An integrative analysis of 5HTT-mediated mechanism of hyperactivity to non-threatening voices

Abstract

The tonic model delineating the serotonin transporter polymorphism's (5-HTTLPR) modulatory effect on anxiety points towards a universal underlying mechanism involving a hyper-or-elevated baseline level of arousal even to non-threatening stimuli. However, to our knowledge, this mechanism has never been observed in non-clinical cohorts exhibiting high anxiety. Moreover, empirical support regarding said association is mixed, potentially because of publication bias with a relatively small sample size. Hence, how the 5-HTTLPR modulates neural correlates remains controversial. Here we show that 5-HTTLPR short-allele carriers had significantly increased baseline ERPs and reduced fearful MMN, phenomena which can nevertheless be reversed by acute anxiolytic treatment. This provides evidence that the 5-HTT affects the automatic processing of threatening and non-threatening voices, impacts broadly on social cognition, and conclusively asserts the heightened baseline arousal level as the universal underlying neural mechanism for anxiety-related susceptibilities, functioning as a spectrum-like distribution from high trait anxiety non-patients to anxiety patients.

Fig. 1. Flow-chart of participant selection for the study.

This study assessed state and trait anxiety (STAI) in three hundred thirty-four healthy volunteers, aged between 19–63 (mean ± SD: 27.1 ± 9.9, 155 males) years old. Subsequently, we genotyped the 5-HTTLPR and recorded the event-related potentials (ERPs) in one hundred eighty-eight of them, aged between 19–46 (23.4 ± 3.5, 92 males) years old, and which were included in the data analysis. One hundred and twenty-eight of these exhibited high trait anxiety scores (STAI-T ≥ 41), and which were used for the path analysis. Finally, 13 of them went ahead to participate in the neuropharmacological testing.

Fig. 2. Association between the 5-HTTLPR and emotional MMN.

a Subtracting neutral ERPs from fearful and angry ERPs determines fearful and angry MMN, respectively. b L/L homozygotes (4.7 ± 0.44 μv) exhibit stronger MMN than S allele carriers (LS: 2.99 ± 0.25 μv; SS: 2.96 ± 0.24 μv), irrespective of the deviant type. There are significant interactions among the deviant type, coronal site, and genotype (F_{4, 368} = 3.43, P = 0.01, ηp² = 0.036, (1–β) ≈ 100%). Post hoc analyses indicate that, at F4 and C4 electrodes, fearful and angry MMN are comparable in the S/S group (2.98 ± 0.24 vs. 2.94 ± 0.23 μV: t₈₆ = 0.24, P = 0.81), but significantly different in the L/L (5.4 ± 0.73 vs. 4.56 ± 0.71: t₂₄ = 2.80, P = 0.01) and L/S (3.13 ± 0.26 vs. 2.71 ± 0.25: t₇₄ = 1.97, P = 0.053), with larger amplitudes in fearful than angry MMN.

Fig. 3. Fearful MMN as a function of neutral and fearful ERPs.

a Fearful MMN amplitudes are positively correlated with fearful ERP (r₁₈₇ = 0.47, P < 0.001), and negatively correlated with neutral ERP (r₁₈₇ = −0.57, P < 0.001). Fisher r-to-z transformation confirmed that both of fearful and neutral ERPs independently contribute to fearful MMN (Δz = 11.1, P < 0.01). Larger fearful MMN are ascribed to increased fearful as well as to reduced neutral ERP amplitudes. b The same pattern emerges in an independent dataset (n = 30, 16 males). The negative emotionality (fearful vs. neutral) in the amygdala varies as a function of neutral and fearful face processing in a previously collected dataset. The amygdala reactivity to explicitly perceived emotionality (fearful vs. neutral) is positively correlated with the response to fearful faces (r = 0.49, P = 0.006) but negatively correlated with the response to neutral faces (r = −0.73, P < 0.001). Fisher r-to-z transformation confirms that both of the responses to fearful and neutral faces independently contribute to explicitly perceived emotionality (Δz = 5.38, P < 0.01). Meanwhile, the amygdala reactivity to fearful facial expressions (explicit fear vs. neutral faces) is closely associated with MMN to fearful vocal expressions (fearful vs. neutral voices) (r = 0.58, P = 0.008).

Fig. 4. Neutral ERPs and fearful MMN in each genotype in individuals with high trait anxiety.

The S/S (neutral ERP: 0.37 ± 0.45 μV; fearful MMN: 3.1 ± 0.42 μV) exhibit larger neutral ERP and weaker fearful MMN than the L/S (−0.01 ± 0.49; 3 ± 0.46 μV) and L/L (−2.23 ± 0.99; 4.9 ± 0.93 μV).

Fig. 5. Lorazepam impacts on hyperreactivity to neutral non-threatening voices in non-clinical individuals with high trait anxiety.

While there is no significant effect of acute lorazepam treatment on neither fearful MMN (F_{1, 12} = 1.89, P = 0.19) nor angry MMN (F_{1, 12} = 0.87, P = 0.37), acute lorazepam administration significantly reduces the neutral ERP amplitudes in individuals with high trait anxiety (lorazepam vs. placebo: −0.15 ± 0.33 vs. 0.23 ± 0.31 μV; F_{1, 12} = 5.00, P = 0.045, ηp² = 0.294, (1−β) ≈ 98.91%).