Risk Gene Variants for Nicotine Dependence in the
CHRNA5–CHRNA3–CHRNB4 Cluster Are Associated
With Cognitive Performance
Georg Winterer,1,2* Kirstin Mittelstrass,3Ina Giegling,4Claudia Lamina,3,5Christoph Fehr,6
Hermann Brenner,7Lutz P. Breitling,7Barbara Nitz,3Elke Raum,7Heiko M€ uller,7J€ urgen Gallinat,8
Andreas Gal,9Katharina Heim,10Holger Prokisch,10Thomas Meitinger,10Annette M. Hartmann,11
Hans-J€ urgen M€ oller,11Christian Gieger,3H-Erich Wichmann,3Thomas Illig,3Norbert Dahmen,6
and Dan Rujescu11
1Cologne Center for Genomics (CCG), University of Cologne, Cologne, Germany
2Institute of Neurosciences & Medicine, Helmholtz Center Juelich, Juelich, Germany
3Institute of Epidemiology, Helmholtz Zentrum Munich, German Research Center for Environmental Health, Neuherberg, Germany
4Division of Molecular and Clinical Neurobiology, Department of Psychiatry, Ludwig-Maximilians-University, Munich, Germany
5Division of Genetic Epidemiology, Department of Medical Genetics, Molecular and Clinical Pharmacology, Innsbruck Medical University,
6Department of Psychiatry, University of Mainz, Mainz, Germany
7Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (Helmholtz), Heidelberg, Germany
8Department of Psychiatry, Charit? e University Medicine Berlin, Campus Mitte, Berlin, Germany
9Institute for Human Genetics, University of Hamburg Medical Center Eppendorf, Hamburg, Germany
10Institute of Human Genetics, Technical University of Munich and Helmholtz Center Munich, Munich, Germany
11Institute of Medical Informatics, Biometry and Epidemiology, Ludwig-Maximilians-University, Munich, Germany
Received 12 April 2010; Accepted 4 August 2010
Recent studies strongly support an association of the nicotinic
CHRNB4 with nicotine dependence (ND). However, the precise
genotype–phenotype relationship is still unknown. Clinical and
epidemiological data on smoking behavior raise the possibility
that the relevant gene variants may indirectly contribute to
the development of ND by affecting cognitive performance
in some smokers who consume nicotine for reasons of
‘‘co nition enhancement.’’ Here, we tested seven single nucleo-
Additional Supporting Information may be found in the online version of
Grant sponsor: BMBF; Grant number: 01EB0113; Grant sponsor: DFG;
Grant number: Wi2997/2-2.
The authors declare that they have no competing interests.
biobanking KORA: HEW, CG, TI, BN; biobanking nicotine dependence
replication studies: DR, GW, ND, CF, JG, IG, HJM, HB, (NCOOP), HB,
DR, ER and LB (ESTHER), biobanking ‘‘cognition study’’: DR, IG,
genotyping: TI, KM; genotyping in the replication studies: DR, AR,
AMH, TI, KM, statistical analysis: CG, CL; gene expression studies: HP,
KH, TM; cognitive measurements: IG, DR, AMH, HJM; manuscript
writing: GW, DR, AM, IG, KM, CL, ND, TI, HB, LB.
Georg Winterer and Kirstin Mittelstraa ˚ contributed equally to this work.
Georg Winterer, M.D., Ph.D., Cologne Center for Genomics (CCG),
University of Cologne, Weyertal 115b, 50931 Cologne, Germany.
Published online 30 September 2010 in Wiley Online Library
How to Cite this Article:
C, Fehr C, Brenner H, Breitling LP, Nitz B,
Raum E, M€ uller H, Gallinat J, Gal A, Heim K,
Prokisch H, Meitinger T, Hartmann AM,
Dahmen N, Rujescu D. 2010. Risk Gene
Variants for Nicotine Dependence in the
CHRNA5–CHRNA3–CHRNB4 Cluster Are
Associated With Cognitive Performance.
Am J Med Genet Part B 153B:1448–1458.
? 2010 Wiley-Liss, Inc.
tide polymorphisms (SNPs) rs684513, rs637137, rs16969968,
CHRNA5–CHRNA3–CHRNB4 gene cluster for association with
expected, we found all SNPs being associated with ND in three
independent cohorts (KORA, NCOOP, ESTHER) comprising
5,561 individuals. In an overlapping sample of 2,186 subjects
we found three SNPs (rs16969968, rs1051730, rs3743078) being
associated with cognitive domains from the Wechsler-Adult-
Intelligence Scale (WAIS-R)—most notably in the performance
subtest ‘‘object assembly’’ and the verbal subtest ‘‘similarities.’’
In a refined analysis of a subsample of 485 subjects, two of these
task performance/Continuous Performance Test. Furthermore,
two CHRNA5 risk alleles (rs684513, rs637137) were associated
with CHRNA5 mRNA expression levels in whole blood in a
subgroup of 190 subjects. We here report for the first time an
cognition possibly mediating in part risk for developing ND.
The observed phenotype–genotype associations may depend on
altered levels of gene expression. ? 2010 Wiley-Liss, Inc.
Key words: CHRNA5–CHRNA3–CHRNB4; single nucleotide
polymorphisms; gene expression; cognition; n-back task; nico-
Tobacco smoking is the leading preventable cause of death and
Organization (WHO) estimates that 150million people will die
from their tobacco use in the next 25 years. In Established Market
Economies, nicotine addiction accounts for 12.2% of the disability
as measured by DALYs (The World Health report 2002, http://
www.who.int). Most smokers consume tobacco daily and are
physically dependent on nicotine, the primary addictive compo-
nent of cigarette smoke [Benowitz, 2008].
Twin, family segregation and adoption studies showed that
genetic factors strongly contribute to substance abusing behavior
including nicotine dependence (ND) [Lessov-Schlaggar et al.,
2008]. It is currently estimated that the additive genetic effects
account for 50–70% of the variance in liability to smoking related
because of the likely heterogeneity of ND comparable to other
consideration. Reviewing the literature, Sullivan and Kendler
 pointed out, for instance, that the genetic factors that
predispose to smoking initiation (SI) appear to overlap substan-
tiallybutnotcompletely withthoseforNDorsmoking persistence
(SP). Others argued that some heritable personality traits such as
reward dependence and impulsivity may indirectly predispose to
et al., 2009].
