNPs across the CNR1 gene were
examined in psychiatric patients treated with antipsychotic medications. The CNR1 rs806378
polymorphism was found to be associated with weight gain in patients of European ancestry
treated with either clozapine or olanzapine. Carriers of the rs806378 T-allele (CT + TT
genotype) gained more weight (5.96%), than the CC homozygote group (2.76%, p=0.008)
(Tiwari et al., 2010). In another paper, five NPY polymorphisms were analyzed in our
antipsychotic treated sample set, and the NPY rs16147 was significantly associated with weight
change (p=0.012). Patients who received either clozapine or olanzapine and carrying the C-allele
gained significantly more weight with CC+CT gaining 5.82%±5.6 compared to individuals with
TT-genotype who gained on average 2.25%±4.8 (p=0.001) (Tiwari et al., 2013).
Our group has also investigated the glucagon-like peptide gene (GLP-1), a critical endogenous
glucose regulator. It has been postulated that GLP-1 inhibits AMPK activity, resulting in reduced
122
food intake (Burmeister et al., 2013) in mice. Recently, a GLP-1 analog, liraglutide, has been
demonstrated to treat weight gain in Type 2 Diabetes patients (Flint et al., 2013). Our group
investigated four GLP-1 (GCG) gene polymorphisms in AWIG, and found that the rs13429709
‘C’-allele was associated with this phenotype (p = 0.002) (Brandl et al., 2014).
Recently, our group has also examined the orexin receptors, orexin receptor 1 (OX1R or
HCRTR1) and orexin 2 receptor (OX2R or HCRTR2). Both receptors are expressed in the
hypothalamus, and have established roles in feeding behaviour (Sakurai et al., 1998). In our
sample, the OX2R rs3134701 and rs4142972 SNPs were nominally associated with AIWG
(Tiwari et al., submitted).
Finally, our group has also been highly involved in examining the role of mitochondrial genes in
AIWG. The mitochondria organelle is a pimary energy source for neurons, and has several
distinct functions. It has been reported that clozapine and olanzapine may impact mitochondrial
function. For instance, clozapine has been proposed to inhibit activity of complex I, and may be
involved in oxidation of mitochondrial specific proteins (Baig et al., 2010). Our group found that
two mitochondrial system genes, including the NADH dehydrogenase (ubiquinone) Fe-S protein
1, 75KDa (NDUFS1), and the translocator protein-18 kDa (TSPO), were associated with AIWG.
The NDUFS1 gene is among the most conserved mitochondrial genes, and is a part of the
complex I system. Complex I protein expression has been reported to be down regulated in
postmortem brains of SCZ patients (Maurer et al., 2001; Prabakaran et al., 2004). The TSPO
gene is involved in biosynthesis of steroids, and may also have a role in general weight
regulation (Giannaccini et al., 2011). The TSPO ligand, PK1195, has been reported to lower lipid
123
accumulation in the liver, and free and LDL cholesterol, as well as blood glucose levels in
zebrafish (Gut et al., 2013). The TSPO gene has been associated with weight gain in our samples
assessed for AIWG (Pouget et al, submitted).
We incorporated genotypic information from individual’s SNPs to determine risk for AIWG
development. The mode of inheritance for each SNP was determined using post hoc pair-wise
comparison of mean AIWG among the genotypic groups. The total risk score for each individual
was the sum of the individual scores for each genotype of the SNPs included in the model. Based
on the post hoc comparisons, genotypes were re-coded as ‘0’ = no risk or as ‘1’ = at-risk for
AIWG. The total risk score for each individual was determined by adding genotypes across the
nine loci leading to a score ranging from 0 (no-risk) to 9 (highest risk). Finally, the score was
then entered as a factor in an ANCOVA analysis in order to determine the cumulative effect of
these variants on AIWG, with baseline weight and duration of treatment as covariates in the
model. In our model evaluation of the nine SNPs, individuals who carried from zero up to three
risk variants gained the least amount of weight (0.81% ± 3.7) compared to individuals with four
or more risk alleles. This nine SNP model can explain 68.1% of the variance in AIWG (United
States Patent 61946003, filed 2014). If this genetic model can predict AIWG in other samples, it
could be a clinically useful tool for physicians to utilize in the future to determine which
medications are most beneficial to individual patients. The nine SNP model comprised in our
laboratory requires replication in an independent sample set in order for validation of its
predictive utility (Full details of model described in our US patent 61946003, 2014).
124
6.4 Postulated Mechanisms of Antipsychotic Induced Weight Gain
Combination models such as the one described above provide evidence that the interaction
across several genes may be a way to determine which individuals are predisposed to gaining
weight when treated with antipsychotic medication. Despite these models, the exact mechanism
of combined genetic effects in AIWG is still not well understood. The mechanisms by which a
subset of patients exhibit substantial weight gain after second generation antipsychotic treatment
remains to be determined in terms of the detailed biological mechanisms.
Full understanding of AIWG is extremely complicated, since this phenotype is multi-faceted and
governed by several factors. The complexity is well illustrated by the association of various
SNPs reported in the literature that are described as being contributory to the variance of AIWG.
Some considerations for discerning a primary mechanism include the potential overlap between
several gene systems, and that signal transduction pathways stemming from primary
antipsychotic medication receptor targets results in virtually thousands of possible candidates in
AIWG risk. Recently, with the availability of databases such as the ENCODE, the role of
enhancers, amplifiers, regulators, and transcription factors are becoming extremely important to
examine in molecular genetic studies. The finding that the RNA molecule microRNA-137 has
an influence in schizophrenia risk (Ripke et al., 2011; Lett et al., 2014; Ripke et al., 2014),
demonstrates that factors targets other than the classic dopamine and serotonin system gene
variants are important to investigate in order to further understand genetic mechanisms. It is
important to consider that even if a receptor has not been identified in the receptor binding
profile for antipsychotic medications, that this does not exclude the receptor from having a
potential regulatory effect of AIWG (Panariello et al., 2011). Once example of this situation is
125
the mechanism of the HR1 mediated AMPK phosphorylation in mice treated with clozapine
(Kroeze et al., 2003). It has also been shown in animal models that AMPK activity is mediated
by the melanocortin pathway. Administration of a melanocortin agonist (MT-II) was shown to
overcome high-fat diet-induced leptin resistance in mice, and alter AMPK activity (Masuzaki et
al., 2009). More recently, an in vivo study examined whether α-MSH affects AMPK via direct
intracellular signaling cascades. The α-MSH was shown to result in dephosphorylation of
AMPK, and this is mediated by enzymes including extracellular-signal-regulated kinases (ERK -
s1/2) (Damm et al., 2012). The authors proposed that further study of mechanisms could lead to
development of drugs for the treatment of general obesity. Whether antipsychotic medications
alter AMPK phosphorylation via melanocortin mediation remains to be determined. In terms of
genetic studies, a study conducted by our group (Souza et al., 2011), detected no association
between AMPK subunit polymorphisms and melanocortin SNPS in AIWG. Nonetheless, altered
AMPK activity due to antipsychotic administration could be one of many potential mechanisms
that could be a reason for the onset of AIWG.
Finally, an additional topic of importance in the scope of this research is the understudied aspect
of the metabolic syndrome which may develop shortly after antipsychotic treatment. The
prevalence of type 2 diabetes in schizophrenia patients has been reported to be three times higher
than the general population, even prior to treatment (Nasrallah et al., 2008). Thus, it appears that
some schizophrenia patients may have a genetic predisposition toward diabetes, adding to the
complexity of examining altered metabolic indices associated with antipsychotic treatment
(Tschoner et al., 2007). The development of type 2 diabetes may occur as a result of the weight
gain. However, within 6 weeks of SGA treatment, hyperglycemia has been reported to occur in a
126
subset of patients (Tschoner et al., 2007). The effects of antipsychotics on glucose dysregulation
per se have been proposed to be due to the inhibitory effects of muscarinic and serotonergic
antagonism upon insulin secretion (Hahn et al., 2011). Therefore, the onset of type 2 diabetes
may be independent of the substantial weight gain effect, and be a regulated by a distinct
mechanism.
6.5 Limitations and Considerations
6.5.1 Sample Characteristics
Though our studies indicate a role of the melanocortinergic system in AIWG, some limitations
should be considered for the interpretation of our results. Firstly, our sample was comprised of
patients with distinct genetic ancestries. Differing genetic ancestries are a consideration within
our studies, as allele frequencies can vary greatly across different populations. As a result of the
varying allele frequencies between ethnic populations, the risk of disease and susceptibilities to
developing disease may vary greatly (http://depts.washington.edu/cgph/pdf/PopStrat.pdf). In
order to account for the differing ethnicities in our studies, we separated our total sample based
on self-reported ancestry. Participants were asked to disclose their ancestry, and this was how
ethnicity was determined for our studies. In the future, genetically determined ancestry will be
more technically accurate. For instance, genetic markers can be assessed across individuals in
order to determine a population’s genetic structure. The ancestry informative marker (AIM)
panel examines a set of 62 markers polymorphisms across the genome that have been selected to
highlight differences in allele frequencies, among populations from different geographical
127
regions (Bauchet et al., 2007). This method is considered more advantageous as it determines
ancestry based on individuals’ genetic information as opposed to self-report.
Another consideration for our analyses is that within our samples, approximately 40% of patients
in our sample have been previously treated with other antipsychotic medications. Notably, the
findings across the three studies were conducted in the same clinical samples. In the context of
AIWG studies, previous antipsychotic treatment can be limitation. If a patient was previously
treated with antipsychotic medication prior to baseline measurement of weight for a study, it is
difficult to ascertain what portion of their overall weight gain is due to the antipsychotic under
investigation. However, in our analyses, a sub-set of patients were treated with a second
generation antipsychotic medication for the first time (Sample B) (described in Masellis al.,1998;
Tiwari et al., 2010).
6.5.2 Sample Size and Power
Our current total sample of psychiatric patients assessed for AIWG (n= 266) has more than 80%
power to detect a genotypic relative risk of as low as 3.6%, if we genotype polymorphisms with
minor allele frequencies of 20% and set the critical p-value at α = 0.05. Our sample size is well
characterized for weight gain assessments, and is one of the largest in the world. We had
adequate statistical power to provide meaningful results. Moreover, our experiments across our
papers began with specific hypotheses and used evidence from the pharmacology, appetite
biology and literature. However, our investigation of the melanocortin system should be
considered exploratory and novel, and should be investigated in independent sample sets.
128
Given that our total sample was divided based upon ethnicity for our sub-analyses, a number of
our statistical analyses were conducted in pools of relatively low sample sizes (n = 104). Thus,
for our sub-sample of patients of European ancestry treated with olanzapine or clozapine, our
sample size was reduced from 266 patients to approximately 104 patients. The issue with a
limited sample size is that genotypic group sizes become very small, and the comparisons
between groups become difficult to interpret. Furthermore, an additional issue in genetic
association studies is that low minor allele frequencies may result in very low group sizes. In
fact, in our analyses of the MC3R and MC4R genes, in order to assess various types of genotypic
models (e.g. dominant vs. recessive), we were required to merge genotypic groups with the
heterozygote group to have large enough sample sizes to have meaningful comparisons. The low
sample size for the European ancestry group treated with risk medications was further
problematic when we examined gene-gene interaction between MC3R x MC4R in Chapter 5.
We detected no gene-gene interaction between MC43R x MC4R, and this may be due to lack of
power from small group sizes.
Often, samples are not large enough to detect the small effect sizes expected in complex traits,
and the inability to replicate the same allelic associations across separate samples points to
common limitations among many psychiatric genetic studies. For instance, whether our samples
were too low in size to detect an effect for the POMC, CART, or AGRP genes, or they have no
influence on the development of AIWG, is difficult to determine. It is important to acknowledge
that our collaborative group (Malhotra et al., 2012), recently demonstrated that effects can be
detected across relatively low sample sizes. The MC4R rs489693 SNP associated with AIWG
was identified in a patient group consisting of 139 subjects and was replicated in three
129
independent studies. These were also relatively small, with (73, 40, and 92 subjects). In an
overall meta-analysis, the effect size for the weight gain was 3 and the p value was 2 x 10-12
(Malhotra et al., 2012). Thus, the importance of MC4R in weight gain in AIWG is further
highlighted, as well as the ability to detect large effect sizes in small samples. Future studies with
larger sample sizes may be able to further understand the role of such genes in the cause of
AIWG.
6.6 Correction for Multiple Testing
An additional consideration for our findings regarding the MC4R and MC3R genes is a caveat
that is often seen in psychiatric genetic studies – the possibility that a statistically significant
result may be a false positive due to multiple testing. This is important to consider in the context
our studies, given that the SNPs across the melanocortinergic genes, in addition to many other
SNPs, were investigated in the same local samples. Despite the hypothesis driven and
exploratory nature of our studies, the issue of correction for multiple testing must be considered.
For our studies investigating melanocortin system genes in AIWG, the statistical threshold was
set at p = 0.05, the general accepted value. The p-value of 0.05 is a probability which indicates
that five tests out of 100 conducted would show a positive finding, even if the null hypothesis
were true. Simply put, the higher the number of performed tests, the higher the probability of
discovering false positives. The process of correcting for multiple testing is a fine balance, as
correction that is too stringent may not allow for observing true positive associations within a
dataset (Perneger et al., 1998). For the purposes of our studies, either Bonferroni (Abdi, 2007) or
SNPSpD (Nyholt, 2004), were utilized to correct for multiple testing. Bonferroni correction is
130
appropriate for genetic association studies when the SNPs are fully independent, that is, not in
linkage disequilibrium, or are not correlated with each other. The Bonferroni correction divides
the significance threshold (p = 0.05) by the number of conducted tests. Thus, we utilized
Bonferroni correction for the investigation of the MC4R gene (Chapter 2), and for the POMC,
AGRP, and CART (Chapter 3) genes. In terms of studying the MC3R gene (Chapter 4), there
was relatively high correlation across the 10 SNPs, and the rs3746619 and rs3827103 were in
complete linkage disequilibrium. Thus, utilizing the SNPSpD (Nyholt, 2004) method was
appropriate in this case.
Overall, the melanocortin system genes examined within this thesis were relatively unexplored in
terms of previous studies investigating AIWG. Prior to our studies, the leptin and leptin receptor
genes were explored by various research groups in order to determine whether certain variations
of these genes explained risk for development of AIWG (reviewed in Brandl et al., 2012).
However, a limited number of studies have investigated the melanocortin receptors in the context
of AIWG. One common inconsistency observed in AIWG studies include a conundrum in which
opposite alleles are associated with conferred risk for a phenotype.
6.7 Replication in Independent Sample Sets
An additional obstacle commonly encountered in psychiatric pharmacogenetic studies is that
associations between SNPs and a given phenotype are typically difficult to replicate, or the same
allele is not associated with a given phenotype when researchers attempt to replicate the initial
finding (Lin et al., 2007). An in-depth analytical study by Lin et al., (2007) suggested that the
131
observed effect of a genetic variant may differ between studies due to correlation differences
between the marker SNP and the actual causal variant. The examination of a single locus and a
disease resulting in inconsistent findings may be due to overlooking of other genetic loci effects
as well as environmental factors that could work in tandem with the candidate gene or locus. So
called ‘flip-flop’ associations may be due to a causative variant that is in variable LD with a non-
casual variant – where the non-casual variant is reported to be positively associated with a
phenotype. However, when an independent research group attempts to replicate the association, a
negative association or the previously associated allele may come to have a protective role as
opposed to risk. Additional factors to be considered when observing flip-flop phenomenon in a
dataset include the fact that LD can vary greatly across different ancestral populations. Finally,
sampling variation may also result in opposite allele findings between studies. Statistical models
have indicated that an r2 value as low as 0.3 between markers should still be examined carefully
to explain opposite-allele effects (Lin et al., 2007). Overall, replication samples consisting of a
carefully characterized and specific phenotype, in the case of our research focus –AIWG, are
often difficult to obtain. We were able to collaborate with a research group who examined a
sample of pediatric patients treated with risperidone - Research Units on Pediatric
Psychopharmacology (RUPP) patients - for weight gain measures. Variations within the LEP and
CNR1 genes were associated with AIWG in the RUPP sample (Nurmi et al., 2013). However,
the ‘C’-allele of the MC4R rs489693 was nominally associated with AIWG, which displays an
opposite allele association. Overall, the MC4R rs489693 ‘A’-allele as the risk allele has been
replicated across various samples assessed for AIWG (Malhotra et al., 2012; Czerwensky et al.,
2014). Closer examination of LD patterns between both our local sample and the RUPP sample
were not deemed to be the cause of the A-allele being positive in our sample, and the C-allele in
132
theirs for the MC4R rs489693. This same phenomenon or the “flip-flop phenomenon” was
observed with our MC3R findings and the subsequent attempt to replicate within the RUPP
sample. The opposite-allele effects within the MC3R must be examined more carefully in future
sample sets.
