Modulating factors in MDMA induced neurotoxicity

(1)

Literature thesis

Date 17-01-2020

Student name Sebastiaan van Bruchem Student number 10213929

Email srvanbruchem@gmail.com

Supervisor Prof. dr. Liesbeth Reneman Second assessor Prof. dr. Paul Lucassen

Institution Amsterdam UMC

Department Radiology and Nuclear Medicine

Address Location AMC

Meibergdreef 9 1105 AZ, Amsterdam

Modulating factors in

MDMA induced

(2)

(3)

2

Abstract

MDMA is an illicit drug with widespread use in recreational and therapeutic settings. Because of its widespread use, it is important to gain insight into the possibly damaging effects of MDMA. To date, research into the neurotoxic effects of MDMA has been inconsistent however. Animal studies clearly indicate selective serotonergic neurotoxicity produced by MDMA. Human studies find memory problems in MDMA users. However, research using markers of serotonergic damage in humans give inconsistent results. The discrepancy between animal and human research is most likely caused by the wide variety of factors modulating an individual’s susceptibility to the neurotoxic effects of MDMA. This thesis gives an overview of some of the most important modulating factors.

Toxicity of MDMA is mainly caused by neurotoxic metabolites of MDMA and excessive dopamine influx into serotonergic neurons. Polymorphisms of CYP2D6 and COMT, both enzymes involved in MDMA metabolism, greatly influence susceptibility to the neurotoxic effects of MDMA. CYP1A2, CYP2B6, CYP3A4, and CYP2C19 polymorphisms are minor modulating factors, while there is an indication MAO-B polymorphism is also a modulating factor in MDMA induced neurotoxicity. Women seem to be more susceptible to men with regards to the neurotoxic effects of MDMA, at least at higher doses of MDMA. African American people, and Asian people at higher doses of MDMA, seem less susceptible than Caucasians. Ambient temperature and excessive physical exertion often seen at dance parties exacerbate MDMA induced neurotoxicity. Lastly, most MDMA users are polydrug users and this greatly influences susceptibility to the neurotoxic effects of MDMA. Cannabis and low/moderate doses of alcohol seem protective, while stimulants and psychedelics exacerbate MDMA induced neurotoxicity.

The information presented in this thesis can be valuable in assessing the impact of modulating factors in susceptibility to the neurotoxic effects of MDMA. By correcting for these modulating factors research accuracy can be greatly improved, ultimately leading to an improved understanding of the damaging effects of MDMA. In addition, this information can be used to improve harm-reduction practices for users which is especially relevant for use in therapeutic setting.

(5)

4

Introduction

MDMA (3,4-methylenedioxymethamphetamine) or “ecstasy” is a psychoactive drug which induces euphoria, empathy, increased energy and sometimes mild hallucinations in its users. It was first described in a patent by Merck as a by-product and chemical intermediate in the synthesis of hydrastinine analogues from safrole. The patent for the synthesis of MDMA by Merck stems from 1912. It was not until 1978 that it’s effects were first described in humans (Shulgin & Nichols 1978), though reports of its appearance on the illicit drug market already appeared date back to 1972 (Gaston & Rasmussen 1972). Besides its use as a club drug, in the mid 1970’s Alexander Shulgin began distributing MDMA to psychotherapists recognizing the value of MDMA as an adjunct in psychotherapy (Benzenhöfer & Passie 2010). In the years following the value of MDMA for its use in psychotherapy became more recognized (e.g. Grinspoon & Bakalar 1986, and more recently White 2014; Mithoefer et al. 2018). Its recreational use began to rise rapidly during the 1980’s as well, with the drug being mass produced and sold openly in bars in the US in the mid 1980’s (Pentney 2001). Because of its increasing use and concerns about the safety of MDMA, the DEA decided on an emergency ban on MDMA in the US in 1985. In 1987 the ban was lifted to support research into possible benefits of its use in psychotherapy, but this was reversed in 1988, citing lack of evidence regarding the beneficial effects as an adjunct in psychotherapy and lack of toxicological studies as among their main reasons (Lawn 1988). Despite the ban on MDMA, use of MDMA has spread with the Global Drug Survey 2019 reporting that MDMA is the second most popular used illicit drug after cannabis worldwide (Winstock et al. 2019). The widespread use of MDMA has prompted researchers to gain insight into the possibly damaging effects of MDMA. Most research seems to indicate MDMA is neurotoxic in humans. However, inconsistent research results, partially caused by the high amount of confounding factors in MDMA research, make it harder to come to a convincing conclusion with regards to the neurotoxic effects of MDMA.

Indirect evidence of neurotoxicity: cognitive problems due to MDMA

Use of MDMA is not without its risks. After use, users can experience heavy negative side-effects like anhedonia, lethargy, and depression (Parrott 2014). This could have especially negative consequences when used for therapeutic purposes. There are also concerns of possible neurotoxicity as a result of recreational MDMA use. Early research into the possibly damaging effects of MDMA indicated MDMA users show poorer memory performance compared to controls (Gouzoulis-Mayfrank et al. 2000; Reneman et al. 2000; Reneman et al. 2001b). As part of the Netherlands XTC Toxicity study (NeXT study) (de Win et al. 2005) a prospective study on 188 MDMA-naive subjects likely to use MDMA in the future was performed (Schilt et al. 2007). After a follow up of on average 11 months, 58 subjects had started using MDMA (median use at follow up was 1.5 pill) and were compared to 60 still MDMA-naive subjects from the same cohort matched for age, sex, intelligence, and use of other substances. At the follow-up, the researchers found a significant negative effect of MDMA use on verbal memory compared to controls, with no effect on other cognitive functions. This study thus provides convincing evidence of a causal relationship between MDMA use and subtle poorer memory performance, even following low exposure. Rogers and colleagues (2009) provide a systematic review of 114 research papers that assess the damaging neurocognitive effects of MDMA. They found evidence for significant lower memory performance of MDMA users compared to polydrug controls. The highest negative effect was seen on working memory (Standardized Mean Difference (SMD) –0.391, 95% Confidence Interval (CI) –0.589 to –0.192), delayed verbal memory (SMD –0.377, 95% CI –0.498 to –0.257) and immediate verbal memory (SMD –0.332, 95% CI –0.451

(6)

5 to –0.214). These effects were deemed small by Cohen’s rule of thumb (SMD small = 0.2, medium = 0.5, large = 0.8, Cohen 2013). When comparing MDMA users to drug-naive controls this effect was even larger, with the largest effects on immediate verbal memory (SMD –0.840, 95% CI –0.990 to – 0.690) and delayed verbal memory (SMD –1.037, 95% CI –1.734 to –0.341). However, no dose-response relationship is found in most literature between memory problems and MDMA use. Studies reporting heavier average use of ecstasy did not provide more extreme effect measures than those consisting of lighter users. In addition, lifetime MDMA use is found not to be predictive for the severity of memory problems. Although the prospective NeXT study did observe memory problems as a consequence of MDMA use, the lack of an observed relationship between exposure and outcome in all other included studies make it difficult to infer any causal relation between MDMA use and memory problems.

Direct evidence of MDMA induced neurotoxicity

The exact cause of the memory deficits caused by MDMA is still to be elucidated. The mechanism of action of MDMA is briefly described in box 1; the effects of MDMA are mainly on serotonin, dopamine and norepinephrine. As will be elucidated later on, MDMA appears to be specifically neurotoxic to serotonergic neurons. This serotonergic neurotoxicity could be responsible for the observed memory problems in MDMA users, since serotonin is an important modulator in memory processes (Schmitt et al. 2006). Biomarkers have been developed that specifically target the serotonergic system in order to quantify serotonergic neurotoxicity. These biomarkers include density of the serotonin transporter protein (SERT), serotonin receptors, serotonin (5-HT), the serotonin metabolite 5-hydroxyindoleacetic acid (5-HIAA), and the enzyme responsible for serotonin synthesis tryptophan-hydroxylase (TPH). Lower density of these markers is used as an indicator for loss of serotonergic axons/neurons (Green et al. 2003). This section focuses on neurophysiological evidence for serotonergic neurotoxicity in animals and humans which is assumed to be the cause of the memory deficits seen in MDMA users.

