Abstract
Background
Testosterone has been attributed importance for various aspects of behaviour. The aim of our study was to investigate the potential influence of 2 functional polymorphisms in the amino terminal of the androgen receptor on personality traits in men.
Methods
We assessed and genotyped 141 men born in 1944 recruited from the general population. We used 2 different instruments: the Karolinska Scales of Personality and the Temperament and Character Inventory. For replication, we similarly assessed 63 men recruited from a forensic psychiatry study group.
Results
In the population-recruited sample, the lengths of the androgen receptor repeats were associated with neuroticism, extraversion and self-transcendence. The association with extraversion was replicated in the independent sample.
Limitations
Our 2 samples differed in size; sample 1 was of moderate size and sample 2 was small. In addition, the homogeneity of sample 1 probably enhanced our ability to detect significant associations between genotype and phenotype.
Conclusion
Our results suggest that the repeat polymorphisms in the androgen receptor gene may influence personality traits in men.
Abstract
Contexte
On a attribué de l’importance à la testostérone pour divers aspects du comportement. Notre étude visait à étudier l’influence possible de 2 polymorphismes fonctionnels de la terminaison amine du récepteur de l’androgène sur les traits de personnalité chez l’homme.
Méthodes
Nous avons évalué et génotypé 141 hommes nés en 1944 recrutés dans la population générale. Nous avons utilisé 2 instruments différents : les échelles de la personnalité de Karolinska et le questionnaire sur le tempérament et le caractère. Pour la répétition, nous avons évalué de la même façon 63 hommes recrutés dans un groupe d’étude en psychiatrie médicolégale.
Résultats
Dans l’échantillon recruté au sein de la population, on a établi un lien entre la longueur des répétitions du récepteur de l’androgène et les traits névrotiques, l’extraversion et l’autotranscendance. On a retrouvé le lien avec l’extraversion dans l’échantillon d’origine indépendante.
Limites
La taille de nos 2 échantillons était différente : l’échantillon 1 était de taille modérée et l’échantillon 2, de taille modeste. En outre, l’homogénéité de l’échantillon 1 a probablement amélioré notre capacité de détecter des liens significatifs entre le génotype et le phénotype.
Conclusion
Nos résultats indiquent que les polymorphismes répétés dans le gène récepteur de l’androgène peuvent avoir une influence sur des traits de personnalité chez l’homme.
Introduction
Testosterone influences the structure and function of the brain, both during development and in the adult organism, and is a key factor for the sexual differentiation of this organ.1,2 Given this influence, it is reasonable to assume that testosterone may be of importance for interindividual differences in personality traits, not least those displaying sexual dimorphism, such as neuroticism, extraversion and proneness to aggression. Supporting this notion, studies suggest that androgen levels may correlate with such traits.2–10
The actions of androgens are mediated mainly by androgen receptors (ARs), which are widely distributed throughout the human brain. The AR gene (AR) is located on chromosome Xq11–12 and composed of 8 exons.11,12 Exon 1, encoding the amino-terminal domain, contains 2 polymorphic trinucleotide repeats: one CAG repeat encoding a polyglutamine stretch13 and one GGN repeat encoding a polyglycine stretch.14
The polymorphic polyglutamine stretch appears to be of importance in the function of the receptor as a transcription factor, tentatively by influencing the interaction between the receptor and various coactivators (relatively short fragments improving this interaction and hence increasing receptor responsiveness).15–20 The importance of the polymorphic polyglycine stretch for AR function is less explored. Recent in vitro studies indicate that polyglycine repeat length is associated with receptor responsiveness, but the exact nature of these effects is not fully clarified.21–25 In line with these data, a large number of studies have reported associations between short CAG or GGN repeats and increased risk of prostate cancer and benign prostate hyperplasia26–29 and decreased risk of infertility in men.30–32 Although investigations on the influence of AR repeat polymorphisms on human behaviour have been sparse, associations with personality,33–37 externalizing behaviour38 and depression39 have been suggested.
In the present study, we investigated the extent to which the AR repeat polymorphisms are related to personality traits in 2 independent samples of men assessed with 2 personality questionnaires, the Karolinska Scales of Personality (KSP)35,40,41 and the Temperament and Character Inventory (TCI).42
Methods
Participants
We recruited sample 1 from the general population, and we recruited sample 2 from a forensic psychiatry clinic. Sample 1 comprised men living in Gothenburg, Sweden, who were born during the first 6 months of 1944. They had previously been recruited by means of the population register for a study on metabolic risk factors and anthropometry; for details see a previous report by Rosmond and colleagues.43 We assessed their personalities with the KSP and TCI. Sample 2 comprised perpetrators of severe violent and sexual crimes who had been consecutively referred by the court for pretrial forensic psychiatric investigation at the Department of Forensic Psychiatry in Gothenburg and who had accepted to participate in a research project.44 In addition to lifetime file reviews and collateral information, the clinical workup for this group included the Structured Clinical Interview45 for Axis I and Axis II DSM-IV46 disorders, assessments for childhood-onset neuropsychiatric disorders, the KSP and TCI.
All participants provided written informed consent. The Ethics Committee, University of Gothenburg, Sweden approved our study protocol.
Personality assessments
The KSP consists of 135 items forming 15 subscales and measuring relatively stable personality traits.35,40,41 These subscales are often classified into 4 factors covering different dimensions of temperament: neuroticism, psychoticism, nonconformity and extraversion.35,47–49 Interindividual variations in several of the KSP scales have been shown to be partly hereditary.50 To adjust for age and sex, we standardized the KSP personality test scores using normative data (T scores) to have an expected mean of 50 (standard deviation [SD] 10). The TCI is based on a self-administered true/false questionnaire and designed to assess personality along 4 dimensions of temperament (novelty seeking, harm avoidance, reward dependence and persistence) and along 3 dimensions of character (self-directedness, cooperativeness and self-transcendence).51 In the present study, we used a Swedish 238-item translation of the TCI.42 As with the KSP, we standardized the TCI scores using normative data (T scores).
Genotyping
We isolated genomic DNA using the QIAamp DNA Blood Mini Kit (QIAGEN). We amplified the 2 different regions by polymerase chain reaction (PCR).