6.8 Future Directions
6.8.1 Measures of Weight Change
In the studies presented across Chapters 3,4 &5, our phenotype was measured as percentage
weight change. More specifically, AIWG was measured as the change in weight between
baseline and after a minimum of six weeks of being treated with antipsychotic medication. We
were able to use this measure of weight change in our studies, given that this information was
consistently available across the clinical data in our samples. In the general literature describing
AIWG, the US Food and Drug Administration (FDA) reports a 7% threshold for assessing
weight gain. Patients who gain more than 7% weight after the initiation of antipsychotic
treatment are considered to have gained a significant amount of weight. We chose to examine a
quantitiate phenotype (percentage change in weight), as this approach is more a more
informative analysis compared to examining a categorical phenotype.
Body mass index (BMI) is the most common method to measure weight in general obesity
studies (reviewed in Section 1.6). Assessing BMI is a relatively inexpensive and reliable method
for screening weight changes, and has been shown to be correlated with direct measures of body
fat (Mei et al., 2002; Garrow et al., 1985). In addition to measuring body fat, waist circumference
(abdominal weight) has also been implicated as an important measure in obesity studies.
Abdominal weight has been reported to be an indicator for metabolic diseases (Rosito et al.,
133
2008). Alternative methods of measuring weight include measuring body fatness via skinfold
thickness measurements (with calipers), underwater weighing, bioelectrical impedance, dual-
energy x-ray absorptiometry (DXA), and isotope dilution (Prentice et al., 2001). Some
considerations for utilizing alternative measures to BMI are the costliness, as well as the
potential difficulty to replicate across studies at different sample sites. Overall, BMI has a strong
correlation to body fatness, and is considered to be a reliable indicator of weight. Specifically
measuring adiposity could also be a more precise way to measure AIWG in future
pharmacogenetic studies.
6.8.2 Candidate Gene Study vs. Genome Wide Association Study
Limitations of the MC3R and MC4R findings may partially be attributed to the candidate gene
study approach. Our studies primarily utilized a candidate gene approach, which has some
limitations. Studying one gene at a time may lead to ambiguous results and non-replication
across samples. Genome wide association studies (GWAS) have become an additional way to
investigate genetic associations within pharmacogenetic studies. The GWAS is a more recent
methodology in comparison to the candidate gene approach. The GWAS essentially consists of
genotyping of hundreds of thousands of SNPs on array chips, and then using that large number
of tests as a correction for association with a phenotype (Pearson et al., 2008: Manolio et al,
2010). A GWAS typically examines numerous polymorphisms and determines associations with
a given trait. One of the cited advantages for using GWAS is that this methodology utilizes a
hypothesis free approach. The GWAS thus also allows for identification of novel gene
associations with a phenotype (Pearson et al., 2008; Manolio et al, 2010).
134
In terms of AIWG, the findings from GWAS studies are mostly limited and open to
interpretation in terms of the results. Previously, Adkins et al., (2011), conducted a genome wide
association study on 738 schizophrenia patients assessed in the Clinical Antipsychotic Trial of
Intervention Effectiveness (CATIE) study. In this study, 492 900 SNPs were analyzed across
twelve weight regulation and related metabolic variables, including a blood lipid panel, glucose,
hemoglobin A1c, blood pressure and heart rate. Mixed models were used to predict glucose and
triglycerides with fasting time, and output the residuals as fasting time-adjusted measures.
Several genes including the meis homeobox 2 (MEIS2), protein kinase cAMP-dependent,
regulatory, type II, beta. (PRKAR2B), G protein-coupled receptor 98 (GPR98), formin
homology 2 domain containing 3 (FHOD3), ring finger protein 144A (RNF144A), astrotactin 2
(ASTN2), SRY-related HMG-box (SOX5) and activating transcription factor 7 interacting
protein 2 (ATF7IP2) were associated with antipsychotic induced metabolic side-effects. There
are several limitations of this study, primarily due to the fact that CATIE was not designed to
assess genetics of weight gain measures. The primary caveat as outlined by the authors included
that patients were not antipsychotic naïve, and that the assessed sub-sample with available DNA
and GWAS data had lower symptom severity. The authors attempted to adjust for these potential
confounders using mixed modeling. However, it is difficult to extrapolate from the findings in
this study, given that the initial study design was not ideal for assessing weight gain. Also, the
blood levels of the medication were not checked and thus adherence was not confirmed. Another
example of utilizing the GWAS method was presented in our collaborative study conducted with
the Malhotra et al., (2012) research group, and discussed above. The patients assessed in the
discovery sample were antipsychotic naïve and the association was detected across separate
135
samples, illustrating that GWAS can be effective in the study of complex phenotypes in small
and well characterized populations.
However, there are number of limitations of the GWAS approach that should be considered.
Typically, a large sample size is needed to conduct a GWAS experiment, and the number of tests
require a heavy correction penalty for multiple testing (Manolio et al, 2010). The GWAS
approach is valuable for decting novel variants. However, the functional effects of these novel
hits are not characterized in these studies. The unknown function or method of regulation of
genetic variants is currently a prominent barrier in genetic association studies. Recently,
expression quantitative trait loci (eQTLs), which act as regulatory elements of genetic
expression, have been highlighted in the literature. A combination of varying methodologies,
including GWAS, genome sequencing, the study of eQTLs (Lappalainen et al., 2013), and use of
novel databases such as the Encyclopedia of DNA Elements (ENCODE) project are beginning to
provide some answers.
6.8.3 Remote Regulatory Site Databases
The US National Human Genome Research Institute initiated a novel and large undertaking –
and launched the ENCODE project – a large public research consortium. The primary objective
of the ENCODE project was to discover functional, regulatory elements of the human genome.
One of the key, critical findings of the ENCODE project was that molecular biological functions
could be attributed to over 80% of the genome. The primary mechanism or function of these
regulatory regions was the mediation of expression level of coding DNA. This finding was
136
revolutionary in the molecular genetic field, due to the novel understanding that coding genes
located within the human genome may be regulated by multiple regulatory sites, and that these
sites may be located very far from the gene itself. Thus, the ENCODE findings provided further
steps towards understanding the functional roles of SNPs which are located within intergenic
regions. Recently, the ENCODE database has been incorporated into newer platforms and
provide additional information regarding gene regulation (e.g. HaploReg v2:
http://www.broadinstitute.org/mammals/haploreg/haploreg.php, and Brain Cloud:
http://braincloud.jhmi.edu/).
The ENCODE project and its methodology and aims are relevant to the research in this thesis.
For instance, the MC4R rs489693 was significantly associated with AIWG. Yet, this SNP is
located a considerable distance, 188,000 base pairs, away from the MC4R gene, and its
regulatory function is unknown. Projects such as ENCODE may help to determine whether the
MC4R rs489693 is part of an enhancer site, or perhaps the rs489693 is in LD with a SNP which
is the true mediator of a biological function, for instance, a SNP that might truly regulate MC4R
expression. Additionally, we were unable to uncover whether the MC4R rs8087522 ‘A’-allele
regulated transcription factor binding, despite our investigation using electrophoretic mobility
shift assay.In a recent study conducted by Evans et al., (2014), MC4R SNPs reported to be
associated with obesity were checked against expression quantitative trait loci (eQTLs), to
determine whether MC4R SNPs are associated with MC4R expression levels. The authors of this
study proposed that the transcriptional insulator, CCCTC-binding factor (CTCF), can bind to the
intergenic MC4R segment (that also contains rs489693). The authors suggested that intra-
chromosomal interactions regulated by CTCF could potentially bring gene variants in close
137
proximity to the MC4R promoter (Evans et al., 2014). Nonetheless, further exploration is needed
to uncover potential functional regulators of the MC4R gene.
6.8.4 Whole Genome Sequencing
In order to uncover additional variants, whole genome sequencing may be a complimentary
methodology. High throughput sequencing allows for thorough examination of the genome, and
can aid in identifying rare variants. This methodology allows for an organism’s entire genome to
be read at one time. A variety of different platforms provide sequencing platforms, and this
approach provides a vast amount of genetic information. The decreasing cost of sequencing
methodology has made this a plausible technique to collect genetic information (El-Metwally et
al., 2013). The 1000 Genomes Project was initiated in 2008, whole genome sequencing was
performed on a set of humans for various ancestral backgrounds (Durbin et al., 2010). This
method remains promising in terms of studying the MC4R and MC3R genes. However, it is
important to consider that collecting genetic information on whole genomes from individuals
results in the compilation of an enormous dataset. Theoretically, whole genome sequencing
could provide information six billion nucleotides, or three billion bas pairs, in one individual’s
DNA to researchers. The bioinformatic and analysis processes for determining associations
between genetic information provided from sequencing requires enormous resources, and may
not be feasible for all laboratories to implement at this point in time. Whole genome sequencing
findings are advantageous for discovery of novel variants. However, with the increasingly
common use of databases such as 1000 Genomes (www.1000Genomes.org), a repository of very
rare variants has become available for researchers to investigate. However, the clinical utility of
very rare variants in psychiatric pharmacogenetics is questionable. For instance, if a mutation or
138
rare variant is determined to be causative of AIWG in one individual out of a thousand, this
finding would be difficult to translate to other patients.
6.8.5 Epigenetics
Novel methods that have been utilized in other fields of psychiatric genetics include studying
epigenetic mechanisms. However, utilizing epigenetic approaches have not yet been widely
investigated in AIWG studies. Epigenetic studies in pharmacogenetics warrant further
investigation due and may provide novel information for AIWG mechanisms. Epigenetic
changes primarily occur in 1 of 2 ways, 1) DNA methylation, or 2) histone modification
(reviewed in Jaenisch, & Bird, 2003, reviewed in Meaney et al., 2010). DNA methylation
primarily refers to a methyl group being added to a DNA nucleotide, which may a result in
changes in gene expression. Histone modification refers to alteration of chromatin structure via
acetylation of the histone protein, which may affect gene transcription (reviewed in Meaney et
al., 2010). The CYP 1A2 enzyme is the major metabolizer of clozapine and olanzapine, the two
antipsychotic medications with the highest propensity for weight gain in patients (Ghotbi et al.,
2009). The CYP 1A2 has been reported to have the expected inverse correlation between DNA
methylation and CPA 1A2 mRNA (Ghotbi et al., 2009). A long term, high fat diet has been
reported to influence the methylation of the MC4R exon in mice (Widiker et al., 2010). These
findings, among many others, are suggestive that epigenetic analysis could be informative in the
field of pharmaco-epigenetics in the future. However, most epigenetic studies have been
explored in cancer chemotherapy research, and are only beginning to be conducted in the field of
psychiatry (Ptak & Petronis, 2010).
139
6.8.6 Environmental Factors
A common limitation across pharmacogenetic/weight gain studies include the exclusion of
potential environmental factors which may influence weight. For instance, measures of diet,
smoking and food intake are rarely assessed in many studies examining the effects of AIWG.
Thus, the relative contribution of such environmental factors to the development of AIWG is
unknown (Panariello et al., 2010).
Environmental Treatment Factors
A randomized clinical trial conducted by Wu et al., (2008), reported that lifestyle intervention
and additional medication off-set the weight gain effects of antipsychotic medication treatment.
Treatment with metformin, an anti-diabetic drug, and lifestyle intervention have resulted in better
outcome for patients. The lifestyle intervention was administered by providing information from
a psychoeducational program which highlighted information about proper nutrition, exercise and
behavioural techniques. In this study, lifestyle intervention alone, metformin treatment alone,
and the combination of lifestyle intervention and metformin treatment all resulted in decreased
measures of BMI, insulin levels, waist circumference, and insulin resistance index (IRI) compared
to the placebo group. Furthermore, the combination of lifestyle intervention and metformin were
more efficacious in weight reduction in comparison to either lifestyle intervention alone or
metformin treatment alone. The findings from this study clearly demonstrate that environmental
factors can affect weight gain, and that factor such as exercise, diet and additional treatments
should be carefully considered in future studies.
140
6.9 Concluding Remarks
The investigation of antipsychotic induced side-effects, including weight gain, is a relatively new
sub-field within psychiatric genetics. Until recently, the most likely candidate genes that were
reported to be associated with antipsychotic induced weight are SNPs across the dopamine,
serotonin and leptin genes. Overall, the reported associations between various SNPs and AIWG
were relatively inconsistent across the literature. Furthermore, with each SNP explaining a small
amount of variance in AIWG (~1-2%), the next step forward became utilizing statistical
modeling to determine whether a combination of SNPS contributed in an additive manner to the
onset of AIWG. It has been proposed that a combination of accounting for age, baseline BMI,
5HT2C and leptin promoter gene variants accounted for 30% of the variance of short term
AIWG (Reynolds et al., 2006). However, this model has not been successfully tested in an
independent sample set in order to validate its predictive potential. AIWG is a complex trait, and
likely has several different contributing genes contributing a small effect. Our patent includes
eight genes (nine SNPs) in an additive model (US patent 61946003, 2014). Nonetheless, it is
likely that some genes have a higher influence on the development of this phenotype over others.
Overall, the findings presented in this thesis are novel and unique, and may be a basis for clinical
genetic testing in the future. The MC4R locus has generated much interest. We collaborated with
colleagues at Zucker Hillside Hospital, New York to determine an association between the
MC4R gene and AIWG. Our collaborators conducted a genome wide association and found that
MC4R rs489693 was significantly associated with weight gain in a pediatric sample treated with
second generation antipsychotic medications. Our group was able to replicate this finding in our
schizophrenia samples. Replication of the MC3R in independent sample sets is warranted, in
order to further understand whether this gene has a role in AIWG. The MC3R opposite allele
141
was reported to be associated with antipsychotic weight gain the RUPP sample set, illustrating
the difficulty of replication with complex phenotypes. The advancements of ENCODE and
related databases, whole genome sequencing, and epigenetic studies may further help in
understanding the mechanism as to why AIWG and its associated metabolic abnormalities occur
within a large subset of patients.
142
References
Abdi, H. (2007) Bonferroni and Sidak corrections for multiple comparisons. In Encyclopedia of
Measurement and Statistics (Salkind, N.J., ed.). Thousand Oaks: Sage, pp. 103–107.
Ananth J, Parameswaran S, Gunatilake Burgoyne K, Sidhom T. (2004).Neuroleptic malignant syndrome
and atypical antipsychotic drugs.J Clin Psychiatry. 65(4):464-70.
Arch JR. (2005). Central regulation of energy balance: inputs, outputs and leptin resistance. Proc Nutr
Soc.64: 39–46.
Argyropoulos G, Rankinen T, Neufeld DR, Rice T, Province MA, Leon AS, et al. (2002). A
polymorphism in the human agouti-related protein is associated with late-onset obesity. J Clin
Endocrinol Metab 87:4198–4202.
Arrizabalaga M, Larrarte E, Margareto J, Maldonado-Martín S, Barrenechea L, Labayen I. (2014).
