Hotheads by nature

Share on FacebookShare on Google+Email this to someoneTweet about this on Twitter


If some guy spilt your beer by accident, would you punch him in the face? If he was unapologetic, you might at least consider it – you might in fact feel a pretty strong urge to do it. What stops you? Or, if you’re the type who acts on those urges, what doesn’t stop you? New research has found a mutation in one gene that may contribute to these differences in temperament.

Self-control is the ability to inhibit an immediate course of action in the pursuit of a longer-term goal or to consciously override a base urge. Some people show far more inhibitory control than others. This trait is very stable – indeed, inhibitory control in children, which can be assessed using the famous “marshmallow test”, is predictive of their score on scales of impulsivity as adults. (The marshmallow test must go down as one of the cruellest experiments in psychology – it involves asking four-year olds not to eat a lovely yummy marshmallow for five minutes, after which they will be given another one to go with it if they have resisted. The videos of these poor kids as they struggle to resist this urge are priceless). Impulsivity is also partly heritable – that is, more closely related people are more similar in this trait.

This is generally true of all personality traits, suggesting they are influenced by genetic variation. However, the specific genes involved are almost entirely unknown. Indeed, a recent study that failed to find any such genes was interpreted by many (e.g., 1, 2) as evidence that either personality was not really genetic or that measures of personality traits were effectively meaningless. In fact, this was a gross misinterpretation of the results of this study. What these researchers did was look for common genetic variants that were associated with differences in personality traits, across a sample of over 5,000 people. Common variants are ancient differences at specific positions in the DNA code, where some proportion of the population carries one base, say a “C”, and the rest carry another base, say an “A”. There are millions of such variable positions across the human genome. Most of them do not do anything – they do not affect the sequence of a protein or how much of it is made. And, it seems, none of them affects personality significantly.


This does NOT mean that these traits are not affected by genetic variation. The genome-wide association analysis could not detect rare variants – ones that only a few people in the population carry. These are mutations that have arisen in the much more recent past and which have been passed on to only a small proportion of the population. In general, such mutations are far more likely to affect a protein and have some influence on the observable traits of an organism (its phenotype). Why? Because usually such effects are not very positive and natural selection pretty rapidly weeds them out – if a variant becomes common it is usually because it does not have any effect. (Not always, but usually).

So, how can these rare variants be found? Well, advances in sequencing technologies now make it possible to sequence the entire genetic code of a person or determine the entire sequence of a specific gene or genes across large numbers of people. This approach will pick up all the genetic differences, whether they are rare or common. This is what researchers from the National Institutes of Health and from Helsinki have done in a new study that led to the identification of a mutation in the Finnish population that apparently affects impulsivity.

They started with the hypothesis that this trait might be affected by variation in genes involved in the synthesis or signalling pathways of the neuromodulators dopamine and serotonin. These molecules act in the brain to alter the responsiveness of neurons to other signals – they set the tone, the internal context that helps determine how the organism will respond to various stimuli at any given moment. Differences in these pathways may also explain why different people will respond differently to the same stimulus (like that guy spilling your pint). There is a good deal of pharmacological evidence implicating these pathways in mood and temperament, as well as some prior genetic evidence for a couple specific genes.

To look for variation specifically affecting impulsivity, the researchers sequenced fourteen genes involved in the dopamine and serotonin pathways in a sample of the most impulsive people they could find – prisoners who had been convicted of violent, spontaneous crimes. All of these subjects had one of several psychiatric diagnoses that specifically include impulsive behaviour as a core symptom: borderline personality disorder, antisocial personality disorder or intermittent explosive disorder.

The scientists found one mutation that had never been seen in any other population – in the gene HTR2B, which encodes a receptor for serotonin. The mutation completely abolishes the production of the protein, so that people who carry one copy of this mutant version of the gene have only half the normal amount of the receptor protein. The mutant version was found to be greatly over-represented (7.5% frequency) among a set of 228 violently impulsive subjects, compared to 295 controls from the general population (1.2%). Among family members of the violent offenders who carried the mutation there was also an increased rate of the psychiatric disorders listed above, specifically in those relatives who also inherited the mutation.

These findings therefore suggest that this mutation increases the risk of this kind of violent, impulsive behaviour. It must only be one factor, however, as most of the 1% in the Finnish population who carry it are not violent criminals. Being male and alcohol abuse are two other likely risk factors. Almost all of the violent impulsive cases had committed crimes under the influence of alcohol, mostly unpremeditated “disproportionate reactions to minor irritations”. (Note the difference with psychopaths, who show much more cold-blooded and goal-directed violence). Two-thirds had also attempted suicide at least once, with an average of over 3 attempts.