We performed the PCR of the CAG repeat in the AR in a total volume of 15 μl containing about 50 ng DNA, 0.625 U AmpliTaq DNA polymerase (Applied Biosystems) and 0.2 μM each of the following primers: 5′-GTGCGCGAAGT-GATCCAG A-3′ and 5′-GTTTCCTCATCCAGGACCAGGTA-3′.52 We fluorescently labelled the forward primer with 6-FAM. We added nucleotides promoting the nontemplated addition of adenine by Taq DNA polymerase to the 5′ end of the reverse primer.53 We performed thermal cycling in an Applied Biosystems GeneAmp PCR System 9700 with the following temperature profile: 95°C for 15 minutes followed by 35 cycles of 95°C for 30 seconds, 57°C for 30 seconds and 72°C for 30 seconds, with a final incubation at 72°C for 7 minutes.
We performed the PCR of the GGN repeat in the AR in a total volume of 15 μL containing about 50 ng DNA, 0.75 U ThermalAce DNA polymerase (Invitrogen) and 0.2 μM each of the following primers: 5′-TCCTGGCACACTCTCTTCAC -3′ and 5′-GCCAGGGTACCACACATCAGGT -3′ .14 We fluorescently labelled the forward primer with HEX. We performed thermal cycling in an Applied Biosystems GeneAmp PCR System 9700 with the following temperature profile: 98°C for 3 minutes followed by 35 cycles of 98°C for 30 seconds, 62°C for 30 seconds and 72°C for 30 seconds, with a final incubation at 72°C for 10 minutes.
We analyzed the fluorescently labelled DNA fragments by size with automated capillary electrophoresis using an ABI PRISM 310 Genetic Analyzer (Applied Biosystems).
Sequencing
To verify the actual number of repeats corresponding to a specific fragment length, we sequenced PCR products from 5 individuals. To this end, we amplified fragments containing the CAG or GGN repeats using the PCR primers described previously (with the exception of the forward primer for the CAG repeat being 5′-GTGCGCGAAGTGATCCAGA-3′) with 1U of HotStarTaq polymerase (QIAGEN)/20 μL reaction and an annealing temperature of 58°C. We sequenced the PCR products using the BigDye Terminator Cycle Sequencing Kit (v3.1, Applied Biosystems). We then subjected samples to electrophoresis using an ABI 3730 genetic analyzer (Applied Biosystems).
Statistical analysis
Prompted by previous studies suggesting an inverse relation between the number of CAG repeats in the AR and receptor activity, we grouped all alleles into 2 groups of roughly equal size using the median as cut-off; we defined those with 21 repeats or fewer as short (S), whereas we defined those with 22 repeats or more as long (L). In the same way, we defined a short GGN repeat as 23 repeats or fewer and a long repeat as 24 repeats or more.
Based on previous studies,48,49 we grouped the 15 subscales of KSP into 4 factors (extraversion, including impulsiveness and monotony avoidance; neuroticism, including somatic anxiety, psychic anxiety, muscular tension, psychasthenia, inhibition of aggression, guilt and socialization; nonconformity, including indirect aggression, verbal aggression, irritability and social desirability; and psychoticism, including detachment and suspicion). This grouping was undertaken by calculating the mean of all items within one factor, using 100-X for the 2 scales (socialization and social desirability) displaying negative loading in relation to the other scales within that factor.49 In case a certain factor displayed a significant association with genotype, the possible relations between genotype and the different subscales within this factor were further analyzed.
We assessed associations between AR genotypes on the one hand and KSP and TCI scores on the other using 2-way analysis of variance (ANOVA) with CAG and GGN polymorphisms as categorical explanatory variables, each having 2 levels: S and L. To measure the proportion of variance in phenotypes explained by the combined effect of the 2 genetic variants, we calculated R2 values for the 2-way ANOVA models. For all analyses on sample 1, we applied 2-sided tests. For the analyses on sample 2, aiming to replicate the findings observed in sample 1, we used 1-sided tests. We used SPSS software, version 15.0 (SPSS Inc.) for the 2-way ANOVA.
We corrected for multiple testing for the analyses of sample 1 by means of permutation tests (10 000 permutations), the 4 KSP factors and the 7 TCI dimensions. The reported p values are uncorrected.
We determined the extent of linkage disequilibrium between the AR repeat polymorphisms by means of χ2 tests; the sum of squared D values divided by the product of the corresponding allele frequencies multiplied by the number of haplotypes in the sample is χ2n–1,m–1, where n and m are the number of alleles for the 2 loci. We pooled alleles with frequencies less than 1%.
Results
Participants
Sample 1 comprised 275 men aged 59 years at the time of personality assessment. We obtained genotype data and useful KSP and TCI questionnaires from 141 of these men. Sample 2 comprised 100 perpetrators of severe violent and sexual crimes (92 men and 8 women). We obtained genotype data and useful KSP questionnaires from 63 of the men and TCI questionnaires from 50 men. The age of these participants varied from 17 to 76 (mean 35) years.
The allele distribution of the AR repeat polymorphisms, as displayed in Figure 1, was similar to that reported in other Swedish populations.31,35 The 2 polymorphisms were not in linkage disequilibrium. The AR genotype (S/L) frequencies did not differ between sample 1 and sample 2 (p > 0.1 for both polymorphisms). The KSP and TCI scores in the 2 samples, divided by AR genotypes, are shown in Table 1 and Table 2, respectively.
Fig. 1.
Allele distributions of (A) CAG and (B) GGN alleles in the AR gene in samples 1 and 2.
Table 1.