(7)

6

Neurotoxicity of MDMA in animals

Research into the neurotoxic effects of MDMA stems back almost four decades, with the first paper on MDMA induced neurotoxicity by Schmidt and colleagues (1986) reporting lasting depletions in 5-HT and 5-HIAA in rats after administration of MDMA indicating serotonergic neurotoxicity. Most animal research on MDMA induced neurotoxicity uses mice, rats, and nonhuman primates. Mice show dopaminergic neurotoxicity as a consequence of MDMA administration (e.g. Granado et al. 2008), an effect not seen in other animals (Moratalla et al. 2017). In rats and nonhuman primates MDMA seems to induce specific serotonergic neurotoxicity (Schmidt et al. 1986; Ricaurte et al. 1988) as evidenced by structural damage to serotonergic nerve fibers. The serotonergic neurotoxicity seen in animals is accompanied by reductions in brain 5-HT, 5-HIAA, SERT density, 5-HT receptors (5-HT1A and 5-HT2A), TPH, and a reduction in cerebrospinal fluid 5-HIAA (Green et al. 2003).

There has been some critique on the use of animal studies and the applied allometric scaling in making conclusions with regards to possible neurotoxic effects of MDMA in humans (de la Torre & Farré 2004b). Animal studies often use high doses of MDMA (10-20 mg/kg) to correct for the higher metabolism of these animals. However, this allometric scaling has not been validated for MDMA. Research indicates animals given an intraperitoneal injection of MDMA at a dose comparable to humans (2.0 mg/kg) also produced an MDMA Cmax comparable to humans (210 ± 108 ng/mL compared to 292 ± 76 ng/mL in humans at a dose of 1.6 mg/kg) (Green et al. 2012). In addition, MDMA produces comparable neurochemical, endocrine, and behavioral actions in rats and humans at equivalent doses. This indicates the allometric scaling used in most animal research is flawed and results most likely do not give accurate information with regards to the neurotoxic effects of MDMA in humans. However, serotonergic neurotoxicity has even been observed in squirrel monkeys using a dose regimen that is comparable to human users (Mechan et al. 2006). In conclusion, animal studies clearly indicate MDMA can be toxic to serotonergic neurons, but it is unclear whether these results are translatable to humans.

Neurotoxicity of MDMA in humans

In order to assess the neurotoxic profile of MDMA in humans, some biomarkers used in animals cannot be used due to their invasive nature. SERT density has been used most frequently as a biomarker in non-invasive neuroimaging studies using PET and SPECT imaging to detect serotonergic neurotoxicity caused by MDMA use. The reduction in SERT is thought to signify loss of serotonergic synapses/axons and possibly loss of serotonergic neurons, since SERT is a structural component of the serotonergic axon. SERT reduction has been validated as a measure for serotonergic damage in animals (de Win et al. 2004). Early research indeed found a reduction in SERT density in MDMA users compared to controls (McCann et al. 1998; Reneman et al. 2001a), providing preliminary evidence that the serotonergic neurotoxicity induced by MDMA in animals also occurs in humans. In a recent meta-analysis on 10 imaging studies on SERT density in MDMA users, a significant reduction in SERT density in MDMA users compared to controls was also observed (Müller et al. 2019). This reduction in SERT density was apparent in 8 of the 13 studied brain areas, specifically the anterior cingulate, posterior cingulate, hippocampus, occipital lobe, parietal lobe, temporal lobe, and the thalamus. No significant reductions were observed in the caudate, frontal lobe, insula, midbrain, and putamen. Strikingly, the authors found no relation between lifetime MDMA use and severity of SERT reduction. This is in accordance to the lack of dose-response relationship between observed memory problems and MDMA use. However, recovery of SERT density has been observed in humans in several studies (e.g. Buchert et al. 2006; Selvaraj et al. 2009; Erritzoe et al. 2011) with Erritzoe and colleagues

(8)

7 estimating SERT density would recover fully after about 200 days of abstinence. Reneman and colleagues (2001b) found that while abstinent female MDMA users showed recovery of SERT density, memory problems in this group persisted. Since the memory problems observed in MDMA users seem permanent and are even observed to worsen after abstinence, the validity of SERT density in estimating MDMA induced neurotoxicity has been questioned. It is possible the reduction in SERT signifies an adaptive response as a consequence of serotonin depletion without loss of serotonergic axons. Another hypothesis explaining the recovery in SERT density after MDMA use is related to the ability of serotonergic axons to undergo regenerative sprouting after MDMA use. Indeed, recovery of serotonergic axons is observed in animals after neurotoxic doses of MDMA (Fischer et al. 1995; Hatzidimitriou et al. 1999). However, the observed reinnervation patterns of these axons appear abnormal, possibly explaining the lack of recovery of memory function.

Other imaging studies have focused on density of the 5-HT2A (serotonin) receptor as a measure of MDMA induced neurotoxicity. Research finds a decrease in 5-HT2A receptor binding a short period after use, followed by increased 5-HT2A receptor binding after abstinence which is thought to be a compensatory mechanism as a result of serotonin depletion (Reneman et al. 2000; Reneman et al. 2002b; Di Iorio et al. 2012; Urban et al. 2012). Interestingly, Reneman and colleagues (2000) found a correlation between 5-HT2A density and severity of memory problems. Additionally, Di Iorio and colleagues (2012) find a relationship between lifetime MDMA use and 5-HT2A density. Further research is warranted because of the relatively low sample sizes used in the aforementioned studies, but 5-HT2A density appears to be a promising biomarker in assessing MDMA induced neurotoxicity and these studies seem to indicate serotonergic neurotoxicity as a consequence of MDMA use happens in humans as well.

Thesis aim

There is some ambiguity with regards MDMA induced serotonergic neurotoxicity in humans. Decades of animal work clearly indicate dose-dependent serotonergic neurotoxicity induced by MDMA. In line with this body of animal work, in humans, reduced SERT densities have been observed in addition to effects on verbal and working memory. Intriguingly, and in contrast to the animal work, the effects of MDMA on SERT density and memory function are not related to cumulative MDMA consumption. The discrepancy between the effects of MDMA observed in animal research and in human research might be caused by other factors that influence an individual’s susceptibility to the neurotoxic effects of MDMA. Insight into these modulating factors in MDMA induced neurotoxicity can help improve research assessing the neurotoxic effects of MDMA, especially improving accuracy of research by correcting for important modulating factors. Potentially, this information can also be used to improve harm-reduction practices for MDMA users in both a recreational and a therapeutic setting.

A lot of modulating factors exist that obscure research into MDMA induced neurotoxicity. This problem is exacerbated further by a lack of uniformity with regards to the variables for which study groups are matched (Rogers et al. 2009). Known factors influencing MDMA induced neurotoxicity include age, sex, ethnicity, diet, genes, polydrug use, and environment. This thesis aims to assess the impact of some of the most important modulating factors based on earlier research and from a mechanistic point of view. As will be elucidated on in a later section, the neurotoxic effects of MDMA appear to be caused by its metabolites rather than MDMA itself. Therefore, this thesis will focus on the metabolism of MDMA, as well as factors like genetic polymorphisms, ethnicity, and sex that

(9)

8 modulate activity of the enzymes involved in the metabolism of MDMA. In addition, MDMA users are often polydrug users and use MDMA at dance parties with high ambient temperature; both modulating factors in MDMA induced neurotoxicity.

