Available via license: CC BY-NC-ND
Content may be subject to copyright.
Monoamine oxidase A genotype is associated with gang membership and
weapon use
Kevin M. Beaver
a,
⁎, Matt DeLisi
b
, Michael G. Vaughn
c
, J.C. Barnes
a
a
College of Criminology and Criminal Justice, Florida State University, Tallahassee, FL 32306-1127, USA
b
Department of Sociology, Iowa State University, Ames, IA 50111, USA
c
School of Social Work, Division of Epidemiology, School of Public Health, Department of Public Health Studies, Saint Louis University,
St. Louis, MO 63103, USA
Abstract
Context: A functional polymorphism in the promoter region of the monoamine oxidase A (MAOA) gene has been found to be associated
with a broad range of antisocial phenotypes, including physical violence. At the same time, it is well known that gang members represent
some of the most serious violent offenders. Even so, no research has ever examined the association between MAOA and gang membership.
Objectives: The aim of this study is to examine the association between MAOA and gang membership and between MAOA and weapon use.
Design: We examined the effects of MAOA by using a molecular genetic association research design.
Setting: A nonclinical sample was used in this study.
Participants: Participants were drawn from the National Longitudinal Study of Adolescent Health (1155 females, 1041 males).
Main Outcome Measures: The outcome measures of this study are gang membership and weapon use.
Results: The low MAOA activity alleles conferred an increased risk of joining a gang and using a weapon in a fight for males but not for
females. Moreover, among male gang members, those who used weapons in a fight were more likely to have a low MAOA activity allele
when compared with male gang members who do not use weapons in a fight.
Conclusions: Male carriers of low MAOA activity alleles are at risk for becoming a gang member and, once a gang member, are at risk for
using weapons in a fight.
© 2009 Published by Elsevier Inc.
1. Introduction
The low-activity alleles of a functional polymorphism in
the promoter region of the monoamine oxidase A (MAOA)
gene confer an increased risk to developing a range of
antisocial phenotypes [1,2]. To date, research has linked the
low-activity MAOA alleles to various psychopathologies,
maladaptive behaviors, cognitive dysfunctions, and criminal
behaviors. Brunner et al [3] identified Brunner syndrome,
which is an X-linked disorder characterized by impulsivity,
heightened aggressiveness, mild mental retardation, and
serious criminal behaviors including arson and sexual assault
caused by MAOA deficiency. Samochowiec et al [4] tested
whether length variation of the 30-bp repeat of the MAOA
polymorphism was associated with variation in antisocial
behavior and alcohol dependence using a clinical sample of
488 German males including 59 alcoholics with antisocial
personality disorder. Prevalence of the low-activity 3-repeat
allele was significantly higher among the 59 antisocial
alcoholics compared with 185 controls (51% vs 35%, P=
.031) and compared with 244 alcoholics without antisocial
personality disorder (51% vs 32%, P= .0008). The authors
concluded that the low-activity 3-repeat allele conferred
increased susceptibility to antisocial behavior. Based on data
from a sample of 41 autistic males, Cohen et al [5] found that
males with the low-activity MAOA genotype demonstrated
more severe autistic behaviors and had lower IQ than peers
with the high-activity MAOA gene.
Caspi et al [1] used data from a New Zealand birth cohort
to study gene-environment interplay involved in the
relationship between childhood maltreatment and adult
antisocial behavior. They found that males who possessed
low-activity MAOA alleles and who had been maltreated
were significantly likely to evince conduct disorder,
A
vailable online at www.sciencedirect.com
Comprehensive Psychiatry xx (2009) xxx –xxx
www.elsevier.com/locate/comppsych
⁎Corresponding author. Tel.: +1 850 644 9180; fax: +1 850 644 9614.
E-mail address: kbeaver@fsu.edu (K.M. Beaver).
0010-440X/$ –see front matter © 2009 Published by Elsevier Inc.
doi:10.1016/j.comppsych.2009.03.010
ARTICLE IN PRESS
antisocial personality, and violent, antisocial behavior.
Subsequent validation studies affirmed the gene-environ-
ment interaction (maltreatment × MAOA) among white but
not African American youths [6], whereas others reported
null effects based on a community sample [7]. Recently,
Kim-Cohen et al [2] reported confirmatory evidence of a link
between MAOA-maltreatment and psychiatric symptoms
and conducted a meta-analysis that further implicated
MAOA as a candidate gene for antisocial phenotypes.