Preliminary findings on the influence of FTO rs9939609 and MC4R rs17782313 polymorphisms
on resting energy expenditure, leptin and thyrotropin levels in obese non-morbid
premenopausal women. J Physiol Biochem.70(1):255-62.
Allison DB, Mentore JL, Heo M, et al. (1999).Antipsychotic-induced weight gain: a
comprehensive research synthesis. Am J Psychiatry. 11,1686-96.
Almoguera B, Riveiro-Alvarez R, Lopez-Castroman J, Dorado P, Vaquero-Lorenzo C,
Fernandez-Piqueras J, Llerena A, Abad-Santos F, Baca-García E, Dal-Ré R, Ayuso C;
Spanish Consortium of Pharmacogenetics Research in Schizophrenia. (2013).
Association of common genetic variants with risperidone adverse events in a Spanish
143
schizophrenic population. Pharmacogenomics J. 13(2):197-204.
Andreasen CH, Stender-Petersen KL, Mogensen MS, Torekov SS, Wegner L, Andersen G, et al. (
(2008). Low physical activity accentuates the effect of the FTO rs9939609 polymorphism on
body fat accumulation. Diabetes. 57(1):95-101.
Atalayer D, Robertson KL, Haskell-Luevano C, et al. (2010). Food demand and meal size in mice with
single or combined disruption of melanocortin type 3 and 4 receptors. Am J Physiol Regul Integr
Comp Physiol. 298(6), R1667-74.
Atmaca M, Kuloglu M, Tezcan E, Ustundag B, Semercioz A. (2003). Serum leptin levels in patients with
premature ejaculation before and after citalopram treatment. BJU Int. (3):252-4.
Baig MR, Navaira E, Escamilla MA, Raventos H, Walss-Bass C.Clozapine treatment causes oxidation of
proteins involved in energy metabolism in lymphoblastoid cells: a possible mechanism for
antipsychotic-induced metabolic alterations.J Psychiatr Pract. 2010 Sep;16(5):325-33.
Baker M, Gaukrodger N, Mayosi BM, Imrie H, Farrall M, Watkins H, Connell JM, Avery PJ, Keavney B.
(2005). Association between common polymorphisms of the proopiomelanocortin gene and
body fat distribution: a family study. Diabetes 54: 2492-2496.
Bakker SC, Hoogendoorn ML, Selten JP, Verduijn W, Pearson PL, Sinke RJ, Kahn RS(2004).
Neuregulin 1: genetic support for schizophrenia subtypes.Mol Psychiatry.
(12):1061-3.
Baptista, T., M. A. Parada & L. Hernandez. (1987). Long-term administration of some
144
antipsychotic drugs increases body weight and feeding in rats: are D2 dopamine
receptors involved? Pharmacol. Bio- chem. Behav. 27, 399405.
Baptista T, Hernandez L, Hoebel BG. (1990). Systemic sulpiride increases dopamine metabolites in the
lateral hypothalamus.Pharmacol Biochem Behav. 37(2):227-9.
Barrett JC, Fry B, Maller J, Daly MJ. (2005). Haploview: analysis and visualization of LD and haplotype
maps. Bioinformatics. 21: 263–265
Bartlett, J. M. S. Stirling, D. (2003). A Short History of the Polymerase Chain Reaction. PCR Protocols
226. pp. 3–6.
Basile VS, Masellis M, Potkin SG, Kennedy JL. (2002).Pharmacogenomics in schizophrenia: the quest
for individualized therapy. Hum Mol Genet. 11(20):2517-30.
Bauchet M, McEvoy B, Pearson LN, Quillen EE, Sarkisian T, Hovhannesyan K, et al. (2007). Measuring
European population stratification with microarray genotype data. Am J Hum Genet.
80(5):948-56.
Beckers S, Zegers D, de Freitas F, Mertens IL, Van Gaal LF, Van Hul W. (2011). Association study of
MC4R with complex obesity and replication of the rs17782313 association signal. Mol Genet
Metab. 103(1):71-5.
Beaudry G, Zekki H, Rouillard C, Lévesque D. (2004). Clozapine and dopamine D3 receptor antisense
reduce cocaine- and amphetamine-regulated transcript expression in the rat nucleus
accumbens shell. Synapse. 15;51(4):233-40.
145
Bellman.(1961).Adaptive control Processes, p.94, Princeton University Press, NJ.
Benjamini Y, Drai D, Elmer G, Kafkafi N, Golani I. (2001). Controlling the false discovery rate in
behavior genetics research.125(1-2):279-84.
Biebermann H, Kühnen P, Kleinau G, Krude H. (2012). The neuroendocrine circuitry controlled by
POMC, MSH, and AGRP. Handb Exp Pharmacol. (209):47-75.
Bienertova-Vasku JA, Spinarova L, Bienert P, Vasku A. (2009). Association between variants in the
genes for leptin, leptin receptor, and proopiomelanocortin with chronic heart failure in the Czech
population. Heart Vessels 24: 131-137.
Bird, A. (2007). "Perceptions of epigenetics". Nature 447 (7143): 396–8.
Blum K, Noble EP, Sheridan PJ, Montgomery A, Ritchie T, Jagadeeswaran P, Nogami H, Briggs
AH, Cohn JB. (1990). Allelic association of human dopamine D2 receptor gene in
alcoholism.JAMA. 263(15):2055-60.
Bogardus C. (2009).Missing heritability and GWAS utility. Obesity (Silver Spring). 17(2):209-10.
Bonilla C, Panguluri RK, Taliaferro-Smith L, Argyropoulos G, Chen G, Adeyemo AA, et al. (2006).
Agouti-related protein promoter variant associated with leanness and decreased risk for diabetes
in West Africans. Int J Obes (London) 30: 715–721.
Brady LS, Smith MA, Gold PW, Herkenham M. (1990). Altered expression of hypothalamic
neuropeptide mRNAs in food-restricted and food-deprived rats. Neuroendocrinology 52: 441-
146
447.
Brandl EJ, Frydrychowicz C, Tiwari AK, Lett TA, Kitzrow W, Büttner S, Ehrlich S, Meltzer HY,
Lieberman JA, Kennedy JL, Müller DJ, Puls I. (2012). Association study of polymorphisms in
leptin and leptin receptor genes with antipsychotic-induced body weight gain. Prog
Neuropsychopharmacol Biol Psychiatry. 38(2):134-41.
Brandl EJ, Tiwari AK, Chowdhury NI, Zai CC, Lieberman JA, Meltzer HY, et al. (2014).
Genetic variation in the GCG and in the GLP1R genes and antipsychotic-induced weight gain.
Pharmacogenomics. 15(4):423-31.
Brönner G, Sattler AM, Hinney A, Soufi M, Geller F, Schäfer H, et al. (2006).The 103I variant of the
melanocortin 4 receptor is associated with low serum triglyceride levels.J Clin Endocrinol
Metab.;91(2):535-8.
Brown S, Inskip H, Barraclough B. (2000).Causes of the excess mortality of schizophrenia.
Br J Psychiatry. 177:212-7.
Brown, AS. (2006). Prenatal Infection as a Risk Factor for Schizophrenia. Schizophr Bull. 32 (2):
200-202.
Brüne M. (2004). Schizophrenia-an evolutionary enigma? Neurosci Biobehav Rev. 28(1):41-53.
Brunton, L, Chabner, B, Knollman, B (2010). Goodman and Gilman's The Pharmacological Basis of
Therapeutics (12th ed.). McGraw Hill Professional. pp. 417–455.
Buckland PR, Hoogendoorn B, Guy CA, Smith SK, Coleman SL, O’Donovan MC. (2005) Low gene
147
expression conferred by association of an allele of the 5-HT2C receptor gene with
antipsychotic- induced weight gain. Am J Psychiatry.162: 613–615.
Burmeister MA, Ayala J, Drucker DJ, Ayala JE. (2013).Central glucagon-like peptide 1 receptor-induced
anorexia requires glucose metabolism-mediated suppression of AMPK and is impaired by central
fructose.Am J Physiol Endocrinol Metab. 304(7):E677-85
Calarge CA, Ellingrod VL, Zimmerman B, Acion L, Sivitz WI, Schlechte JA. (2009). Leptin gene –
2548G/A variants predict risperidone-associated weight gain in children and adolescents.
Psychiatr Genet. 19(6):320-7.
Callaghan RC, Boire MD, Lazo RG, McKenzie K, Cohn T. (2009).Schizophrenia and the incidence of
cardiovascular morbidity: a population-based longitudinal study in Ontario, Canada.
Schizophr Res.115: 325–332.
Calle ML, Urrea V, Malats N, Van Steen K. (2010). mbmdr: an R package for exploring gene-gene
interactions associated with binary or quantitative traits.Bioinformatics. 1;26(17):2198-9.
Calton MA, Ersoy BA, Zhang S, Kane JP, Malloy MJ, Pullinger CR, et al.. (2009). Association of
functionally significant Melanocortin-4 but not Melanocortin-3 receptor mutations with severe
adult obesity in a large North American case-control study. Hum Mol Genet. 18(6):1140-7.
Carpenter CL, Wong AM, Li Z, Noble EP, Heber D. (2013). Association of dopamine D2
receptor and leptin receptor genes with clinically severe obesity.Obesity (Silver Spring).
21(9):E467-73.
148
Cauchi S, Stutzmann F, Cavalcanti-Proença C, Durand E, Pouta A, Hartikainen AL, et al. (2009).
Combined effects of MC4R and FTO common genetic variants on obesity in European general
populations. J Mol Med (Berl). 87(5):537-46.
Cattaert T, Calle ML, Dudek SM, Mahachie John JM, Van Lishout F, et al. (2011).Model-based
multifactor dimensionality reduction for detecting epistasis in case-control data in the presence of
noise.Ann Hum Genet.75(1):78-89.
Chagnon YC, Bureau A, Gendron D, Bouchard RH, Mérette C, Roy MA, Maziade M. (2007).Possible
association of the pro-melanin-concentrating hormone gene with a greater body mass index
as a side effect of the antipsychotic olanzapine. Am J Med Genet B Neuropsychiatr Genet.
144B(8):1063-9.
Challis BG, Yeo GS, Farooqi IS, Luan J, Aminian S, Halsall DJ, et al.(2000). The CART gene and human
obesity: mutational analysis and population genetics. Diabetes 49: 872-875.
Chen AS, Metzger JM, Trumbauer ME, Guan XM, Yu H, Frazier EG, et al. (2000). Role of the
melanocortin-4 receptor in metabolic rate and food intake in mice. Transgenic Res.
9(2):145-54.
Cheung CY, Tso AW, Cheung BM, Xu A, Ong KL, Fong CH, Wat NM, Janus ED, Sham PC, Lam KS.
(2010). Obesity susceptibility genetic variants identified from recent genome-wide association
studies: implications in a chinese population. J Clin Endocrinol Metab. 95(3):1395-403.
Chowdhury NI, Remington G, Kennedy JL. (2011). Genetics of antipsychotic-induced side
effects and agranulocytosis. Curr Psychiatry Rep.13(2):156-65.
149
Chowdhury NI, Souza RP, Tiwari AK, Brandl EJ, Sicard M, Meltzer HY, Lieberman JA, Kennedy JL,
Müller DJ.(2014). Investigation of melanocortin system gene variants in antipsychotic-induced
weight gain. World J Biol Psychiatry. 15(3):251-8.
Chowdhury NI, Tiwari AK, Souza RP, Zai CC, Shaikh SA, Chen S, Liu F, Lieberman JA,
Meltzer HY, Malhotra AK, Kennedy JL, Müller DJ. (2013). Genetic association study between
antipsychotic-induced weight gain and the melanocortin-4 receptor gene. Pharmacogenomics J.
13(3):272-9.
Cieslak J, Majewska KA, Tomaszewska A, Skowronska B, Fichna P, Switonski M. (2013). Common
polymorphism (81Val>Ile) and rare mutations (257Arg>Ser and 335Ile>Ser) of the MC3R
gene in obese Polish children and adolescents.Mol Biol Rep. (12):6893-8.
Concato,J., Feinstein,A.R. and Holford,T.R. (1996). The risk of determining risk with multivariable
models. Ann. Int. Med., 118, 201–210.
Cone RD. (2005) Anatomy and regulation of the central melanocortin system. Nat Neurosci. 8(5):571-8.
Comuzzie AG, Funahashi T, Sonnenberg G, Martin LJ, Jacob HJ, Black AE, et al.. (2001). The genetic
basis of plasma variation in adiponectin, a global endophenotype for obesity and the metabolic
syndrome. J Clin Endocrinol Metab;86(9):4321-5.
Correll CU. (2011). Safety and tolerability of antipsychotic treatment in young patients with
schizophrenia. J Clin Psychiatry. 72(8) e26.
Corvin AP, Morris DW, McGhee K, Schwaiger S, Scully P, Quinn J, Meagher D, Clair DS,
150
Waddington JL, Gill M. (2004). Confirmation and refinement of an 'at-risk' haplotype for
schizophrenia suggests the EST cluster, Hs.97362, as a potential susceptibility gene at the
Neuregulin-1 locus. Mol Psychiatry. 9(2):208-13.
Croteau-Chonka DC, Marvelle AF, Lange EM, Lee NR, Adair LS, Lange LA, Mohlke
KL. (2011). Genome-wide association study of anthropometric traits and evidence of
interactions with age and study year in Filipino women. Obesity (Silver Spring). 2011
19(5):1019-27.
Cui, H & Lutter M. (2013). The expression of MC4Rs in D1R neurons regulates food intake and
locomotor sensitization to cocaine. Genes Brain Behav. 12(6):658-65.
Czerwensky F, Leucht S, Steimer W. (2013).MC4R rs489693: a clinical risk factor for second generation
antipsychotic-related weight gain? Int J Neuropsychopharmacol. 16(9):2103-9.
Dale R. Nyholt. (2004). A Simple Correction for Multiple Testing for Single-Nucleotide
Polymorphisms in Linkage Disequilibrium with Each Other. Am J Hum Genet. 74(4),
765–769.
Damm E, Buech TR, Gudermann T, Breit A. (2012). Melanocortin-induced PKA activation inhibits
AMPK activity via ERK-1/2 and LKB-1 in hypothalamic GT1-7 cells.Mol Endocrinol.
26(4):643-54.
Davis, KL, Charney, D, Coyle, JT, Nemeroff, C (2002). Neuropsychopharmacology—The fifth
generation of progress. Philadelphia: Lippincott Williams & Wilkins
151
Davoodi N, Kalinichev M, Korneev SA, Clifton PG. (2009). Hyperphagia and increased meal size are
responsible for weight gain in rats treated sub-chronically with olanzapine. Psychopharmacology
(Berl) 203: 693-702.
De Luca V, Müller DJ, de Bartolomeis A, Kennedy JL.(2007). Association of the HTR2C gene
and antipsychotic-induced weight gain: a meta-analysis. Int J Neuropsychopharmacol.
10(5):697–704.
del Giudice EM, Santoro N, Cirillo G, D'Urso L, Di Toro R, Perrone L. Mutational screening of the
CART gene in obese children: identifying a mutation (Leu34Phe) associated with reduced
resting energy expenditure and cosegregating with obesity phenotype in a large family. (2001).
Diabetes. 50(9):2157-60.
De Sousa KR, Tiwari AK, Giuffra DE, Mackenzie B, Zai CC, Kennedy JL. (2013). Age at
onset of schizophrenia: cannabis, COMT gene, and their interactions.Schizophr Res.
151(1-3):289-90.
Dohrenwend BP, Levav I, Shrout PE, Schwartz S, Naveh G, Link BG, Skodol AE, Stueve A.
(1992).Socioeconomic status and psychiatric disorders: the causation-selection issue.
21;255(5047):946-52.
Duan J, Wainwright MS, Comeron JM, Saitou N, Sanders AR, Gelernter J, Gejman PV. (2003).
Synonymous mutations in the human dopamine receptor D2 (DRD2) affect mRNA stability and
synthesis of the receptor.Hum Mol Genet. 12(3):205-16.
