Serotonin Transporter Genetic Variation and the Response of the Human Amygdala

See allHide authors and affiliations

Science  19 Jul 2002:
Vol. 297, Issue 5580, pp. 400-403
DOI: 10.1126/science.1071829


A functional polymorphism in the promoter region of the human serotonin transporter gene (SLC6A4) has been associated with several dimensions of neuroticism and psychopathology, especially anxiety traits, but the predictive value of this genotype against these complex behaviors has been inconsistent. Serotonin [5- hydroxytryptamine, (5-HT)] function influences normal fear as well as pathological anxiety, behaviors critically dependent on the amygdala in animal models and in clinical studies. We now report that individuals with one or two copies of the short allele of the serotonin transporter (5-HTT) promoter polymorphism, which has been associated with reduced 5-HTT expression and function and increased fear and anxiety-related behaviors, exhibit greater amygdala neuronal activity, as assessed by BOLD functional magnetic resonance imaging, in response to fearful stimuli compared with individuals homozygous for the long allele. These results demonstrate genetically driven variation in the response of brain regions underlying human emotional behavior and suggest that differential excitability of the amygdala to emotional stimuli may contribute to the increased fear and anxiety typically associated with the short SLC6A4 allele.

The elucidation of underlying biological mechanisms that contribute to individual differences in both normal and abnormal behavior remains a crucial and largely unmet challenge. Advances in both molecular genetics and noninvasive functional neuroimaging, however, have begun to address this issue, particularly in regards to affective behaviors, such as fear and anxiety, which exhibit considerable individual variability (1). In humans, two common alleles, the short (s) and long (l), in a variable repeat sequence of the promoter region of the serotonin transporter gene (5-HTTLPR) have been differentially associated with anxiety-related behavioral traits in healthy subjects (2, 3). Likewise, both positron emission tomography and functional magnetic resonance imaging (fMRI) studies have revealed links between the physiological responses of brain regions, such as the prefrontal cortex and amygdala, and individual differences in affect and temperament (4, 5).

At the physiological level, the serotonin transporter promoter polymorphism alters both SLC6A4 transcription and level of serotonin transporter function. Cultured human lymphoblast cell lines homozygous for the l allele have higher concentrations of 5-HTT mRNA and express nearly twofold greater 5-HT reuptake compared with cells possessing either one or two copies of the s allele, which may act dominantly (2). Nearly identical differences in 5-HTT binding levels in human brain were detected between individuals with the l/l versus l/s or s/s genotypes, using in vivo SPECT imaging (6) and postmortem ligand binding (7). At the behavioral level, individuals carrying the s allele are slightly more likely to display abnormal levels of anxiety (2, 8–10), acquire conditioned fear responses (11), and develop affective illness (3) compared with those homozygous for the l allele. These behavioral and personality differences most likely reflect relative differences in 5-HTT expression and subsequent levels of synaptic 5-HT, a potent modulator of emotional behavior (12). Not surprisingly, however, the relation betweenSLC6A4 genotype and subjective measures of emotion and personality has been weak and inconsistent (8,13–16), likely reflecting the vagueness and subjectivity of the behavioral measurements, but also raising concern that the relation may be spurious (17).

Although the potential influence of a genetic variation in 5-HTT function on human anxiety and fear behavior has been corroborated in studies of 5-HTT knockout mice (18), the underlying neurobiological correlates of this functional relation are unknown. Recent human (11) and animal studies (19) have revealed abnormal fear conditioning, a phenomenon dependent on the amygdala (20), to be associated with 5-HTT function, suggesting that this structure may be critical in mediating the effects of 5-HT on emotional behavior. Because the physiological response of the amygdala during the processing of fearful stimuli may be more objectively measurable than the subjective experience of emotionality, a functional polymorphism in SLC6A4 may have a more obvious impact at the level of amygdala biology than at the level of individual responses to questionnaires or ratings of emotional symptoms.

We used fMRI to explore the relation between this functional polymorphism and the response of the amygdala to fearful stimuli in two independent cohorts of healthy volunteers (21). We hypothesized that individuals with either one or two copies of the s allele, who presumably have relatively lower 5-HTT function and expression and relatively higher levels of synaptic 5-HT and who have been associated in earlier studies with more anxiety and fearfulness, would exhibit a greater amygdala response than those homozygous for the l allele, who presumably have lower levels of synaptic 5-HT and have been reported to be less anxious and fearful.

In each cohort, subjects were divided into two equal groups on the basis of their 5- HTTLPR genotype (22): individuals with either one or two copies of the s allele (s group) and individuals homozygous for the l allele (l group). The two groups were matched for age, gender, and IQ in each cohort (23). During fMRI, all subjects completed a blocked paradigm in which a sensorimotor control task was interleaved with an emotion task that required subjects to match the affect (angry or afraid) of one of two faces to that of a simultaneously presented target face (Fig. 1). Such perceptual processing of facial expressions has been shown to effectively and consistently engage the amygdala (24, 25).

Figure 1

fMRI paradigm. Two blocks of an emotion task were interleaved with three blocks of a sensorimotor control task. (A) During the emotion task, subjects viewed a trio of faces and selected one of two faces (bottom) that expressed the same emotion as the target face (top). The identity of all three faces was always different. Each emotion block consisted of six images, three of each gender and target affect (angry or afraid) all derived from a standard set of pictures of facial affect (42), presented sequentially for 5 s. (B) During the sensorimotor control, the subjects viewed a trio of simple geometric shapes (circles, vertical and horizontal ellipses) and selected one of two shapes (bottom) identical to the target shape (top). Each control block consisted of six different images presented sequentially for 5 s. Subject performance (accuracy and reaction time) was monitored during all scans.