Karolinska Scales of Personality and Temperament and Character Inventory scores* in a normal population of men (sample 1) grouped into carriers of short and long alleles of the CAG and GGN repeat polymorphisms of the AR, respectively
| Polymorphism; mean (SD)
|
||||
|---|---|---|---|---|
| Test measure | CAG-S | CAG-L | GGN-S | GGN-L |
| KSP | (n = 59–66) | (n = 72–75) | (n = 7 8–83) | (n = 53–58) |
| Neuroticism | 50.57 (7.06) | 47.43 (6.67) | 47.85 (6.50) | 50.32 (7.50) |
| Somatic anxiety | 53.44 (9.00) | 48.27 (8.36) | 49.24 (8.39) | 52.61 (9.55) |
| Psychic anxiety | 48.90 (8.93) | 45.85 (8.73) | 46.48 (8.58) | 48.32 (9.37) |
| Muscular tension | 54.13 (9.74) | 49.90 (9.74) | 50.30 (9.49) | 54.03 (10.23) |
| Psychasthenia | 50.31 (11.47) | 46.09 (10.51) | 47.78 (11.09) | 48.31 (11.25) |
| Inhibition of aggression | 45.11 (9.43) | 44.65 (10.48) | 44.60 (10.12) | 45.24 (9.87) |
| Guilt | 48.49 (7.35) | 45.94 (9.51) | 45.78 (8.70) | 49.02 (8.33) |
| Socialization | 46.38 (10.93) | 48.66 (10.53) | 49.24 (9.80) | 45.26 (11.66) |
| Psychoticism | 46.93 (9.20) | 47.93 (9.69) | 47.96 (8.94) | 46.78 (10.16) |
| Nonconformity | 48.64 (6.30) | 47.09 (6.39) | 47.96 (6.58) | 47.61 (6.13) |
| Extraversion | 54.57 (9.69) | 51.27 (8.05) | 52.36 (8.18) | 53.47 (10.04) |
| Impulsiveness | 51.99 (10.77) | 49.68 (8.74) | 49.90 (8.37) | 52.01 (11.45) |
| Monotony avoidance | 57.15 (10.73) | 52.86 (11.51) | 54.83 (11.23) | 54.93 (11.55) |
| TCI | (n = 67) | (n = 74) | (n = 8 2) | (n = 59) |
| Temperament | ||||
| Novelty seeking | 51.52 (11.98) | 52.00 (9.10) | 52.01 (10.15) | 51.44 (11.11) |
| Harm avoidance | 49.70 (10.78) | 47.70 (9.68) | 48.34 (9.49) | 49.08 (11.25) |
| Reward dependence | 49.73 (10.76) | 47.93 (10.69) | 48.44 (10.55) | 49.27 (11.02) |
| Persistence | 52.43 (10.16) | 49.89 (10.88) | 51.20 (10.45) | 50.97 (10.86) |
| Character | ||||
| Self-directedness | 48.43 (10.78) | 50.65 (10.56) | 49.72 (10.52) | 49.42 (10.99) |
| Cooperativeness | 47.24 (8.52) | 48.19 (10.92) | 46.65 (9.94) | 49.25 (9.56) |
| Self-transcendence | 51.76 (9.31) | 46.26 (9.29) | 47.32 (8.62) | 51.03 (10.68) |
| Self-transcendence 1 | 54.24 (11.99) | 48.22 (10.11) | 49.95 (10.14) | 52.64 (12.90) |
| Self-transcendence 2 | 51.60 (9.28) | 49.61 (9.84) | 49.09 (8.99) | 52.59 (10.11) |
| Self-transcendence 3 | 49.30 (8.96) | 44.47 (9.15) | 45.55 (8.96) | 48.46 (9.68) |
KSP = Karolinska Scales of Personality; L = long; S = short; SD = standard deviation; TCI = Temperament and Character Inventory.
We standardized all personality test scores using normative data (T scores)
Table 2.
Karolinska Scales of Personality and Temperament and Character Inventory scores* in a forensic psychiatric population (sample 2) grouped into carriers of short and long alleles of the CAG and GGN repeat polymorphisms of the AR, respectively
| Polymorphism; mean (SD)
|
||||
|---|---|---|---|---|
| Test measure | CAG-S | CAG-L | GGN-S | GGN-L |
| KSP | (n = 30) | (n = 20) | (n = 34) | (n = 16) |
| Neuroticism | 63.08 (12.46) | 57.35 (11.72) | 59.87 (12.84) | 62.72 (11.49) |
| Somatic anxiety | 67.87 (18.85) | 58.85 (14.77) | 63.56 (17.99) | 65.75 (17.70) |
| Muscular tension | 70.23 (17.91) | 58.35 (14.77) | 66.44 (17.94) | 63.44 (17.19) |
| Psychasthenia | 62.27 (14.69) | 55.35 (12.92) | 57.35 (12.69) | 64.06 (16.73) |
| Socialization | 21.63 (17.28) | 34.20 (17.64) | 27.15 (18.96) | 25.63 (17.45) |
| Psychoticism | 56.73 (9.74) | 54.43 (12.60) | 55.18 (11.79) | 57.16 (8.94) |
| Nonconformity | 53.28 (9.04) | 49.04 (7.25) | 50.77 (9.15) | 53.31 (7.09) |
| Extraversion | 60.70 (13.70) | 50.73 (12.51) | 55.81 (12.52) | 58.63 (17.02) |
| Impulsiveness | 63.70 (17.04) | 55.15 (14.03) | 58.24 (16.09) | 64.63 (16.41) |
| Monotony avoidance | 57.70 (13.86) | 46.30 (13.09) | 53.38 (12.25) | 52.63 (19.02) |
| TCI | (n = 36) | (n = 27) | (n = 45) | (n = 18) |
| Temperament | ||||
| Novelty seeking | 52.08 (10.31) | 50.48 (10.54) | 52.36 (10.27) | 49.00 (10.47) |
| Harm avoidance | 61.06 (14.22) | 59.85 (15.18) | 58.73 (14.99) | 65.06 (12.59) |
| Reward dependence | 47.67 (10.27) | 48.67 (10.31) | 48.22 (10.57) | 47.78 (9.55) |
| Persistence | 47.86 (9.64) | 50.00 (9.98) | 49.73 (10.34) | 46.39 (7.90) |
| Character | ||||
| Self-directedness | 38.50 (15.25) | 40.37 (13.38) | 40.56 (15.66) | 36.17 (10.32) |
| Cooperativeness | 42.58 (17.85) | 45.19 (13.85) | 43.51 (15.14) | 44.17 (19.02) |
| Self-transcendence | 58.97 (15.01) | 53.41 (14.49) | 57.04 (14.73) | 55.44 (15.82) |
KSP = Karolinska Scales of Personality; L = long; S = short; SD = standard deviation; TCI = Temperament and Character Inventory.
We standardized all personality test scores using normative data (T scores)
Karolinska Scales of Personality data
In sample 1, 2-way ANOVA revealed associations between AR polymorphisms and neuroticism and extraversion, but not psychoticism or nonconformity (Table 3). Neuroticism was associated with the CAG repeat only (Fig. 2), whereas extraversion was associated with the CAG repeat and with the interaction between the 2 repeat polymorphisms (Fig. 3). The association with extraversion survived correction for multiple testing (pcorrected = 0.009), whereas that with neuroticism did not (pcorrected = 0.15).