There are several reasons also to consider individual differences
in cognitive performance as a factor mediating ND. On one hand,
animal experiments and investigations in humans both suggested
that nicotine is a powerful enhancer of cognitive capabilities—
particularly in smokers [Ernst et al., 2001a,b; Kumari et al., 2003;
Newhouse et al., 2004; Thiel et al., 2005; Stolerman et al., 2009].
Nicotine may thus serve as a positive reinforcer by facilitating
cognitive performance [Rezvani and Levin, 2001; Kumari et al.,
2003]. On the other hand, cognitive capabilities and intelligence,
strong genetic control [Winterer and Goldman, 2003]. Thus,
genetic and environmental influences were shown to impact dif-
ferentially on cognition over time, with shared environmental
influences predominating early in life, but dissipating to near zero
by adulthood. By then heritability of intelligence is well above 0.5
and Goldman, 2003; Deary et al., 2009]. Accordingly, it is conceiv-
associations with ND) is explained by a mechanism where some
nicotine-dependent subjects initiate and sustain ND to improve
their cognitive capabilities—in particular in an environmental
context of high cognitive demand. Evidence for this notion origi-
nally comes from clinical studies. For instance, it has been argued
that some clinical populations with cognitive deficits, for example,
patients with schizophrenia or attention deficit disorder (ADD),
may smoke or initiate smoking for reasons of ‘‘self-medication’’
2008]. In some patients, this risk factor may play out already at a
a relationship between the development of ND and cognitive
capabilities in clinical populations also appears to apply to the
of 1,265 children Fergusson et al.  found a highly significant
correlation between intelligence quotient (IQ) during childhood
The correlation remained significant after adjustment for child-
prevalence may have several reasons such as people with low IQ
in social networks where smoking is more concentrated. However,
the observed association was much stronger than the relationship
correlation was virtually absent—perhaps because alcohol and
most illicit drugs do not have cognition enhancing properties.
Further support for a relationship between smoking initiation and
intelligence or cognitive performance comes from a study of Yakir
et al.  who investigated 328 undergraduate females aged
they concluded that a neurocognitive profile characterized by
be one of the factors that predispose young women who initiate
support for the relationship between smoking and intelligence. In
strongest predictors for smoking initiation [Young and Rogers,
basis of their long-term follow-up investigation of a large Scottish
cohort that the observed inverse relationship between intelligence
and smoking explains in part the association between high child-
hood IQ scores and reduced rates of total mortality in adulthood.
Whether the inverse relationship between IQ and mortality also
WINTERER ET AL.
refers to lung cancer is currently debated [Deary et al., 2003;
Hemmingsson et al., 2006; Batty et al., 2007b].
Recently, considerable evidence has accumulated suggesting the
involvement of the nicotinic acetylcholine receptor CHRNA5–
CHRNA3–CHRNB4 gene cluster on chromosome 15q24 in ND
and smoking severity. Direct evidence in this regard both comes
from candidate gene investigations [Saccone et al., 2007; Bierut
2007; Berrettini et al., 2008; Thorgeirsson et al., 2008, 2010; Liu
et al., 2010; The Tobacco and Genetics Consortium, 2010]. These
studies have been remarkably consistent so far, although recent
locus with lung cancer which were not found to be necessarily
et al., 2008].
The precise relationship between ND and variants on the
CHRNA5–CHRNA3–CHRNB4 gene cluster still needs to be eluci-
dated. Two previous studies reported that common susceptibility
and protective haplotypes are associated with SI, that is, with ND
severity insubjectswhobegandailysmoking atorbeforetheageof
16, an exposure period that results in a more severe form of adult
ND [Schlaepfer et al., 2008; Weiss et al., 2008]. However, Thor-
geirsson et al.  did not find genomewide association with SI.
subjects no association between smoking cessation success and
seven SNPs from the 15q24 locus [Breitling et al., 2009]. Together,
these studies may suggest that for some yet unknown reasons the
 provides some preliminary evidence that cognitive capabil-
ities are perhaps relevant in this regard. They investigated a study
sample consisting of 100 female college students, current or past
smokers, and 144 who had never smoked. All performed a com-
puterized neurocognitive test battery and were genotyped for 39
the gene cluster on chromosome 15q24. Here, significant associ-
ations were found with omission errors on the continuous perfor-
mance task (CPT) which measures sustained attention and
vigilance, with immediate face recognition and with the STROOP
test—a measure of selective attention, cognitive flexibility, and
In the present explorative study, we primarily addressed the
question whether genetic variations in the CHRNA5–CHR-
CHRNA3–CHRNB4 cluster, which have been previously implicat-
ed in ND, are associated with cognitive performance. Besides, we
also sought to replicate these earlier association findings with ND
and whether related genotype-dependent mRNA expression
changes in a recent post mortem study [Wang et al., 2009] are also
found in whole blood.
MATERIALS AND METHODS
Three independent cohorts KORA, NCOOP, and ESTHER com-
prising 5,561 subjects (¼combined sample) were investigated for
genetic association with ND (Supplemental Table I). A detailed
sample description (Note: the KORA sample was part of the
recently published genome-wide association meta-analyses [Liu
smokers (>20cigarettes/day). Therefore, the ESTHER sample was
also used for testing genetic association with SI. In the NCOOP
sample (Munich sample), an overlapping sample with data on
cognitive ability as measured with WAIS-R [Tewes, 1991] of 2,186
subjects wasavailable (seebelow) whichallowedtestingforgenetic
association with cognitive phenotypes. Out of this sample, addi-
tional refined neuropsychological analyses were conducted in a
subsample of 485 subjects and tested for association. Association
with gene expression was investigated in a KORA F3 subsample of
190 subjects (Gene expression sample) (see Supplemental sample
studies have been approved by the local ethics committees. Across
was based on the availability of information. ‘‘Regular-smokers’’
were defined as persons smoking at least 1cigarette/day. Persons
smoking ?20cigarettes/day were defined as ‘‘heavy-smokers.’’
