First Time View on Human Metabolome Changes after a Single Intake of 3,4-Methylenedioxymethamphetamine in Healthy Placebo-Controlled Subjects

This randomized, double-blind, placebo-controlled crossover study (n=15) investigated changes in endogenous plasma metabolites following a single intake of MDMA (125 mg) and found an overall increase in oxidative stress indicated by the metabolic ratio of methionine-sulfoxide over methionine.

Abstract

“Introduction: 3,4-Methylenedioxymethamphetamine (MDMA; “ecstasy”) is widely consumed recreationally. Little is known about its effects on the human metabolome. Mapping biochemical changes after drug exposure can complement traditional approaches by revealing potential biomarkers of organ toxicity or discovering new metabolomic features in a time- and dose-dependent manner.

Methods: We aimed to analyze for the first time plasma samples from a randomized, double-blind, placebo-controlled crossover study in healthy adults to explore changes in endogenous plasma metabolites following a single intake of MDMA. Plasma samples from 15 subjects taken at four different time points were analyzed with the commercially available AbsoluteIDQ kit (Biocrates).

Results: Time series analysis revealed a total of nine metabolites, which showed a significant concentration change after MDMA administration compared with placebo. Paired t tests of the single time points showed statistically significant concentration changes mainly of glycerophospholipids and the metabolic ratio of methionine-sulfoxide over methionine.

Discussion: Changes of this metabolic ratio may be indicative for changes in systemic oxidative stress levels, and the increased amount of glycerophospholipids could be interpreted as an upregulation of energy production. Baseline samples within the experimental study design were crucial for evaluation of metabolomics data as interday individuality within subjects was high otherwise resulting in overestimations of the findings.”

Authors: Martina I. Boxler, Matthias E. Liechti, Yasmin Schmid, Thomas Kraemer & Andrea E. Steuer

Summary

■ INTRODUCTION

The recreational psychoactive drug 3,4-methylenedioxymethamphetamine (MDMA; “ecstasy”) releases serotonin, norepinephrine, and dopamine presynaptically by interacting with the corresponding membrane transporters. It also produces sympathomimetic toxicity and an acute endocrine stress response.

Metabolomics research focuses on high-throughput identification of small molecular weight molecules. It may complement traditional approaches to drug abuse research by revealing potential biomarkers of organ toxicity, discovering new metabolites in time- and dose-dependent manner and different pharmacodynamic targets, and giving insights about the pathways implicated in the mechanism of action, adverse effects, and variability of the drug response.

We analyzed plasma samples from young healthy adults in a randomized, double-blind, placebo-controlled crossover study to explore changes in plasma metabolites in response to a single intake of MDMA and compare these results to metabolic changes in the same subjects after placebo ingestion.

Clinical Study and Sample Collection

The metabolomics study used plasma from two sessions of a double-blind, placebo-controlled crossover study. The wash-out periods between the sessions were at least 10 days, and participants were under surveillance and served a standardized meal at 12:30 p.m. The clinical study was conducted at the University Hospital of Basel in Switzerland and was approved by the Ethics Committee and the Swiss Agency for Therapeutic Products.

Sample Preparation and Analysis

A targeted metabolomics approach was used to analyze 188 metabolites in human blood plasma using the AbsoluteIDQ p180 kit (Biocrates, Life Sciences AG, Innsbruck, Austria). The nomenclature of the lipid metabolites is as follows: acylcarnitines, biogenic amines, acylcarnitines, carbohydrates, and phospho-and sphingolipids.

Plasma samples were randomized and transferred onto a 96-well kit plate containing the stable isotope labeled internal standards (IS). The samples were derivatized with 5% phenylisothiocyanante reagent and extracted with 5 mM ammonium acetate in methanol and centrifuged through the filter plate (2 min, 500g).

Metabolites from the AA and biogenic amine groups were quantified by multiple reaction monitoring (MRM) and flow injection analysis (FIA-MS/MS) using the MetIDQ software package. Three kits were used to measure all samples from 15 out of 16 participants.

Statistical Methods

Data were normalized to median, log-transformed, and checked for normality. 19 metabolic ratios were calculated, and statistical tests were performed using the open source statistical software R and the webtool metaboanalyst 3.0.23.

Paired fold change analysis was calculated from nontransformed plasma concentrations, and statistically significant within-person differences were identified using paired t tests and two-way repeated measure analysis of variance.

After evaluation of different scaling methods, autoscaling was applied prior to principal component analysis (PCA) and partial least-squares discriminant analysis (PLS-DA). PLS-DA was used to investigate whether the MDMA intake led to significant metabolic changes 3 h, 8 h, or 24 h after consumption.

■ RESULTS

We profiled 188 metabolites in blood plasma of 15 study participants from a controlled MDMA administration study to humans. Six metabolites were excluded from further statistical analysis due to high CV.