Table 3.
Results* from 2-way analyses of variance of the effect of the CAG and GGN repeat polymorphisms on Karolinska Scales of Personality and Temperament and Character Inventory scores in a normal population of men (sample 1)
|
p value; polymorphism
|
|||||
|---|---|---|---|---|---|
| Test measure | CAG | GGN | CAG × GGN | R2, % | ANOVA p value |
| KSP | |||||
| Neuroticism | 0.015 | 0.08 | 0.44 | 7.6 | 0.018 |
| Somatic anxiety | 0.001 | 0.06 | 0.14 | 12.0 | 0.001 |
| Psychic anxiety | 0.07 | 0.35 | 0.90 | 3.6 | 0.20 |
| Muscular tension | 0.02 | 0.06 | 0.54 | 7.3 | 0.02 |
| Psychasthenia | 0.04 | 0.99 | 0.98 | 3.6 | 0.20 |
| Inhibition of aggression | 0.82 | 0.74 | 0.93 | 0.1 | 0.98 |
| Guilt | 0.16 | 0.06 | 0.98 | 4.9 | 0.09 |
| Socialization | 0.23 | 0.04 | 0.14 | 5.7 | 0.06 |
| Psychoticism | 0.61 | 0.53 | 0.98 | 0.6 | 0.85 |
| Nonconformity | 0.13 | 0.55 | 0.77 | 1.8 | 0.49 |
| Extraversion | 0.01 | 0.7 | 0.004 | 9.1 | 0.004 |
| Impulsiveness | 0.12 | 0.29 | 0.04 | 5.1 | 0.07 |
| Monotony avoidance | 0.006 | 0.74 | 0.006 | 8.8 | 0.005 |
| TCI | |||||
| Temperament | |||||
| Novelty seeking | 0.94 | 0.78 | 0.06 | 2.7 | 0.29 |
| Harm avoidance | 0.34 | 0.80 | 0.40 | 1.5 | 0.55 |
| Reward dependence | 0.44 | 0.76 | 0.36 | 1.4 | 0.59 |
| Persistence | 0.12 | 0.74 | 0.44 | 2.0 | 0.44 |
| Character | |||||
| Self-directedness | 0.22 | 0.99 | 0.88 | 1.1 | 0.68 |
| Cooperativeness | 0.44 | 0.10 | 0.91 | 2.2 | 0.38 |
| Self-transcendence | 0.001 | 0.06 | 0.44 | 10.8 | 0.001 |
| Self-transcendence 1 | 0.002 | 0.33 | 0.19 | 8.8 | 0.005 |
| Self-transcendence 2 | 0.39 | 0.05 | 0.64 | 4.0 | 0.13 |
| Self-transcendence 3 | 0.003 | 0.15 | 0.49 | 8.4 | 0.007 |
KSP = Karolinska Scales of Personality; TCI = Temperament and Character Inventory.
p values are 2-sided. For p values corrected for multiple testing, see the Results section. Cell sizes for analysis of neuroticism: CAG-SxGGN-S: 31, CAG-SxGGN-L: 28, CAG-LxGGN-S: 47, CAG-LxGGN-L: 25, nonconformity: CAG-SxGGN-S: 31, CAG-SxGGN-L: 33, CAG-LxGGN-S: 49, CAG-LxGGN-L: 24, etxraversion: CAG-SxGGN-S: 33, CAG-SxGGN-L: 33, CAG-LxGGN-S: 50, CAG-LxGGN-L: 25, all TCI dimensions: CAG-SxGGN-S: 34, CAG-SxGGN-L: 33, CAG-LxGGN-S: 48, CAG-LxGGN-L: 26.
Fig. 2.
Carriers of short CAG alleles display higher neuroticism scores than carriers of long CAG repeat alleles in men recruited from the general population (sample 1). No significant interaction between the repeat polymorphisms is seen. For cell sizes, see Table 3.
Fig. 3.
Two-way interaction effect between the CAG and GGN repeat polymorphisms on extraversion scores in men recruited from the general population (sample 1); carriers of a short GGN allele show no effect of the CAG repeat on extraversion scores, whereas carriers of a long GGN allele show higher extraversion scores when also carrying a short CAG repeat allele. For cell sizes, see Table 3.
Within the extraversion factor, the 2-way ANOVA revealed significant associations between the CAG × GGN interaction and both subscales, impulsiveness and monotony avoidance, carriers of the short CAG allele displaying higher scores when also carrying a long GGN repeat allele (Table 3). As shown in Table 3, the CAG main effect was also significant for the subscale monotony avoidance.
Although the results did not remain significant after correction for multiple testing, it is noteworthy that within the neuroticism factor 2-way ANOVA revealed significant associations for somatic anxiety and muscular tension. As shown in Table 3, the CAG repeat was associated with the neuroticism factor and the somatic anxiety, psychasthenia and muscular tension subscales. For the GGN repeat polymorphism, these associations did not reach significance; moreover, there was no significant interaction between the 2 repeats for any of these items. However, in contrast to the CAG repeat, the GGN repeat was significantly associated with the socialization subscale (Table 3). We observed no significant associations for the other neuroticism subscales (i.e., psychic anxiety, inhibition of aggression or guilt).
In sample 2, the effect of the interaction between the 2 repeats on the extraversion factor and the monotony avoidance subscale observed in sample 1 could be replicated, as could the associations between the CAG repeat and these traits (Table 4). Moreover, in sample 2 the 2-way ANOVA model showed significant associations between the CAG repeat and the impulsiveness subscale (Table 4).
Table 4.
Results* from 2-way analyses of variance of the effect of the CAG and GGN repeat polymorphisms on Karolinska Scales of Personality and Temperament and Character Inventory scores in a forensic psychiatric population (sample 2)
|
p value; polymorphism
|
|||||
|---|---|---|---|---|---|
| Test measure | CAG | GGN | CAG × GGN | R2, % | ANOVA p value |
| KSP | |||||
| Neuroticism | 0.6 | 0.35 | 0.30 | 6.5 | 0.19 |
| Somatic anxiety | 0.02 | 0.45 | 0.15 | 8.7 | 0.12 |
| Muscular tension | 0.003 | 0.10 | 0.09 | 16.0 | 0.02 |
| Psychasthenia | 0.07 | 0.11 | 0.41 | 9.6 | 0.1 |
| Socialization | 0.001 | 0.24 | 0.01 | 20.4 | 0.07 |
| Extraversion | 0.001 | 0.38 | 0.01 | 21.5 | 0.005 |
| Impulsiveness | 0.007 | 0.34 | 0.01 | 18.2 | 0.01 |
| Monotony avoidance | 0.0005 | 0.15 | 0.04 | 21.0 | 0.006 |
| TCI | |||||
| Character | |||||
| Self-transcendence | 0.07 | 0.27 | 0.37 | 4.1 | 0.24 |
KSP = Karolinska Scales of Personality; TCI = Temperament and Character Inventory.