Dudbridge F. (2002). Pedigree disequilibrium tests for multilocus haplotypes.Genet Epidemiol. 25: 115–
152
121.
Durbin, RM, Abecasis, GR, Altshuler, R, Auton, GA, Brooks, DR, Durbin, A et al. (2010). "A map of
human genome variation from population-scale sequencing". Nature 467 (7319): 1061–1073.
Donald F, Behan SHC. 2012. Obesity and overweight fact sheet N°311.
http://www.who.int/mediacentre/factsheets/fs311/en.
Douet V, Chang L, Pritchett A, Lee K, Keating B, Bartsch H, Jernigan TL, Dale A, Akshoomoff
N, Murray S, Bloss C, Kennedy DN, Amaral D, Gruen J, Kaufmann WE, Casey BJ,
Sowell E, Ernst T. (2014).Schizophrenia-risk variant rs6994992 in the neuregulin-1 gene
on brain developmental trajectories in typically developing children. Transl Psychiatry.
2014.
Dwivedi OP, Tabassum R, Chauhan G, Kaur I, Ghosh S, Marwaha RK, Tandon N, Bharadwaj
D. (2013). Strong influence of variants near MC4R on adiposity in children and adults: a cross-
sectional study in Indian population. J Hum Genet. 58(1):27-32.
Dušátková L, Zamrazilová H, Sedláčková B, Včelák J, Hlavatý P, Aldhoon Hainerová I,
Et al. (2013). Association of obesity susceptibility gene variants with metabolic syndrome and
related traits in 1,443 czech adolescents. Folia Biol (Praha). 59(3):123-33.
Eaton, W., C.-Y. Chen (2006). Epidemiology. Textbook of Schizophrenia. J. Lieberman, S.
Stroup and D. Perkins. Arlington, VA, American Psychiatric Association.
Ehrlich S, Leopold K, Merle JV, Theophil I, Haag W, Lautenschlager M, Schaefer M.(2012).Trajectories
of agouti-related protein and leptin levels during antipsychotic-associated weight gain in patients
153
with schizophrenia. J Clin Psychopharmacology. 32(6):767-72.
Echwald SM, Sorensen TI, Andersen T, Hansen C, Tommerup N, Pedersen O. (1999). Sequence variants
in the human cocaine and amphetamine-regulated transcript (CART) gene in subjects with early
onset obesity. Obes Res 7: 532-536.
Ellingrod VL, Perry PJ, Ringold JC, et al. (2005). Weight gain associated with the −759C/T
polymorphism of the 5HT2C receptor and olanzapine. Am J Med Genet B
Neuropsychiatr Genet. (1):76–78.
El-Metwally S, Hamza T, Zakaria M, Helmy M. (2013). Next-generation sequence assembly: four stages
of data processing and computational challenges. PLoS Comput Biol. 2013
9(12):e1003345.
Elmquist JK, Elias CF, Saper CB. (1999).From lesions to leptin: hypothalamic control of food intake
and body weight. Neuron22(2):221-32.
Espinoza CA, Ren B. Mapping higher order structure of chromatin domains. Nat Genet 2011; 43: 615–
616.
Farooqi IS, Keogh JM, Yeo GS, et al. (2003). Clinical spectrum of obesity and mutations in the
melanocortin 4 receptor gene. N Engl J Med. 348(12), 1085–1095.
Farooqi S, O'Rahilly S. (2006). Genetics of obesity in humans. Endocr Rev. 27(7):710-18.
Feng N, Young SF, Aguilera G, Puricelli E, Adler-Wailes DC, Sebring NG, Yanovski JA. (2005).
Co-occurrence of two partially inactivating polymorphisms of MC3R is associated with
154
pediatric-onset obesity. Diabetes54(9):2663-7.
Fernø J, Varela L, Skrede S, Vázquez MJ, Nogueiras R, Diéguez C, Vidal-Puig A, Steen VM, López
M.Olanzapine-induced hyperphagia and weight gain associate with orexigenic hypothalamic
neuropeptide signaling without concomitant AMPK phosphorylation. (2011).PLoS One.
6(6):e20571.
Flint A, Kapitza C, Zdravkovic M. (2013).The once-daily human GLP-1 analogue liraglutide impacts
appetite and energy intake in patients with type 2 diabetes after short-term treatment. Diabetes
Obes Metab. 15(10):958-62.
Frayling TM, Timpson NJ, Weedon MN, Zeggini E, Freathy RM, Lindgren CM, et al. (2007). A common
variant in the FTO gene is associated with body mass index and predisposes to childhood and
adult obesity. Science. 316:889-893.
Fried MG. (1989). "Measurement of protein-DNA interaction parameters by electrophoresis mobility shift
assay". Electrophoresis 10, 366-376.
Friedman JM, Halaas JL. (1998).Leptin and the regulation of body weight in mammals.
Nature. 395(6704):763-70.
Gauderman WJ, Morrison JM. (2007). QUANTO 1.1: a computer program for power and sample size
calculations for genetic-epidemiology studies. http://hydra.usc.edu/gxe.
Garrow JS and Webster J. Quetelet's index (W/H2) as a measure of fatness. (1985). International
Journal of Obesity 1985;9:147–153.
155
Gebhardt S, Theisen FM, Haberhausen M, Heinzel-Gutenbrunner M,Wehmeier PM, Krieg JC et al.
(2010). Body weight gain induced by atypical antipsychotics: an extension of the monozygotic
twin and sib pair study. J Clin Pharm Ther .35: 207–211.
Geller F, Reichwald K, Dempfle A, Illig T, Vollmert C, et al. (2004) Melanocortin-4 receptor gene
variant I103 is negatively associated with obesity. Am J Hum Genet 74: 572–581.
Georgieva L, Moskvina V, Peirce T, Norton N, Bray NJ, Jones L, et al. Convergent evidence that
oligodendrocyte lineage transcription factor 2 (OLIG2) and interacting genes influence
susceptibility to schizophrenia. (2006).Proc Natl Acad Sci U S A. 103(33):12469-74.
Ghotbi R, Gomez A, Milani L, Tybring G, Syvänen AC, Bertilsson L, Ingelman-
Sundberg M, Aklillu E. Allele-specific expression and gene methylation in the control of
CYP1A2 mRNA level in human livers. Pharmacogenomics J. 2009 Jun;9(3):208-17.
Giannaccini G, Betti L, Palego L, Pirone A, Schmid L, Lanza M, et al. (2011). Serotonin transporter
(SERT) and translocator protein (TSPO) expression in the obese ob/ob mouse. BMC Neurosci.
7;12:18.
Girault EM, Alkemade A, Foppen E, Ackermans MT, Fliers E, Kalsbeek A. (2012).Acute peripheral
but not central administration of olanzapine induces hyperglycemia associated with
hepatic and extra-hepatic insulin resistance. PLoS One. 7(8).
Godlewska BR, Olajossy-Hilkesberger L, Ciwoniuk M, et al. (2009).Olanzapine-induced weight
gain is associated with the −759C/T and −697G/C polymorphisms of the HTR2C gene.
Pharmacogenomics J. 9(4):234–241.
156
Gonçalves VF, Zai CC, Tiwari AK, et al. 2013 (in press). A Hypothesis Driven Association
Study of 28 Nuclear-Encoded Mitochondrial Genes With Antipsychotic-Induced Weight
Gain in Schizophrenia. Neuropsychopharmacology.
González-Pinto A, Mosquera F, Palomino A, Alberich S, Gutiérrez A, Haidar K, et al. (2010). Increase in
brain-derived neurotrophic factor in first episode psychotic patients after treatment with
atypical antipsychotics. Int Clin Psychopharmacol. 25(4):241-5.
Goodpaster BH., Wolfe, R., Kelley, DE. (2002). Measuring body fat distribution and content in humans.
Curr Opin Clin Nutr Metab Care. 5(5):481-7. Review.
Gottesman, II and D. L. Wolfgram (1991). Schizophrenia genesis: The origins of madness,
Freeman New York.
Graham M, Shutter JR, Sarmiento U, Sarosi I, Stark KL. 1997. Overexpression of Agrt leads to
obesity in transgenic mice. Nat Genet 17:273–274.
Green AI, Patel JK, Goisman RM, et al. (2000). Weight gain from novel antipsychotic drugs: need
for action. Gen Hosp Psychiatry. 22, 24–35.
Granell S, Mohammad S, Ramanagoudr-Bhojappa R, Baldini G. (2010). Obesity- linked variants of
melanocortin-4 receptor are misfolded in the endoplasmic reticulum and can be rescued
to the cell surface by a chemical chaperone. Mol Endocrinol.24: 1805–1821.
Grant SF, Bradfield JP, Zhang H, Wang K, Kim CE, Annaiah K, et al. (2009). Investigation of the locus
near MC4R with childhood obesity in Americans of European and African ancestry. Obesity
157
(Silver Spring). 17(7):1461-5.
Green, MJ, Matheson, SL, Shepherd, AC, Weickert, S, Carr, V. (2011). Brain-derived neurotrophic factor
levels in schizophrenia: a systematic review with meta-analysis. Molecular Psychiatry.16,
960–972.
Gregoor JG, van der Weide J, Loovers HM, van Megen HJ, Egberts TC, Heerdink ER. (2011).
Polymorphisms of the LEP, LEPR and HTR2C gene: obesity and BMI change in patients
using
antipsychotic medication in a naturalistic setting. Pharmacogenomics. 12(6):919-23.
Gunes A, Melkersson KI, Scordo MG, Dahl M-L. (2009).Association between HTR2C and HTR2A
polymorphisms and metabolic abnormalities in patients treated with olanzapine or
clozapine. J Clin Psychopharmacol . 29: 65–68.
Gut P, Baeza-Raja B, Andersson O, Hasenkamp L, Hsiao J, Hesselson D, et al. (2013).Whole-organism
screening for gluconeogenesis identifies activators of fasting metabolism. Nat Chem Biol.
9(2):97-104.
Hagg S, Soderberg S, Ahren B, Olsson T, Mjorndal T (2001). Leptin concentrations are
increased in subjects treated with clozapine or conventional antipsychotics. J Clin
Psychiatry 62: 843–848.
Hahn M, Chintoh A, Giacca A, Xu L, Lam L, Mann S, et al, (2011).Atypical antipsychotics and effects of
muscarinic, serotonergic, dopaminergic and histaminergic receptor binding on insulin secretion
in vivo: an animal model.Schizophr Res. 131(1-3):90-5.
158
Hahn LW, Ritchie MD, Moore JH. (2003). Multifactor dimensionality reduction software for detecting
gene-gene and gene-environment interactions. Bioinformatics. 19(3):376-82.
Hainer V, Zamrazilová H, Spálová J, Hainerová I, Kunesová M, Aldhoon B, Bendlová B. (2008). Role of
hereditary factors in weight loss and its maintenance. Physiol Res. 57.
Hardy R, Wills AK, Wong A, Elks CE, Wareham NJ, Loos RJ, Kuh D, Ong KK. (2010). Life course
variations in the associations between FTO and MC4R gene variants and body size. Hum Mol
Genet. 1;19(3):545-52.
Haskell-Luevano C, Monck EK. (2001).Agouti-related protein functions as an inverse agonist at a c
constitutively active brain melanocortin-4 receptor. Regul Pept. 99(1):1-7.
Hall J, Whalley HC, Job DE, Baig BJ, McIntosh AM, Evans KL, Thomson PA, Porteous DJ,
Cunningham-Owens DG, Johnstone EC, Lawrie SM. (2006). A neuregulin 1 variant associated
with abnormal cortical function and psychotic symptoms. Nat Neurosci. 2006 (12):1477-8.
He S, Ye Z, Dobbelaar PH, Bakshi RK, Hong Q, Dellureficio JP et al. (2010). Discovery of highly potent
and efficacious MC4R agonists with spiroindane N-Me-1,2,4-triazole privileged structures for
the treatment of obesity. Bioorg Med Chem Lett. 20: 6524–6532.
Heard-Costa NL, Zillikens MC, Monda KL, Johansson A, Harris TB, Fu M, Haritunians T, et al. (2009).
NRXN3 is a novel locus for waist circumference: a genome-wide association study from the
CHARGE Consortium. PLoS Genet. 5(6):e1000539.
Heid IM, Vollmert C, Hinney A, Döring A, Geller F, Löwel H, et al. (2005). KORA Group. Association
of the 103I MC4R allele with decreased body mass in 7937 participants of two population based
159
surveys. J Med Genet. 42(4):e21.
Heiman ML1, Witcher DR. (2006). Ghrelin in obesity. Metab Syndr Relat Disord. Spring;4(1):37-42.
Hertel P, Nomikos GG, Schilström B, Arborelius L, Svensson TH. (1997). Risperidone dose-dependently
increases extracellular concentrations of serotonin in the rat frontal cortex: role of alpha 2-
adrenoceptor antagonism. Neuropsychopharmacology. 17(1):44-55.
Hertel P, Nomikos GG, Svensson TH. (1999)The antipsychotic drug risperidone interacts with auto- and
hetero- receptors regulating serotonin output in the rat frontal cortex. Neuropharmacology.
38(8):1175-84.
Hill MJ, Reynolds GP. (2007). 5-HT2C receptor gene polymorphisms associated with antipsychotic drug
action alter promoter activity. Brain Res.1149: 14–17.
Hirvonen MM, Laakso A, Någren K, Rinne JO, Pohjalainen T, Hietala J. (2009).C957T polymorphism of
dopamine D2 receptor gene affects striatal DRD2 in vivo availability by changing the receptor
affinity.Synapse. (10):907-12.
Hoffstedt J, Eriksson P, Mottagui-Tabar S, et al. (2002) A polymorphism in the leptin gene
promoter region (-2548 g/a) influences gene expression and adipose tissue secretion of
leptin. Horm Metab Res. 34, 355-359.
Holland, P. M.; Abramson, R. D.; Watson, R.; Gelfand, D. H. (1991). Detection of specific polymerase
chain reaction product by utilizing the 5'----3' exonuclease activity of Thermus aquaticus DNA
polymerase. Proceedings of the National Academy of Sciences of the United States of
160
America 88 (16): 7276–7280.
Horikoshi M, Hara K, Ohashi J, Miyake K, Tokunaga K, Ito C, Kasuga M, Nagai R, Kadowaki T. A
(2006).Polymorphism in the AMPK alpha2 subunit gene is associated with insulin resistance and
type 2 diabetes in the Japanese population.Diabetes. 55(4):919-23.
Horstmann A, Kovacs P, Kabisch S, Boettcher Y, Schloegl H, Tönjes A, et al. (2013). Common genetic
variation near MC4R has a sex-specific impact on human brain structure and eating behavior.
PLoS One. 16;8(9):e74362.
Ho-Urriola J, Guzmán-Guzmán IP, Smalley SV, González A, Weisstaub G, Domínguez-Vásquez P, et al.
(2014). Melanocortin-4 receptor polymorphism rs17782313: association with obesity and
eating in the absence of hunger in Chilean children. Nutrition. 30(2):145-9.
Hunter RG, Philpot K, Vicentic A, Dominguez G, Hubert GW, Kuhar MJ. (2004) CART in feeding
and
obesity. Trends Endocrinol Metab. 15(9):454-9.
Humbert-Claude M1, Davenas E, Gbahou F, Vincent L, Arrang JM. (2012). Involvement of histamine
receptors in the atypical antipsychotic profile of clozapine: a reassessment in vitro and in
vivo. Psychopharmacology (Berl). 220(1):225-41.
Huszar D, Lynch CA, Fairchild-Huntress V, Dunmore JH, Fang Q, Berkemeier LR et al. (1997).
Targeted disruption of the melanocortion-4 receptor results in obesity in mice. Cell.88: 131–141.
Hong CJ, Liou YJ, Bai YM, Chen TT, Wang YC, Tsai SJ. (2010).Dopamine receptor D2 gene is
associated with weight gain in schizophrenic patients under long-term atypical antipsychotic
161
treatment. Pharmacogenetics Genomics.20: 359–366.
