Genes, Brain and Behavior (2010) 9: 910–917 doi: 10.1111/j.1601-183X.2010.00631.x
Replicated association of the NR4A3 gene with
smoking behaviour in schizophrenia and in bipolar
disorder
G. Novak†,‡,§, C. C. Zai¶, M. Mirkhani§,
S. Shaikh¶, J. B. Vincent††, H. Meltzer‡‡,
J. A. Lieberman§§, J. Strauss¶,¶¶,
D. Le´vesque∗∗∗, J. L. Kennedy∗,¶
and B. Le Foll§,∗∗,†††
†Neuroscience Research Department, Centre for Addiction and
Mental Health, ‡Department of Pharmacology, University of
Toronto, §Translational Addiction Research Laboratory, Centre
for Addiction and Mental Health and University of Toronto,
¶Neurogenetics Section, Neuroscience Research Department,
Centre for Addiction and Mental Health, and ††Molecular
Neuropsychiatry and Development Laboratory of the Psychiatric
Neurogenetics Section, Neuroscience Research Department,
Centre for Addiction and Mental Health, Toronto, Ontario,
Canada, ‡‡Psychiatric Hospital, Vanderbilt University, Nashville,
TN, and §§Department of Psychiatry, Columbia University
College of Physicians and Surgeons and the New York State
Psychiatric Institute, New York, NY, USA, ¶¶Department of
Psychiatry, University of Toronto, Ontario, ***Faculty of
Pharmacy, University of Montreal, Montre´al, Quebec, and
†††Family and Community Medicine, Psychiatry, Pharmacology,
Institute of Medical Sciences, Toronto, Canada
*Corresponding author: Corresponding author for questions
regarding genetic analysis: J. L. Kennedy, CAMH, 250 College
Street R-30, Toronto, Ontario, Canada M5T 1R8. E-mail:
james kennedy@camh.net
∗∗Corresponding author: Dr B. Le Foll, CAMH, 33 Russell Street,
Toronto, Ontario, Canada M5S 2S1. E-mail: bernard lefoll@
camh.net
Schizophrenia and bipolar disorder are associated with
dopamine neurotransmission and show high comorbid-
ity with tobacco dependence. Recent evidence indicates
that the family of the NR4A orphan nuclear recep-
tors, which are expressed in dopamine neurons and
in dopaminoceptive brain areas, may play a role in
dopamine-mediated effects. We have, therefore, anal-
ysed the association of six single nucleotide poly-
morphisms (SNPs) within the three genes belonging
to the NR4A orphan nuclear receptor family, NR4A1
(rs2603751, rs2701124), NR4A2 (rs12803, rs834835) and
NR4A3 (rs1131339, rs1405209), with the degree of smok-
ing in a sample of 204 unrelated schizophrenia patients,
which included 126 smokers and 78 non-smokers. SNPs
within the NR4A3 gene (rs1131339 and rs1405209) were
significantly associated with heavy smoking in this
cohort, using a stepwise analysis of the escalated num-
ber of cigarettes smoked per day (P = 0.008 and 0.006,
respectively; satisfying the Nyholt significance thresh-
old of 0.009, an adjustment for multiple testing). We
then repeated the association analysis of the NR4A3
markers (rs1131339 and rs1405209) in a larger cohort
of 319 patients with bipolar disorder, which included
167 smokers and 152 non-smokers. We have replicated
the positive association with smoking of the NR4A3
SNP rs1131339 in this group (P = 0.04), providing an
important confirmation of the involvement of the NR4A3
gene in nicotine addiction in patients with mental health
disease, a population significantly at risk for nicotine
addiction.
Keywords: Association, bipolar disorder, dopamine, nicotine
addiction, NOR-1, NR4A3, Nur, schizophrenia, smoking, SNP
Received 15 February 2010, revised 28 May 2010 and
17 July 2010, accepted for publication 17 July 2010
Nicotine addiction is comorbid with a number of neuropsy-
chiatric disease states, in particular schizophrenia (Hughes
et al. 1986) and bipolar disorder (Diaz et al. 2009), contribut-
ing additional morbidity to an already susceptible population.