Only items yielding significant associations in sample 1 are included; p values are 1-sided. Cell sizes for the for the analysis of neuroticism and extraversion: CAG-SxGGN-S: 19, CAG-SxGGN-L: 11, CAG-LxGGN-S: 15, CAG-LxGGN-L: 5, self-transcendence: CAG-SxGGN-S: 24, CAG-SxGGN-L: 12, CAG LxGGN-S: 21, CAG-LxGGN-L: 6.
On the other hand, analysis of association between AR genotypes and the neuroticism factor did not reach significance in sample 2 (p = 0.06). However, the significant associations observed in sample 1 between the CAG repeat and the somatic anxiety and muscular tension neuroticism subscales could be replicated (Table 4). With respect to the socialization subscale, the outcome differed from that observed in sample 1; whereas the effect of the CAG repeat and the interaction between the 2 repeats were significant in sample 2, the GGN main effect was not.
As in sample 1, we observed no associations in sample 2 between AR genotypes and psychoticism (2-tailed p values: CAG p = 0.67; GGN p = 0.55; CAG × GGN p = 0.69; R2 = 2.0%; model p = 0.82) or nonconformity (2-tailed p values: CAG p = 0.06; GGN p = 0.68; CAG × GGN = 0.27; R2 = 9.8%; model p = 0.19).
Temperament and Character Inventory data
When we performed 2-way ANOVA to assess possible associations between the 2 AR repeat polymorphisms and the 7 dimensions of personality assessed using TCI, we found that only the self-transcendence dimension was significantly associated with AR genotype (Table 3 and Fig. 4). The association between the CAG repeat and self-transcendence remained significant after correction for multiple testing (pcorrected = 0.006). Within this dimension, the CAG repeat was significantly associated with the self-transcendence 1 (self-forgetful v. self-conscious experiences) and self-transcendence 3 (spiritual acceptance v. material rationalism) subscales (Table 3), with participants carrying short CAG repeat alleles displaying higher scores. In contrast, the GGN repeat was significantly associated only with the self-transcendence 2 subscale. The interaction between the 2 repeats was neither significant for the dimension nor for the subscales (Table 3 and Fig. 4).
Fig. 4.
Carriers of a short CAG allele display higher self-transcendence scores than do carriers of a long CAG repeat allele in men recruited from the general population (sample 1). There is no interaction between the 2 AR repeat polymorphisms. For cell sizes, see Table 3.
In sample 2, there was a tendency for an association between self-transcendence and the AR CAG repeat similar to that seen in sample 1; however, this association did not reach statistical significance (p = 0.07).
Discussion
Our data suggest that men recruited from the normal population carrying an AR with relatively short CAG repeats display higher scores on the personality factor extraversion (as assessed using KSP) than those carrying an AR with relatively long CAG repeats, and that this association is dependent on the length of the GGN repeat. That this observation is not a chance finding gains support from the fact that it could be replicated in an independent sample.
Our conclusion that androgens may contribute to interindividual differences in behaviour supports previous investigations that suggested testosterone levels2,3,7,10 and AR variants34,38 are related to personality traits such as sensation seeking, impulsiveness and other externalizing behaviours in men. With respect to associations between AR polymorphisms and personality traits related to extraversion, the literature is not unanimous. On the one hand, at least partly in line with our data, recent studies suggest that men displaying short CAG repeats are at higher risk for committing severe violent and sexual crimes than those with long CAG repeats.54,55 On the other hand, 2 studies report associations between short CAG repeat alleles and psychoticism,36,37 which we did not observe in our samples, and between short GGN repeats and high scores on scales related to aggression, hostility and impulsivity.34 Moreover, a previous study assessing the possible influence of the CAG repeat in the AR on personality traits, as assessed using the KSP, failed to detect any associations that remained significant after correction for multiple comparisons.35 Differences in outcomes among studies may, for example, be related to variations in race or other characteristics of the studied populations and to discrepancies in the assessments of the phenotype.
As described, there is a negative correlation between CAG repeat length and AR responsiveness. Although less established, several investigations show a positive correlation between GGN repeat length and AR activity or protein amount.23–25 Since androgens promote extraversion-related traits, our observation that participants carrying a short CAG repeat and a long GGN repeat were those displaying the highest extraversion scores is thus in line with these functional data. Further studies on how these 2 polymorphisms interact with respect to AR activity in brain are warranted, as are studies of sufficient sample size to permit an analysis of the effect of different allele combinations on behaviour.
In our sample recruited from the general population, the CAG polymorphism was associated with neuroticism (as assessed using KSP). This association did not remain significant after correction for multiple testing, but there was a similar tendency in the sample from the forensic psychiatry group. With respect to 2 of the KSP neuroticism subscales (somatic anxiety and muscular tension), the findings obtained in sample 1 could be replicated in sample 2: short CAG repeats were significantly associated with high scores in both populations. These data support previous reports suggesting that CAG repeat length may affect interindividual variation in anxiety traits35 and indirectly influence the risk for depression in men.39
The significant association between CAG genotype of the AR gene and self-transcendence in sample 1 is noteworthy. The self-transcendence dimension consists of 3 subscales representing different aspects of religious behaviour, subjective experience and individual world view. This dimension is reported to be the most stable TCI dimension over time, and it is 1 of the 2 TCI dimensions showing the largest interindividual variability.42 In the initial presentation of the TCI, self-transcendence — as well as the other character traits — was thought to be influenced mainly by environmental factors.51 However, recent twin studies challenge this view.56,57 Also supporting genes to be of importance for the TCI character traits are previous studies revealing associations between various polymorphisms and these traits, including self-transcendence.33,58–60 The notion that AR may influence religious behaviour and self-transcendence is reasonable in view of the fact that these aspects of the phenotype display marked sex differences.42,57 Moreover, this notion is supported by a previous report by Comings and colleagues33 in which a multivariate analysis comprising 59 different genes revealed an association between the AR repeat polymorphisms and TCI-assessed self-transcendence.