Hyman SE. (2007).Can neuroscience be integrated into the DSM-V? Nat Rev Neurosci. 8(9):725-32.
Hosmer, DW., Lemeshow, S (2000). Applied logistic regression. Wiley. New York.
Inayama Y, Yoneda H, Sakai T, Ishida T, Nonomura Y, Kono Y, Takahata R, Koh J, Sakai J,
Takai A, Inada Y, Asaba H. (1996). Positive association between a DNA sequence
variant in the serotonin 2A receptor gene and schizophrenia. Am J Med Genet .
67(1):103-5.
Jääskeläinen A, Schwab U, Kolehmainen M, Kaakinen M, Savolainen MJ, Froguel P, et al. (2013).Meal
frequencies modify the effect of common genetic variants on body mass index in adolescents of
the northern Finland birth cohort 1986. PLoS One. 8(9):e73802.
Jaenisch, R.; Bird, A. (2003). "Epigenetic regulation of gene expression: how the genome integrates
intrinsic and environmental signals". Nature Genetics. (3s): 245–254.
Jamshidi N, Taylor DA (2001). Anandamide administration into the ventromedial hypothalamus
stimulates appetite in rats.Br J Pharmacol. 134: 1151–1154.
Jassim, G., Ferno, J., Theisen, F.M., Haberhausen, M., Christoforou, A., Havik, B et al. (2011).
Association study of energy homeostasis genes and antipsychotic-induced weight gain in
patients with schizophrenia. Pharmacopsychiatry, 44,pp. 15–20.
Jaworski JN, Jones DC. (2006).The role of CART in the reward/reinforcing properties of
162
psychostimulants. Peptides.27(8):1993-2004.
Jean A, Conductier G, Manrique C, Bouras C, Berta P, Hen R et al. (2007). Anorexia induced by
activation of serotonin 5-HT4 receptors is mediated by increases in CART in the nucleus
accumbens. Proc Natl Acad Sci U S A. 104(41):16335-40.763-770.
Jin H, Meyer JM, Mudaliar S, Jeste DV. (2008). Impact of atypical antipsychotic therapy on leptin,
ghrelin, and adiponectin. Schizophr Res. 100(1-3):70-85.
Johnson DE, Yamazaki H, Ward KM, Schmidt AW, Lebel WS, Treadway JL, Gibbs EM,
Zawalich WS, Rollema H. (2005). Inhibitory effects of antipsychotics on carbachol-
enhanced insulin secretion from perifused rat islets: role of muscarinic antagonism in
antipsychotic-induced diabetes and hyperglycemia. Diabetes. (5):1552-8.
Jones B, Basson BR, Walker DJ, Crawford AM, Kinon BJ. (2001). Weight change and atypical
antipsychotic treatment in patients with schizophrenia.J Clin Psychiatry. 62 Suppl 2:41-4.
Kalinichev M, Rourke C, Daniels AJ, Grizzle MK, Britt CS, Ignar DM, Jones DN. (2005).
Characterisation of olanzapine-induced weight gain and effect of aripiprazole vs
olanzapine on body weight and prolactin secretion in female rats.Psychopharmacology
(Berl). 182(2):220-31.
Kane JM, Honigfeld G, Singer J, et al. (1988). Clozapine in treatment-resistant schizo-
phrenics. Psychopharmacol Bull;24(1):62–7.
Kaplan, R. M. (2008). "Being Bleuler: the second century of schizophrenia." Australas
Psychiatry 16(5): 305-11.
163
Kapur S, Seeman P. (2001). Does fast dissociation from the dopamine d(2) receptor explain the
action of atypical antipsychotics?: A new hypothesis. Am J Psychiatry. 158(3):360-9.
Keller MC, Visscher PM, Wray NR. (2012). Estimating the proportion of variation in
susceptibility to schizophrenia captured by common SNPs.Nat Genet. 44(3):247-50.
Kennedy, Jl., Farrer., LA., Andreasen, N.C., Mayeux,R., St GeorgeHyslop, P. (2003). The Genetics of
Adult-Onset Neuropsychiatric Disease: Complexities and Conundra? Science. Vol. 302 no.
5646 pp. 822-826
Kim SF, Huang AS, Snowman AM, Teuscher C, Snyder SH. (2007). From the Cover: Antipsychotic
drug-induced weight gain mediated by histamine H1 receptor-linked activation of hypothalamic
AMP-kinase . Proc Natl Acad Sci U S A. 104(9):3456-9.
Kishi, T, Matsuda, Y, Nakamura, H, Iwata, N (2013). "Blonanserin for schizophrenia: Systematic review
and meta-analysis of double-blind, randomized, controlled trials". Journal of Psychiatric Research
47 (2): 149–54.
Komossa, K, Depping, AM, Gaudchau, A, Kissling, W, Leucht, S (2010). "Second-generation
antipsychotics for major depressive disorder and dysthymia." (PDF). The Cochrane Database of
Systematic Reviews (12): CD008121.
Komossa, K, Rummel-Kluge, C, Hunger, H, Schmid, F, Schwarz, S, Duggan, L; Kissling, W, Leucht, S
(2010). Komossa, Katja, ed. "Olanzapine versus other atypical antipsychotics for schizophrenia".
Cochrane Database of Systematic Reviews (3): CD006654.
164
Koylu EO, Couceyro PR, Lambert PD, Ling NC, DeSouza EB, Kuhar MJ. (1997). Immunohistochemical
localization of novel CART peptides in rat hypothalamus, pituitary and adrenal gland. J
Neuroendocrinol. 1997 Nov;9(11):823-33.
Krabbendam L, van Os J. (2005). Schizophrenia and urbanicity: a major environmental
influence--conditional on genetic risk.Schizophr Bull. 31(4):795-9.
Kraus T, Haack M, Schuld A, Hinze-Selch D, Kühn M, Uhr M, Pollmächer T. (1999). Body weight and
leptin plasma levels during treatment with antipsychotic drugs. Am J Psychiatry. 156(2):312-4.
Kring SI, Holst C, Toubro S, Astrup A, Hansen T, Pedersen O, Sørensen TI. (2010).Common variants
near MC4R in relation to body fat, body fat distribution, metabolic traits and energy
expenditure. Int J Obes (Lond). 34(1):182-9.
Kroeze WK, Hufeisen SJ, Popadak BA, Renock SM, Steinberg S, Ernsberger P, Jayathilake K,
Meltzer HY, Roth BL. (2003). H1-histamine receptor affinity predicts short-term weight gain for
typical and atypical antipsychotic drugs. Neuropsychopharmacology. 28(3):519-26.
Kuzman MR, Medved V, Bozina N, Hotujac L, Sain I, Bilusic H. (2008).The influence of 5-
HT(2C) and MDR1 genetic polymorphisms on antipsychotic-induced weight gain in
female schizophrenic patients. Psychiatry Res. 160(3):308–315.
Lahiri DK, Nurnberger JI. (1991).A rapid non-enzymatic method for the preparation of HMW
DNA from blood for RFLP studies. Nucleic Acids Res. 19: 5444.
165
Lane HY, Liu YC, Huang CL, Chang YC, Wu PL, Lu CT, Chang WH. (2006).Risperidone-related weight
gain: genetic and nongenetic predictors.J Clin Psychopharmacol. 26(2):128-34.
Lee SH, DeCandia TR, Ripke S, Yang J; Schizophrenia Psychiatric Genome-Wide Association Study
Consortium (PGC-SCZ); International Schizophrenia Consortium (ISC); Molecular Genetics of
Schizophrenia Collaboration (MGS), Sullivan PF, Goddard ME, Keller MC, Visscher PM, Wray
NR (2012).Estimating the proportion of variation in susceptibility to schizophrenia captured by
common SNPs. Nat Genet. 44(3):247-50.
Lee SJ, Verma S, Simonds SE, Kirigiti MA, Kievit P, Lindsley SR et al. (2013). Leptin stimulates
neuropeptide Y and cocaine amphetamine-regulated transcript coexpressing neuronal activity in
the dorsomedial hypothalamus in diet-induced obese mice. J Neurosci. 18;33(38):15306-17.
Lencz T, Robinson DG, Napolitano B, Sevy S, Kane JM, Goldman D, Malhotra AK. (2010). “DRD2
promoter region variation predicts antipsychotic-induced weight gain in first episode
schizophrenia.” Pharmacogenetics and Genomics. 20(9):569-72.
Leibowitz SF, Alexander JT. (1998). Hypothalamic serotonin in control of eating behavior, meal
size, and body weight. Biol Psychiatry. 44(9):851-64.
Leinninger GM, Jo YH, Leshan RL, Louis GW, Yang H, Barrera JG, (2009).Leptin acts via leptin
receptor-expressing lateral hypothalamic neurons to modulate the mesolimbic dopamine system
and suppress feeding.Cell Metab. 10(2):89-98.
Lett TA, Chakavarty MM, Felsky D, Brandl EJ, Tiwari AK, Gonçalves VF, Rajji TK, Daskalakis ZJ,
Meltzer HY, Lieberman JA, Lerch JP, Mulsant BH, Kennedy JL, Voineskos AN. (2013). The
genome-wide supported microRNA-137 variant predicts phenotypic heterogeneity within
166
schizophrenia. Mol Psychiatry. (4):443-50.
Lett TA, Wallace TJ, Chowdhury NI, Tiwari AK, Kennedy JL, Müller DJ. (2012).
Pharmacogenetics of antipsychotic-induced weight gain: review and clinical implications.
Mol Psychiatry. 17(3):242-66.
Leucht, S, Corves, C, Arbter, D, Engel, RR, Li, C, Davis, JM (2009).
"Second-generation versus first-generation antipsychotic drugs for schizophrenia: A meta-
analysis". The Lancet 373 (9657): 31–41.
Leucht, S, Cipriani, A, Spineli, L, Mavridis, D, Örey, D, Richter, F et al (2013). "Comparative efficacy
and tolerability of 15 antipsychotic drugs in schizophrenia: A multiple-treatments meta-analysis".
The Lancet 382 (9896): 951–62.
Li J, Ji L. (2005). Adjusting multiple testing in multilocus analyses using the eigenvalues of a
correlation matrix. Heredity 95:221-227
Li WD, Joo EJ, Furlong EB, Galvin M, Abel K, Bell CJ, Price RA. (2000). Melanocortin 3 receptor
(MC3R) gene variants in extremely obese women. Int J Obes Relat Metab Disord. 24(2):206-10.
Li S, Zhao JH, Luan J, Luben RN, Rodwell SA, Khaw KT, Ong KK, Wareham NJ, Loos RJ. (2010).
Cumulative effects and predictive value of common obesity-susceptibility variants
identified by genome-wide association studies. Am J Clin Nutr. 91(1):184-90.
Lieberman JA, Stroup TS, McEvoy JP, et al. (2005). Effectiveness of antipsychotic drugs in
patients with chronic schizophrenia. N Engl J Med. (12),1209-23.
167
Lim CT, Kola B, Korbonits M. (2010). AMPK as a mediator of hormonal signalling. Journal of
Molecular Endocrinology. 44, 87-97.
Lin PI, Vance JM, Pericak-Vance MA, Martin ER. (2007). No gene is an island: the flip-
flopphenomenon. Am J Hum Genet.;80(3):531-8.
Lipska BK, Khaing ZZ, Weickert CS, Weinberger DR. (2001).BDNF mRNA expression in rat
hippocampus and prefrontal cortex: effects of neonatal ventral hippocampal damage and
antipsychotic drugs. Eur J Neurosc14(1):135-44.
Liu G, Zhu H, Lagou V, Gutin B, Barbeau P, Treiber FA, Dong Y, Snieder H. (2010). Common variants
near melanocortin 4 receptor are associated with general and visceral adiposity in
European- and African-American youth. J Pediatr. 156(4):598-605.
Lombard Z, Crowther NJ, van der Merwe L, Pitamber P, Norris SA, Ramsay M. (2012). Appetite
regulation genes are associated with body mass index in black South African adolescents:
a genetic association study. BMJ Open. 21;2(3).
Loos RJ, Lindgren CM, Li S, Wheeler E, Zhao JH, Prokopenko I, Inouye M, Freathy RM, Attwood AP et
al., (2008). Common variants near MC4R are associated with fat mass, weight and risk of
obesity. Nat Genet. 40(6):768-75.
Lubrano-Berthelier C, Durand E, Dubern B, Shapiro A, Dazin P, Weill J, et al. (2003). Intracellular
retention is a common characteristic of childhood obesity-associated MC4R mutations.
Hum Mol Genet. 12(2):145-53.
168
Macciardi F, Owen MJ. (1996). Association between schizophrenia and T102C
polymorphism of the 5-hydroxytryptamine type 2a-receptor gene. European Multicentre
Association Study of Schizophrenia (EMASS) Group. Lancet. 11;347(9011):1294-6.
Marcadenti A, Fuchs FD, Matte U, Sperb F, Moreira LB, Fuchs SC. (2013). Effects of FTO RS9939906
and MC4R RS17782313 on obesity, type 2 diabetes mellitus and blood pressure in patients with
hypertension.Cardiovasc Diabetol. 2013. 12;12:103.
Malhotra AK, Correll CU, Chowdhury NI, Müller DJ, Gregersen PK, Lee AT, et al. (2012).
Association between common variants near the melanocortin 4 receptor gene and severe
antipsychotic drug-induced weight gain. Arch Gen Psychiatry. 69(9):904-12.
Mammès O, Betoulle D, Aubert R, Giraud V, Tuzet S, Petiet A, Colas-Linhart N, Fumeron F.
(1998).Novel polymorphisms in the 5' region of the LEP gene: association with leptin levels and
response to low-calorie diet in human obesity. Diabetes. 47(3):487-9.
Mammès O, Aubert R, Betoulle D, Péan F, Herbeth B, Visvikis S, Siest G, Fumeron F. (2001).LEPR
gene polymorphisms: associations with overweight, fat mass and response to diet in women.
Eur J Clin Invest. 31(5):398-404.
Manolio TA; Guttmacher, Alan E.; Manolio, Teri A. (2010). "Genomewide association studies and
assessment of the risk of disease". N. Engl. J. Med. 363 (2): 166–76.
Marcelis M, Takei N, van Os J. (1999). Urbanization and risk for schizophrenia: does the effect
operate before or around the time of illness onset?Psychol Med. 29(5):1197-203.
169
Marcus R. Munafò,Nicholas J. Timpson, Sean P. David, Shah Ebrahim, and Debbie A. (2009). Lawlor
Association of the DRD2 gene Taq1A polymorphism and smoking behavior: A meta-
analysis and new data. Nicotine Tob Res. 11(1): 64–76.
Marzullo G, &Fraser FC. (2005). Similar rhythms of seasonal conceptions in neural tube defects
and schizophrenia: a hypothesis of oxidant stress and the photoperiod.Birth Defects Res
A Clin Mol Teratol. (1):1-5.
Masellis M, Basile V, Meltzer HY, Lieberman JA, Sevy S, Macciardi FM, et al. (1998). Serotonin
subtype 2receptor genes and clinical response to clozapine in schizophrenia patients.
Neuropsychopharmacology. 19(2):123-32.
Masellis, Tiwari AK, Brandl EJ, Weber C, Likhodi O, Zai CC, Hahn MK, et al. (2013). Association of a
functional polymorphism in neuropeptide Y with antipsychotic-induced weight gain in
schizophrenia patients. J Clin Psychopharmacol. 33(1):11-7.
Masuzaki H, Tanaka T, Ebihara K, Hosoda K, Nakao K. (2009).Hypothalamic melanocortin signaling
and leptin resistance--perspective of therapeutic application for obesity-diabetes
syndrome. Peptides. 30(7):1383-6.
Maurer I, Zierz S, Möller H. (2001). Evidence for a mitochondrial oxidative phosphorylation defect in
brains from patients with schizophrenia.Schizophr Res. 48(1):125-36.