One hypothesis for the pleiotropic effects of MAOA is
that it affects the regulation of emotion and cognition in the
limbic system. For instance, Meyer-Lindenberg et al [8]
found that carriers of the low-activity MAOA polymorph-
ism (2, 3, or 5 repeats) showed 8% reductions in gray matter
volumes in the amygdala, cingulated gyrus, insula, and
hypothalamus compared with carriers of the high-activity
MAOA (3.5 or 4 repeats). Functional magnetic resonance
imaging analyses showed that carriers of the low-activity
MAOA had increased amygdala arousal and diminished
reactivity of the regulatory prefrontal cortex particularly
among males. Although MAOA has been associated with a
range of antisocial phenotypes, no prior study has linked it
to gang membership. This study assessed the role of MAOA
in predicting gang membership—and the use of weapons
for violent means while involved in gangs—among a
sample of youths.
2. Methods
2.1. Data
This study uses data drawn from the genetic subsample of
the National Longitudinal Study of Adolescent Health (Add
Health). Detailed information about the Add Health data has
been published elsewhere [9,10]. Briefly, the Add Health is a
prospective and nationally representative sample of Amer-
ican youths. Data collection efforts began in 1994 when
more than 90 000 students completed self-report surveys
during regular school hours. A subsample of 20 745
participants and 17 700 of their primary caregivers were
reinterviewed in their homes. Approximately 1 to 2 years
later, a second round of interviews was conducted with
14 738 of the respondents. The third and final wave of
interviews were completed between 2001 and 2002 with
15 197 participants. Most of the respondents were young
adults at the time that the wave 3 surveys were administered.
All of the participants provided voluntary consent to be
included in the Add Health study.
Embedded within the Add Health study is a subsample of
participants who were genotyped. To be eligible to be
included in the DNA subsample, respondents had to have a
sibling who was also participating in the Add Health study.
In total, 2574 respondents submitted buccal cells to be
genotyped for a number of genetic polymorphisms that are
involved in neurotransmission. Overall, the Add Health data
span nearly 7 years of human development and contain
phenotypic, genotypic, and environmental measures.
3. Measures
3.1. Monoamine oxidase A
The MAOA gene is a polymorphic gene that is found on
the X chromosome at location Xp11.23-11.4 [11]. The
polymorphism arises from a 30 base pair variable number
of tandem repeats upstream in the 5′regulatory region of
the gene. The Add Health participants were genotyped for
this polymorphism using a variant of the assay developed
previously [12]. Primer sequences were as follows:
forward, 5′ACAGCCTGACCG-TGGAGAAG-3′(fluores-
cently labeled); and reverse, 5′-GAACGTGACGCTC-
CATTCGGA-3′. This assay produced polymerase chain
reaction (PCR) products of 291 (2-repeat allele), 321 (3-
repeat allele), 336 (3.5-repeat allele), 351 (4-repeat allele),
and 381 (5-repeat allele) base pairs. The genotypes were
scored independently by 2 different raters.
In line with previous researchers analyzing the Add
Health data [13], the MAOA gene was divided into 2 groups:
a low MAOA activity group and a high MAOA activity
group. The low MAOA activity group consisted of the 2-
repeat allele and the 3-repeat allele, whereas the high MAOA
activity group consisted of the 3.5-repeat allele, the 4-repeat
allele, and the 5-repeat allele. Using this nomenclature,
42.3% of males had a low MAOA activity allele and 57.7%
of males had a high MAOA activity allele. For females,
17.4% were homozygous for the low MAOA activity allele,
44.7% were heterozygous, and 37.9% were homozygous for
the high MAOA activity allele. Descriptive information for
MAOA and all of the variables used in the analyses are
presented in Table 1.
Table 1
Descriptive statistics for the add health study variables
MAOA (males), no. (%)
Low MAOA 440 (42.3)
High MAOA 601 (57.7)
MAOA (females), no. (%)
Low MAOA/Low MAOA 201 (17.4)
Low MAOA/High MAOA 516 (44.7)
High MAOA/High MAOA 438 (37.9)
Gang member, no. (%)
Yes 77 (3.5)
No 2119 (96.5)
Weapon use, no. (%)
Yes 58 (2.6)
No 2138 (97.4)
Sex, no. (%)
Male 1041 (47.4)
Female 1155 (52.6)
Race, no. (%)
White 1484 (67.6)
African American 383 (17.4)
Other 329 (15.0)
Age, mean (SD) 16.47 (1.69)
2K.M. Beaver et al. / Comprehensive Psychiatry xx (2009) xxx–xxx
ARTICLE IN PRESS
3.2. Gang member
During wave 2 interviews, respondents were asked
whether they had been initiated into a named gang within
the past 12 months. This item was coded dichotomously,
where 0 indicates not a gang member and 1 indicates a gang
member. Overall, 3.5% of the sample indicated they were a
gang member, with 5.2% of males and 2% of females
indicating they were gang members. These prevalence
estimates are similar to those garnered in other longitudinal
samples of youth [14].