Limitations
When interpreting our observation that certain associations were significant in sample 1 but only close to significant in sample 2, it should be taken into consideration that sample 2 was small and that, being a forensic psychiatry population, this sample probably differs in various aspects from the population-based sample. Moreover, sample 2 was less homogeneous than sample 1 in terms of age. Indeed, the homogeneity with respect to sex, age and race characterizing sample 1 probably enhanced our ability to detect significant associations between genotype and phenotype, thereby compensating for the fact that this population was also moderate in size.
Conclusion
Our study lends support to the notion that the CAG repeat of the AR is associated with KSP-assessed extraversion and that this association is influenced by the GGN repeat in the same gene. We also provide preliminary evidence of an association between the CAG repeat and KSP-assessed neuroticism and between the same repeat and TCI-assessed self-transcendence.
Acknowledgements
Inger Oscarsson and Gunilla Bourghardt are gratefully acknowledged for their excellent technical assistance. The study was supported by grants from the Swedish Science Council (8668), Torsten and Ragnar Söderberg’s Foundation and the Swedish BrainPower consortium.
Footnotes
Competing interests: None declared.
Contributors: Drs. Westberg, Rosmond, Anckarsäter and Eriksson designed the study. Drs. Westberg, Landén, Annerbrink, Melke, Rosmond, Holm, Anckarsäter and Eriksson acquired the data, which Drs. Westberg, Henningsson, Melke, Nilsson, Anckarsäter and Eriksson analyzed. Drs. Westberg and Eriksson wrote the article. All authors reviewed the article and gave final approval for publication.
References
- 1.Pfaff DW. Morphological changes in the brains of adult male rats after neonatal castration. J Endocrinol. 1966;36:415–6. doi: 10.1677/joe.0.0360415. [DOI] [PubMed] [Google Scholar]
- 2.Rubinow DR, Schmidt PJ. Androgens, brain, and behavior. Am J Psychiatry. 1996;153:974–84. doi: 10.1176/ajp.153.8.974. [DOI] [PubMed] [Google Scholar]
- 3.Gerra G, Avanzini P, Zaimovic A, et al. Neurotransmitters, neuroendocrine correlates of sensation-seeking temperament in normal humans. Neuropsychobiology. 1999;39:207–13. doi: 10.1159/000026586. [DOI] [PubMed] [Google Scholar]
- 4.Maras A, Laucht M, Gerdes D, et al. Association of testosterone and dihydrotestosterone with externalizing behavior in adolescent boys and girls. Psychoneuroendocrinology. 2003;28:932–40. doi: 10.1016/s0306-4530(02)00119-1. [DOI] [PubMed] [Google Scholar]
- 5.Olweus D, Mattsson A, Schalling D, et al. Testosterone, aggression, physical, and personality dimensions in normal adolescent males. Psychosom Med. 1980;42:253–69. doi: 10.1097/00006842-198003000-00003. [DOI] [PubMed] [Google Scholar]
- 6.Rowe R, Maughan B, Worthman CM, et al. Testosterone, antisocial behavior, and social dominance in boys: pubertal development and biosocial interaction. Biol Psychiatry. 2004;55:546–52. doi: 10.1016/j.biopsych.2003.10.010. [DOI] [PubMed] [Google Scholar]
- 7.Stalenheim EG, Eriksson E, von Knorring L, et al. Testosterone as a biological marker in psychopathy and alcoholism. Psychiatry Res. 1998;77:79–88. doi: 10.1016/s0165-1781(97)00143-1. [DOI] [PubMed] [Google Scholar]
- 8.van Bokhoven I, van Goozen SH, van Engeland H, et al. Salivary testosterone and aggression, delinquency, and social dominance in a population-based longitudinal study of adolescent males. Horm Behav. 2006;50:118–25. doi: 10.1016/j.yhbeh.2006.02.002. [DOI] [PubMed] [Google Scholar]
- 9.Aluja A, Torrubia R. Hostility-aggressiveness, sensation seeking, and sex hormones in men: re-exploring their relationship. Neuropsychobiology. 2004;50:102–7. doi: 10.1159/000077947. [DOI] [PubMed] [Google Scholar]
- 10.King AC, Errico AL, Parsons OA. Eysenck’s personality dimensions and sex steroids in male abstinent alcoholics and nonalcoholics: an exploratory study. Biol Psychol. 1995;39:103–13. doi: 10.1016/0301-0511(94)00966-2. [DOI] [PubMed] [Google Scholar]
- 11.Chang CS, Kokontis J, Liao ST. Molecular cloning of human and rat complementary DNA encoding androgen receptors. Science. 1988;240:324–6. doi: 10.1126/science.3353726. [DOI] [PubMed] [Google Scholar]
- 12.Lubahn DB, Joseph DR, Sullivan PM, et al. Cloning of human androgen receptor complementary DNA and localization to the X chromosome. Science. 1988;240:327–30. doi: 10.1126/science.3353727. [DOI] [PubMed] [Google Scholar]
- 13.La Spada AR, Wilson EM, Lubahn DB, et al. Androgen receptor gene mutations in X-linked spinal and bulbar muscular atrophy. Nature. 1991;352:77–9. doi: 10.1038/352077a0. [DOI] [PubMed] [Google Scholar]
- 14.Sleddens HF, Oostra BA, Brinkmann AO, et al. Trinucleotide (GGN) repeat polymorphism in the human androgen receptor (AR) gene. Hum Mol Genet. 1993;2:493. doi: 10.1093/hmg/2.4.493. [DOI] [PubMed] [Google Scholar]
- 15.Chamberlain NL, Driver ED, Miesfeld RL. The length and location of CAG trinucleotide repeats in the androgen receptor N-terminal domain affect transactivation function. Nucleic Acids Res. 1994;22:3181–6. doi: 10.1093/nar/22.15.3181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Tut TG, Ghadessy FJ, Trifiro MA, et al. Long polyglutamine tracts in the androgen receptor are associated with reduced trans-activation, impaired sperm production, and male infertility. J Clin Endocrinol Metab. 1997;82:3777–82. doi: 10.1210/jcem.82.11.4385. [DOI] [PubMed] [Google Scholar]