So, does this mutation really affect the personality trait of impulsivity specifically, or is that just one component of a wider and more severe phenotype? The authors did look for effects on cognitive measures across a large Finnish twin sample, identifying significant effects on working memory in males, but do not report a test of association with impulsivity as a trait in this sample. We shall therefore have to wait to see if that more general association holds.

Their case is supported by observations in mice which carry mutations in the same gene – mice with both copies of this gene mutated score higher on a range of test used to measure impulsivity (yes, mice can be more or less impulsive). Also, the protein encoded by the HTR2B gene, the serotonin receptor 5-HT2B, is the target for the mood-altering drug ecstasy (3,4-methylene-dioxymethamphetamine, MDMA). When this drug binds the 5-HT2B receptor it induces serotonin release in the brain and a subsequent chain of events including dopamine release in the reward area of the brain.

These data naturally lead to the idea that the mutation found in this study has its effect by altering the amount of this receptor protein in the adult brain, thereby altering the tone of serotonin signalling. There is an alternative hypothesis, however, which is that the brain develops differently due to this mutation. There is good reason to think this may be the case as it is known that serotonin plays important roles in brain wiring at early stages of neural development. More on that possibility in a later post.

Whether the mechanism is acute or developmental, these findings emphasise the importance of rare variants – which may occur only in one population, in one kindred or family, or even in a single individual – in determining an individual’s phenotype.

Verweij KJ, Zietsch BP, Medland SE, Gordon SD, Benyamin B, Nyholt DR, McEvoy BP, Sullivan PF, Heath AC, Madden PA, Henders AK, Montgomery GW, Martin NG, & Wray NR (2010). A genome-wide association study of Cloninger’s temperament scales: implications for the evolutionary genetics of personality. Biological psychology, 85 (2), 306-17 PMID: 20691247

Bevilacqua L, Doly S, Kaprio J, Yuan Q, Tikkanen R, Paunio T, Zhou Z, Wedenoja J, Maroteaux L, Diaz S, Belmer A, Hodgkinson CA, Dell’osso L, Suvisaari J, Coccaro E, Rose RJ, Peltonen L, Virkkunen M, & Goldman D (2010). A population-specific HTR2B stop codon predisposes to severe impulsivity. Nature, 468 (7327), 1061-6 PMID: 21179162

8 Comments

  1. Dick et al and Viding et al also found no significant SNPs for conduct disorder and antisocial personality disorder. However, this is hardly the totality of common genetic variation. Most of the candidate gene research is focusing on promoter tandem repeats in dopamine- and serotonin- affecting genes. For instance, 5-HTTLPR was just redeemed as an important factor in depression predisposition after a new meta-analysis discredited a biased 2009 paper. It is a not a gene at all, but rather a promoter for SLC6A4.

  2. Great article.

    23AndMe tests 3 SNP’s on the HTR2B gene, they are:
    rs17586405 A or G
    rs10194776 C or T
    rs4973377 A or G

    Are any of these 3 SNP’s relevant to this new finding, if so which and what are the implicated values?

    Thanks in advance!

  3. The particular mutation they identified was present only in the Finnish population. Other populations or individuals may have other mutations that may be linked in certain cases to common SNPs, including the ones you mention. However, the variants a person has at those SNPs, by themselves, are unlikely to be informative.

  4. But can you reveal what that Finnish SNP was?

  5. The mutation is referred to as Q20* (meaning it changes the glutamine residue at position 20 of the protein to a stop codon).

  6. 5-HT2B is not the primary target of MDMA, which is in fact the serotonin transporter. However, it is possible that activation of 5-HT2B in conjunction with this, either by the serotonin released through SERT or by MDMA having weak agonist activity by itself, contributes to the effects of blocking serotonin uptake.

    Also, wasn’t 5-HT2B the one blamed for fenfluramine-induced heart valve defects? I wonder if there is any valvular phenotype to this SNP.

  7. Hmm, I found this, which supports my thinking…

    http://www.ncbi.nlm.nih.gov/pubmed/12761331

  8. MDMA certainly has additional targets including the serotonin transporter and other serotonin receptors. However, mice which are mutant in the HTR2B gene show no acute hyperactive response or sensitisation to MDMA, indicating that the 5HT2B receptor is indeed a crucial target.

    http://www.jneurosci.org/cgi/content/full/28/11/2933

    http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0007952

Leave a Reply

a