(10)

9

Metabolism of MDMA

The neurotoxic effects of MDMA depend on its metabolism which is mainly executed by the liver. Phase 1 metabolism of MDMA is primarily executed by cytochrome p450 (CYP) enzymes and catechol-O-methyltransferase (COMT) (de la Torre et al 2004a). MDMA can be O-demethylated to dihydroxymethamphetamine (HHMA) (see figure 1-B) or N-demethylated to MDA (figure 1-A). CYP2D6 is mainly responsible for O-demethylation and CYP2B6 is mainly responsible for N-demethylation, though other enzymes are involved in these reactions as well (figures 2-A and 2-B). O-demethylation happens one order of magnitude faster than N-deamination (Kreth et al. 2000) hence HHMA is formed in higher quantities than MDA. MDA can also be O-demethylated to its respective metabolite dihydroxyamphetamine (HMA) (figure 1-B). After demethylation, COMT catalyzes methylation of HHMA and HHA to hydroxymethoxymethamphetamine (HMMA) and hydroxymethoxyamphetamine (HMA) respectively (figure 1-C). The methylated compounds will undergo phase 2 metabolism resulting in their respective sulphate or glucuronide conjugates before being excreted by the kidneys. In an alternative route, HHMA and HHA can be oxidized to their respective ortho-quinones (Hiramatsu et al. 1990). These quinone intermediates can form glutathione (GSH) adducts (figure 1-D) which are the primary candidates for causing neurotoxicity associated with MDMA use, as will be elucidated on in the next section.

Figure 1, Metabolism of MDMA to MDA (A), subsequent demethylation (B) and methylation (C) of metabolites, and formation of glutathionyl conjugates of HHMA and HHA via their quinone intermediates (D). Based on de la Torre & Farré 2004b, Meyer et al. 2008.

In a minor metabolic pathway, MDMA and its metabolites MDA, HHMA, HMA, HMMA, and HMA can undergo deamination followed by conjugation to glycine before being excreted. Additionally, HHMA and HMA can be ring-hydroxylated to trihydroxymethamphetamine (THMA) 2,4,5-trihydroxyamphetamine (THA) respectively, though this also constitutes a minor pathway in metabolism of MDMA.

(11)

10

Causes of MDMA neurotoxicity

MDMA itself does not appear to be neurotoxic as evidenced by the fact that MDMA administered directly into the dorsal raphe or median raphe of Sprague-Dawley rats does not produce reductions in 5-HT and its metabolite 5-HIAA (Paris & Cunningham 1992). Also, MDMA perfusion into the hippocampus of Dark Agouti rats in concentrations matching those reached by systemic MDMA administration (5, 10, and 15 mg/kg) did not cause long-term decreases in brain 5-HT, 5-HIAA, and DA content (Esteban et al. 2001). Research indicates MDMA needs to be bio-activated by hepatic metabolism for neurotoxicity to occur (Capela et al. 2009). Direct administration of HHA to the rat brain does not produce neurotoxicity (McCann et al. 1991), therefore it is thought peripherally formed glutathione conjugates of HHMA and HHA, (HHMA, 2,5-bis-(GSH)-HHMA, 5-(GSH)-HHA, and 2,5-bis-(GSH)-HHA (see figure 1), cause neurotoxicity. Indeed, HHMA and HHA glutathione conjugates have been proven to be neurotoxic in Wistar rat cortical neuronal cultures (Capela et al. 2007). Direct administration of 2,5-bis-(GSH)-HHMA to Sprague-Dawley rats caused long-term 5-HT depletion indicating lasting serotonergic damage (Miller et al. 1997). Peripherally formed glutathione adducts of MDMA metabolites have also been observed to be present in the rat brain after MDMA administration (Jones et al. 2005). Inside the neuron, quinones like HHMA and HHA conjugated to glutathione undergo redox-cycling which produces a disproportionate amount of reactive oxygen species (ROS) and reactive nitrogen species (RNS) (Monks & Lau 1997). High amounts of oxidative stress damages cellular components and will ultimately result in cell death (Droge 2002). High levels of ROS and RNS have indeed been implicated in neurotoxicity caused by MDMA (Colado et al. 1997). Inside the brain these glutathione adducts are metabolized to their corresponding N-acetylcysteine adducts which have been proven to be selective serotonergic neurotoxins as well (Bai et al. 1999). The N-acetylcysteine HHMA and HMA conjugates have been identified in human urine providing supporting evidence this form of neurotoxicity also happens in humans (Perfetti et al. 2009). In conclusion, MDMA is metabolized by the liver in various steps to its thioether metabolites (figure 1-D). These metabolites are transported into the brain where they undergo redox cycling. As a result, oxidative stress within the neuron increases dramatically causing damage and ultimately cell death. Another possible mechanism of MDMA neurotoxicity has been postulated by Sprague and colleagues (1998). MDMA causes the release serotonin resulting in subsequent serotonin depletion. MDMA also activates the 5-HT2A receptor which results in increased dopamine synthesis and release. Dopamine is then taken up into the serotonergic neuron by SERT where it is metabolized by monoamine oxidase B (MAO-B). This metabolism in turn causes the formation of ROS and RNS which leads to cellular damage and eventually cell death. In support of this hypothesis, Hrometz and colleagues (2004) found that human choriocarcinoma cells exposed to MDMA and dopamine exhibited strongly reduced cell viability which was blocked by either a SERT inhibitor or an MAO-B inhibitor. Administration of an MAO-B inhibitor to Wistar rats reversed MDMA induced increases in oxidative stress markers (Alves et al. 2007). Also, attenuated serotonergic neurotoxicity is observed in MAO-B knockout mice and MAO-B knockdown rats (Fornai et al. 2001; Falk et al. 2002).

Lastly, it has been hypothesized increased oxidation of neurotransmitters within the presynaptic neuron adds to MDMA induced neurotoxicity (Capela et al. 2009). Also, minor metabolites of MDMA, THMA and THA, appear to be neurotoxic in the rat brain as well (Elayan et al. 1993). However, these metabolites are unlikely to be involved in MDMA mediated neurotoxicity since these metabolites also confer dopaminergic toxicity (Elayan et al. 1992) not seen with MDMA.

(12)

11 In conclusion, two main mechanisms appear to be involved in MDMA induced neurotoxicity. MDMA is converted to the metabolites HHMA and HHA. These metabolites are conjugated to glutathione and subsequently taken up into the serotonergic neuron. Inside the neuron these metabolites undergo redox cycling producing a disproportionate amount of ROS. In addition, serotonin depletion enables dopamine to be taken up into the serotonergic neuron. MAO-B catalyzes dopamine metabolism inside the serotonergic neuron which again produces ROS. The high level of oxidative stress caused by increased ROS production leads to damage to serotonergic neurons and ultimately cell death.

(13)

12

Genetic susceptibility

Figure 1 gives an overview of the enzymes involved in the metabolism of MDMA. The activity of these enzymes can be highly variable between individuals. Since the enzymes are involved in MDMA induced neurotoxicity, individual differences in enzyme activity are an important modulating factor in MDMA induced neurotoxicity. This section gives an overview of the factors involved in determining enzymatic activity and the implications of these differences in activity.

CYP2D6

The hepatic CYP2D6 enzyme is responsible for phase 1 metabolism of a large variety of pharmaceutical and recreational drugs. A large number of polymorphisms of the CYP2D6 gene exist with a total of 74 allele variants known to date (Zhou 2009). The allele variants can cause loss of function, reduced function, normal function or increased function, though these effects can be substrate-dependent. The genetic polymorphisms result in four discernable phenotypes: a poor metabolizer (PM), an intermediate metabolizer (IM), an extensive metabolizer (EM), and an intensive metabolizer (IM) phenotype. In general, most people possess the EM phenotype; for example 70-80% of Caucasians possess the EM phenotype (Sachse et al. 1997). Polymorphism of the CYP2D6 gene is the main determinant of CYP2D6 functions, with factors such as diet, age, and sex playing a relatively minor role (Steiner et al. 1985). Therefore, polymorphism of the CYP2D6 gene can have a big influence on the pharmacokinetics of a drug and therefore also its toxicity.

