Opposite effects of dopamine and serotonin on resting-state networks: review and implications for psychiatric disorders

Abstract

Alterations in brain intrinsic activity-as organized in resting-state networks (RSNs) such as sensorimotor network (SMN), salience network (SN), and default-mode network (DMN)-and in neurotransmitters signaling-such as dopamine (DA) and serotonin (5-HT)-have been independently detected in psychiatric disorders like bipolar disorder and schizophrenia. Thus, the aim of this work was to investigate the relationship between such neurotransmitters and RSNs in healthy, by reviewing the relevant work on this topic and performing complementary analyses, in order to better understand their physiological link, as well as their alterations in psychiatric disorders. According to the reviewed data, neurotransmitters nuclei diffusively project to subcortical and cortical regions of RSNs. In particular, the dopaminergic substantia nigra (SNc)-related nigrostriatal pathway is structurally and functionally connected with core regions of the SMN, whereas the ventral tegmental area (VTA)-related mesocorticolimbic pathway with core regions of the SN. The serotonergic raphe nuclei (RNi) connections involve regions of the SMN and DMN. Coherently, changes in neurotransmitters activity impact the functional configuration and level of activity of RSNs, as measured by functional connectivity (FC) and amplitude of low-frequency fluctuations/temporal variability of BOLD signal. Specifically, DA signaling is associated with increase in FC and activity in the SMN (hypothetically via the SNc-related nigrostriatal pathway) and SN (hypothetically via the VTA-related mesocorticolimbic pathway), as well as concurrent decrease in FC and activity in the DMN. By contrast, 5-HT signaling (via the RNi-related pathways) is associated with decrease in SMN activity along with increase in DMN activity. Complementally, our empirical data showed a positive correlation between SNc-related FC and SMN activity, whereas a negative correlation between RNi-related FC and SMN activity (along with tilting of networks balance toward the DMN). According to these data, we hypothesize that the activity of neurotransmitter-related neurons synchronize the low-frequency oscillations within different RSNs regions, thus affecting the baseline level of RSNs activity and their balancing. In our model, DA signaling favors the predominance of SMN-SN activity, whereas 5-HT signaling favors the predominance of DMN activity, manifesting in distinct behavioral patterns. In turn, alterations in neurotransmitters signaling (or its disconnection) may favor a correspondent functional reorganization of RSNs, manifesting in distinct psychopathological states. The here suggested model carries important implications for psychiatric disorders, providing novel and well testable hypotheses especially on bipolar disorder and schizophrenia.