Never-smoking controls have smoked less than 100 cigarettes in
The cognition sample (Table I) was specifically included in this
study as part of an in-depth investigation to address the question
whether the CHRNA5–CHRNA3–CHRNB4 locus is associated
with cognitive performance. 2,186 unrelated healthy volunteers of
TABLE I. Description of the Cognition Sample
Males, N (%)
Females, N (%)
Mean age in years (range)
Never-smokers, N (%)
Ex-smokers, N (%)
Current smokers, N (%)
Mean IQ (SD)—total
Mean IQ (SD)—never-smokers
Mean IQ (SD)—ex-smokers
Mean IQ (SD)—current smokers
Mean verbal IQ (SD)—total
Mean verbal IQ (SD)—never-smokers
Mean verbal IQ (SD)—ex-smokers
Mean verbal IQ (SD)—current smokers
Mean performance IQ (SD)—total
Mean performance IQ (SD)—never-smokers
Mean performance IQ (SD)—ex-smokers
Mean performance IQ (SD)—current smokers
No significant (P<0.05) differences of IQ-scales between groups (also see text).
1450 AMERICAN JOURNAL OF MEDICAL GENETICS PART B
German descent (i.e., both parents were German) were randomly
selected from the general population of Munich, Germany, and
contacted by mail. Of all subjects, 52.0% were females and 48.0%
were males with a mean age of 52.2 (SD 16.0) years (range 19–79).
To exclude subjects with neuropsychiatric disorders or subjects
who had first-degree relatives with neuropsychiatric disorders,
further screenings were conducted before the volunteers were
enrolled in the study. The sample overlaps with the sample of the
Ludwig-Maximilians-University (LMU) from the NCOOP study
(seesupplementary sample description). Inbothoverlappingsam-
ples, all data have been collected in anidentical way. The cognition
sample comprises 1,075 (49%) never-smokers, 701 (32%) ex-
smokers, and 410 (19%) current smokers including 99 heavy
smokers with a consumption of ?20cigarettes/day. The level of
education in the study subjects was estimated on the basis of their
3: high educational level, i.e., general qualification for university
entrance (41.5%)]. All subjects underwent full cognitive measure-
ment using the German version of the WAIS-R which was admin-
istered and scored according to the standardized procedures
outlined in the manual [Tewes, 1991; Rujescu et al., 2002, 2003].
It allowed calculating Full Scale IQ, and has the possibility to
calculate Verbal IQ (consisting of the subtests Information, Digit
Span, Vocabulary, Arithmetics, Comprehension, Similarities) and
Performance IQ (consisting of the subtests Picture Completion,
Picture Arrangement, Block Design, Object Assembly, Digit Sym-
A subsample of 485 subjects, who were willing to participate in
logical testing (see supplemental test description). Of all subjects,
48.5% were females and 51.5% were males with a mean age of 49.0
(SD 13.7) years (range 20–72). This sample included 230 (47.4%)
never-smokers, 160 (33.0%) ex-smokers, and 95 (19.6%) current
smokers including 16 heavy smokers.
For details see supplemental genotyping description. The same
seven SNPs were genotyped across all samples of the study.
The selection of SNPs from CHRNA5 (rs684513, rs637137,
rs16969968), CHRNA3 (rs578776, rs1051730, rs3743078) and
CHRNB4 (rs3813567) was based on the study of Saccone et al.
. When our study was planned, these SNPs were reported to
be associated with ND.
Gene Expression Analysis
Whole blood transcript levels of the CHRNA5, CHRNA3, and
CHRNB4 genes were measured with whole genome expression
profiles available. For details see supplemental gene expression
For the combined sample, haploview 3.2 (http://www.broad.
mit.edu/mpg/haploview/) was used to generate a linkage disequi-
librium (LD) map (r2and D0), to test for departure from
Hardy–Weinberg equilibrium (HWE) at a significance level of
P<0.05 and to test for differences between cases and controls for
alleles and genotypes in single SNPs as well as in haplotypes.
Furthermore, tests for associations using multimarker haplotypes
Global and single haplotype significance and odds ratios were
calculated. For association analyses of single SNPs with age at SI
respectively, both adjusted for gender. For association analyses of
age, gender, education and smoking status (never-smokers,
ex-smokers, current smokers with <20cigarettes/day vs. heavy
smokers with ?20cigarettes/day) were added as covariates into
the linear regression analyses. To compare quantitative traits
between the haplotypes, the R software package ‘‘haplo.score’’ was
used integrating the covariates gender, age, education, smoking
the main scores of 8 further tests (see above) were entered into a
relation between these measurements. Five factors could be
extracted (see Supplemental Table II) which explained a total
variance of 67.9%. These factors were tested for association
with the SNPs which were associated with the WAIS-R tests using
subset of the KORA sample six out of the seven SNPs, that is,
from CHRNA5 and CHRNA3 (rs684513, rs637137, rs16969968,
rs578776, rs1051730, rs3743078) were included. Linear regression
analyses were conducted with age, gender, and smoking status as
covariates as appropriate. Additional statistical analyses for group
comparisons of quantitative measures (IQ, gene expression) were
conducted with an analysis of variance (ANOVA) or t-tests as
A significant association with heavy smoking (range of P values:
0.0046–0.000075) for all seven tested SNPs (three in CHRNA5;
three in CHRNA3, one in CHRNB4) was observed (Table II). Also
regular smoking status was associated with the same alleles, but
the analysis did not give additional information (not depicted).
The following alleles showed a higher risk for smoking
rs578776–rs1051730–rs3743078–rs3813567), with consistent and
homogeneous odds ratios (OR) across all studies, as also shown by
the Q-test of heterogeneity (all P values >0.54).
Furthermore, LD-analysis showed a block spanning over gen-
otyped SNPs in the CHRNA5–CHRNA3 region, with D0>0.9 (see
Supplemental Fig. 1). For SNPs with similar minor allele frequen-
cies within this region, the correlation coefficient r2is also high
(between 0.59 and 0.99). Only SNP rs3813567 (CHRNB4) falls out
WINTERER ET AL.
Results from haplotype analyses support the single SNP
analyses, but do not give additional information on certain
high risk or protective allele-combinations (see Supplemental
Table III). The risk haplotype [C–T–A–C–T–C–T (rs684513–
this haplotype with the 7-SNP-haplotype including all protective
alleles [combined sample: P¼0.0000248; haplotype frequencies:
KORA: 0.13, P¼0.0107; NCOOP: 0.12, P¼0.0153; ESTHER:
0.122, P¼0.0166)] in the comparison of heavy-smokers with
Additional analyses in the ESTHER sample of heavy smokers
(age of SI: 18.8, SD 4.2, range 7–50 years) revealed that none of the
seven CHRNA5–CHRNA3–CHRNB4 SNPs (or haplotypes) was
This negative finding was obtained both when age of SI was
dichotomized (<16 vs. ?16 years) or when age of SI was treated
as a continuous variable or when a recessive model rather than an
additive model was used instead.