Since CYP2D6 is one of the main enzymes responsible for MDMA metabolism, different CYP2D6 phenotypes have a big influence on the pharmacokinetics of MDMA. For example, subjects with the PM phenotype had a higher Cmax (maximum concentration) and AUC (area under the curve) of MDMA after MDMA administration compared to subjects with the EM phenotype (Steuer et al. 2016). Also, PM subjects had a lower Cmax and AUC of MDMA’s metabolites HHMA and HMMA. Another study found people with the CYP2D6 *1/*1 genotype (two functional alleles resulting in an EM phenotype) have a higher Cmax and AUC of MDMA’s main metabolites HHMA and HMMA than people with the

CYP2D6 *4/*4 genotype (two non-functional alleles) (de la Torre et al. 2005). Also, inhibition of

CYP2D6 by either paroxetine or bupropion in people with an EM phenotype results in a PM phenotype and subsequent higher MDMA Cmax and AUC and a lower Cmax and AUC for HHMA, HMMA, and HMA (Segura et al. 2005; Steuer et al. 2016).

The neurotoxic effects of MDMA are probably dependent on bio-activation, most importantly by CYP2D6. Higher CYP2D6 activity leads to a higher Cmax and AUC of HHMA and HMA. Since these metabolites are implicated in MDMA induced neurotoxicity via their thioether conjugates, high CYP2D6 activity could lead to a higher susceptibility to the neurotoxic effects of MDMA. Indeed, a high CYP2D6 activity phenotype seems to be related to a lower performance on verbal fluency tests (Cuyás et al. 2011) indicating possibly increased neurotoxic effects of MDMA in this group. Conversely, since the CYP2D6 PM phenotype leads to a lower amount of toxic metabolites being formed this phenotype might be protective against MDMA induced neurotoxicity (Perfetti et al. 2009).

De la Torre and colleagues found that MDMA has non-linear pharmacokinetics, probably via auto-inhibition of CYP2D6. This auto-inhibition might be caused by MDMA forming a complex with CYP2D6 rendering it inactive. Recovery of CYP2D6 enzymatic activity is slow (~280 hours) and depends on its de novo synthesis. Yang and colleagues (2006) found that MDMA could inactivate most hepatic

(14)

13 CYP2D6 within an hour. Also, repeated MDMA administration causes a disproportionately high MDMA Cmax and AUC and a disproportionately low HMMA and HMA Cmax and AUC as expected by simple dose accumulation (Farré et al. 2004; Farré et al. 2015). This clearly indicates MDMA inhibits its own metabolism after repeated doses in humans. Because of the ability of MDMA to inhibit its own metabolism, de la Torre and colleagues (2012) hypothesized that regardless of genetic polymorphism, with a high enough dose all MDMA users will exhibit a functional CYP2D6 PM phenotype. As a result, CYP2D6 polymorphism is a genetic modulating factor in the susceptibility to MDMA induced neurotoxicity in low to moderate doses, but at high doses this effect is abolished due to auto-inhibition of metabolism by MDMA.

Other cytochrome p450 enzymes

Although CYP2D6 activity is strongly correlated to the generation of neurotoxic MDMA metabolites, the enzyme is responsible for about 30% of total metabolism of MDMA (Segura et al. 2005). As can be seen in figure 1, multiple other CYP enzymes are involved in the bio-activation of MDMA. CYP1A2, CYP2B6, CYP3A4, and CYP2C19 all have significant MDMA O-demethylation activity (see figure 1). In addition, CYP1A2 and CYP2B6 also exhibit significant MDMA N-demethylation activity. This section gives a brief overview of factors influencing activity of the aforementioned enzymes.

No polymorphisms of CYP1A2 have been identified that cause reduced function. However, one single-nucleotide polymorphism (SNP) at rs762551 has been identified: a transversion form C to A (Sachse et al. 1999). The CYP1A2 rs762551 A/A polymorphism appears to be inducible by smoking. Vizeli and colleagues (2017) found that smokers who possessed this polymorphism had a higher MDA Cmax after MDMA administration, without any influence on the concentration of MDMA and HMMA or the subjective effects. Also, Yubero-Lahoz and colleagues (2012) found that CYP1A2 activity was raised after MDMA administration. Unfortunately they did not look at the genetic polymorphism of CYP1A2 in their subjects.

A multitude of allele variants and SNP’s of CYP2B6 have been identified making this one of the most polymorphic CYP genes in humans (Zanger et al. 2007). Subjects with the CYP2B6 rs3745274 T/T genotype causing reduced activity had a higher MDMA Cmax, though the plasma levels of MDA and HMMA were unaffected. Interestingly, the effects of CYP2B6 genotype on MDMA concentration appeared later in time. Vizeli and colleagues hypothesized this could indicate that MDMA metabolism by CYP2B6 might become more important after CYP2D6 has been inactivated by MDMA. As with CYP2D6, CYP3A4 activity is mainly influenced by genes (Özdemir et al. 2000; Lee & Goldstein 2005). Genetic variants in CYP3A4 causing amino-acid sequence changes in the CYP3A4 enzyme are rare and most polymorphisms influence expression of the enzyme (Wang et al. 2011). CYP3A4 activity has been correlated to the generation of neurotoxic metabolites of MDMA (Antolino-Lobo et al. 2010) and could therefore be an important factor in MDMA mediated neurotoxicity. However, CYP3A4 activity gets inhibited by MDMA (Yubero Lahoz et al. 2011; Jamshidfar et al. 2017) and inhibition of CYP3A4 does not completely prevent the neurotoxic effects of MDMA (Antolino-Lobo et al. 2011). Therefore the impact of CYP3A4 activity on MDMA induced neurotoxicity seems relatively small.

Multiple allele variants of CYP2C19 have been identified that cause loss of function or gain of function (Desta et al. 2002; Hicks et al. 2013). Vizeli and colleagues (2017) found that reduced activity

(15)

14 of CYP2C19 caused a higher Cmax of MDA after MDMA administration. However, lowered CYP2C19 activity did not influence the concentration of MDMA and HMMA or the subjective effects of MDMA. It appears genetics have a strong influence on CYP2B6, CYP3A4, and CYP2C19 activity. In contrast, only one SNP has been identified in CYP1A2 that can influence its activity. Differences in activity of the aforementioned enzymes seem to have a mild effect on MDMA metabolism compared to differences in CYP2D6 activity. Lowered CYP2C19 and CYP3A4 activity, and higher CYP1A2 activity could shift the metabolic route of MDMA from O-demethylation (figure 1-B) to N-demethylation (figure 1-A) causing higher formation of MDA. MDA appears to be more neurotoxic than MDMA (O’Hearn et al. 1988) and so the aforementioned changes in enzymatic activity could result in more neurotoxicity. However, lower CYP2C19 activity and higher CYP1A2 activity did not influence HMMA plasma concentrations so the influence of altered activity of these enzymes on the metabolism of MDMA seems comparatively small. In conclusion, activity of the aforementioned enzymes can have a small influence on MDMA induced neurotoxicity but CYP2D6 activity is a far more important modulating factor in susceptibility to the neurotoxic effects of MDMA.