McCarthy S, Mottagui-Tabar S, Mizuno Y, Sennblad B, Hoffstedt J, Arner P et al. (2005). Complex
HTR2C linkage disequilibrium and promoter associations with body mass index and serum
170
leptin. Human Genetics. 117: 545–557.
McEvoy JP, Meyer JM, Goff DC, Nasrallah HA, Davis SM, Sullivan L et al. (2005). Prevalence of the
metabolic syndrome in patients with schizophrenia:baseline results from the Clinical
Antipsychotic Trials of Intervention Effectiveness (CATIE) schizophrenia trial and
comparison with national estimates from III NHANES. Schizophr Res. 8: 19–32.
McGrath, J., S. Saha, et al. (2008). "Schizophrenia: a concise overview of incidence,
prevalence, and mortality." Epidemiol Rev.30: 67-76.
McIntyre RS, McCann SM, Kennedy SH. (2001).Antipsychotic metabolic effects: weight gain, diabetes
mellitus, and lipid abnormalities.Can J Psychiatry. 46(3):273-81.
McNally, K. (2009). "Eugene Bleuler's four As." HistPsychol 12(2): 43-59.
Mei Z, Grummer-Strawn LM, Pietrobelli A, Goulding A, Goran MI, Dietz WH. (2002). Validity of body
mass index compared with other body-composition screening indexes for the assessment of
body fatness in children and adolescents. American Journal of Clinical Nutrition. 7597–985.
Meltzer HY, Matsubara S, Lee JC. (1989). Classification of typical and atypical antipsychotic
drugs on the basis of dopamine D-1, D-2 and serotonin2 pKi values. J Pharmacol Exp
Ther. 251(1):238-46.
Mejía-Benítez A, Klünder-Klünder M, Yengo L, Meyre D, Aradillas C, Cruz E, et al. (2013).Analysis of
the contribution of FTO, NPC1, ENPP1, NEGR1, GNPDA2 and MC4R genes to obesity in
Mexican children. BMC Med Genet. 1;14:21.
171
Mencarelli M, Walker GE, Maestrini S, Alberti L, Verti B, Brunani A. (2008). Sporadic mutations in
melanocortin receptor 3 in morbid obese individuals. Eur J Hum Genet.16(5):581-6.
Messias, E., Chuan-Yu,C., EatonW.W. (2007). Epidemiology of Schizophrenia: Review of Findings and
Myths. Psychiatr Clin North Am. 30(3): 323–338.
Meyer JM, Nasrallah HA, McEvoy JP, Goff DC, Davis SM, Chakos M et al. (2005).The Clinical
Antipsychotic Trials of Intervention Effectiveness (CATIE) Schizophrenia Trial: clinical
comparison of subgroups with and without the metabolic syndrome. Schizophr Res.
80: 9–18.
Meyer JM, Simpson GM. (1997). From chlorpromazine to olanzapine: a brief history of antipsychotics.
Psychiatr Serv. 48(9):1137-9.
Meyer J, Loh C, Leckband SG, Boyd JA, Wirshing WC, Pierre JM et al. (2006). Prevalence of the
metabolic syndrome in veterans with schizophrenia. J Psychiatr Pract.12: 5–10.
Meyre D, Delplanque J, Chevre JC, Lecoeur C, Lobbens S, et al. (2009) Genome-wide association study
for early-onset and morbid adult obesity identifies three new risk loci in European
populations. Nat Genet 41: 157–159
Michaud EJ, Mynatt RL, Miltenberger RJ, Klebig ML, Wilkinson JE, Zemel MB, et al. (1997). Role of
the agouti gene in obesity. J Endocrinol 155:207–209.
Minokoshi Y, Alquier T, Furukawa N, Kim YB, Lee A, Xue B, Mu J (2004). AMP-kinase regulates food
intake by responding to hormonal and nutrient signals in the hypothalamus.Nature.
172
428(6982):569-74.
Monteleone P, Bifulco M, Maina G, Tortorella A, Gazzerro P, Proto MC, et al. (2010).Investigation of
CNR1 and FAAH endocannabinoid gene polymorphisms in bipolar disorder and major
depression. Pharmacol Res.61(5):400-4.
Moons T, Claes S, Martens GJ, Peuskens J, Van Loo KM, Van Schijndel JE, De Hert M, van Winkel
R.vClock genes and body composition in patients with schizophrenia under treatment with
antipsychotic drugs. Schizophr Res. 2011 (2-3):187-93.
Mulder H, Cohen D, Scheffer H, Gispen-de Wied C, Arends J, Wilmink FW, et al. (2009).HTR2C gene
polymorphisms and the metabolic syndrome in patients with schizophrenia: a replication study.
J Clin Psychopharmacol. 29(1):16-20.
Müller DJ, Kennedy JL. (2006).Genetics of antipsychotic treatment emergent weight gain in
schizophrenia. Pharmacogenomics. 6):863-87.
Müller DJ, Zai CC, Sicard M, Remington E, Souza RP, Tiwari AK, et al. (2012). Systematic analysis
of dopamine receptor genes (DRD1-DRD5) in antipsychotic-induced weight gain.
Pharmacogenomics J.12(2):156–64.
Mutombo PB, Yamasaki M, Hamano T, Isomura M, Nabika T, Shiwaku K. (2014). MC4R rs17782313
gene polymorphism was associated with glycated hemoglobin independently of its effect on BMI
in Japanese: the Shimane COHRE study. Endocr Res. 39(3):115-9.
Minokoshi Y, Kim YB, Peroni OD, Fryer LG, Müller C, Carling D, Kahn BB. (2002). Leptin stimulates
173
fatty-acid oxidation by activating AMP-activated protein kinase. Nature. 2002 Jan
17;415(6869):339-43.
Millar JK, Christie S, SempleCA, Porteous DJ. (2000).Chromosomal location and genomic
structure of the human translin-associated factor X gene (TRAX; TSNAX) revealed by
intergenic splicing to DISC1, a gene disrupted by a translocation segregating with
schizophrenia.Genomics. 67(1):69-77.
Miller DD, Ellingrod VL, Holman TL, Buckley PF, Arndt S. Clozapine-induced weight gain
associated with the 5HT2C receptor −759C/T polymorphism. (2005).Am J Med Genet B
Neuropsychiatr Genet. (1):97–100.
Miraglia Del Giudice E, Cirillo G, Nigro V, Santoro N, D'Urso L, Raimondo P.(2002). Low frequency of
melanocortin-4 receptor (MC4R) mutations in a Mediterranean population with early-onset
obesity. Int J Obes Relat Metab Disord26(5):647-51.
Miyaki K, Sutani S, Kikuchi H, Takei I, Murata M, Watanabe K, et al. (2005). Increased risk of obesity
resulting from the interaction between high energy intake and the Trp64Arg polymorphism of
the beta3-adrenergic receptor gene in healthy Japanese men. J Epidemiol. 15(6):203-10.
Morris JA, Kandpal G, Ma L, Austin CP. (2003).DISC1 (Disrupted-In-Schizophrenia 1) is a
centrosome-associated protein that interacts with MAP1A, MIPT3, ATF4/5 and NUDEL:
regulation and loss of interaction with mutation.Hum Mol Genet. (13):1591-608.
Müller DJ, Kennedy JL. (2006). Genetics of antipsychotic treatment emergent weight gain in
174
schizophrenia.Pharmacogenomics. 7(6):863-87.
Müller DJ, Klempan TA, De Luca V, Sicard T, Volavka J, Czobor P et al. (2005). The
SNAP-25 gene may be associated with clinical response and weight gain in
antipsychotic treatment of schizophrenia. Neurosci Lett. 379: 81–89.
Munafò MR, Thiselton DL, Clark TG, Flint J. (2006). Association of the NRG1 gene and
schizophrenia: a meta-analysis. Mol Psychiatry. 11(6):539-46.
Murphy KC.(2002). Schizophrenia and velo-cardio-facial syndrome. 359:426-430.
Mutschler J, Abbruzzese E, Wiedemann K, von der Goltz C, Dinter C, Mobascher A, et al. (2013).
Functional polymorphism in the neuropeptide Y gene promoter (rs16147) is associated with
serum leptin levels and waist-hip ratio in women. Ann Nutr Metab. 62(4):271-6.
Murray, R, Lewis SW. (1987). Is schizophrenia a neurodevelopmental disorder? Br Med J (Clin Res Ed).
295(6600): 681–682.
Nasrallah, HA. (2008).Atypical antipsychotic-induced metabolic side effects: insights from receptor-
binding profiles. Mol Psychiatry. 13(1):27-35.
Neel JV (1962). "Diabetes Mellitus: A "Thrifty" Genotype Rendered Detrimental by "Progress"?". Am. J.
Hum. Genet. 14 (4): 353–62.
Nguyen CT, Rosen JA, Bota RG. (2012).Aripiprazole partial agonism at 5-HT2C: a comparison of weight
gain associated with aripiprazole adjunctive to antidepressants with high versus low
175
serotonergic activities. Prim Care Companion CNS Disord. 14(5).
Nijenhuis WA, Oosterom J, Adan RA. (2001). AgRP(83-132) acts as an inverse agonist on the human-
melanocortin-4 receptor. Mol Endocrinol. 2001 Jan;15(1):164-71.
Norris RA, Kern MJ. (2001). The identification of Prx1 transcription regulatory domains provides a
mechanism for unequal compensation by the Prx1 and Prx2 loci. J Biol Chem 2001; 276: 26829–
26837.
Nurmi EL, Spilman SL, Whelan F, Scahill LL, Aman MG, McDougle, et al. (2013). Research Units on
Pediatric Psychopharmacology Autism Network).Moderation of antipsychotic-induced weight
gain by energy balance gene variants in the RUPP autism network risperidone studies. Transl
Psychiatry. 5;3:e274.
Nyholt DR. (2004). A simple correction for multiple testing for single-nucleotide polymorphisms in
linkage disequilibrium with each other. Am J Hum Genet.74(4):765-9.
Nutt, DJ. (1994) Putting the 'A' in atypical: does alpha-2 adrenoceptor antagonism account for
the therapeutic advantage of new antipsychotics? J Psychopharmacol., 8, 193-5.
Noble EP, Blum K, Khalsa ME, Ritchie T, Montgomery A, Wood RC, Fitch RJ, Ozkaragoz T,
Sheridan PJ, Anglin MD, et al. (1993). Allelic association of the D2 dopamine receptor
gene with cocaine dependence.Drug Alcohol Depend.33(3):271-85.
Nyholt DR. (2004). A simple correction for multiple testing for single-nucleotide polymorphisms in
linkage disequilibrium with each other.Am J Hum Genet. 74(4):765-9.
176
Obregón AM, Diaz E, Santos JL.J Physiol Biochem. (2012). Effect of the melanocortin-3 receptor
Thr6Lys and Val81Ile genetic variants on body composition and substrate oxidation in
Chilean obese children.68(1):71-6.
Ollmann MM, Wilson BD, Yang YK, Kerns JA, Chen Y, Gantz I, Barsh GS. (1997). Antagonism of
central melanocortin receptors in vitro and in vivo by agouti-related protein. Science 278: 135-
138.
Opgen-Rhein C, Brandl EJ, Müller DJ, et al. (2010). Association of HTR2C, but not LEP or INSIG2,
genes with antipsychotic-induced weight gain in a German sample. Pharmacogenomics.
11(6):773–780.
Orkunoglu-Suer FE, Harmon BT, Gordish-Dressman H, Clarkson PM, Thompson PD, Angelopoulos TJ,
et al. (2011). MC4R variant is associated with BMI but not response to resistance training in
young females. Obesity (Silver Spring).;19(3):662-6.
Ortega-Azorín C, Sorlí JV, Asensio EM, Coltell O, Martínez-González MÁ, Salas-Salvadó J, et al.
(2012).Associations of the FTO rs9939609 and the MC4R rs17782313 polymorphisms with type
2 diabetes are modulated by diet, being higher when adherence to the Mediterranean diet
pattern is low. Cardiovasc Diabetol. 6;11:137.
Onrust, SV & McClellan, Karen (2001). "Perospirone". CNS Drugs 15 (4): 329–37; discussion 338.
Owen, M. J., N. M. Williams, et al. (2004). "The molecular genetics of schizophrenia: new
findings promise new insights." Mol Psychiatry 9(1): 14-27.
177
Panariello F, De Luca V, de Bartolomeis A.(2011).Weight gain, schizophrenia and antipsychotics: new
findings from animal model and pharmacogenomic studies. Schizophr Res Treatment.
2011:459284.
Parada, M. A,, L. Hernandez & B. G. Hoebel (1988). Sulpiride injections in the lateral
hypothalamus induce feeding and drinking in rats. Pharmacol. Biochem. Behav. 30, 917-
923.
Park YM, Chung YC, Lee SH, Lee KJ, Kim H, Byun YC, (2006). HJ.Weight gain associated with
the alpha2a-adrenergic receptor -1,291 C/G polymorphism and olanzapine treatment. Am J Med
Genet B Neuropsychiatr Genet.141B(4):394-7.
Pearson TA, Manolio TA (2008). "How to interpret a genome-wide association study". JAMA 299 (11):
1335–44.
Peduzzi,P., Concato,J., Kemper,E., Holford,T.R. and Feinstein,A.R. (1996). A simulation study of the
number of events per variable in logistic regression analysis. J. Clin. Epidemiol., 49, 1373–1379.
Perneger TV. (1998). What's wrong with Bonferroni adjustments.BMJ. 316(7139):1236-8.
Petry CJ, López-Bermejo A, Díaz M, Sebastiani G, Ong KK, de Zegher F, Dunger DB, Ibáñez L (2010).
Association between a common variant near MC4R and change in body mass index
develops by two weeks of age. Horm Res Paediatr. 73(4):275-80.
Polesskaya OO& Sokolov BP. (2002). Differential expression of the "C" and "T" alleles of the
178
5-HT2A receptor gene in the temporal cortex of normal individuals and schizophrenics.
J Neurosci Res. 67(6):812-22.
Popp J, Leucht S, Heres S, Steimer W. (2009). DRD4 48 bp VNTR but not 5-HT 2C Cys23Ser receptor
polymorphism is related to antipsychotic-induced weight gain. Pharmacogenomics J. 9(1):71-7.
Prabakaran S, Swatton JE, Ryan MM, Huffaker SJ, Huang JT, Griffin JL, et al. (2004).Mitochondrial
dysfunction in schizophrenia: evidence for compromised brain metabolism and oxidative stress.
Mol Psychiatry. 9(7):684-97, 643.
Prentice AM and Jebb SA. (2001). Beyond Body Mass Index. Obesity Reviews. 2(3): 141–7.
Ptak C, &Petronis A. (2010). Epigenetic approaches to psychiatric disease. Dialogues Clin Neurosci.
12(1):25-35.
Price and Gottesman (1991). Body fat in identical twins reared apart: roles for genes and
environments. Behav Genet, 21 (1991), pp. 1–7.
Qi L, Cho YA. (2008). Gene-environment interaction and obesity. Nutr Rev. 66(12):684-94.
Qu D, Ludwig DS, Gammeltoft S, Piper M, Pelleymounter MA, Cullen MJ, Mathes WF, et al. (1996).A
role for melanin-concentrating hormone in the central regulation of feeding behaviour. Nature.
21;380(6571):243-7.
Ravinet Trillou C, Delgorge C, Menet C, Arnone M, Soubrie P. (2004). CB1 cannabinoid receptor
knockout in mice leads to leanness, resistance to diet-induced obesity and enhanced leptin
179
sensitivity.Int J Obes Relat Metab Disord. 28 : 640–648.
Ray WA, Chung CP, Murray KT, Hall K, Stein CM. (2009). Atypical antipsychotic drugs and the risk of
sudden cardiac death. N Engl J Med 360: 225-235.
Reich, DE (2001). "Linkage disequilibrium in the human genome". Nature (Nautre Publishing Group)
411: 199–204.