3.3. Weapon use
During wave 2 interviews, respondents were asked a
number of questions pertaining to their involvement in
serious, physical violence. One of these variables—whether
the respondent had used a weapon in a fight—was identified
as being indicative of gang activity. This item was coded
dichotomously, where 0 indicates respondent did not use a
weapon in a fight and 1 indicates respondent did use a
weapon in a fight.
3.4. Control variables
Two control variables were included in the models to help
prevent model misspecification. First, race was included in
the analyses to help control for population stratification
effects. During wave 1 interviews, respondents were asked to
self-identify their racial status. In the present analyses, race
was coded with a series of dummy variables measuring
whether the respondent was white, African American, or
other. Second, age was coded as a continuous variable
measured in years.
4. Statistical analyses
The analysis for this study proceeds in a number of
interlocked steps. First, logistic regression models will be
calculated to determine whether MAOA is associated with
gang membership and with weapon use in a fight. Second,
logistic regression models will be calculated to determine
whether MAOA is able to distinguish between gang
members who used a weapon in a fight from those gang
members who did not use a weapon in a fight. These models
will be confined only to gang members. Given that MAOA is
located on the X chromosome all of the analyses will be
conducted separately for males and females. Last, the DNA
subsample contains nested data, where more than 1 person
from the same household is included in the sample. As a
result, some of the observations lacked independence and
thus could artificially deflate standard errors. This problem
was corrected by randomly removing 1 twin from each MZ
twin pair from the analyses [13] and by estimating all of the
statistical models using Huber/White standard errors [15].
5. Results
Tab le 2 contains the results of the binary logistic
regression models predicting gang membership and weapon
use for females. As can be seen, MAOA was unrelated to
both gang membership and weapon use. Race was unrelated
to both dependent variables, whereas age maintained an
inverse relationship with gang membership and weapon use.
Tab le 3 contains the results of the binary logistic
regression models predicting gang membership and weapon
use for males. In contrast to the female models, MAOA had a
statistically significant and positive effect on gang member-
ship and on weapon use. Stated differently, males with the
low MAOA genotype, compared with males with the high
MAOA genotype, were 1.94 times more likely to be gang
members, and they were also 1.82 times more likely to have
used a weapon in a fight.
The next set of binary logistic regression models
examined the association between MAOA and weapon use
among gang members. The results of these models are
presented in Table 4. As can be seen in the left-hand side of
Table 2
Logistic regression models predicting gang membership and weapon use
among females (n = 1155)
Gang member Weapon use
bSE Odds ratio bSE Odds ratio
MAOA −0.05 0.32 0.950 −0.24 0.43 0.787
White −0.39 0.64 0.677 −0.83 1.17 0.469
African American 0.03 0.71 1.04 1.24 1.10 3.47
Age −0.59⁎0.13 0.375 −0.47⁎0.16 0.627
Huber/White standard errors presented.
⁎Pb.05 level, 2-tailed.
Table 3
Logistic regression models predicting gang membership and weapon use
among males (n= 1041)
Gang member Weapon use
bSE Odds ratio bSE Odds ratio
MAOA 0.66⁎0.30 1.94 0.60⁎0.30 1.82
White −1.07⁎0.35 0.342 −0.22 0.40 0.802
African American −0.14 0.38 0.872 −0.34 0.50 0.715
Age −0.08 0.07 0.927 −0.10 0.09 0.902
Huber/White standard errors presented.
⁎Pb.05 level, 2-tailed.
Table 4
Logistic regression models predicting weapon use among gang members
Females (n = 23) Males (n = 54)
bSE Odds ratio bSE Odds ratio
MAOA −2.22 1.21 0.108 1.47⁎0.71 4.37
White −4.51⁎2.26 0.011 1.19 0.78 3.28
African American −1.35 2.54 0.257 −0.35 0.89 0.706
Age 0.61 0.51 1.85 0.47⁎0.22 1.61
Huber/White standard errors presented.
⁎Pb.05 level, 2-tailed.
3K.M. Beaver et al. / Comprehensive Psychiatry xx (2009) xxx–xxx
ARTICLE IN PRESS
the table, MAOA was unrelated to weapon use for female
gang members. For male gang members, however, a
different set of findings emerged. Male gang members who
carried low MAOA activity alleles were 4.37 times more
likely to use a weapon when compared with male gang
members who carried high MAOA activity alleles.