Catechol-O-Methyltransferase

Catechol-O-Methyltransferase (COMT) catalyzes the conversion of HHMA and HHA to HMMA and HMA respectively. Since HHMA and HHA can be converted to neurotoxic thioether metabolites, COMT activity could be protective against MDMA induced neurotoxicity. In this case, COMT will catalyze the conversion of HHMA and HHA to HMMA and HMA before HHMA and HHA can oxidize and get converted to neurotoxic metabolites. Two COMT isoforms are known: the short soluble S-COMT isoform found throughout the body and the long membrane bound MB-S-COMT mostly found in the brain. One important functional polymorphism is the Val158Met polymorphism which results in a 40% decrease of enzyme activity in vivo (Chen et al. 2004). It is therefore expected that people possessing one or two Val158Met alleles will exhibit slower conversion of HHMA and HHA to HMMA and HMA respectively. Lower COMT activity could decrease the rate at which HHMA and HHA are converted to HMMA and HMA and increase the rate of formation of neurotoxic metabolites. Also, COMT is involved in breaking down dopamine (Weinshilboum et al. 1999) and higher dopamine levels might be involved in MDMA induced serotonergic neurotoxicity (Sprague et al. 1998). Therefore, lower COMT activity might result in higher dopamine levels even more exacerbating MDMA induced neurotoxicity.

Antolino-Lobo and colleagues (2010) found that MDMA caused cytotoxicity in cells expressing both CYP2D6 and COMT. This cytotoxicity was significantly greater when cells were exposed to both MDMA and COMT inhibitor, providing evidence for the protective effect of COMT activity on MDMA induced toxicity. Pharmacological inhibition of COMT also exacerbated MDMA induced neurotoxicity in rats as determined by brain 5-HT and 5-HIAA levels (Herndon et al. 2014). Also, COMT val/met and COMT met/met (possessing either one or two Val158Met alleles respectively) mice were more sensitive to MDMA-induced neurotoxicity compared to COMT val/val mice. The hypothesis that the COMT Val158Met polymorphism leads to increased production of neurotoxic thioether metabolites is substantiated further by the fact that after administration of MDMA, the recovery of the toxic MDMA metabolite N-Ac-5-Cys-HHMA is two times higher in human subjects possessing the COMT met/met polymorphism compared to subjects possessing the COMT val/val polymorphism (Perfetti et al. 2009). This difference was not significant however (p < 0.1), possibly owing to a low sample size (n = 14). Lastly, COMT met-allele carriers seem more susceptible to the sustained negative effects of

(16)

15 MDMA on verbal learning compared to COMT val-carriers (Schilt et al. 2009). Because of the auto-inhibiting effect of MDMA on CYP2D6 and other cytochrome p450 enzymes, it has been hypothesized that at higher or repeated doses often seen in users COMT polymorphism plays a more important role in MDMA mediated neurotoxicity than cytochrome p450 enzymes (Perfetti et al. 2009). In conclusion, people possessing one or two COMT Val158Met alleles have an increased risk to MDMA induced neurotoxicity.

Monoamine Oxidase-B

Dopamine influx into serotonergic neurons might be an additional mechanism of MDMA induced neurotoxicity. After MDMA induces serotonin depletion, dopamine is taken up by the serotonergic neuron where it is metabolized by B (Sprague et al. 1998). Several polymorphisms of the

MAO-B gene exists, the most common being a G/A substitution on intron 13. Since this SNP is located on

an intron, no change in MAO-B activity has been observed as a result of this polymorphism (Pivac et al. 2007). On the other hand, the polymorphism does seem to influence expression of MAO-B (Jakubauskiene et al. 2012). The MAO-B ‘A’ allele enhances splicing and thereby expression of MAO-B protein. Since MAO-B is involved in MDMA induced neurotoxicity, and MAO-B inhibitors completely abolish these neurotoxic effects, it seems likely that genetic polymorphisms of MAO-B modulate the neurotoxic effects of MDMA. However, to date no human research has looked at MAO-B polymorphism as a modulating factor in MDMA induced neurotoxicity so this remains speculation. In conclusion, there is clear evidence CYP2D6 and COMT polymorphisms play an important modulating role in MDMA induced neurotoxicity. CYP1A2, CYP2B6, CYP3A4, and CYP2C19 polymorphisms are assumed to have a relatively low influence on MDMA induced neurotoxicity since altered activity of these enzymes have a small impact on MDMA pharmacokinetics. Lastly, there are indications MAO-B polymorphisms could influence MDMA induced neurotoxicity, though no human research has been done to date to support this assumption.

Sex

Research into gender-related differences in susceptibility to the neurotoxic effects of MDMA is scarce. Female rats seem to be more vulnerable to the memory impairment effects of MDMA than male rats (Asl et al. 2015). In humans, gender-related differences in serotonergic markers have also been observed as a result of MDMA use. Female MDMA users show a greater reduction in SERT binding compared to males (Reneman et al. 2001; Buchert et al. 2004). McCann and colleagues (1994) also found a greater reduction in CSF 5-HIAA and homovanillic acid (HVA), both markers of serotonergic function, in female MDMA users compared to male users. However, no gender differences have been observed in memory function between male and female MDMA users (Reneman et al. 2006). In contrast, Bolla and colleagues (1998) found that males had greater memory impairment than females after abstinence from MDMA. This could be caused by a greater ability to recover from MDMA induced neurotoxicity in women compared to men. Reneman and colleagues (2001) found that after abstinence SERT density in women did not differ significantly from control. This recovery of SERT density was not observed in men. The negative effect on memory by MDMA use also seems to reverse in women but not in men (Price et al. 2014). The greater recovery seen in women might be caused by the effects of female hormones on the serotonergic system. For example, the female estrogen estradiol-17β has a positive modulatory effect on SERT expression (McQueen et al. 1997). A clear difference in susceptibility to the neurotoxic effects of MDMA between sexes has not yet been found. On the other hand, since the damage from MDMA stems

(17)

16 mostly from its metabolism, differences in metabolism could cause a different risk profile for women compared to men.

Research into sex-related differences in CYP2D6 activity has been ambiguous. For example, Hägg and colleagues (2001) found that females have a slightly lower baseline CYP2D6 activity compared to men while Llerena and colleagues (1996) found no differences between genders. These results are not surprising considering variability of CYP2D6 activitty is mostly determined by genetics (Steiner et al. 1985). Some research has suggested that women have a slightly lower CYP2D6 baseline activity for MDMA metabolism compared to men (Yubero-Lahoz et al. 2011). Pardo-Lozano and colleagues (2012) did not find discernable differences in pharmacokinetic parameters between men and women. Kolbrich and colleagues (2008) found a higher Cmax and AUC for MDMA and MDA after MDMA administration in women compared to men. This effect was only significant at the lower dose MDMA (1mg/kg) and lost significance at the higher dose MDMA (1.6 mg/kg). This result was reproduced by Hartman and colleagues (2014) and it was hypothesized women experience CYP2D6 auto inhibition by MDMA at a lower dose than men. Indeed, Yubero-Lahoz and colleagues (2011) found that when dosing men and women (all CYP2D6 EM phenotype) 1.5 mg/kg MDMA, 100% of women exhibited a PM functional phenotype, while 67% of men exhibited a PM functional phenotype. In conclusion, CYP2D6 baseline activity appears to be marginally lower for women than men. However, MDMA is able to inhibit CYP2D6 in women at a much lower dose than in men. Because of this, it seems plausible that women are more protected for the neurotoxic effects of MDMA, at least with regards to CYP2D6 mediated toxicity and at lower doses of MDMA.

Sex also seems to influence activity of other enzymes implicated in MDMA induced neurotoxicity. Estrogen is a regulator of COMT activity and COMT activity have been found to be lower in females in blood (Weinshilboum et al. 1999) and postmortem brain tissue (Chen et al. 2004). This lowered activity was independent of COMT polymorphisms. Snell and colleagues (2002) found that females have a higher MAO-B platelet activity compared to males. Platelet MAO-B activity has been determined to be a reliable indicator of brain MAO-B activity (Bench et al. 1991). Since a lower COMT activity and a higher MAO-B activity seem to confer an increased risk in the emergence of MDMA induced neurotoxicity, it seems plausible women are more susceptible to the damaging effects of MDMA than men, at least at higher doses of MDMA when both men and women exhibit a CYP2D6 PM phenotype.