Renström F, Payne F, Nordström A, Brito EC, Rolandsson O, Hallmans Get al; GIANT Consortium.
(2009).Replication and extension of genome-wide association study results for obesity in 4923
adults from northern Sweden. Hum Mol Genet. 15;18(8):1489-96.
Remington G, Mamo D, Labelle A, Reiss J, Shammi C, Mannaert E, Mann S, Kapur S. (2006).
A PET study evaluating dopamine D2 receptor occupancy for long-acting injectable
risperidone.Am J Psychiatry. 163(3):396-401.
Reynolds GP, Hill MJ, Kirk SL. (2006).The 5-HT2C receptor and antipsychotic induced weight
gain - mechanisms and genetics. J Psychopharmacol. 4 Suppl,15-8.
Reynolds GP, Zhang Z-J, Zhang X-B. (2002). Association of antipsychotic drug-induced weight
gain with a 5-HT2C receptor gene polymorphism. Lancet. 359(9323):2086–2087.
Reynolds GP. (2012). Histamine and antipsychotic drug-induced weight gain. J Psychopharmacol.
26(12):1608-9.
Rigoli L, Munafò C, Di Bella C, Salpietro A, Procopio V, Salpietro C. (2010). Molecular analysis of the
180
CART gene in overweight and obese Italian children using family-based association methods.
Acta Paediatr. 99(5):722-6.
Ripke S,Sanders AR, Kendler KS, Levinson DF, Sklar P, Holmans PA et al. Genome-wide association
study identifies five new schizophrenia loci.Schizophrenia Psychiatric Genome-Wide Association
Study (GWAS) Consortium. Nat Genet. 2011 Sep 18;43(10):969-76.
Ripke S, Neale BM, Corvin A, Walters JT, Farh KH, Holmans PA, (2014).Biological insights from 108
schizophrenia-associated genetic loci. Schizophrenia Working Group of the Psychiatric Genomics
Consortium. Nature. 511(7510):421-7.
Rodrigues AR, Almeida H, Gouveia AM. (2013). Alpha-MSH signalling via melanocortin 5
receptor promotes lipolysis and impairs re-esterification in adipocytes. Biochim Biophys
Acta. (7),1267-75.
Roerig JL, Steffen KJ, Mitchell JE. (2011).Atypical antipsychotic-induced weight gain: insights
into mechanisms of action. CNS Drugs. 25(12):1035-59.
Rosito GA, Massaro JM, Hoffmann U, Ruberg FL, Mahabadi AA, Vasan RS, et al. (2008). Pericardial
fat, visceral abdominal fat, cardiovascular disease risk factors, and vascular calcification in a
community-based sample: the Framingham Heart Study.Circulation. 117(5):605-13.
Roth BL, Berry SA, Kroeze WK, Willins DL, Kristiansen K. (1998).Serotonin 5-HT2A receptors:
molecular biology and mechanisms of regulation. Crit Rev Neurobiol. 12(4):319-
38. Review.
181
Roth BL, Ciaranello RD, Meltzer HY. (1992).Binding of typical and atypical antipsychotic
agents to transiently expressed 5-HT1C receptors.J Pharmacol Exp Ther. 260(3):1361-5.
Roth BL, Craigo SC, Choudhary MS, Uluer A, Monsma FJ Jr, Shen Y, Meltzer HY, Sibley DR.
J Pharmacol Exp Ther. (1994). Binding of typical and atypical antipsychotic agents to 5-
hydroxytryptamine-6 and 5-hydroxytryptamine-7 receptors. 268(3):1403-10.
Ruano G, Goethe JW, Caley C, Woolley S, Holford TR, Kocherla M, Windemuth A, de Leon J. (2007).
Physiogenomic comparison of weight profiles of olanzapine- and risperidone-treated patients.
Mol Psychiatry. 12:474–82
Rummel-Kluge C, Komossa K, Schwarz S, et al. (2010). Head-to-head comparisons of metabolic
side effects of second generation antipsychotics in the treatment of schizophrenia: a
systematic review and meta-analysis. Schizophr Res. 123(2-3), 225-33.
Sánchez E, Rubio VC, Cerdá-Reverter JM. (2009) .Characterization of the sea bass melanocortin 5
receptor: a putative role in hepatic lipid metabolism. J Exp Biol. Pt 23, 3901-10.
Santini F, Maffei M, Pelosini C, Salvetti G, Scartabelli G, Pinchera A. (2009). Melanocortin-4 receptor
mutations in obesity. Adv Clin Chem. 48:95-109.
Santoro N, Perrone L, Cirillo G, Raimondo P, Amato A, Brienza C, Del Giudice EM. (2007). Effect of
the melanocortin-3 receptor C17A and G241A variants on weight loss in childhood
obesity.Am J Clin Nutr. 85(4):950-3.
Savontaus E, Conwell IM, Wardlaw SL. (2002). Effects of adrenalectomy on AGRP, POMC, NPY and
182
CART gene expression in the basal hypothalamus of fed and fasted rats. Brain Res 958: 130-138.
Saiz PA, Susce MT, Clark DA, Kerwin RW, Molero P, Arranz MJ, de Leon J. (2008). An investigation of
the alpha1A-adrenergic receptor gene and antipsychotic-induced side-effects. (2008). Hum
Psychopharmacol. 23(2):107-14.
Sakurai T, Amemiya A, Ishii M et al. (1998). "Orexins and orexin receptors: a family of hypothalamic
neuropeptides and G protein-coupled receptors that regulate feeding behavior." Cell 92 (4):
573–85.
Schiöth HB, Haitina T, Ling MK, et al. (2005). Evolutionary conservation of the structural,
pharmacological and genomic characteristics of the melanocortin receptors subtypes.
Peptides. 26,1886–1900.
Seeman P, Lee T, Chau-Wong M, Wong K. (1976).Antipsychotic drug doses and
neuroleptic/dopamine receptors.Nature. 261(5562):717-9.
Seeman P, Corbett R, Van Tol HH. (1997).Atypical neuroleptics have low affinity for dopamine
D2 receptors or are selective for D4 receptors. Neuropsychopharmacology. 16(2):93-110.
Sejima E, Yamauchi A, Nishioku T, Koga M, Nakagama K, Dohgu S et al., (2011). A role for
hypothalamic AMP-activated protein kinase in the mediation of hyperphagia and weight gain
induced by chronic treatment with olanzapine in female rats. Cell Mol Neurobiols;31(7):985-9.
Sentissi, O., Epelbaum, J., Olié, J., Poirier, M. (2008).Leptin and Ghrelin Levels in Patients With
Schizophrenia During Different Antipsychotics Treatment: A Review Schizophr Bull.
34(6): 1189–1199.
183
Shimada M, Tritos NA, Lowell BB, Flier JS, Maratos-Flier E. (1998). Mice lacking melanin-
concentrating hormone are hypophagic and lean. Nature. 396(6712):670-4.
Shing EC, Tiwari AK, Brandl EJ, Zai CC, Lieberman JA, Meltzer HY, et al. (2014).Fat mass- and
obesity-associated (FTO) gene and antipsychotic-induced weight gain: an association study.
Neuropsychobiology. 69(1):59-63.
Sicard MN, Zai CC, Tiwari AK, et al. (2010).Polymorphisms of the HTR2C gene and
antipsychotic-induced weight gain: an update and meta-analysis. Pharmacogenomics.
(11):1561–1571.
Sickert L, Müller DJ, Tiwari AK, Shaikh S, Zai C, De Souza R, De Luca V, et al (2009). Association of
the alpha 2A adrenergic receptor -1291C/G polymorphism and antipsychotic-induced weight gain
in European-Americans. Pharmacogenomics. 10(7):1169-76.
Silvestre JS, Prous J. (2005).Research on adverse drug events. I. Muscarinic M3 receptor binding
affinity could predict the risk of antipsychotics to induce type 2 diabetes.Methods Find
Exp Clin Pharmacol. (5):289-304.
Souza RP, Tiwari AK, Chowdhury NI, Ceddia RB, Lieberman JA, Meltzer HY, Kennedy JL, Müller DJ.
(2012). Association study between variants of AMP-activated protein kinase catalytic and \
regulatory subunit genes with antipsychotic-induced weight gain. J Psychiatr Res.
46(4):462-8.
Speliotes EK, Willer CJ, Berndt SI, Monda KL, Thorleifsson G, Jackson AU, Lango Allen H,Lindgren.
184
(2010). Association analyses of 249,796 individuals reveal 18 new loci associated with body
mass index. Nat Genet. 42(11):937-48.
Srivastava A, Mittal B, Prakash J, Narain VS, Natu SM, Srivastava N. (2014). Evaluation of MC4R
[rs17782313, rs17700633], AGRP [rs3412352] and POMC [rs1042571] Polymorphisms with
Obesity in Northern India. Oman Med J. 29(2):114-8.
Srisai D, Gillum MP, Panaro BL, Zhang XM, Kotchabhakdi N, Shulman GI et al. (2011).Characterization
of the hyperphagic response to dietary fat in the MC4R knockout mouse. Endocrinology.152:
890–902.
Spencer-Jones NJ, Ge D, Snieder H, Perks U, Swaminathan R, Spector TD, Carter ND, O'Dell
SD. (2006). AMP-kinase alpha2 subunit gene PRKAA2 variants are associated with total
cholesterol, low-density lipoprotein-cholesterol and high-density lipoprotein-cholesterol
in normal women. J Med Genet.43(12):936-42.
Santini F, Maffei M, Pelosini C, Salvetti G, Scartabelli G, Pinchera A. (2009).Melanocortin-4 receptor
mutations in obesity. Adv Clin Chem. 48:95-109.
Santos JL, De la Cruz R, Holst C, Grau K, Naranjo C, Maiz A, et al; NUGENOB Consortium. (2011).
Allelic variants of melanocortin 3 receptor gene (MC3R) and weight loss in obesity: a
randomised trial of hypo-energetic high- versus low-fat diets. PLoS One. 6(6):e19934.
Sentissi, O., Epelbaum, J Olié, JP., Poirier, M. (2008). Leptin and Ghrelin Levels in Patients With
Schizophrenia During Different Antipsychotics Treatment: A Review. Schizophr Bull.34(6):
1189–1199.
Smith AJ, Humphries SE (January 2009). "Characterization of DNA-binding proteins using multiplexed
185
competitor EMSA". J. Mol. Biol. 385 (3): 714–7.
Stanley BG, Kyrkouli SE, Lampert S, Leibowitz S. (1986).Neuropeptide Y chronically injected into the
hypothalamus: a powerful neurochemical inducer of hyperphagia and obesity. Peptides.
7:1189–1192.
Stefansson H, Rujescu D, Cichon S, Pietiläinen OP, Ingason A, Steinberg S et al. (2008).Large
recurrent microdeletions associated with schizophrenia.Nature. 11;455(7210):232-
Stefansson H, Sigurdsson E, Steinthorsdottir V, Bjornsdottir S, Sigmundsson T, Ghosh S, et
al.(2002). Neuregulin 1 and susceptibility to schizophrenia.Am J Hum Genet. 71(4):877-92.
Stefansson H, Sarginson J, Kong A, Yates P, Steinthorsdottir V, Gudfinnsson E, Gunnarsdottir
S, Walker N, Petursson H, Crombie C, Ingason A, Gulcher JR, Stefansson K, St Clair D.
(2003). Association of neuregulin 1 with schizophrenia confirmed in a Scottish population.
Am J Hum Genet.;72(1):83-7.
Stone, M. H. (2006). History of Schizophrenia and Its Antecedents. Textbook of Schizophrenia.
J. Lieberman, S. Stroup and D. Perkins. Arlington, VA, American Psychiatric Publishing,
Inc.
Stunkard AJ, Foch TT, Hrubec Z. (1986). A twin study of human obesity. JAMA. 256(1):51-4.
Stunkard AJ, Sørensen TI, Hanis C, Teasdale TW, Chakraborty R, Schull WJ, Schulsinger F. (1986). An
adoption study of human obesity. N Engl J Med. 314(4):193-8.
186
Stutzmann F, Cauchi S, Durand E, Calvacanti-Proença C, Pigeyre M, Hartikainen AL, et al. (2009).
Common genetic variation near MC4R is associated with eating behaviour patterns in European
populations. Int J Obes (Lond). 33(3):373-8.
Sulem P, Gudbjartsson DF, Stacey SN, et al. (2007). Genetic determinants of hair, eye and skin
pigmentation in Europeans. Nat Genet. 39(12),1443-52.
Sull JW, Lee M, Jee SH. (2013). Replication of genetic effects of MC4R polymorphisms on body mass
index in a Korean population. Endocrine. 44(3):675-9.
Sullivan, P. F., K. S. Kendler, et al. (2003). "Schizophrenia as a complex trait: evidence from
a meta-analysis of twin studies." Arch Gen Psychiatry 60(12): 1187-92.
Sullivan, P.F., O’Donovan, Mick., Purcell, S., Neale, B., Wray., N., Daly, M. Proceedings of the
XXIst World Congress of Psychiatric Genetics; 2012. October 21. Hamburg, Germany.
Susser E, Neugebauer R, Hoek HW, Brown AS, Lin S, Labovitz D, Gorman JM. (1996).Schizophrenia
after prenatal famine. Further evidence.Arch Gen Psychiatry. 53(1):25-31.
Susser E, Miller M, Valencia E, Colson P, Roche B, Conover S. (1996).Injection drug use and
risk of HIV transmission among homeless men with mental illness.Am J Psychiatry.
153(6):794-8.
Sutton BS, Langefeld CD, Williams AH, Norris JM, Saad MF, Haffner SM, Bowden DW. (2005).
Association of proopiomelanocortin gene polymorphisms with obesity in the IRAS family study.
Obes Res 13: 1491-1498
187
Sørensen TI, Price RA, Stunkard AJ, Schulsinger F. (1989). Genetics of obesity in adult adoptees
and their biological siblings.BMJ. 298(6666):87-90.
Tan W, Wang Y, Gold B, Chen J, Dean M, Harrison PJ, Weinberger DR, Law AJ. (2007). Molecular
cloning of a brain-specific, developmentally regulated neuregulin 1 (NRG1) isoform and
identification of a functional promoter variant associated with schizophrenia. J
Biol Chem. 282(33):24343-51.
Tao YX, Segaloff DL. (2004). Functional characterization of melanocortin-3 receptor variants identify a
loss-of-function mutation involving an amino acid critical for G protein-coupled receptor
activation. J Clin Endocrinol Metab. 89(8):3936-42.
Tao L, Zhang Z, Chen Z, Zhou D, Li W, Kan M, et al., (2012). Common variant near the melanocortin 4
receptor is associated with low-density lipoprotein cholesterol and total cholesterol in the
Chinese Han population. Mol Biol Rep. 39(6):6487-93.
Tang JX, Chen WY, He G, Zhou J, Gu NF, Feng GY, He L. (2004).Polymorphisms within 5' end
of the Neuregulin 1 gene are genetically associated with schizophrenia in the Chinese
population. Mol Psychiatry. 9(1):11-2.
Taylor, D, Paton, C, Kapur, S (2012). The Maudsley Prescribing Guidelines (12th ed.). Informa
Healthcare. pp. 12–152, 173–196, 222–235.
Templeman LA, Reynolds GP, Arranz B, San L. (2005). Polymorphisms of the 5-HT2C receptor and
leptin genes are associated with antipsychotic drug-induced weight gain in Caucasian
188
subjects with a first-episode psychosis. Pharmacogenet Genomics. 15(4):195–200.
Ternouth A, Brandys MK, van der Schouw YT, Hendriks J, Jansson JO, Collier D, Adan RA. (2011).
Association study of POMC variants with body composition measures and nutrient choice. Eur J
Pharmacol 660: 220-225.
Theisen,F. M., Linden, A, König,I. R., Martin,M., Remschmidt, H., Hebebrand, J. (2003).