Metabolite Pool Size Changes

All baseline-corrected metabolites were tested with two-way repeated measure ANOVA (within subject), and FDR for multiple testing corrections was applied. Nine metabolites showed a significant concentration change between the two sessions, 23 metabolites showed a significant concentration change between the sessions and the time points, and 88 metabolites showed no changes.

Effect of MDMA Intake at Different Time Points

Time points of both sessions were compared using PCA, PLS-DA, and paired t test. The results showed that there was no clear clustering between the MDMA and placebo intake at any of the measured time points.

The concentrations of 19 metabolites were statistically different between the two sessions before the intake of MDMA or placebo, indicating presence of high interday variabilities.

The concentration of 64 metabolites changed significantly between the two sessions, mainly glycerophospholipids PC aa, PC ae, Asn, and the metabolic ratio Met-SO/Met. PLS-DA showed no significant separation of the two sessions.

A total of 64 metabolites showed significant differences between MDMA and placebo intake. PLS-DA was able to separate the two groups with R2 = 0.88 and Q2 = 0.75 showing a good model.

Paired t test revealed 30 metabolites to be significantly different between the sessions, although cross-validation testing showed a relatively good data fit of the PLS-DA model.

MDMA and placebo intake resulted in a similar concentration of 42 metabolites, but the concentration of 30 metabolites had a VIP value >1.5, and the PLS-DA model showed no complete separation between the two sessions.

There is limited information on the influence of drug of abuse consumption on the metabolome. However, the first study with MDMA in humans is pretty rare.

We analyzed plasma samples from a highly controlled clinical study on MDMA to reveal metabolic changes. We found that the dose administered was relatively high compared to single, recreational MDMA doses, and that sufficient sample stability can be assumed based on previously published studies.

Our data set was multilevel and contained multiple types of variation. A combination of univariate and multivariate statistical methods was used to detect as many metabolite changes as possible. Multivariate analysis using PCA showed no clear clustering of the two sessions at any time point after the MDMA or placebo intake. PLS-DA was used to identify components that would separate the two sessions, but could not clearly separate the two sessions on the corrected data set.

Many AA as well as taurine, creatinine, and kynurenine showed significant concentration changes between time points, but Trp did not. This indicates that Trp changes are time-dependent and independent of MDMA intake.

The metabolic ratio Met-SO/Met showed significant changes in systemic oxidative stress at time points 1 and 2. This ratio is a key indicator of oxidative stress and is correlated with various pathologic conditions such as Alzheimer’s disease or biological aging. Similar changes of Met after MDMA consumption were reported elsewhere. The levels of Met-SO in plasma samples of abstinent and ecstasy user groups did not differ significantly, and a negative correlation with the cumulative dose and consumption in the last 12 months was found.

After MDMA consumption, the amount of glycerophospholipids increased, which could be associated with increased energy production. The increased activity and cardiac stress could be compensated by -oxidation of fatty acids in muscle, heart, and liver.

Two lysoPCs showed baseline-corrected significant concentration increases 8 h after MDMA intake, which is conciliable with MDMA consumption. Lyso-phosphatidylcholine is a risk factor for vascular diseases, such as acute coronary syndrome, coronary artery disease, and acute aorta dissection.

Serotonin, a metabolite of MDMA, showed no significant changes in any statistical analyses of the different sessions and time points.

Targeted metabolomics measures only a defined set of metabolites, typically focusing on pathways of interest. Untargeted metabolomics allows searching for affected metabolites of for example inflammatory or hormonal pathways which would be of interest.

The metabolome is sensitive to both internal (such as sex, age, and genetics) and external factors (such as diet, lifestyle, and analytical procedures). The participants of this study were in the same age range and gender was balanced, which reduced confounding factors. Metabolomics studies often lack a suitable study design, such as zero/start point-samples for the comparison before and after treatment, and a minimum of three to five replicates with a preference of biological replication over technical replication.

Using baseline-corrected data, we aimed to overcome metabolic differences between the session days of the participant, and showed that a suitable “zero sample” is very important for the outcome of statistical calculations.

Metabolic ratios were increased, oxidative stress was increased, and many metabolites changed in concentration time dependently, but were not strongly altered by a single intake of MDMA.

*S Supporting Information

The authors thank Dr. Sandra Staeheli, Dr. Michael Poetzsch, Emma Louise Kessler, MD, Jakob Danbon, Yulia Kulagina, Jean Garret, and Simon Hediger for their support and helpful discussion, and the Swiss National Science Foundation for funding.

■ ABBREVIATIONS

MDMA, 3,4-methylenedioxymethamphetamine, dopamine, 5-HT, norepinephrine, lyso-phosphatidylcholine, methionine sulfoxide, ROS, reactive oxygen species, Msr, methionine sulfoxide reductase, and FIA were measured in liquid chromatography.