Current research into sex-related differences in susceptibility to the neurotoxic effects of MDMA remains inconclusive. However, there is reason to assume women are more susceptible when looking at the mechanism behind MDMA induced neurotoxicity. There does not appear to be a big difference in CYP2D6 function between men and women. However, women undergo CYP2D6 auto inhibition by MDMA at a lower dose than men. Since MDMA users frequently take higher doses or take redoses, it is expected most users, men and women, exhibit a functional PM phenotype during use. Therefore COMT and MAO-B activity could be bigger contributors to MDMA induced neurotoxicity. COMT activity seems lower and MAO-B activity higher in women, both contributing factors into the development of neurotoxicity. The difference in enzyme activity seen between men and women could render women more susceptible to the neurotoxic effects of MDMA. This would explain the higher reduction in serotonergic markers observed in women MDMA users compared to male users.

(18)

17 In conclusion, with regards to CYP2D6 activity women seem less susceptible to the neurotoxic effects of MDMA, while lower COMT and higher MAO-B activity seen in women result in a higher susceptibility to the neurotoxic effects. Since most users exhibit a functional CYP2D6 PM phenotype, in most situations women are supposed to be more susceptible to the neurotoxic effects of MDMA than men.

Ethnicity

Ethnicity can have an influence on the frequency of certain polymorphisms, including the ones discussed in this paper. For example CYP2D6 polymorphism frequency seems to be dependent on ethnicity (Zhou 2009). In Caucasians the frequency of the CYP2D6 PM, IM, EM, and UM phenotypes is 5-10%, 10-17%, 70-80%, and 3-5% respectively (Sachse et al. 1997). The PM phenotype is almost absent in the Asian population; PM frequency for Chinese, Korean, and Japanese people is 0%, 0.5%, and 0.7% respectively (Sohn et al. 1991). An unusually high amount of the CYP2D6 UM phenotype has been found in black Ethiopians: ~16% (Aklillu et al. 1996). In Caucasians the frequency of non-functional and reduced function alleles is 29%, while almost 50% in Africans and African Americans (Bradford 2002). There is some evidence these ethnic differences in CYP2D6 polymorphism frequencies impact MDMA pharmacokinetics. After administration of MDMA, African Americans have a higher MDMA Cmax and AUC (Kolbrich et al. 2008) and a lower HMMA AUC (Hartman et al. 2014). This data indicates reduced MDMA metabolism by CYP2D6 in African-Americans, possibly reducing the risk of MDMA induced neurotoxicity.

Differences in COMT Val158Met polymorphism frequency between ethnic groups have also been observed. In Caucasians, Southwest Asians, and Turkish population the frequency of COMT val/val, COMT val/met, and COMT met/met have been found to be 25%, 50%, and 25% respectively (McLeod et al. 1998; Kocabaş et al. 2001). The frequency of the COMT met/met polymorphism seems lower however in Asian and African populations. For example 6% of Japanese people, 10% of Chinese people, and 9% of Kenyan people possess the COMT met/met polymorphism (Kunugi et al. 1997; Li et al 1997; McLeod et al. 1998). Since the COMT val/met and COMT met/met polymorphisms resulting in less active COMT are associated with increased neurotoxicity, Asian and African people seem more protected against MDMA induced neurotoxicity because of the low COMT met/met frequency in these ethnic groups.

In conclusion, the previously mentioned research indicates that African American people have a lowered risk of MDMA induced neurotoxicity compared to Caucasian people. African Americans have a higher frequency of non-functional and reduced function CYP2D6 alleles and a lower frequency of the COMT met/met polymorphism. This would result in lower CYP2D6 activity and higher COMT activity, both being protective against MDMA induced neurotoxicity. For the Asian group the risk is more difficult to determine. This group has a low frequency of the CYP2D6 PM polymorphism, but also a lower frequency of the COMT met/met polymorphism. The higher average CYP2D6 activity confers a higher risk to MDMA induced neurotoxicity while the lower frequency of the COMT met/met polymorphism could be protective against MDMA induced neurotoxicity. Because higher doses of MDMA result in a CYP2D6 PM phenotype regardless of genetic polymorphism, it is assumed Asians are less susceptible to the neurotoxic effects of MDMA, at least at higher doses frequently consumed by MDMA users.

(19)

18

Environmental

Another component that seems to be involved in MDMA mediated neurotoxicity is hyperthermia. MDMA is often used at crowded dance parties with a relatively high ambient temperature. Combined with high physical exertion this makes users especially susceptible to the hyperthermic effects of MDMA. The hyperthermia induced by MDMA seems to be primarily caused by an increase in catecholamines and thyroid activation (Capela et al. 2009). For example, the D1 antagonist SCH 23390 completely reverses MDMA induced hyperthermia in rats (Mechan et al. 2002). However, 5-HT2A antagonists have also been observed to prevent MDMA induced hyperthermia in rats and rabbits (Blessing et al. 2003; Shioda et al. 2008). 5-HT2A mediated hyperthermia could be due to an increase in dopamine release after 5-HT2A agonism. In addition, activation of spinal 5-HT2A receptors increases cutaneous vasoconstriction limiting heat loss via the skin. The increase in catecholamines, vasoconstriction, physical exertion, and high ambient temperature all contribute to MDMA induced hyperthermia.

A strong correlation between changes in body temperature as a result of MDMA use and severity of neurotoxicity has been found in animals (Malberg & Seiden 1998). For example, low ambient temperatures (10°C - 11°C) appear to cause hypothermia in rats administered MDMA which results in a lack of observed neurotoxicity (Drafters 1994; Broening et al. 1995). In contrast, high ambient temperatures (>24°C) appear to exacerbate MDMA induced neurotoxicity (Sanchez et al. 2004). Though hyperthermia is involved in MDMA mediated neurotoxicity, the effect seems to be additive instead of causative. This is evidenced by the fact that compounds that protect against MDMA mediated neurotoxicity like memantine and fluoxetine do not influence MDMA induced hyperthermia (Chipana et al. 2008; Sanchez et al. 2001). The primary mechanism of MDMA induced neurotoxicity involves its reactive metabolites resulting in an increase in ROS. Hyperthermia observed with MDMA use most likely results in an additional increase in oxidative stress exacerbating neurotoxicity (Colado et al. 1998; Colado et al. 1999).

In a controlled environment, hyperthermia as a result of MDMA use has not been observed in humans at relatively low doses (1-1.7 mg/kg, Grob et al. 1995; Mas et al. 1999; Kolbrich et al. 2008). However, at a higher dose of 2 mg/kg Freedman and colleagues (2005) observed a 0.4°C increase in body temperature at 18°C ambient temperature (p < 0.02) and a 0.6°C increase in body temperature at 30°C ambient temperature (p < 0.001). Interestingly, there was no significant difference between the temperature increases at low and high ambient temperature. This could mean that club-goers who consume MDMA are not necessarily more susceptible to the hyperthermic effects of MDMA due to high ambient temperature. However, most research finds that the increase in body temperature of MDMA users at dance parties is higher (> 1°C) than that of MDMA users in controlled environments (Parrott 2012). In addition, Parrott and colleagues (2006) found that MDMA users who reported to be dancing ‘all the time’ while on MDMA had higher Prospective Memory Questionnaire long-term problem scores, indicating possible neurotoxic effects, compared to users who danced sometimes or frequently.

In conclusion, it seems that the combination of high ambient temperature combined with high physical exertion can exacerbate MDMA induced hyperthermia and possibly MDMA induced neurotoxicity.