Spectrum of binge eating symptomatology in patients treated with clozapine and
Olanzapine. Journal of Neural Transmission. Volume 110, Issue 1, pp 111-121.
Thomsen M, Dahl M, Tybjærg-Hansen A, Nordestgaard BG. (2012).β2-adrenergic receptor
Thr164Ile polymorphism, obesity, and diabetes: comparison with FTO, MC4R, and
TMEM18 polymorphisms in more than 64,000 individuals. J Clin Endocrinol Metab.
97(6):E1074-9.
Thorleifsson G, Walters GB, Gudbjartsson DF, Steinthorsdottir V, Sulem P, Helgadottir A, et al. (2009).
Genome-wide association yields new sequence variants at seven loci that associate with
measures of obesity. Nat Genet. 41(1):18-24.
Timpson NJ, Harbord R, Davey Smith G, Zacho J, Tybjaerg-Hansen A, Nordestgaard BG. (2009).Does
greater adiposity increase blood pressure and hypertension risk?: Mendelian randomization
using the FTO/MC4R genotype. Hypertension.54(1):84-90.
Tiwari, A, inventor; Compositions and Methods for Treatment and Prevention of Antipsychotic Induced
Weight Gain. US patent 42803-50801. February 28, 2014.
189
Tiwari AK, Brandl EJ, Weber C, Likhodi O, Zai CC, Hahn MK, et al. (2013). Association of a functional
polymorphism in neuropeptide Y with antipsychotic-induced weight gain in schizophrenia
patients. J Clin Psychopharmacol. 33(1):11-7.
Tiwari AK, Zai CC, Likhodi O, Lisker A, Singh D, Souza RP, Batra P. (2010). A common polymorphism
in the cannabinoid receptor 1 (CNR1) gene is associated with antipsychotic-induced weight
gain in Schizophrenia. Neuropsychopharmacology. 35(6):1315-24.
Torrey EF, Miller J, Rawlings R, Yolken RH. (1997). Seasonality of births in schizophrenia and
bipolar disorder: a review of the literature.Schizophr Res.28(1):1-38. Review.
Towler DA, Rutledge SJ, Rodan GA. (1994). Msx-2/Hox 8.1: a transcriptional regulator of the rat
osteocalcin promoter. Mol Endocrinol. 11: 1484–1493.
Tsai SJ, Ouyang WC, Hong CJ.(2002).Association for serotonin transporter gene variable number
tandem repeat polymorphism and schizophrenic disorders. Neuropsychobiology.
45(3):131-3.
Tschoner A, Engl J, Laimer M, Kaser S, Rettenbacher M, Fleischhacker WW, Patsch JR, Ebenbichler CF.
(2007). Metabolic side effects of antipsychotic medication. Int J Clin Pract. 61(8):1356-70.
Umbricht DS, Pollack S, Kane JM. (1994) Clozapine and weight gain. J Clin Psychiatry 55,157–
160.
Ujike H, Nomura A, Morita Y, Morio A, Okahisa Y, Kotaka T, Kodama M, Ishihara T, Kuroda
190
S. (2008).Multiple genetic factors in olanzapine-induced weight gain in schizophrenia patients: a
cohort study. J Clin Psychiatry. (9):1416-22.
Vaisse C, Clement K, Guy-Grand B, Froguel P. (1998).A frameshift mutation in human MC4R is
associated with a dominant form of obesity. Nat Genet. 20(2):113-4.
Valladares M, Domínguez-Vásquez P, Obregón AM, Weisstaub G, Burrows R, Maiz A, Santos JL.
(2010).Melanocortin-4 receptor gene variants in Chilean families: association with childhood
obesity and eating behavior. Nutr Neurosci. 13(2):71-8.
van Erp TG, Guella I, Vawter MP, Turner J, Brown GG, McCarthy G, et al. (2014).Schizophrenia miR-
137 locus risk genotype is associated with dorsolateral prefrontal cortex hyperactivation.Biol
Psychiatry. 75(5):398-405.
van Os J, Bak M, Hanssen M, Bijl RV, de Graaf R, Verdoux H. (2002).Cannabis use and
psychosis: a longitudinal population-based study.Am J Epidemiol. 15;156(4):319-27.
van Os, J., S. Kapur (2009). "Schizophrenia." Lancet 374(9690): 635-45.
Van Tol HH, Bunzow JR, Guan HC, Sunahara RK, Seeman P, Niznik HB, Civelli O. (1991).
Cloning of the gene for a human dopamine D4 receptor with high affinity for the
antipsychotic clozapine.Nature.350(6319):610-4.
Valette M, Poitou C, Le Beyec J, Bouillot JL, Clement K, Czernichow S. (2012). Melanocortin-4
receptor mutations and polymorphisms do not affect weight loss after bariatric surgery.
PLoS One. 7(11):e48221.
191
Vasan SK, Fall T, Neville MJ, Antonisamy B, Fall CH, Geethanjali F, et al. (2012). Associations of
variants in FTO and near MC4R with obesity traits in South Asian Indians. Obesity (Silver
Spring). 20(11):2268-77.
Vasseur F, Guerardel A, Barat-Houari M, Cottel D, Amouyel P, Froguel P, Helbecque N. (2007). Impact
of a CART promoter genetic variation on plasma lipid profile in a general population. Mol Genet
Metab 90: 199-204.
Vehof J, Risselada AJ, Al Hadithy AF, Burger H, Snieder H, Wilffert B, et al. (2011). Association of
genetic variants of the histamine H1 and muscarinic M3 receptors with BMI and HbA1c values
in patients on antipsychotic medication. Psychopharmacology (Berl).216(2):257-65.
Volavka J, Czobor P, Sheitman B, Lindenmayer JP, Citrome L, McEvoy JP et al. (2002). Clozapine,
olanzapine, risperidone, and haloperidol in the treatment of patients with chronic
schizophrenia and schizoaffective disorder. Am J Psychiatry. S159: 255–262.
Vong L, Ye C, Yang Z, Choi B, Chua S Jr, Lowell BB. (2011). Leptin action on GABAergic neurons
prevents obesity and reduces inhibitory tone to POMC neurons. Neuron. 71(1):142-54.
Wade, J., Milner, J., Krondl, M. (1981) Evidence for a physiological regulation of food selection and
nutrient intake in twins.Am J Clin Nutr. 34(2): 143-147
Walker RM, Christoforou A, Thomson PA, McGhee KA, Maclean A, Mühleisen TW,
Strohmaier J, Nieratschker V, Nöthen MM, Rietschel M, Cichon S, Morris SW, Jilani O,
Stclair D, Blackwood DH, Muir WJ, Porteous DJ, Evans KL. (2010).Association analysis
of Neuregulin 1 candidate regions in schizophrenia and bipolar disorder. Neurosci Lett.
30;478(1):9-13.
192
Walley AJ1, Blakemore AI, Froguel P. (2006). Genetics of obesity and the prediction of risk for health.
Hum Mol Genet. 15 Spec No 2:R124-30.
Wang Y, Lobstein T. (2006). Worldwide trends in childhood overweight and obesity.Int J Pediatr
Obes.1: 11–25.
Wang D, Ma J, Zhang S, Hinney A, Hebebrand J, et al. (2010). Association of the MC4R V103I
polymorphism with obesity: a Chinese case-control study and meta-analysis in 55,195
individuals. Obesity (Silver Spring) 18: 573–579.
Wehmeier PM, Gebhardt S, Schmidtke J, Remschmidt H, Hebebrand J, Theisen FM. (2005). Clozapine:
weight gain in a pair of monozygotic twins concordant for schizophrenia and mild mental
retardation. Psychiatry Res 133: 273-276.
Weston-Green K1, Huang XF, Deng C. (2012). Alterations to melanocortinergic, GABAergic and
cannabinoid neurotransmission associated with olanzapine-induced weight gain.PLoS One.
7(3):e33548.
Weston-Green K, Huang XF, Deng C. (2011).Olanzapine treatment and metabolic dysfunction: a dose
response study in female Sprague Dawley rats. Behav Brain Res. 217(2):337-46.
Weston-Green K1, Huang XF, Deng C. (2013). Second generation antipsychotic-induced type 2
diabetes: a role for the muscarinic M3 receptor.CNS Drugs. 27(12):1069-80.
Weston-Green K, Huang XF, Han M, Deng C. The effects of antipsychotics on the density of
193
cannabinoid receptors in the dorsal vagal complex of rats: implications for olanzapine-induced
weight gain. Int J Neuropsychopharmacol. 2008;11:827–835.
Widiker S, Karst S, Wagener A, Brockmann GA. (2010). High-fat diet leads to a decreased methylation
of the Mc4r gene in the obese BFMI and the lean B6 mouse lines.J Appl Genet. 51(2):193-7.
Willer CJ, Speliotes EK, Loos RJ, Li S, Lindgren CM, Heid IM, Berndt SI, Elliott AL, Jackson AU.
(2009). Six new loci associated with body mass index highlight a neuronal influence on body
weight regulation. Nat Genet. 41(1):25-34.
Wong AH, Van Tol HH.Schizophrenia: from phenomenology to neurobiology.(2003).
Neurosci Biobehav Rev. (3):269-306.
Williams NM, Preece A, Spurlock G, Norton N, Williams HJ, Zammit S, O'Donovan MC, Owen
MJ. (2003). Support for genetic variation in neuregulin 1 and susceptibility to
schizophrenia. Mol Psychiatry. 8(5):485-7.
Williams J, Spurlock G, McGuffin P, Mallet J, Nöthen MM, Gill M, Aschauer H, Nylander PO,
Wong AH, Van Tol HH. (2003).The dopamine D4 receptors and mechanisms of
antipsychotic atypicality. Prog Neuropsychopharmacol Biol Psychiatry.
27(7):1091-9. Review.
World Health, O. (1992). "The ICD-10 classification of mental and behavioural disorders:
Clinical descriptions and diagnostic guidelines." Geneva, Switzerland: World Health
Organisation.
194
Wu RR, Zhao JP, Jin H, Shao P, Fang MS, Guo XF, He YQ, Liu YJ, Chen JD, Li LH. (2008).Lifestyle
intervention and metformin for treatment of antipsychotic-induced weight gain: a randomized
controlled trial.JAMA. 299(2):185-93.
Xi B, Chandak GR, Shen Y, Wang Q, Zhou D. (2012). Association between common polymorphism near
the MC4R gene and obesity risk: a systematic review and meta-analysis. PLoS One.
7(9):e45731.
Xu AW, Kaelin CB, Morton GJ, Ogimoto K, Stanhope K, Graham J, et al. (2005). Effects of
hypothalamic neurodegeneration on energy balance. PLoS Biol. 3(12):e415.
Xu Y, Elmquist JK, Fukuda M. (2011).Central nervous control of energy and glucose balance: focus on
the central melanocortin system.Ann N Y Acad Sci. 1243:1-14.
Xu L, Janssen D, van der Knaap N, Roubos EW, Leshan RL, Myers MG Jr,et al. (2014).Integration of
stress
and leptin signaling by CART producing neurons in the rodent midbrain centrally projecting
Edinger-Westphal nucleus. Front Neuroanat. 3;8:8.
Yang CW, Li CI, Liu CS, Bau DT, Lin CH, Lin WY, et al. (2013). The joint effect of cigarette smoking
and polymorphisms on LRP5, LEPR, near MC4R and SH2B1 genes on metabolic syndrome
susceptibility in Taiwan. Mol Biol Rep. 40(1):525-33.
Yang JZ, Si TM, Ruan Y, Ling YS, Han YH, Wang XL, et al (2003).Association study of
neuregulin 1 gene with schizophrenia. Mol Psychiatry. 8(7):706-9.
Yang YK, Harmon CM. (2003).Recent developments in our understanding of melanocortin system in the
195
regulation of food intake.Obes Rev. 4(4):239-48.
Yeo GS, Farooqi IS, Aminian S, Halsall DJ, Stanhope RG, O'Rahilly S. (1998). A frameshift mutation in
MC4R associated with dominantly inherited human obesity. Nat Genet. 20(2):111-2.
Yilmaz Z, Davis C, Loxton NJ, Kaplan AS, Levitan RD, Carter JC, Kennedy JL. (2014). Association
between MC4R rs17782313 polymorphism and overeating behaviors. Int J Obes (Lond). (In
Press).
Yuan X, Yamada K, Ishiyama-Shigemoto S, Koyama W, Nonaka K. Identification of polymorphic loci in
the promoter region of the serotonin 5-HT2C receptor gene and their association with obesity and
type II diabetes. Diabetologia 2000; 43: 373–376.
Xu Y, Jones JE, Lauzon DA, Anderson JG, Balthasar N, Heisler LK, Zinn AR, Lowell BB, Elmquist JK.
(2010).A serotonin and melanocortin circuit mediates D-fenfluramine anorexia. J Neurosci.
3;30(44):14630-4.
Young EH, Wareham NJ, Farooqi S, Hinney A, Hebebrand J, et al. (2007). The V103I polymorphism of
the MC4R gene and obesity: population based studies and meta-analysis of 29 563 individuals.
Int J Obes (Lond) 31: 1437–1441.
Zhang ZJ, Yao ZJ, Mou XD, Chen JF, Zhu RX, Liu W, Zhang XR, Sun J, Hou G. (2003). Association of
2548G/A functional polymorphism in the promoter region of leptin gene with antipsychotic
agent-induced weight gain. Zhonghua Yi Xue Za Zhi. 83(24):2119-23.
Zai GC, Zai CC, Chowdhury NI, Tiwari AK, Souza RP, Lieberman JA, et al. (2012).The role of brain-
derived
196
neurotrophic factor (BDNF) gene variants in antipsychotic response and antipsychotic-induced
weight gain. Prog Neuropsychopharmacol Biol Psychiatry. 39(1):96-101.
Zegers D, Beckers S, Hendrickx R, Van Camp JK, Van Hoorenbeeck K, Desager KN.(2013). Endocrine.
44(2):386-90. Prevalence of rare MC3R variants in obese cases and lean controls. Obesity (Silver
Spring)19 (1), 152–159.
Zhao J, Bradfield JP, Zhang H, Sleiman PM, Kim CE, Glessner JT. (2011). Role of BMI-associated
loci identified in GWAS meta-analyses in the context of common childhood obesity in European
Americans.Obesity (Silver Spring). 19(12):2436-9.
Zhao X, Xi B, Shen Y, Wu L, Hou D, Cheng H, Mi J. (2014). An obesity genetic risk score is associated
with metabolic syndrome in Chinese children. Gene.535(2):299-302.
Zhang XY, Zhou DF, Wu GY, Cao LY, Tan YL, Haile CN, Li J, Lu L, (2008). BDNF levels and
genotype are associated with antipsychotic-induced weight gain in patients with chronic
schizophrenia. Neuropsychopharmacology. 33(9):2200-5.
Zheng F, Kim YJ, Chao PT, Bi S. (2013).Overexpression of neuropeptide Y in the dorsomedial
hypothalamus causes hyperphagia and obesity in rats. Obesity (Silver Spring).21(6):1086-92.
Ziotopoulou M, Erani DM, Hileman SM, Bjorbaek C, Mantzoros CS. (2000). Unlike leptin, ciliary
neurotrophic factor does not reverse the starvation-induced changes of serum corticosterone
and hypothalamic neuropeptide levels but induces expression of hypothalamic inhibitors of
leptin signaling. Diabetes 49: 1890-1896.
197
Zlatohlavek L, Vrablik M, Motykova E, Ceska R, Vasickova L, Dlouha D, Hubacek JA. (2013).FTO and
MC4R gene variants determine BMI changes in children after intensive lifestyle intervention.Clin
Biochem. 46(4-5):313-6.
Zobel DP, Andreasen CH, Grarup N, Eiberg H, Sørensen TI, Sandbaek A, (2009). Variants near MC4R
are associated with obesity and influence obesity-related quantitative traits in a population of
middle-aged people: studies of 14,940 Danes. Diabetes. 58(3):757-64.
198