In the subsequent analysis of the cognition sample, we found no
statistically significant effects of smoking status but only statistical
trend findings with slightly lower IQ values in current smokers on
the three major WAIS-R scales: IQ (P¼0.085), Verbal IQ
Thus, the subsequent linear regression analyses to assess asssoci-
ation with intelligence were conducted with the covariates age,
TABLE II. Logistic Regression Results on the Outcome ‘‘Heavy-Smokers’’
Minor alleleMajor alleleMAFP (hetero-geneity)a
0.86 (0.77, 1.06)
0.82 (0.67, 0.99)
0.88 (0.76, 1.04)
0.86 (0.77, 0.95)
0.83 (0.68, 1.01)
0.78 (0.65, 0.95)
0.83 (0.72, 0.97)
0.82 (0.74, 0.91)
1.18 (1.00, 1.40)
1.26 (1.08, 1.48)
1.14 (1.00, 1.28)
1.18 (1.09, 1.29)
0.80 (0.67, 0.96)
0.88 (0.74, 1.06)
0.86 (0.75, 1.00)
0.85 (0.77, 0.94)
1.18 (1.00, 1.40)
1.26 (1.08, 1.47)
1.16 (1.01, 1.30)
1.19 (1.09, 1.29)
0.83 (0.68, 1.01)
0.81 (0.67, 0.98)
0.84 (0.73, 0.98)
0.83 (0.75, 0.92)
0.77 (0.63, 0.94)
0.80 (0.66, 0.98)
0.88 (0.75, 1.03)
0.83 (0.75, 0.92)
Adjusted for age and sex using an additive genetic model. Combined odds ratios (OR) and corresponding P values from fixed effect models are also shown.aQ-Test.
1452 AMERICAN JOURNAL OF MEDICAL GENETICS PART B
gender, education, and smoking status. We first calculated the
SNPs with various WAIS-R (sub-)scales (Table III). After correc-
tion for multiple testing, we saw significant associations of the
verbal subscale ‘‘Similarities’’ with the two SNPs rs16969968
(b¼?0.510,CI¼?0.845 to ?0.175)and rs1051730 (b¼?0.501,
?0.501, CI¼?0.836 to ?0.166) and also of the performance
subscale ‘‘Object Assembly’’ with rs3743078 (b¼0.797, CI¼
1.260–0.333). Haplotype analyses pointed in the same direction
but provided no further information (overall P-value >0.05). The
directionality of the observed associations of the three SNPs with
IQ-subscales matches in part the opposite associations with ND.
the similarities subtest. However, for rs3743078 the risk allele for
ND is associated with better perfromance in the object assembly
test. Interaction analyses between smoking status and cognition
were not conducted due to insufficient sample size of (heavy)
These three significant SNPs were further tested in an extended
in a subsample of 485 subjects. Five factors obtained by a principle
Interestingly, two SNPs (rs1051730 and rs16969968) were associ-
ated with factor 2 (Table IV) which primarily includes n-back task
measures and also the D0measure of the Continuous Perfomance
significant correlation was observed between factor 2 and the IQ-
Finally, we tested the question whether any of the seven tested
SNPsareassociatedwith geneexpression.Expression of CHRNA5,
CHRNA3, and CHRNB4 transcripts did not significantly differ
we found two SNPs from CHRNA5 (rs684513, rs637137) being
associated with CHRNA5 transcript expression levels in whole
VII, Supplemental Fig. 2) with both SNPs being in strong LD
(Supplemental Fig. 1). Inclusion of smoking status as covariate in
the linear regression analyses (in addition to gender and age) or
choosing a dominant instead of an additive model hardly changed
the results. Associations were no longer significant for SNPs
(rs684513,rs637137) when a recessive model was used. There were
no significant differences in CHRNA3, CHRNA5, and CHRNB4
expression levels for the three CHRNA3 SNPs.
strong evidence that SNPs (rs684513, rs637137, rs16969968,
rs578776, rs1051730, rs3743078, rs3813567) in the 15q24 region
are associated with ND in three independent cohorts comprising
a recent genomewide meta-analysis [Liu et al., 2010; Thorgeirsson
with ND in all cohorts including same allelic directionality and
similar ORs. Furthermore, the associated SNPs showed the same
allelic directionality as SNPs or their proxies in the above-
mentioned studies further supporting these findings. However,
contrary to our expectations from earlier candidate gene studies
[Schlaepfer et al., 2008; Weiss et al., 2008] but in line with a recent
TABLE III. P Values of the Linear Regression Analysis of the Seven SNPs on WAIS-R Subscales Using a Dominant Model
in n¼2,186 Subjects
Digit symbol coding
b estimates for significant SNPs are given in text.aSignificantafter correction for multiple testing (Bonferroni), that is, based on11subscales andone LD block because ofstrongLD (see Supplemental
WINTERER ET AL.
GWA meta-analysis of Thorgeirsson et al. , we obtained no
evidence for an association with age of smoking initiation. Sample
heterogeneity may account for these different findings and our
study differed from previous work [Schlaepfer et al., 2008] regard-
during which SI occurred. In fact, the retrospective assessment of
age of smoking initiation in later adulthood may be taken into
account as a potential source of error—in particular because
smoking is initiated mostly within a narrow age range between
14–17 years [Khuder et al., 2008]. Alternatively, one might also
consider the possibility that SI is only indirectly—and therefore
loosely—associated with variants from the 15q24 region being
mediated through an association with cognitive performance.