(20)

19

Drug-drug interactions

MDMA users are often polydrug users, with a UK survey finding most MDMA users also use amphetamines, cannabis, cocaine, and LSD (Winstock et al. 2001). Some of these drugs have been proven to be neurotoxic. As a result, polydrug use obscures research into MDMA induced neurotoxicity. In addition, besides being neurotoxic on their own, some drugs might exacerbate MDMA induced neurotoxicity when consumed concurrently with MDMA (Gouzoulis-Mayfrank & Daumann 2006). Other drugs have even been found to be protective against MDMA induced neurotoxicity. This section gives an overview of drugs commonly used concurrently with MDMA and known interactions between these drugs.

Caffeine

Even though caffeine is generally considered safe for consumption, there are indications that combined use of caffeine with MDMA can increase the neurotoxic effects of MDMA. For example, Downey and colleagues (2010) found that MDMA combined with caffeine caused increased cytotoxicity in vitro compared to MDMA alone. In rats, caffeine appeared to increase the hyperthermic effects of MDMA (Vanattou et al. 2010) and this effect seemed independent on caffeine dose (5 and 10 mg/kg, McNamara et al. 2006). These hyperthermic effects seem to be mediated by dopamine and adenosine A1 and A2 receptors. Caffeine also exacerbated MDMA induced 5-HT and 5-HIAA depletion. Camarasa and colleagues (2006) hypothesized caffeine could increase neurotoxicity by increasing dopamine release and subsequent dopamine oxidation in serotonergic neurons (Sprague et al. 1998). No human studies have been performed to date into the interaction between caffeine and MDMA, but it seems plausible the hyperthermic effect of caffeine can have an additive effect on MDMA induced neurotoxicity.

Tobacco

MDMA is often used in combination with tobacco. For example, one study reported 64% of students who used MDMA used it in combination with tobacco (Barrett et al. 2006). The possible interaction between nicotine and MDMA is not well understood. Since nicotine increases oxidative stress it seems likely concurrent use of nicotine with MDMA increases the risk of neurotoxicity compared to MDMA use alone (Parrot 2003). Indeed, Budzynska and colleagues (2018) found that co-administration of MDMA and nicotine to Swiss mice decreased total antioxidant status (sum of extracellular endogenous and food-derived antioxidants, Miller et al. 1993), decreased GPx activity, and raised malondialdehyde levels compared to MDMA alone. This indicates that co-administration of nicotine with MDMA increases oxidative stress compared to MDMA administration alone which could possibly lead to increased neurotoxicity. However, the value of mice research is limited since MDMA has a different neurotoxic profile in mice, showing mainly dopaminergic damage instead of serotonergic damage, compared to rats and humans (Granado et al. 2008; Capela et al. 2009). Since nicotine also releases dopamine it could contribute to dopamine mediated serotonergic neurotoxicity (Sprague et al. 1998), though the dopamine release caused by nicotine seems insignificant compared to other drugs (Tsukada et al. 2002). In conclusion, not enough research has been done to date to elucidate the interaction between nicotine and MDMA, but there is an indication there could be a harmful interaction between the two compounds.

(21)

20

Cannabis

Most MDMA users also use cannabis (Winstock et al. 2001). Cannabis use can have negative effects on attention and memory but these effects appear to be reversible and not indicative of neurotoxicity (Gouzoulis-Mayfrank & Daumann 2006). Some in vitro research indicates one of the active compounds in cannabis, Δ9-tetrahydrocannabinol (THC), can be neurotoxic to hippocampal neurons (Chan et al. 1998) while more recent research indicates THC could actually have neuroprotective effects on hippocampal neurons (e.g. Gilbert et al. 2007). There is also an indication that cannabis can protect against MDMA induced neurotoxicity. Morley and colleagues (2004) found that cannabinoids are able to partially protect against the 5-HT depleting effects of MDMA. This could be related to the hypothermic effects of cannabis, as well as antioxidant effects of THC and cannabidiol (CBD) (Hampson et al. 1998). In conclusion, cannabis seems to be protective against MDMA induced neurotoxicity.

Alcohol

Alcohol has been observed to increase the Cmax of MDMA in rats (Hamida et al. 2009), as well as in humans; a dose of 0.8 g/kg causes a rise in MDMA Cmax of 13% in humans (Hernandez-Lopez et al. 2002). Alcohol attenuates the effects of MDMA on hyperthermia and water retention in humans at a dose of 2-3 beverages (Dumont et al. 2010). The hypothermic effect of alcohol is also seen in rodents at 23°C ambient temperature, but this effect is abolished at 32°C ambient temperature (Cassel et al. 2007). Ros-Simó and colleagues (2012) also found that besides its hypothermic effects, alcohol also protected against MDMA induced neuroinflammation in mice. In contrast, pre-exposing rats to a binge regimen ethanol for four days increased MDMA induced cognitive deficits and serotonergic neurotoxicity determined by 5-HT concentration and SERT density (Izco et al. 2007). Also, the ethanol binge regimen did not cause the hypothermic effect observed with low doses of ethanol. It appears alcohol might increase MDMA induced neurotoxicity at high repeated doses. However, at low doses the hypothermic effects of alcohol might have a neuroprotective effect. In addition, consumption of a few alcoholic drinks could mitigate the risk of MDMA induced hypernatremia, a condition that could be fatal to MDMA users when combined with excessive water consumption (Kalantar-Zadeh et al. 2006). In conclusion, alcohol in moderate doses could have a protective effect on MDMA induced neurotoxicity, while higher repeated doses exacerbate MDMA induced neurotoxicity.

Stimulants

Concurrent use of stimulants like amphetamine, methamphetamine, and cocaine with MDMA is high (Winstock et al. 2001). Evidence suggests these stimulants can be neurotoxic to dopaminergic neurons (Tung et al. 2017; Moratalla et al. 2017; Pereira et al. 2015). Reneman and colleagues (2002a) found evidence of dopaminergic neurotoxicity in MDMA users who also used amphetamine, while subjects who only used MDMA had no signs of dopaminergic neurotoxicity. It is not clear however if the combined use of MDMA with other stimulants results in a synergistic effect on neurotoxicity in humans. A synergistic effect between MDMA and stimulants on cytotoxicity in vitro has been observed (Da Siva et al. 2013). Cocaine appears to have an additive effect on MDMA induced ROS formation (Peraile et al. 2013). In addition, rats given a combination of methamphetamine and MDMA showed greater serotonergic and dopaminergic toxicity and hyperthermia compared to methamphetamine or MDMA alone (Clemens et al. 2004; Clemens et al. 2005). It seems stimulants can exacerbate MDMA induced neurotoxicity via increased hyperthermia. Also, since MDMA induced neurotoxicity seems to be partially related to excessive dopamine release,

(22)

21 the additional dopamine release by stimulant drugs might add further to MDMA induced neurotoxicity. Even though no human studies have been performed to date that elucidate the interaction between stimulants and MDMA with regards to neurotoxicity, there is enough reason to assume the combination exacerbates MDMA induced neurotoxicity.

Psychedelics

Most psychedelics (e.g. magic mushrooms, LSD, mescaline) do not appear to possess any neurotoxic effects (Johnson et al. 2008). However, there is evidence that combined use of MDMA and psychedelics can exacerbate MDMA induced neurotoxicity. For example, combined administration of LSD and MDMA to rats resulted in an increase in serotonergic neurotoxicity compared to MDMA administration alone as measured by brain SERT density (Armstrong et al. 2004). In addition, the psychedelic drugs DOI and 5-MeO-DMT also exacerbate MDMA induced neurotoxicity (Gudelsky et al. 1994). Psychedelics mainly work via activation of the 5-HT2A receptor. As mentioned before, activation of the 5-HT2A receptor appears to be involved in MDMA induced neurotoxicity via increased dopamine release and increased hyperthermic effects. Indeed, the additive effects of DOI and 5-MeO-DMT on MDMA induced neurotoxicity seem to be mediated by increased dopamine release (Gudelsky et al. 1994). In conclusion, even though psychedelics do not appear to be neurotoxic on their own, combined use with MDMA can exacerbate MDMA induced neurotoxicity via increased activation of the 5-HT2A receptor.