The results of the binary logistic regression models
revealed that MAOA was able to distinguish between male
gang members who did and did not use a weapon in a fight.
As a result, the next step was to examine this association in
greater detail by plotting the percentage of gang members
using a weapon in a fight by MAOA genotype. Fig. 1
displays the results and shows that there were 21 male gang
members with the high MAOA genotype. Of these gang
members, 81% had not used a weapon in a fight, whereas
19% had used a weapon in a fight. In contrast, there were 33
male gang members with the low MAOA genotype. Of these
gang members, 55% had not used a weapon in a fight,
whereas 45% had used a weapon in a fight. A χ
2
statistic
(χ
2
= 3.924, df =1,Pb.05) was calculated and confirmed
that the association between MAOA and weapon use among
male gang members was statistically significant.
6. Comment
An impressive amount of empirical research has demon-
strated that the low MAOA activity alleles are associated
with a range of antisocial phenotypes, including serious
physical violence and criminal behavior. This study extended
this line of research and examined the relationship between
MAOA and gang membership and between MAOA and
weapon use. The results of the analyses revealed that male
carriers of the low MAOA activity alleles were more likely to
join gangs than were males who possessed the high MAOA
activity alleles. In addition, the low MAOA activity alleles
also increased the odds of using a weapon in a fight for
males. Also of interest was that MAOA was able to
distinguish between male gang members who used weapons
in a fight and male gang members who did not use weapons
in a fight. This finding is of particular interest because it
indicates that variation in violence among gang members
may be partially circumscribed by genotype.
Also of importance was that there was no association
between MAOA and gang membership and MAOA and
weapon use for females. This finding was not unexpected
because MAOA is X-linked, and thus, the criminogenic
effects of the MAOA genotype are thought to be strongest
for males. In addition, the low base rate of physical violence
for females in the Add Health sample necessarily reduced
substantially the statistical power needed to detect a
statistically significant effect for MAOA. Future research
would benefit by examining the association between MAOA
and antisocial phenotypes in females as well as males.
With these findings in mind, it is important to point out
that there are a number of limitations that need to be
addressed before any definitive conclusions can be drawn.
First, the measure of gang membership was based on self-
reports, thereby bringing into question the validity and
reliability of this measure. It should be noted, however, that
prior researchers have used self-reports as a way to gauge
gang membership, and this measurement strategy is
considered acceptable [16-18]. Second, the measure of
weapon use was based on a single, self-report measure.
Whether MAOA would be related to other forms of extreme
violence remains an open empirical question that should be
addressed. Third, the DNA subsample of the Add Health
data is not nationally representative, which necessarily calls
into question whether the findings reported in this study
would be generalizable to the general population. It should
be pointed out, however, that the genotype frequencies in the
Add Health are comparable to those found in other samples
[13]. Still, this does not rule out the possibility that the DNA
sample analyzed in this study differs in significant ways from
the larger, nationally representative sample of the Add
Health study. As a result, replication studies addressing these
limitations are needed to test the robustness and general-
izability of the findings from this study.
Despite these limitations, it should be noted that this is the
first study to demonstrate a link between MAOA and gang
membership. Although the precise mechanisms leading from
MAOA to gang membership are unknown, it is likely the
result of a gene-environment correlation [19]. It is possible, for
instance, that male carriers of low MAOA activity alleles are
attracted to violence and thus seek out gangs to join. Likewise,
it is also possible that the most violent adolescents are
recruited to join certain gangs. Although the study of gangs
has largely proceeded as a sociological phenomenon, this
investigation shows that gang formation and activity, like most
antisocial behaviors, involves gene-environment interplay.
Fig. 1. The relationship between MAOA and weapon use among male gang
members (n = 54). χ
2
= 3.924, df =1,Pb.05.
4K.M. Beaver et al. / Comprehensive Psychiatry xx (2009) xxx–xxx
ARTICLE IN PRESS
Acknowledgment
This research uses data from Add Health, a program
project designed by J. Richard Udry, Peter S. Bearman, and
Kathleen Mullan Harris, and funded through grant P01-
HD31921 from the Eunice Kennedy Shriver National
Institute of Child Health and Human Development, with
cooperative funding from 17 other agencies. Special
acknowledgment is due to Ronald R. Rindfuss and Barbara
Entwisle for assistance in the original design. Persons
interested in obtaining data files from Add Health should
contact Add Health, Carolina Population Center, 123 W.