- 17.Callewaert L, Christiaens V, Haelens A, et al. Implications of a polyglutamine tract in the function of the human androgen receptor. Biochem Biophys Res Commun. 2003;306:46–52. doi: 10.1016/s0006-291x(03)00902-1. [DOI] [PubMed] [Google Scholar]
- 18.Irvine RA, Ma H, Yu MC, et al. Inhibition of p160-mediated coactivation with increasing androgen receptor polyglutamine length. Hum Mol Genet. 2000;9:267–74. doi: 10.1093/hmg/9.2.267. [DOI] [PubMed] [Google Scholar]
- 19.Buchanan G, Yang M, Cheong A, et al. Structural and functional consequences of glutamine tract variation in the androgen receptor. Hum Mol Genet. 2004;13:1677–92. doi: 10.1093/hmg/ddh181. [DOI] [PubMed] [Google Scholar]
- 20.Wang Q, Udayakumar TS, Vasaitis TS, et al. Mechanistic relationship between androgen receptor polyglutamine tract truncation and androgen-dependent transcriptional hyperactivity in prostate cancer cells. J Biol Chem. 2004;279:17319–28. doi: 10.1074/jbc.M400970200. [DOI] [PubMed] [Google Scholar]
- 21.Ding D, Xu L, Menon M, et al. Effect of GGC (glycine) repeat length polymorphism in the human androgen receptor on androgen action. Prostate. 2005;62:133–9. doi: 10.1002/pros.20128. [DOI] [PubMed] [Google Scholar]
- 22.Lundin KB, Giwercman A, Dizeyi N, et al. Functional in vitro characterisation of the androgen receptor GGN polymorphism. Mol Cell Endocrinol. 2007;264:184–7. doi: 10.1016/j.mce.2006.11.008. [DOI] [PubMed] [Google Scholar]
- 23.Brockschmidt FF, Nothen MM, Hillmer AM. The two most common alleles of the coding GGN repeat in the androgen receptor gene cause differences in protein function. J Mol Endocrinol. 2007;39:1–8. doi: 10.1677/JME-06-0072. [DOI] [PubMed] [Google Scholar]
- 24.Gao T, Marcelli M, McPhaul MJ. Transcriptional activation and transient expression of the human androgen receptor. J Steroid Biochem Mol Biol. 1996;59:9–20. doi: 10.1016/s0960-0760(96)00097-0. [DOI] [PubMed] [Google Scholar]
- 25.Werner R, Holterhus PM, Binder G, et al. The A645D mutation in the hinge region of the human androgen receptor (AR) gene modulates AR activity, depending on the context of the polymorphic glutamine and glycine repeats. J Clin Endocrinol Metab. 2006;91:3515–20. doi: 10.1210/jc.2006-0372. [DOI] [PubMed] [Google Scholar]
- 26.Chang BL, Zheng SL, Hawkins GA, et al. Polymorphic GGC repeats in the androgen receptor gene are associated with hereditary and sporadic prostate cancer risk. Hum Genet. 2002;110:122–9. doi: 10.1007/s00439-001-0662-6. [DOI] [PubMed] [Google Scholar]
- 27.Clark PE, Irvine RA, Coetzee GA. The androgen receptor CAG repeat and prostate cancer risk. Methods Mol Med. 2003;81:255–66. doi: 10.1385/1-59259-372-0:255. [DOI] [PubMed] [Google Scholar]
- 28.Nelson KA, Witte JS. Androgen receptor CAG repeats and prostate cancer. Am J Epidemiol. 2002;155:883–90. doi: 10.1093/aje/155.10.883. [DOI] [PubMed] [Google Scholar]
- 29.Roberts RO, Bergstralh EJ, Farmer SA, et al. Polymorphisms in genes involved in sex hormone metabolism may increase risk of benign prostatic hyperplasia. Prostate. 2006;66:392–404. doi: 10.1002/pros.20362. [DOI] [PubMed] [Google Scholar]
- 30.Ferlin A, Bartoloni L, Rizzo G, et al. Androgen receptor gene CAG and GGC repeat lengths in idiopathic male infertility. Mol Hum Reprod. 2004;10:417–21. doi: 10.1093/molehr/gah054. [DOI] [PubMed] [Google Scholar]
- 31.Ruhayel Y, Lundin K, Giwercman Y, et al. Androgen receptor gene GGN and CAG polymorphisms among severely oligozoospermic and azoospermic Swedish men. Hum Reprod. 2004;19:2076–83. doi: 10.1093/humrep/deh349. [DOI] [PubMed] [Google Scholar]
- 32.Yong EL, Loy CJ, Sim KS. Androgen receptor gene and male infertility. Hum Reprod Update. 2003;9:1–7. doi: 10.1093/humupd/dmg003. [DOI] [PubMed] [Google Scholar]
- 33.Comings DE, Gade-Andavolu R, Gonzalez N, et al. A multivariate analysis of 59 candidate genes in personality traits: the temperament and character inventory. Clin Genet. 2000;58:375–85. doi: 10.1034/j.1399-0004.2000.580508.x. [DOI] [PubMed] [Google Scholar]
- 34.Comings DE, Muhleman D, Johnson JP, et al. Parent–daughter transmission of the androgen receptor gene as an explanation of the effect of father absence on age of menarche. Child Dev. 2002;73:1046–51. doi: 10.1111/1467-8624.00456. [DOI] [PubMed] [Google Scholar]
- 35.Jonsson EG, von Gertten C, Gustavsson JP, et al. Androgen receptor trinucleotide repeat polymorphism and personality traits. Psychiatr Genet. 2001;11:19–23. doi: 10.1097/00041444-200103000-00004. [DOI] [PubMed] [Google Scholar]
- 36.Loehlin JC, Medland SE, Montgomery GW, et al. Eysenck’s psychoticism and the X-linked androgen receptor gene CAG polymorphisms in additional Australian samples. Pers Individ Dif. 2005;39:661–7. [Google Scholar]
- 37.Turakulov R, Jorm AF, Jacomb PA, et al. Association of dopamine-B-hydroxylase and androgen receptor gene polymorphisms with Eysenck’s P and other personality traits. Pers Individ Dif. 2004;37:191–202. [Google Scholar]