During the past decade, several genes have been implicated in
has been related to dopaminergic and cholinergic neurotransmis-
sion including COMT (catechol-O-methyltransferase), DRD2
(dopamine receptor D2), the two muscarinic receptor genes
CHRM1 and CHRM2 and the nicotinic receptor gene CHRNA4
[e.g., Egan et al., 2001; Comings et al., 2003; Gallinat et al., 2003;
Winterer and Goldman, 2003; Espeseth et al., 2006; Gosso et al.,
to gene cluster involved in cholinergic, that is, nicotinic neuro-
transmission, supporting associations between SNPs from the
CHRNA5–CHRNA3–CHRNB4 locus and several IQ (sub-)-
cales—mostnotably with theverbalIQ subscale ‘‘Similarities’’and
theperformance IQsubscale ‘‘ObjectAssembly.’’ Accordingly,our
who reported associations of CHRNA5 variants (rs684513,
rs601079, rs680244) with performance on various cognitive tests
(Stroop, Immediate face Recognition, Continuous Performance
and a pear alike?’’). Measuring the ability ‘‘Object Assembly’’
requires a subject to put a set of pieces of puzzle together to form
those alleles that are associated with lower cognitive performance
values areassociated with increased risk for ND. Furthermore, two
of these three significant SNPs showed an association with a
Principal Component Factor 2 which primarily included n-back
task measures and the D0(Sensitivity) measure of the Continuous
Performance Test (see supplemental Table II). Carriers of the ND
risk allele of the SNPs rs16969968 and rs1051730 showed lower
performance on this factor. The n-back task engages the working
memory and attention system in maintaining and updating infor-
has been shown to physiologically activate the working memory
TABLE IV. P-Values of the Linear Regression Analysis of the
Three Significant SNPs on Further Cognitive Domains Using a
Dominant Model in n¼485 Subjects
Wechsler memory subscales and D2 test, negative factor loadings on Trail Making Test A and B.
Factor 2: negative factor loadings on n-back tasks (reaction time) and positive loadings on n-
back tasks (correct responses) [we used the n-back task with the numbers 2-4-6-8] and
Continuous Performance Test: Sensitivity D0(see supplementary Table II). Factor 3: Category
Fluency and Word Fluency. Factor 4: positive factor loadings on WCST (perseverative and non-
perseverative errors) and negative loading on WCST (categories completed). Factor 5: negative
factor loadings on Verbal Learning Memory Test (delay and interference), positive loading on
Verbal Learning Memory Test (Immediate). For details see Supplemental Table II.
FIG. 1. Boxplots of Principal Component Analysis factor scores of
Factor 2 ‘‘n-back task performance/Continuous Performance
Test’’ for CHRNA5 SNP rs16969968 (top) and CHRNA3 SNP
rs1051730 (bottom) in n¼485 subjects. Shown are the
median, the 25% and 75% quantiles, ?1.58?interquartile
1454 AMERICAN JOURNAL OF MEDICAL GENETICS PART B
and the IQ-subscale ‘‘Similarities’’ are not significantly correlated
in our sample, the two SNPs rs16969968 and rs1051730 may affect
both cognitive domains independently.
Overall, our findings appear to be compatible with the notion
that genetic variations in the CHRNA5–CHRNA3–CHRNB4 locus
may indirectly increase a subject’s liability to ND for reasons of
experimental studies [Ernst et al., 2001a,b; Kumari et al., 2003;
might be especially the case in heavy smokers who demonstrated
lower cognitive performance in our study which is in accordance
with literature [Fried et al., 2006]. However, the number of heavy
smoking individuals in our studywastoo small to formally test the
hypothesis of an interaction between ND and cognitive perfor-
mance. In case such aninteraction would exist, itneedstobe taken
precede ND but also be a consequence of chronic smoking. Thus,
whereas the acute nicotine effects may serve the purpose of cogni-
tion enhancement, chronic smoking may also have detrimental,
that is, toxic, effects on cognitive performance and impaired brain
function may only reverse after long-term abstinence [Neuhaus
et al., 2006; Ferrea and Winterer, 2009].
We found two SNPs within the CHRNA5 gene (rs684513,
rs637137) being associated with transcript expression levels. How-
ever, these two SNPs were not among those which were—after
correction for multiple testing—significantly associated with cog-
necessarily reflect expressionpatternsin the brain,it is noteworthy
that a recent study [Wang et al., 2009] in post mortem prefrontal
cortex described associations of the seven SNPs from the present
study with transcript expression. Accordingly, this would suggest
that genotype dependent expression of CHRNA5 transcripts in
may be ‘‘false negatives’’ (also see supplemental discussion for
CHRNA5–CHRNA3–CHRNB4 receptor distribution in brain).
Our study does not indicate, however, which specific gene or
which genetic variations are functionally responsible for the ob-
served associations with ND and cognitive performance. Even so,
rs16969968 stands out in several ways. First, it is an attractive
[D] to asparagine [N]) in the a-5 nicotinic acetylcholine receptor
subunit (CHRNA5). In addition, it has been recently shown that
this particular risk allele is decreasing twofold the response to a
nicotine agonist [Bierut et al., 2008]. Interestingly, rs1051730,
which was also associated in several GWAS with ND rsp. with
cognition in the present study, and the functional SNP rs16969968
are not only in full LD but show also high R2¼.99 [Saccone et al.,
2007; Bierut et al., 2008]. This suggests that this functional SNP
could play an important role given that further studies show
evidence for association of this SNP with ND (e.g., Sherva et al.,
2008; Weiss et al., 2008). The potential importance of these two
SNPs is now further highlighted by our finding of consistent
associations across several performance and verbal WAIS IQ-sub-
an analogous way, both SNPs are also associated with the working
credibility to the finding. Moreover, both SNPs show the same
directionality of association with lower performance scores and a
was seen for the subscale ‘‘Object Assembly’’ and rs3743078 which
could indicate that in this particular case the genetic risk for lower
cognitive performance may not mediate risk for ND. If replicated,
this could suggest that the relationship between ND, cognitive
performance and genetic variations in the CHRNA5–CHR-
CHRNA3–CHRNB4 cluster is more complex than originally
thought, that is, that at least two distinct genotype–phenotype
mechanisms are involved [Wang et al., 2009].
In summary, we here provide for the first time evidence that
CHRNA5–CHRNA3–CHRNB4 gene cluster variants—and in par-
ticular the putative functional CHRNA5 variant rs16969968—
could be associated with cognitive performance and that these
associations may indirectly mediate risk for nicotine dependence.
sufficiently powered study population which was specifically col-
lected to study the interaction of smoking-related behavior and
cognitive performance [Mobascher et al., 2010; Brinkmeyer et al.,
submitted; Lindenberg et al., submitted].