Franklin Street, Chapel Hill, NC 27516-2524
(addhealth@unc.edu). No direct support was received from
grant P01-HD31921 for this analysis.
References
[1] Caspi A, McClay J, Moffitt TE, Mill J, Martin J, Craig IW, et al. Role
of genotype in the cycle of violence in maltreated children. Science
2002;297:851-4.
[2] Kim-Cohen J, Caspi A, Taylor A, Williams B, Newcombe R, Craig IW,
et al. Maltreatment, and gene-environment interactions predicting
children's mental health: new evidence and a meta-analysis. Mol
Psychiatry 2006;11:903-13.
[3] Brunner HG, Nelen M, Breakefield XO, Ropers HH, van Oost BA.
Abnormal behavior associated with a point mutation in the structural
gene for monoamine oxidase A. Science 1993;262:578-80.
[4] Samochowiec J, Lesch K-P, Rottman M, Smolka M, Syagailo V,
Okladnova O, et al. Association of a regulatory polymorphism in the
promoter region of the MAOA gene with antisocial alcoholism.
Psychiatry Res 1999;86:67-72.
[5] Cohen IL, Liu X, Schutz C, White BN, Jenkins EC, Brown WT, et al.
Association of autism severity with a monoamine oxidase A functional
polymorphism. Clin Genet 2003;64:190-7.
[6] Widom CS, Brzustowicz LM. MAOA and the “cycle of violence”:
childhood abuse and neglect, MAOA genotype, and risk for violent
and antisocial behavior. Biol Psychiatry 2006;60:684-9.
[7] Huizinga D, Haberstick BC, Smolen A, Menard S, Young SE,
Corley RP, et al. Childhood maltreatment, subsequent antisocial
behavior, and the role of MAOA genotype. Biol Psychiatry 2006;60:
677-83.
[8] Meyer-Lindenberg A, Buckholtz JW, Kolachana B, Hariri AR,
Pezawas L, Blasi G, et al. Neural mechanisms of genetic risk for
impulsivity and violence in humans. Proc Nat Acad Sci U S A 2006;
103:6269-74.
[9] Harris KM, Florey F, Tabor J, Bearman PS, Jones J, Udry JR. The
National Longitudinal Study of Adolescent Health: Research Design.
http://www.cpc.unc.edu/projects/addhealth/design. Accessed Septem-
ber 5, 2008.
[10] Resnick M, Bearman P, Blum R, Bauman K, Harris K, Jones J, et al.
Protecting adolescents from harm: findings from the National Long-
itudinal Study of Adolescent Health. J Am Med Assoc 1997;278:
823-32.
[11] Levy ER, Powell JF, Buckle VJ, Hsu YP, Breakefield XO, Craig IW.
Localization of human monoamine oxidase-A gene to Xp11.23-11.4
by in situ hybridization: implications for Norrie disease. Genomics
1989;5:368-70.
[12] Sabol S, Hus S, Hamer D. A functional polymorphism in the
monoamine oxidase A gene promoter. Hum Genet 1998;103:273-9.
[13] Haberstick BC, Lessem JM, Hopfer CJ, Smolen A, Ehringer MA,
Timberlake D, et al. Monoamine oxidase A (MAOA) and antisocial
behaviors in the presence of childhood and adolescent maltreatment.
Am J Med Genet 2005;135B(1):59-64.
[14] Esbensen F-A, Huizinga D. Gangs, drugs, and delinquency in a survey
of urban youth. Criminology 1993;31:565-90.
[15] Beaver KM, Wright JP, DeLisi M, Walsh A, Vaughn MG, Boisvert D,
et al. A gene x gene interaction between DRD2 and DRD4 is associated
with conduct disorder and antisocial behavior in males. Behav Brain
Funct 2007;3:30.
[16] Bell, KE. Gender and gangs: A quantitative comparison. Crime and
Delinquency. Forthcoming.
[17] DuBois DL, Silverthorn N. Natural mentoring relationships and
adolescent health: evidence from a national study. Am J Public Health
2005;95(3):518-24.
[18] Esbensen F-A, Winfree LT, He N, Taylor TJ. Youth gangs and
definitional issues: when is a gang a gang, and why does it matter? J
Res Crime Delinq 2001;47:105-30.
[19] Jaffee SR, Price TS. Gene-Environment correlations: a review of the
evidence and implications for prevention of mental illness. Mol
Psychiatry 2004;12:432-42.
5K.M. Beaver et al. / Comprehensive Psychiatry xx (2009) xxx–xxx
ARTICLE IN PRESS