Because of its high heritability (Maes et al. 2004), it is pos-
sible that genetic factors affecting predisposition to nicotine
addiction can be identified, providing new targets for effec-
tive treatment, especially in at-risk groups.
The mesocorticolimbic dopamine system plays a key role
in the reinforcement for continued use of nicotine and other
stimulant drugs (Balfour et al. 2000; Silvestri et al. 2004).
This system is composed of neural pathways originating in
the ventral tegmental area (VTA), which project to targets
within the nucleus accumbens (NAc), the caudate putamen
of the ventral striatum (CPU) and the prefrontal cortex (PFC).
The effects of nicotine are mediated by dopamine release
within these target areas, which mediate specific aspects of
nicotine addiction. In particular, the NAc has been implicated
in drug seeking (Fuchs et al. 2004; Wise & Bozarth 1987),
while the PFC and the CPU have been shown to be involved
in mediating drug craving (Childress et al. 1999; Everitt et al.
1999; Maas et al. 1998).
Studies of changes in gene expression, induced by
dopamine receptor (DR) stimulation or antagonism, have
identified factors associated with dopaminergic signalling of
the individual mesocorticolimbic pathways. Among these
factors are members of the NR4A (Nur) family (reviewed by
Willson & Moore 2002), which showed altered expression in
key areas involved in addiction, including the substantia nigra
(SN), the VTA (Maheux et al. 2005), the NAc, as well as the
910 © 2010 The Authors
Genes, Brain and Behavior © 2010 Blackwell Publishing Ltd and International Behavioural and Neural Genetics Society

NR4A3 and smoking in schizophrenia and bipolar disorder
CPU (Werme et al. 2000a). In particular, NR4A3 (NOR1) was
shown to respond to inhibition of the dopaminergic system
(Werme et al. 2000b) and its expression is associated with
reinstatement of drug seeking (Koya et al. 2006), implicating
this gene in the dopaminergic response associated with
drug addiction. Taken together, these data suggest that the
factors belonging to the NR4A family may play an important
role in addiction-associated mesocorticolimbic signalling.
The NR4A (Nurr) factors are a family of orphan nuclear
receptors. The three members of this family, NR4A1 (Nur77),
NR4A2 (Nurr1) and NR4A3 (NOR-1) function as transcription
factors (TFs) and belong to a group of genes which mediate
fast responses to extracellular stimuli (Maruyama et al. 1998;
Sun et al. 2007), with quick termination thereafter (Pirih
et al. 2005). The NR4A TFs play an important function in
the development of dopamine neurons (Zetterstro¨m et al.
1997), as well as in the functioning of the dopaminoceptive
areas in the adult brain, including the NAc and the PFC
(Beaudry et al. 2000; Le´vesque & Rouillard 2007; Werme
et al. 2000a,b; Zetterstro¨m et al. 1996a,b, 1997).
In view of their role in modulation of the dopaminergic
response of the mesocortical system and, hence, the poten-
tial of polymorphisms within these genes to cause alterations
in this response, we have performed an association analysis
of six single nucleotide polymorphisms (SNPs) within the
NR4A genes with smoking behaviour (Table 1). The initial
analysis was performed in a cohort of 204 schizophrenia
patients. We selected the schizophrenia population for our
analysis of the possible association of smoking with polymor-
phisms in the NR4A genes for two reasons. First, as smoking
may be dependent on many different factors, smokers from
the general population would be expected to form a fairly
heterogeneous group. Therefore, we aimed to reduce the
number of contributing factors by focusing on a subgroup of
smokers. This subgroup, in which the dopaminergic system
is to some degree compromised (Lieberman et al. 1994; See-
man 2006), may be more sensitive to effects of further per-
turbations of this system, such as polymorphisms in genes
involved in modulating dopaminergic pathways in reponse to
nicotine (Balfour 2009; Balfour et al. 2000). By selecting such
a population for our analysis, we were hoping to detect risk
alleles in a smaller sample than one required for their detec-
tion in the general population. Second, as heavy smoking is
much more prevalent in patients with schizophrenia (Dalack
et al. 1998), as well as in patients with bipolar disorder
(Grant et al. 2004), we were able to focus on polymorphisms
associated with heavy smoking. Being able to analyse an
equivalent number of heavy smokers would require a much
larger group of individuals from the general population.
