Receptor-Enriched Analysis of functional connectivity by targets (REACT) : A novel, multimodal analytical approach informed by PET to study the pharmacodynamic response of the brain under MDMA
Ottavia Dipasquale, Pierluigi Selvaggi, Mattia Veronese, Anthony S. Gabay,
Federico Turkheimer, Mitul A. Mehta
NeuroImage, 2019, 195, 252–260
A B S T R A C T
One of the main limitations of pharmacological fMRI is its inability to provide a molecular insight into the main effect of compounds, leaving an open question about the relationship between drug effects and haemodynamic response. The aim of this study is to investigate the acute effects of 3,4-methylene-dioxymethamphetamine (MDMA) on functional connectivity (FC) using a novel multimodal method (Receptor-Enriched Analysis of functional Connectivity by Targets – REACT). This approach enriches the resting state (rs-)fMRI analysis with the molecular information about the distribution density of serotonin receptors in the brain, given the serotonergic action of MDMA. Twenty healthy subjects participated in this double-blind, placebo-controlled, crossover study. A high-resolution in vivo atlas of four serotonin receptors (5-HT1A, 5-HT1B, 5-HT2A, and 5-HT4) and its transporter (5-HTT) was used as a template in a two-step multivariate regression analysis to estimate the spatial maps reflecting the whole-brain connectivity behaviour related to each target under placebo and MDMA. Results showed that the networks exhibiting significant changes after MDMA administration are the ones informed by the 5-HTT and 5-HT1A distribution density maps, which are the main targets of this compound. Changes in the 5-HT1A-enriched functional maps were also associated with the pharmacokinetic levels of MDMA and MDMA-induced FC changes in the 5-HT2A-enriched maps correlated with the spiritual experience subscale of the Altered States of Consciousness Questionnaire. By enriching the rs-fMRI analysis with molecular data of voxel-wise distribution of the serotonin receptors across the brain, we showed that MDMA effects on FC can be understood through the distribution of its main targets. This result supports the ability of this method to characterise the specificity of the functional response of the brain to MDMA binding to serotonergic receptors, paving the way to the definition of a new fingerprint in the characterization of new compounds and potentially to a further understanding to the response to treatment.
Keywords,: MDMA, Resting state, fMRI, Functional connectivity, Pharmacological neuroimaging,
Serotonin, Pharmacodynamic response
3,4-methylenedioxymethamphetamine (MDMA) increases levels of serotonin, dopamine and noradrenaline in the brain (Liechti and Vollenweider, 2001). This drug has a history of recreational use because of its positive effects on mood and social interaction, which include feelings of empathy and warmth for others, sociability and a sense of being at peace with the world, as well as euphoria (Rodgers et al., 2006; Sumnall et al., 2006). It can also produce mild visual hallucinatory phenomena, which have been linked to the 5-HT2A receptor, the same receptor targeted by the classic psychedelics (Gouzoulis-Mayfrank et al., 1996; Harris et al., 2002; Liechti and Vollenweider, 2001). While currently classed in the list of Schedule I drugs, indicating no potential medicinal benefits, preliminary evidence of a potential clinical use of MDMA for mental illness (Amoroso and Workman, 2016; Mithoefer et al., 2013) has recently re-ignited interest in its pharmacology, behavioural effects an mechanisms of action.
The serotonergic pharmacology of MDMA is primarily characterised by the inhibition of 5-HT reuptake (Iravani et al., 2000) and the stimulation of 5-HT release in the extracellular space via the serotonin transporter (5-HTT) (Bradbury et al., 2014; Gudelsky and Nash, 1996; Nichols, 1986). Other studies have also shown a weaker affinity of MDMA to 5-HT1A and 5-HT2A receptor sites (Green et al., 2003; Ray, 2010) as well as non-serotonergic effects (Hasler et al., 2009). Some behavioural effects of MDMA such as feelings of empathy, sociability, and interpersonal closeness (Rodgers et al., 2006; Sumnall et al., 2006) are thought to be induced in part by an increase in the levels of the peptide oxytocin (Dumont et al., 2009; Hysek et al., 2012) with 5-HT1A receptor playing an important role in this mechanism (Mottolese et al., 2014; Thompson et al., 2007). .