- 38.Comings DE, Chen C, Wu S, et al. Association of the androgen receptor gene (AR) with ADHD and conduct disorder. Neuroreport. 1999;10:1589–92. doi: 10.1097/00001756-199905140-00036. [DOI] [PubMed] [Google Scholar]
- 39.Seidman SN, Araujo AB, Roose SP, et al. Testosterone level, androgen receptor polymorphism, and depressive symptoms in middle-aged men. Biol Psychiatry. 2001;50:371–6. doi: 10.1016/s0006-3223(01)01148-9. [DOI] [PubMed] [Google Scholar]
- 40.Farde L, Gustavsson JP, Jonsson E. D2 dopamine receptors and personality traits. Nature. 1997;385:590. doi: 10.1038/385590a0. [DOI] [PubMed] [Google Scholar]
- 41.Schalling D, Asberg M, Edman G, et al. Markers for vulnerability to psychopathology: temperament traits associated with platelet MAO activity. Acta Psychiatr Scand. 1987;76:172–82. doi: 10.1111/j.1600-0447.1987.tb02881.x. [DOI] [PubMed] [Google Scholar]
- 42.Brandstrom S, Schlette P, Przybeck TR, et al. Swedish normative data on personality using the Temperament and Character Inventory. Compr Psychiatry. 1998;39:122–8. doi: 10.1016/s0010-440x(98)90070-0. [DOI] [PubMed] [Google Scholar]
- 43.Rosmond R, Lapidus L, Marin P, et al. Mental distress, obesity and body fat distribution in middle-aged men. Obes Res. 1996;4:245–52. doi: 10.1002/j.1550-8528.1996.tb00542.x. [DOI] [PubMed] [Google Scholar]
- 44.Soderstrom H, Sjodin AK, Carlstedt A, et al. Adult psychopathic personality with childhood-onset hyperactivity and conduct disorder: a central problem constellation in forensic psychiatry. Psychiatry Res. 2004;121:271–80. doi: 10.1016/s0165-1781(03)00270-1. [DOI] [PubMed] [Google Scholar]
- 45.First MB, Spitzer RL, Gibbon M, et al. User's Guide for the Structured Clinical Interview for DSM-IV Axis I Disorders — Research Version 20 (SCID-I) New York: NY State Psychiatric Inst, Biometrics Research Department; 1996. [Google Scholar]
- 46.American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington (DC): The Association; 1994. [Google Scholar]
- 47.Gustavsson JP, Weinryb RM, Goransson S, et al. Stability and predicive ability of personality traits across 9 years. Pers Individ Dif. 1997;22:783–91. [Google Scholar]
- 48.Melke J, Westberg L, Nilsson S, et al. A polymorphism in the serotonin receptor 3A (HTR3A) gene and its association with harm avoidance in women. Arch Gen Psychiatry. 2003;60:1017–23. doi: 10.1001/archpsyc.60.10.1017. [DOI] [PubMed] [Google Scholar]
- 49.Westberg L, Melke J, Landen M, et al. Association between a dinucleotide repeat polymorphism of the estrogen receptor alpha gene and personality traits in women. Mol Psychiatry. 2003;8:118–22. doi: 10.1038/sj.mp.4001192. [DOI] [PubMed] [Google Scholar]
- 50.Gustavsson JP, Pedersen NL, Asberg M, et al. Origins of individual differences in anxiety proneness: a twin/adoption study of the anxiety-related scales from the Karolinska Scales of Personality (KSP) Acta Psychiatr Scand. 1996;93:460–9. doi: 10.1111/j.1600-0447.1996.tb10678.x. [DOI] [PubMed] [Google Scholar]
- 51.Cloninger CR, Svrakic DM, Przybeck TR. A psychobiological model of temperament and character. Arch Gen Psychiatry. 1993;50:975–90. doi: 10.1001/archpsyc.1993.01820240059008. [DOI] [PubMed] [Google Scholar]
- 52.Westberg L, Baghaei F, Rosmond R, et al. Polymorphisms of the androgen receptor gene and the estrogen receptor beta gene are associated with androgen levels in women. J Clin Endocrinol Metab. 2001;86:2562–8. doi: 10.1210/jcem.86.6.7614. [DOI] [PubMed] [Google Scholar]
- 53.Brownstein MJ, Carpten JD, Smith JR. Modulation of non-templated nucleotide addition by Taq DNA polymerase: primer modifications that facilitate genotyping. Biotechniques. 1996;20:1004–6. doi: 10.2144/96206st01. [DOI] [PubMed] [Google Scholar]
- 54.Cheng D, Hong CJ, Liao DL, et al. Association study of androgen receptor CAG repeat polymorphism and male violent criminal activity. Psychoneuroendocrinology. 2006;31:548–52. doi: 10.1016/j.psyneuen.2005.11.004. [DOI] [PubMed] [Google Scholar]
- 55.Rajender S, Pandu G, Sharma JD, et al. Reduced CAG repeats length in androgen receptor gene is associated with violent criminal behavior. Int J Legal Med. 2008;122:367–72. doi: 10.1007/s00414-008-0225-7. [DOI] [PubMed] [Google Scholar]
- 56.Ando J, Ono Y, Yoshimura K, et al. The genetic structure of Cloninger’s seven-factor model of temperament and character in a Japanese sample. J Pers. 2002;70:583–609. doi: 10.1111/1467-6494.05018. [DOI] [PubMed] [Google Scholar]
- 57.Kirk KM, Eaves LJ, Martin NG. Self-transcendence as a measure of spirituality in a sample of older Australian twins. Twin Res. 1999;2:81–7. doi: 10.1375/136905299320565942. [DOI] [PubMed] [Google Scholar]
- 58.Nilsson KW, Damberg M, Ohrvik J, et al. Genes encoding for AP-2beta and the serotonin transporter are associated with the personality character spiritual acceptance. Neurosci Lett. 2007;411:233–7. doi: 10.1016/j.neulet.2006.10.051. [DOI] [PubMed] [Google Scholar]
- 59.Comings DE, Gonzales N, Saucier G, et al. The DRD4 gene and the spiritual transcendence scale of the character temperament index. Psychiatr Genet. 2000;10:185–9. doi: 10.1097/00041444-200010040-00006. [DOI] [PubMed] [Google Scholar]
- 60.Ham BJ, Kim YH, Choi MJ, et al. Serotonergic genes and personality traits in the Korean population. Neurosci Lett. 2004;354:2–5. doi: 10.1016/s0304-3940(03)00753-5. [DOI] [PubMed] [Google Scholar]