Analysis of the imaging data (22) revealed a similar and significant bilateral BOLD (blood oxygen level–dependent) response in the amygdala during the emotion task in each cohort. In addition, there were significant responses in the bilateral posterior fusiform gyri, inferior parietal lobules, and frontal eye fields—a network of brain regions implicated in face processing (26). Direct comparisons revealed that the response of the right amygdala was significantly greater in the s carrier group compared with the l homozygous group in each cohort (Figs. 2and 3). The laterality of this amygdala difference is consistent with the proposed general role of right hemisphere brain regions in processing faces, as well as recent reports implicating a specific role for the right amygdala in processing angry and fearful facial expressions (27, 28). The responses of regions within the distributed perceptual network, including the right posterior fusiform gyrus, were also greater in the s groups, possibly reflecting excitatory feedback from the amygdala to object-specific processing regions in an effort to improve recognition and refine behavioral responses (20, 29,30). There were no significant effects of gender or a gender-by-genotype interaction for any of these brain responses. Furthermore, the 5-HTT allele–dependent effect on amygdala excitability did not reflect a nonspecific tendency of s allele carriers to overactivate a complex brain response, as a subsequent fMRI analysis of these same subjects, grouped again by 5-HTT genotype, performing a working memory task revealed no significant group differences in any brain regions involved with the task (31).

Figure 2

Genotype-based parametric comparisons illustrating significantly greater activity in the right amygdala of the s group versus the l group in both the first and second cohort. BOLD fMRI responses in the right amygdala (white circle) are shown overlaid onto an averaged structural MRI in the coronal plane through the center of the amygdala. Talairach coordinates and voxel level statistics (P < 0.05, corrected) for the maximal voxel in the right amygdala for the first and second cohort are as follows:x = 24 mm, y = −8 mm,z = −16 mm; cluster size = 4 voxels; voxel level corrected P value = 0.021; T score = 2.89, andx = 28 mm, y = −4 mm,z = −16 mm; cluster size = 2 voxels; voxel level corrected P value = 0.047; T score = 2.03, respectively.

Figure 3

Effect of 5-HTT genotype on right amygdala activity. Bar graphs represent the mean BOLD fMRI percent signal change in a region of interest (ROI) comprising the entire right amygdala in the s (n = 14) and l (n = 14) groups collapsed across both cohorts. Individual circles represent the activity for each subject in this ROI. Consistent with the statistical parametric maps (Fig. 2), which identified significant voxels within the right amygdala, analysis of variance for the entire amygdala ROI, including voxels that were not differentially activated according to statistical parametric mapping, still revealed significant group differences in the mean (±SEM) BOLD fMRI percent signal change [s group = 0.28 ± 0.08 and l group = 0.03 ± 0.05; F(1,26) = 6.84, P = 0.01].

The genotype groups did not differ in performance (accuracy and reaction time) on the emotion task (32), indicating that general attentional, perceptual, and cognitive phenomena did not contribute to the observed amygdala differences. The lack of a genotype effect on the brain response during the working memory task further supports the conclusion that the results are not driven by nonspecific factors. Moreover, there were no significant group differences in anxiety-like or fear-related traits, as indexed by the Tridimensional Personality Questionnaire (33). However, given the small effect (3 to 4%) of this 5-HTT polymorphism on behavior in previous studies (2), a lack of significant genotype differences in these personality traits is not surprising in view of the considerable individual variability in these measures and our relatively small sample size.

The heightened amygdala response of individuals possessing the s allele most likely reflects increased neuronal excitability leading to larger local field potentials and subsequent increases in the BOLD fMRI signal (34). Relatively increased amygdala neuronal excitability in s carriers may result from the relatively decreased 5-HTT expression and increased available synaptic 5-HT acting on excitatory 5-HT receptor subtypes (35). Such heightened amygdala activity might also reflect partial desensitization of inhibitory 5-HT1A receptors following increased synaptic 5-HT (36). Furthermore, the differential response of the amygdala that we observed in adult subjects may be rooted in early postnatal developmental processes that are critical for establishing emotional behavior and are influenced by serotonergic neurotransmission (37).

Our results directly implicate a genetically determined link between 5-HTT function and the response of brain regions critical for emotion processing. Specifically, individuals carrying the less efficient s allele of the 5-HTT gene promoter exhibit an increased amygdala response to fearful stimuli compared with those homozygous for the l allele. Thus, the increased anxiety and fear associated with individuals possessing the s allele may reflect the hyperresponsiveness of their amygdala to relevant environmental stimuli. Such genetically driven variation in 5-HTT function and subsequent amygdala reactivity may also contribute to previously reported abnormalities of the serotonergic system in depression and suicidal behavior (38). The differences we describe at the neurobiological level were marked in a relatively small sample population in the absence of significant differences in behavioral measures of personality, underscoring the power of a direct assay of brain function (i.e., fMRI) to identify a phenotype related to a functional polymorphism in a gene likely important for human emotion. The application of such techniques appears to provide a unique opportunity to explore and evaluate the functional impact of brain-relevant genetic polymorphisms more rapidly and with greater sensitivity than existing behavioral assessments (39).

  • * To whom correspondence should be addressed at Clinical Brain Disorders Branch, National Institute of Mental Health, 10 Center Drive, Room 4S235, Bethesda, MD 20892–1384, USA. E-mail: weinberd{at}


View Abstract

Navigate This Article