Inverse Relationship Between Serotonin 5-HT_1A Receptor Binding and Anxiety: A [¹¹C]WAY-100635 PET Investigation in Healthy Volunteers

Converging evidence indicates that serotonin (5-HT), in particular the 5-HT_1A receptor, plays a role in modulating anxiety. In animal experiments, “knock-out” mice lacking the gene that regulates the expression of the 5-HT_1A receptor display greater anxiety (1, 2). In the open-field and the elevated-plus-maze tests, these mice exhibited less exploratory activity and greater fear of aversive environments as well as less immobility in the forced-swim test. Furthermore, there is clinical evidence for a role of 5-HT_1A receptors in human anxiety. Partial 5-HT_1A receptor agonists, such as buspirone, are known to exert at least modest antianxiety effects (3).

Through positron emission tomography (PET) with [¹¹C]WAY-100635, 5-HT_1A receptors can be investigated in humans in vivo (4). We present a PET study that explored the relationship between 5-HT_1A binding and the personality trait of anxiety in a group of 19 healthy volunteers.

Method

Nineteen healthy subjects (eight women and 11 men; mean age=34 years, range=22–53) participated in the study. They did not suffer from any axis I psychiatric diagnosis as established by the Structured Clinical Interview for DSM-IV. No participant had a medical illness, a significant head injury, or a history of alcohol or substance dependence. They had not been treated with psychotropic medications within the last 3 months before the study, and female volunteers were not pregnant. All subjects gave their written consent after the experimental procedures had been explained. The study was approved by the Research Ethics Board of the Centre for Addiction and Mental Health and the Department of Psychiatry, University of Toronto.

PET images were obtained by using a PC2048-15B camera (GE Medical Systems, Milwaukee) after injection of 9.6 mCi (SD=0.8) of [carbonyl-¹¹C]WAY-100635. Details of radiolabeling, image acquisition, and data analysis are described elsewhere (5).

Four anatomical regions of interest (dorsolateral prefrontal cortex, anterior cingulate cortex, parietal cortex, and occipital cortex) were delineated on a co-registered magnetic resonance imaging scan and transferred to the PET image. Regional binding potential was calculated to estimate the 5-HT_1A receptor number. The cerebellum was used for a simplified reference tissue model (6) because it is relatively devoid of 5-HT_1A receptors (7). This method proved to be superior to kinetic modeling with arterial data (8). Furthermore, the simplified reference tissue model provided excellent test-retest reproducibility with [¹¹C]WAY-100635 (5).

The self-report version of the Revised NEO Personality Inventory (9) was used to assess personality. It is composed of 240 items and contains five dimensional scales (neuroticism, extraversion, openness to experience, agreeableness, and conscientiousness) that correspond to a five-factor model of personality. Within the neuroticism dimension are six underlying personality trait facets: anxiety, anger-hostility, depression, self-consciousness, impulsiveness, and vulnerability. Revised NEO Personality Inventory results are presented as T scores with a mean of 50 and a standard deviation of 10.

Pearson correlation coefficients (r) were calculated to explore relationships between 5-HT_1A receptor binding potential and Revised NEO Personality Inventory measures. Since an age-dependent decline of cortical 5-HT_1A receptor binding potential had been previously established (5), partial correlation coefficients were calculated in which age was controlled. To test the specificity of the relationship between regional 5-HT_1A binding potential and the trait facet of anxiety, partial correlation coefficients were calculated after both the effects of the five other neuroticism facets and the effects of those five facets combined with age were controlled.

Results

The 5-HT_1A binding potential for the four anatomical regions of interest was as follows: dorsolateral prefrontal cortex: mean=3.1, SD=0.5; anterior cingulate cortex: mean=3.8, SD=0.5; parietal cortex: mean=3.0, SD=0.6; occipital cortex: mean=2.1, SD=0.3. Mean T scores for the Revised NEO Personality Inventory neuroticism dimension and the anxiety facet were 56.2 (SD=4.4) and 50.7 (SD=5.0), respectively.

As seen in Table 1, 5-HT_1A receptor binding potential was negatively correlated with the neuroticism facet of anxiety in all four regions of interest. After the five neuroticism facets other than anxiety were controlled, partial correlation coefficients remained significant for the dorsolateral prefrontal cortex, parietal cortex, and occipital cortex. The combined effects of age and the five other neuroticism personality facets were then statistically removed. The results still revealed a significant partial correlation coefficient for the parietal cortex. None of the reported results survived a Bonferroni correction for multiple comparisons.

Discussion

In this preliminary investigation we found an inverse relationship between anxiety—a facet of the Revised NEO Personality Inventory dimension of neuroticism—and cortical 5-HT_1A receptor binding potential. Hence, we infer that people with lower 5-HT_1A receptor density are more likely to display higher levels of anxiety.

Although none of the presented correlations survived a Bonferroni correction, and a large number of correlations may lead to a higher chance of finding spuriously significant associations, the correlations between 5-HT_1A receptor binding potential and anxiety as measured by the Revised NEO Personality Inventory (correlation range=–0.49 to –0.60) may be regarded as large (10).

Differences in region of interest selection, resolution, and sensitivity of PET scanners as well as the subjects’ age limit the inferences that can be drawn from comparisons of results obtained at different centers that used [¹¹C]WAY-100635 PET. Whereas our mean binding potential value in frontal cortex was comparable to that found in a previous study that used a transient equilibrium analysis (4), 5-HT_1A binding potential in the anterior cingulate cortex was lower than previously published (8). The T scores for personality dimensions and the anxiety facet were within the “normal” range.

Our findings are in line with results from animal studies that have shown greater anxiety in mice lacking 5-HT_1A receptors (1, 2). Furthermore, our results may help explain why partial 5-HT_1A receptor agonists have been found to be clinically effective antianxiety medications (3).

Three twin studies implicated that approximately one-half of the variance specific to anxiety can be explained by heritance (11–13), thus pointing towards a biological substrate modulating this personality trait. Our results are also consistent with recently reported findings that anxiety sensitivity has a significant heritable component (14). The current study cannot determine whether 5-HT_1A receptors are “the missing link” for a genetic susceptibility toward higher anxiety. Further studies exploring a possible genetic and molecular base related to high levels of anxiety seem warranted as further verification of the relation between 5-HT_1A receptors and anxiety.

TABLE 1

Enlarge table