In conclusion, previous research indicates there are interaction effects between MDMA and other often used drugs that can result in either an additive or a protective effect on MDMA induced neurotoxicity. Drugs that appear to be protective against MDMA induced neurotoxicity are cannabis and low/moderate doses of alcohol. On the other hand, high doses of alcohol, stimulants, and psychedelics exacerbate MDMA induced neurotoxicity and some research suggests caffeine and tobacco are able to contribute to the neurotoxic effects of MDMA as well. MDMA is often used in combination with other drugs making this an important modulating factor in research into the neurotoxic effects of MDMA (Gouzoulis-Mayfrank et al. 2006). Therefore it is important to determine the possible interactions between other drugs and MDMA in order to accurately assess the specific neurotoxic effects of MDMA.

(23)

22

Conclusion/recommendations

The aim of this thesis was to provide an overview of the modulating factors that can influence an individual’s susceptibility to the neurotoxic effects of MDMA, in order to better interpret existing as well as design future research. Most of the neurotoxic effects of MDMA emerge via its metabolites HHMA and HHA. After conjugation to glutathione these metabolites undergo redox cycling in serotonergic neurons producing a high amount of ROS and RNS ultimately leading to neurotoxicity. In addition, there is evidence that increased dopamine influx in serotonergic neurons followed by metabolism by MAO-B, as well as the hyperthermic effects of MDMA, add to the generation of ROS and RNS and subsequent neurotoxicity.

The main enzyme involved in the formation of HHMA and HHA, CYP2D6, is highly polymorphic in humans resulting in multiple different phenotypes. People with the lowest (PM) CYP2D6 activity phenotype appear to be less vulnerable to the neurotoxic effects of MDMA since the rate of formation of HHMA and HHA is lower in this group. Following the same logic, people with the highest CYP2D6 activity phenotype appear to be more vulnerable to the neurotoxic effects of MDMA. High COMT activity protects against MDMA induced neurotoxicity by converting the neurotoxic metabolites of MDMA into non-neurotoxic metabolites. Therefore, people possessing the highly active COMT val/val enzyme polymorphism are less susceptible to the neurotoxic effects than people possessing the COMT val/met and COMT met/met polymorphism. Lastly, there is an indication that

MAO-B polymorphisms could influence MDMA induced neurotoxicity, though no empirical evidence

exists to date to support this claim.

There is also an indication that susceptibility to the neurotoxic effects of MDMA differs between ethnic groups and sex. This mostly has to do with differences in frequencies in CYP2D6 and COMT allele variants between ethnic groups, and differences in expression of CYP2D6 and COMT between sexes. Because of a relatively higher frequency of low activity CYP2D6 and high activity COMT phenotypes, African-Americans are less susceptible to the neurotoxic effects of MDMA compared to Caucasians and Asians. In addition, at higher doses that result in a functional CYP2D6 phenotype, Asians also appear to be less susceptible to the neurotoxic effects of MDMA due to the lower frequency of the COMT met/met polymorphism. Women experience auto inhibition of CYP2D6 by MDMA at lower doses of MDMA. Therefore women appear to be less susceptible to the neurotoxic effects of MDMA at low doses. At higher doses that result in a functional PM CYP2D6 phenotype, this difference between sexes is abolished. In this case, women are more susceptible to MDMA induced neurotoxicity due to lower COMT activity and higher MAO-B activity. However, women also show higher recovery of neurotoxic markers compared to men.

Ambient temperature does not appear to have a strong effect on MDMA induced hyperthermia. However, high ambient temperature combined with high physical exertion like dancing leads to a significant increase in body temperature. Since hyperthermia exacerbates MDMA induced neurotoxicity, MDMA users who frequent dance parties have an increased risk of neurotoxicity. In addition to environmental factors, polydrug use also has a significant effect on MDMA induced neurotoxicity, mostly by altering the hyperthermic effects of MDMA. Cannabis and low doses of alcohol can be neuroprotective via their hypothermic effects. Cannabis also has an added antioxidant effect. In contrast, stimulants and psychedelics increase the hyperthermic effects of MDMA and subsequent neurotoxicity. There is also evidence that tobacco and caffeine contribute to the neurotoxic effects of MDMA, either via increased oxidative stress or an increase in hyperthermia.

(24)

23 An overview of the modulating factors in MDMA induced neurotoxicity that have been discussed in this thesis is provided in figure 2 below:

Figure 2, discussed modulating factors in MDMA induced neurotoxicity. Neurotoxicity of MDMA is presumed to occur because of buildup of toxic metabolites and dopamine influx into the serotonergic neuron ultimately leading to increased production of ROS. Cytochrome p450 and COMT polymorphisms influence enzyme activity and thereby formation of neurotoxic metabolites. Interethnic differences in polymorphism frequency have been observed for the aforementioned enzymes. Individual differences in expression and activity of these enzymes also influence formation of metabolites, and sex is an important modulator of activity of these enzymes. Hyperthermia has been observed to both increase the rate of formation of neurotoxic metabolites, as well as directly increasing ROS. Polydrug use can both positively and negatively influence hyperthermia, and can cause additional dopamine influx into the serotonergic neuron. Lastly, MAO-B polymorphisms might modulate the rate at which dopamine is metabolized and the level of ROS produced.

The information presented in this thesis can be helpful in improving research that assesses the neurotoxic effects of MDMA. Insight into the modulating factors of MDMA induced neurotoxicity is important in order to determine for which variables to correct in research. First of all, CYP2D6 and

COMT allele variants are important modulating factors and genetic testing should be applied to a

subject pool to correct for these factors. Also, the influence of MAO-B polymorphisms on MDMA induced neurotoxicity should be assessed in future research. Ethnicity and sex are also important modulating factors; African Americans seem less susceptible to the neurotoxic effects of MDMA than Caucasians, and women seem more susceptible than men. With regards to correcting for the hyperthermic effects of high ambient temperature combined with high physical exertion often seen at dance parties, a method based on research by Parrott and colleagues (2006) might prove valuable. Parrott and colleagues divided subjects in three groups: those who danced ‘all the time’, ‘frequently’, or ‘sometimes’ at parties and found a clear difference between these groups with regards to memory performance. This indicates this method of subdividing subjects might be valuable in correcting for the hyperthermic effects of high physical exertion. Lastly, concurrent use of MDMA with cannabis

(25)

24 and low doses of alcohol can limit the neurotoxic effects of MDMA while stimulant and psychedelic use can exacerbate the neurotoxic effects of MDMA.

This thesis also provides some directions on improving harm-reduction practices for MDMA users. Users should first of all take care when attending dance parties by taking enough breaks in between dancing to cool off. Even though there is evidence low/moderate doses of alcohol and cannabis can be protective against MDMA induced neurotoxicity, there might be an increased risk of detrimental cardiovascular effects as a result of this combination. Therefore more research is necessary to assess the safety of this combination. On the other hand, previous research indicates an interaction effect between stimulants/psychedelics and MDMA exists which could lead to exacerbated serotonergic neurotoxicity. Therefore, users should avoid combining MDMA with stimulant or psychedelic drugs. This thesis has discussed some of the most important modulating factors on MDMA induced neurotoxicity. However, this thesis does not provide information with regards to possible interaction/cumulative effects between the described modulating factors. A combination between, for example, genetic and ethnic modulating factors that increase an individual’s risk to MDMA induced neurotoxicity might work synergistically in exacerbating the neurotoxic effects of MDMA. Future research should therefore also aim at elucidating the interaction effects between these modulating factors.