While MDMA pharmacology has been extensively investigated in terms of its primary targets and affinities, its effect on brain function is less well understood. The investigation of MDMA-induced effects on brain function using resting state fMRI (rs-fMRI) has been focused on specific regions related to social and affective processing, namely the insula, ventromedial prefrontal cortex, hippocampus and amygdala (Carhart-Harris et al., 2015; Walpola et al., 2017). These studies revealed changes in functional connectivity (FC) for circuitry implicated in anxiety and stress-related disorders such as the medial prefrontal cortex, medial temporal regions, amygdala and hippocampus (Carhart-Harris et al., 2015) and a reduction in insula connectivity which correlated with the acute subjective experiences of altered bodily sensations and baseline trait anxiety (Walpola et al., 2017). An independent component analysis (ICA) was performed on the same dataset (Carhart-Harris et al., 2015; Walpola et al., 2017) to evaluate the effects of MDMA on FC between the resting state networks (RSNs). This analysis revealed an increased FC between the executive control network the anterior default mode network (Roseman et al., 2014).
One of the main limitations of pharmacological fMRI is its inability to provide a molecular insight into the main effect of compounds. In fact, this imaging technique relies on the strong assumption that haemodynamic changes can be considered a proxy of altered neurotransmission due to pharmacological agonism or antagonism actions (Attwell and Iadecola, 2002), but the fMRI signal has no intrinsic selectivity to any particular receptor sites. Therefore, the degree to which the fMRI response indexes the action at drug target sites, which could be useful in parsing mechanistic underpinnings, is still an open question.
Here we propose Receptor-Enriched Analysis of functional Connectivity by Targets (REACT), a method that utilises the molecular information about target distribution provided by Positron Emission Tomography (PET) to enrich the pharmacological rs-fMRI analysis. In particular, we tested the degree to which the haemodynamic response to MDMA mirrors the 5-HT receptor density profiles, given its known mixed profile of serotonergic action. We used a publicly available atlas (Beliveau et al., 2017), which provides the 5-HT receptor density profiles for 5-HT1A, 5-HT1B, 5-HT2A and 5-HT4 receptors and the 5-HTT transporter, as target-site distribution maps in a two-step multivariate regression analysis and estimated the spatial maps reflecting the whole-brain connectivity behaviour related to each target, in order to link the knowledge about their distribution with the main effect of MDMA on FC. All the available PET maps were included in the model as we wanted to test the ability of this method to characterise the specificity of the functional response of the brain to MDMA binding to serotonergic receptors. Given MDMA affinity profiles and proposed mediators of MDMA effects at the 5-HTT transporter, the 5-HT1A and 5-HT2A receptor targets (Bradbury et al., 2014; Green et al., 2003; Gudelsky and Nash, 1996; Liechti and Vollenweider, 2001; Nichols, 1986; Ray, 2010; Thompson et al., 2007) we hypothesised that FC informed by these three target maps will be sensitive to MDMA administration.
In addition to the derivation of the spatial maps according to each receptor density distribution, it is possible that the interaction between the targets has a significant influence on FC. For example, both the 5-HTT and the 5-HT1A are important for serotoninergic regulation and their in vivo receptor densities are related both within the brain stem (e.g. raphe nucleus) and at the more distal sites (Bose et al., 2011). Therefore, we set out to explore whether or not overlapping expressions of serotonergic targets are also associated with the main effect of MDMA on FC.
Finally, we investigated whether subjective effects and pharmacokinetic (PK) measures, such as MDMA plasma levels and MDMA-induced increases in the plasma levels of oxytocin, are linked with the pharmacodynamic effects of MDMA in the brain, exploring possible relationships with the 5-HT receptor distribution.