Foundation (DFG) national priority program SPP1226 (http://
ular Effect in the CNS’’ (Br1704/11-1, Da370/5-1, Ga804/1-1,
Ru744/4-1, Wi1316/6-1). Our research was also supported within
the Munich Center of Health Sciences (MC Health) as part of
LMUinnovativ. The expression study was funded by the German
Ministry of Education and Research (BMBF) within the context of
SysMBo. KORA was supported by the German Federal Ministry of
Education and Research (BMBF), the State of Bavaria and the
National Genome Research Network (NGFN). NCOOP was sup-
ported in parts by the University of Mainz, BMBF (#01EB0113),
DFG (Wi2997/2-2), and further BMBF funding is gratefully ac-
knowledged (Project Berlin Neuroimaging Center, #01G00208).
ESTHER was supported by the Baden W€ urttemberg Ministry of
Research, Science and Arts. Genotyping was carried out at the
Genome Analysis Center (GAC) of the Helmholtz Zentrum
M€ unchen (KORA, ESTHER) and at the Genetics Research Centre
line Germany (Munich/Germany, NCOOP). The IRIS study was
within the context of the Baden-W€ urttemberg Research Network
on Addiction, project 01EB0113. The ESTHER baseline examina-
tion was funded by the Baden W€ urttemberg Ministry of Research,
Science and Arts.
Agrawal A, Lynskey MT. 2008. Are there genetic influences on addiction:
Evidence from family, adoption and twin studies. Addiction 103:
Amos CI, Wu X, Broderick P, Gorlov IP, Gu J, Eisen T, Dong Q, Zhang Q,
Gu X, Vijayakrishnan J, Sullivan K, Matakidou A, Wang Y, Mills G,
WINTERER ET AL.
Doheny K, Tsai YY, Chen WV, Shete S, Spitz MR, Houlston RS. 2008.
Genome-wide association scan of tag SNPs identifies a susceptibility
locus for lung cancer at 15q25.1. Nat Genet 40:616–622.
Batra V, Patkar AA, Berrettini WH, Weinstein SP, Leone FT. 2003. The
genetic determinants of smoking. Chest 123:1730–1739.
Batty GD, Deary IJ, Macintyre S. 2007a. Childhood IQ in relation to
risk factors for premature mortality in middle-aged persons: The
Aberdeen Children of the 1950s study. J Epidemiol Community Health
Batty GD, Wennerstad KM, Smith GD, Gunnell D, Deary IJ, Tynelius P,
RasmussenF. 2007b.IQ in earlyadulthood andlater cancer risk:Cohort
study of one million Swedish men. Ann Oncol 18:21–28.
Benowitz NL. 2008. Clinical pharmacology of nicotine: Implications for
understanding, preventing, and treating tobacco addiction. Clin Phar-
macol Ther 83:531–541.
D, Muglia P, Mooser V. 2008. Alpha-5/alpha-3 nicotinic receptor
subunit alleles increase risk for heavy smoking. Mol Psychiatry
Bierut LJ, Madden PA, Breslau N, Johnson EO, Hatsukami D, Pomerleau
OF, Swan GE, Rutter J, Bertelsen S, Fox L, Fugman D, Goate AM,
Hinrichs AL, Konvicka K, Martin NG, Montgomery GW, Saccone NL,
identified in a high-density genome wide association study for nicotine
dependence. Hum Mol Genet 16:24–35.
Bierut LJ, Stitzel JA, Wang JC, Hinrichs AL, Grucza RA, Xuei X, Saccone
NL, Saccone SF, Bertelsen S, Fox L, Horton WJ, Breslau N, Budde J,
Cloninger CR, Dick DM, Foroud T, Hatsukami D, Hesselbrock V,
Johnson EO, Kramer J, Kuperman S, Madden PA, Mayo K, Nurnberger
Edenberg HJ, Rice JP, Goate AM. 2008. Variants in nicotinic receptors
and risk for nicotine dependence. Am J Psychiatry 165:1163–1171.
Bouchard TJ, Jr. 1998. Genetic and environmental influences on adult
intelligence and special mental abilities. Hum Biol 70:257–279.
Breitling LP, Dahmen N, Mittelstrass K, Illig T, Rujescu D, Raum E,
Winterer G, Brenner H. 2009. Smoking cessation and variations in
nicotinic acetylcholine receptor subunits alpha-5, alpha-3, and beta-4
genes. Biol Psychiatry 65:691–695.
Brinkmeyer J,Mobascher A, Musso F, Schmitz M, Wagner M, Gr€ under G,
Spreckelmeyer KN, Wienker T, Diaz-Lacava A, Holler D, Dahmen N,
sensory gating and smoking in the general population. Addiction Biol.
Comings DE, Wu S, Rostamkhani M, McGue M, Lacono WG, Cheng LS,
MacMurray JP. 2003. Role of the cholinergic muscarinic 2 receptor
(CHRM2) gene in cognition. Mol Psychiatry 8:10–11.
follow-up of the Scottish Mental Survey 1932. In: Christen Y, editor.
Deary IJ, Johnson W, Houlihan LM. 2009. Genetic foundations of human
intelligence. Hum Genet 126:215–232.
Doran N, Cook J, McChargue D, Spring B. 2009. Impulsivity and
cigarette craving: Differences across subtypes. Psychopharmacol (Berl)
Egan MF, Goldberg TE, Kolachana BS, Callicott JH, Mazzanti CM, Straub
genotype on frontal lobe function and risk for schizophrenia. Proc Natl
Acad Sci USA 98:6917–6922.
JE, London ED. 2001a. Effect of nicotine on brain activation during
performance of a working memory task. Proc Natl Acad Sci USA
Ernst M, Heishman SJ, Spurgeon L, London ED. 2001b. Smoking history
Espeseth T, Greenwood PM, Reinvang I, Fjell AM, Walhovd KB, Westlye
LT, Wehling E, Lundervold A, Rootwelt H, Parasuraman R. 2006.
volume in healthy middle-aged and older adults. Cogn Affect Behav
II: Childhood intelligence and later outcomes in adolescence and young
adulthood. J Child Psychol Psychiatry 46:850–858.
Ferrea S, Winterer G. 2009. Neuroprotective and neurotoxic effects of
nicotine. Pharmacopsychiatry 42:255–265.