The analysis of an association of two markers within the
NR4A3 gene with smoking behaviour was then repeated in
a larger cohort of 319 patients with bipolar disorder.
Methods and materials
Subjects
Participants in the schizophrenia cohort were from a community-
based sample of 204 unrelated schizophrenia patients (139 males
and 65 females), males included 99 smokers and 40 non-smokers,
females included 27 smokers and 38 non-smokers. Males smoked
on average 20.5 cigarettes per day, females 15.7 cigarettes per day.
The average age of males was 43 and of females 46 years.
Samples in the bipolar disorder cohort were collected under
funding from GlaxoSmithKline, from a population of 500 unrelated
bipolar individuals and 500 unrelated healthy controls matched for
age, gender and ethnicity. Inclusion criteria for bipolar patients
were: (1) diagnosed using Diagnostic and Statistical Manual of
Mental Disorders revision IV (DSMIV) / International Statistical
Classification of Diseases and Related Health Problems published by
the World Health Organization version 10 (ICD 10) Bipolar Disorder
I II (BPAD I II); (2) 18 years old or over; (3) Caucasian, of northern
and western European origin. Exclusion criteria included: (1) use of
intravenous drugs; (2) evidence of mental retardation; (3) related
to an individual already in the study; (4) manias that only ever
occurred in relation to or as a result of alcohol or substance abuse or
dependence and/or medical illness; (5) having any manias as a result
of a non-psychotropic substance. After exclusion of former smokers
and of individuals lacking smoking or demographic information, the
bipolar disorder group consisted of 319 individuals, including 152
non-smokers (94 female and 58 males) and 167 smokers (104
females and 63 males). The average age of smokers was 46.6
and of non-smokers 44.8 years. Subjects were prevalently of self-
reported Caucasian origin, which was reflected in allele frequencies
being identical to frequencies published in the dbSNP database
for Caucasian population (NCBI Single Nucleotide Polymorphism
database http://www.ncbi.nlm.nih.gov/SNP).
In order to ascertain that a stable level of cigarette consumption
has been established, proband smokers were required to be 18 years
or older, have smoked tobacco for at least the last 5 years, and had
consumed an average of 5 cigarettes per day or more for the last
12 months (as per Beuten et al. 2007).
Both groups were recruited through the Centre for Addiction and
Mental Health (CAMH) in Toronto, Canada. The procedures were
approved and monitored by CAMH Research Ethics Board. Written
informed consent was obtained from all participants.
For the schizophrenia cohort, the diagnosis was established by
best-estimate procedures and the clinical information was collected
through a questionnaire addressing smoking status and quantity
smoked.
Table 1: SNP polymorphisms analysed within the NR4A genes
Gene SNP Chromosome Position Location Substitution Substitution
NR4A1 rs2603751 12 14596588 exon 8 3′ UTR G/A
rs2701124 12 14591463 exon 3 Synonymous T/C
NR4A2 rs12803 2 7391343 exon 8 3′ UTR T/G
rs834835 2 7397563 intron 1 Intronic T/C
NR4A3 rs1131339 9 9948507 exon 7 3′ UTR A/G
rs1405209 9 9906750 intron 1 Intronic A/G
For each SNP, the ancestral allele is listed first. Contig reference position is listed according to a reference contig position in the dbSNP
database (NCBI Single Nucleotide Polymorphism database http://www.ncbi.nlm.nih.gov/SNP).
Genes, Brain and Behavior (2010) 9: 910–917 911