NEW YORK (GenomeWeb) – By combining a candidate gene study with a genome-wide association study, researchers led by Tiina Paunio at the Finnish National Institute for Health and Welfare found variants in two genes associated with extremely violent behavior.
Paunio and her colleagues studied more than 500 Finnish repeat violent offenders — including a subset of nearly 140 extremely violent offenders who committed at least 10 violent crimes — as well as non-violent offender and general population controls. As reported in Molecular Psychiatry today, the variants the researchers homed in on near monamine oxidase A and CDH13 were found in the violent offenders, but not controls.
"[O]ur results from the present study indicate that CDH13 and low-activity MAOA are quite specific to violent crime," Paunio and her colleagues said in their paper.
MAOA, the researchers noted, has been linked previously to violent criminal behavior. Nearly 20 years ago, a study of a Dutch cohort found an association between MAOA deficiency and impulsive and aggressive behavior, leading MAOA to be dubbed the "warrior gene." Paunio and her colleagues said that the crimes that cohort committed were "not severe" and that the findings were not replicated and remained controversial.
A separate study linked a stop codon in HTR2B with substance abuse and criminal activity in the past, but whether its association was with substance abuse or with violence was unclear, Paunio and her colleagues said.
Though the researchers noted that criminal behavior is shaped by both genetic and environmental factors, they argued that by focusing on extremely violent offenses it may be possible to tease out genetic factors.
"It is plausible that while research of the genetic background of criminal or violent behavior is hampered by many confounding factors, focusing on extreme phenotypes might yield more robust results," they said.
For their study, Paunio and her colleagues assembled a discovery cohort of 794 people from the 19 largest prisons in Finland. This group included 538 violent offenders, 84 extremely violent offenders, and 215 non-violent offenders. Violent offenses included murder, attempted murder, manslaughter, and attempted manslaughter, while non-violent offenses included drunk driving and drug and property crimes.
The cohort underwent structured interviews to glean information regarding history of substance abuse, childhood maltreatment, and antisocial personality disorder, and to exclude offenders with psychosis.
Through their candidate gene analysis, which examined MAOA and HTR2B, the researchers found that the low-activity MAOA genotype was associated with violent offenses — an odds ratio of 1.71 — and identified an even stronger association among extremely violent offenders — an odds ratio of 2.66. At the same time, they reported no signal linking HTR2B and violent crime in this cohort.
Paunio and her colleagues confirmed this link between the low-activity MAOA genotype and violence in a separate cohort of 114 homicide offenders, though they noted that the odds ratio fell to 1.50 in this group.
Meanwhile, through two GWAS, one of violent offending and one of extremely violent offending, Paunio and her colleagues found five strong, significant associations near the CDH13, PRUNE2, and SPIN1 genes.
While re-genotyping the variants in the wider offender cohort didn't yield any signals, the rs11649622 minor allele near CDH13 replicated in the separate cohort of homicide offenders.
CDH13 codes for T-cadherin, which is involved in neuronal proliferation, migration, and connectivity, and has been implicated in attention deficit/hyperactivity disorder. It, the researchers suggested, may be involved in impulse control as impulsivity is a core feature of ADHD and plays a role in most homicides. Some 80 percent of homicides in Finland are not premeditated, they noted.
Paunio and her colleagues also considered the influence of factors such as early life difficulties, trauma, and alcoholism on violent behaviors. For the low-activity MAOA genotype, neither childhood maltreatment nor gender altered the risk, researchers noted. However, they found a small increase of alcoholism among rs11649622 carriers, and that effect was most pronounced in people who had had troubled childhoods.
Together, the researchers calculated that combined low-activity MAOA and minor CDH13 alleles would yield an odds ratio of 13.45 for violent behavior.
"The results indicate both low monoamine metabolism and neuronal membrane dysfunction as plausible factors in the etiology of extreme criminal violent behavior, and a conservative estimate implies that 5 [percent to] 10 [percent] of all severe violent crime in Finland is attributable to specific MAOA and CDH13 genotypes," Paunio and her colleagues said.