Fried PA, Watkinson B, Gray R. 2006. Neurocognitive consequences of
cigarette smoking in young adults–a comparison with pre-drug perfor-
mance. Neurotoxicol Teratol 28:517–525.
Gallinat J, Bajbouj M, Sander T, Schlattmann P, Xu K, Ferro EF, Goldman
D, Winterer G. 2003. Association of the G1947A COMT (Val(108/
158)Met) gene polymorphism with prefrontal P300 during information
processing. Biol Psychiatry 54:40–48.
Gehricke JG, Loughlin SE, Whalen CK, Potkin SG, Fallon JH, Jamner LD,
Belluzzi JD, Leslie FM. 2007. Smoking to self-medicate attentional and
emotional dysfunctions. Nicotine Tob Res 9(Suppl.4): 523–536.
DI, Posthuma D. 2006. Association between the CHRM2 gene and
intelligence in a sample of 304 Dutch families. Genes Brain Behav
Hemmingsson T, Melin B, Allebeck P, Lundberg I. 2006. The association
years of followup—A prospective observational study among Swedish
males born 1949–51. Int J Epidemiol 35:665–670.
Hung RJ, McKay JD, Gaborieau V, Boffetta P, Hashibe M, Zaridze D,
Liloglou T, Xinarianos G, Cassidy A, McLaughlin J, Liu G, Narod S,
Krokan HE, Skorpen F, Elvestad MB, Hveem K, Vatten L, Linseisen J,
Lagiou P, Trichopoulos D, Holc? atov? a I, Merletti F, Kjaerheim K, Agudo
Healy C, Zelenika D, Boland A, Delepine M, Foglio M, Lechner D,
Matsuda F, Blanche H, Gut I, Heath S, Lathrop M, Brennan P. 2008. A
susceptibility locus for lung cancer maps to nicotinic acetylcholine
receptor subunit genes on 15q25. Nature 452:633–637.
Khuder SA, Price JH, Jordan T, Khuder SS, Silvestri K. 2008. Cigarette
and age at onset. Int J Environ Res Public Health 5:278–289.
Kollins SH, McClernon FJ, Fuemmeler BF. 2005. Association between
smoking and attention-deficit/hyperactivity disorder symptoms in a
population-based sample of young adults. Arch Gen Psychiatry
Krishnan-Sarin S, Reynolds B, Duhig AM, Smith A, Liss T, McFetridge A,
Cavallo DA, Carroll KM, Potenza MN. 2007. Behavioral impulsivity
predicts treatment outcome in a smoking cessation program for adoles-
cent smokers. Drug Alcohol Depend 88:79–82.
Kumari V, Gray JA, Ffytche DH, Mitterschiffthaler MT, Das M, Zachariah
E, Vythelingum GN, Williams SC, Simmons A, Sharma T. 2003. Cogni-
tive effects of nicotine in humans: An fMRI study. Neuroimage
1456 AMERICAN JOURNAL OF MEDICAL GENETICS PART B
of nicotine dependence and pharmacotherapy. Biochem Pharmacol
Lindenberg A, Brinkmeyer J, Dahmen N, Gallinat J, de Millas W,
Mobascher A, Wagner M, Gr€ under G, Spreckelmeyer KN, Clepce M,
Th€ urauf N,Kiefer F,Diaz-LacavaA, Steffens M, Wienker T,Winterer G.
Submitted. The German Multicenter Study on Nicotine Dependence-
—Description of a population-based case–control study.
Liu JZ, Tozzi F, Waterworth DM, Pillai SG, Muglia P, Middleton L,
Berrettini W, Knouff CW, Yuan X, Waeber G, Vollenweider P, Preisig
Lamina C, Polasek O, Zgaga L, Huffman J, Campbell S, Kooner J,
Chambers JC, Burnett MS, Devaney JM, Pichard AD, Kent KM, Satler
Vitart V, Reilly MP, Li M, Qu L, Wilensky R, Matthai W, Hakonarson
HH, Rader DJ, Franke A, Wittig M, Sch€ afer A, Uda M, Terracciano A,
Xiao X, Busonero F, Scheet P, Schlessinger D, St Clair D, Rujescu D,
Abecasis GR, Grabe HJ, Teumer A, V€ olzke H, Petersmann A, John U,
Rudan I, Hayward C, Wright AF, Kolcic I, Wright BJ, Thompson JR,
Worthington J, Thomson W, Eyre S, Barton A, Wellcome Trust Case
Control Consortium, Mooser V, Francks C, Marchini J, 2010. Meta-
analysis and imputation refines the association of 15q25 with smoking
quantity. Nature Genet 42:436–440.
Mobascher A, Winterer G. 2008. The molecular and cellular neurobiology
of nicotine abuse in schizophrenia. Pharmacopsychiatry 41(Suppl.1):
Mobascher A, Brinkmeyer J, Warbrick T, Wels C, Wagner M, Gr€ under G,
Spreckelmeyer KN, Wienker T, Diaz Lacava A, Dahmen N, B€ ottcher M,
Thuerauf N, Clepce M, Kiefer F, De Millas W, Gallinat J, Winterer G.
2010. The P300 event-related potential and smoking—A population-
based case-control study. Int J Psychophysiol 77(2):166–175.
Neuhaus A, Bajbouj M, Kienast T, Kalus P, von Haebler D, Winterer G,
smokers. Psychopharmacol (Berl) 186:191–200.
Newhouse PA, Potter A, Singh A. 2004. Effects of nicotinic stimulation on
cognitive performance. Curr Opin Pharmacol 4:36–46.
first puff and daily cigarette smoking in adolescents. Am J Epidemiol
Rezvani AH, Levin ED. 2001. Cognitive effects of nicotine. Biol Psychiatry
Rigbi A, Kanyas K, Yakir A, Pollak Y, Ben-Asher E, Lancet D, Kertzman S,
Lerer B. 2008. Why do young women smoke? V. Role of direct and
interactive effects of nicotinic cholinergic receptor gene variation on
neurocognitive function. Genes Brain Behav 7:164–172.
Rigbi A, Yakir A, Sarner-Kanyas K, Pollak Y, Lerer B. 2010. Why do young
women smoke? VI. A controlled study of nicotine effects on attention:
pharmacogenetic interactions. Pharmacogenomics J [Epub ahead of
Rujescu D, Meisenzahl EM, Giegling I, Kirner A, Leinsinger G, Hegerl U,
of CSF compartments in healthy volunteers and schizophrenic patients.
Rujescu D, Hartmann AM, Gonnermann C, M€ oller HJ, Giegling I. 2003.
M129V variation in the prion protein may influence cognitive perfor-
mance. Mol Psychiatry 8:937–941.
Breslau N, Johnson EO, Hatsukami D, Pomerleau O, Swan GE, Goate
AM, Rutter J, Bertelsen S, Fox L, Fugman D, Martin NG, Montgomery
GW, Wang JC, Ballinger DG, Rice JP, Bierut LJ. 2007. Cholinergic
study targeting 348 candidate genes with 3713 SNPs. Hum Mol Genet
Savitz J, Solms M, Ramesar R. 2006. The molecular genetics of cognition:
Dopamine, COMT and BDNF. Genes Brain Behav 5:311–328.
Schlaepfer IR, Hoft NR, Collins AC, Corley RP, Hewitt JK, Hopfer CJ,
A3/B4 gene cluster variability as an important determinant of early
alcohol and tobacco initiation in young adults. Biol Psychiatry
Bierut LJ, Neuman RJ, Pomerleau OF. 2008. Association of a single
nucleotide polymorphism in neuronal acetylcholine receptor subunit
alpha 5 (CHRNA5) with smoking status and with ‘pleasurable buzz’
during early experimentation with smoking. Addiction 103:1544–1552.
Stevens VL, Bierut LJ, Talbot JT, Wang JC, Sun J, Hinrichs AL, Thun MJ,
Goate A, Calle EE. 2008. Nicotinic receptor gene variants influence
susceptibility to heavy smoking. Cancer Epidemiol Biomarkers Prev
Stolerman IP, Naylor CG, Mesdaghinia A,Morris HV. 2009.The duration
of nicotine-induced attentional enhancement in the five-choice serial
reaction time task: Lack of long-lasting cognitive improvement. Behav
Sullivan P, Kendler K. 1999. The genetic epidemiology of smoking.
Nicotine Tob Res 1(Suppl.): 51–57.
Tewes U. 1991. HAWIE-R Hamburg–Wechsler Intelligenztest f€ ur
Erwachsene. Revision 1991. Hogrefe: Goettingen.
The Tobacco and Genetics Consortium. 2010. Genome-wide meta-
analyses identify multiple loci associated with smoking behavior. Nat
Thiel CM, Zilles K, Fink GR. 2005. Nicotine modulates reorienting of
visuospatial attention and neural activity in human parietal cortex.
Thorgeirsson TE, Geller F, Sulem P, Rafnar T, Wiste A, Magnusson KP,
Manolescu A, Thorleifsson G, Stefansson H, Ingason A, Stacey SN,
M, Saemundsdottir J, Olafsdottir O, Gudmundsson LJ, Bjornsdottir G,
Kristjansson K, Skuladottir H, Isaksson HJ, Gudbjartsson T, Jones GT,
Mueller T, Gotts€ ater A, Flex A, Aben KK, de Vegt F, Mulders PF, Isla D,
JG, Hansdottir I, Runarsdottir V, Pola R, Lindblad B, van Rij AM,
Dieplinger B, Haltmayer M, Mayordomo JI, Kiemeney LA, Matthiasson
nicotinedependence, lungcancer andperipheralarterial disease.Nature
Sulem P, Rafnar T, Esko T, Walter S, Gieger C, Rawal R, Mangino M,
Prokopenko I, M€ agi R, Keskitalo K, Gudjonsdottir IH, Gretarsdottir S,
Stefansson H, Thompson JR, Aulchenko YS, Nelis M, Aben KK, den
Heijer M, Dirksen A, Ashraf H, Soranzo N, Valdes AM, Steves C,
Diego V, Lezcano V, Garcia-Prats MD, Ripatti S, Perola M, Kettunen J,
Hartikainen AL, Pouta A, Laitinen J, Isohanni M, Huei-Yi S, Allen M,
Krestyaninova M, Hall AS, Jones GT, van Rij AM, Mueller T, Dieplinger
Kiemeney LA, Mayordomo JI, Lindholt JS, Pedersen JH, Franklin WA,
WINTERER ET AL.
WolfH,MontgomeryGW,HeathAC,MartinNG,MaddenPA,Giegling Download full-text
I, Rujescu D, J€ arvelin MR, Salomaa V, Stumvoll M, Spector TD, Wich-
mann HE, Metspalu A, Samani NJ, Penninx BW, Oostra BA, Boomsma
Sequence variants at CHRNB3-CHRNA6 and CYP2A6 affect smoking
behavior. Nat Genet 42:448–453.
WangJC,Cruchaga C,Saccone NL,BertelsenS,LiuP,BuddeJP,DuanW,
Steinbach JH, Stitzel JA, Anderson MW, You M, Stevens VL, Bierut LJ,
Goate AM, COGEND collaborators and GELCC collaborators, 2009.
Risk for nicotine dependence and lung cancer is conferred by mRNA
expression levels and amino acid change in CHRNA5. Hum Mol Genet
Weiss RB, Baker TB, Cannon DS, von Niederhausern A, Dunn DM,
Matsunami N, Singh NA, Baird L, Coon H, McMahon WM, Piper ME,
Fiore MC, Scholand MB, Connett JE, Kanner RE, Gahring LC, Rogers
SW, Hoidal JR, Leppert MF. 2008. A candidate gene approach identifies
the CHRNA5-A3-B4 region as a risk factor for age-dependent nicotine
addiction. PLoS Genet 4:e1000125.
Winterer G, Goldman D. 2003. Genetics of human prefrontal function.
Brain Res Brain Res Rev 43:134–163.
Dahmen N, Weinberger DR. 2006. COMT genotype predicts BOLD
signal and noise characteristics in prefrontal circuits. Neuroimage
J. 2007. Association of attentional network function with exon 5 varia-
tions of the CHRNA4 gene. Hum Mol Genet 16:2165–2174.
S, Lerer B. 2007. Why do young women smoke? III. Attention and
impulsivity as neurocognitive predisposing factors. Eur Neuropsycho-
onset of smoking in high school students. Am J Dis Child 140:257–259.
1458AMERICAN JOURNAL OF MEDICAL GENETICS PART B