- Top of page
The pharmacology of the acute hyperthermia that follows 3,4-methylenedioxymethamphetamine (MDMA, ‘ecstasy’) administration to rats has been investigated.
MDMA (12.5 mg kg−1 i.p.) produced acute hyperthermia (measured rectally). The tail skin temperature did not increase, suggesting that MDMA may impair heat dissipation.
Pretreatment with the 5-HT1/2 antagonist methysergide (10 mg kg−1), the 5-HT2A antagonist MDL 100,907 (0.1 mg kg−1) or the 5-HT2C antagonist SB 242084 (3 mg kg−1) failed to alter the hyperthermia. The 5-HT2 antagonist ritanserin (1 mg kg−1) was without effect, but MDL 11,939 (5 mg kg−1) blocked the hyperthermia, possibly because of activity at non-serotonergic receptors.
The 5-HT uptake inhibitor zimeldine (10 mg kg−1) had no effect on MDMA-induced hyperthermia. The uptake inhibitor fluoxetine (10 mg kg−1) markedly attenuated the MDMA-induced increase in hippocampal extracellular 5-HT, also without altering hyperthermia.
The dopamine D2 antagonist remoxipride (10 mg kg−1) did not alter MDMA-induced hyperthermia, but the D1 antagonist SCH 23390 (0.3 – 2.0 mg kg−1) dose-dependently antagonized it.
The dopamine uptake inhibitor GBR 12909 (10 mg kg−1) did not alter the hyperthermic response and microdialysis demonstrated that it did not inhibit MDMA-induced striatal dopamine release.
These results demonstrate that in vivo MDMA-induced 5-HT release is inhibited by 5-HT uptake inhibitors, but MDMA-induced dopamine release may not be altered by a dopamine uptake inhibitor.
It is suggested that MDMA-induced hyperthermia results not from MDMA-induced 5-HT release, but rather from the increased release of dopamine that acts at D1 receptors. This has implications for the clinical treatment of MDMA-induced hyperthermia.
- Top of page
3,4-Methylenedioxymethamphetamine (MDMA or ‘ecstasy’) is a commonly used recreational drug, often ingested at dance clubs. A major feature of cases presenting with acute MDMA toxicity is hyperthermia, with body temperatures as high as 43°C having been reported (Henry, 1992). It is probable that many of the other, often fatal, toxicological problems that are seen, particularly rhabdomyolysis, disseminated intravenous coagulation and acute renal failure (Brown & Osterloh, 1987; Henry et al., 1992; Screaton et al., 1992) result from the hyperthermia.
Clinically, the major treatment for the hyperthermia is to decrease body temperature which has been achieved by surrounding the body with ice. Pharmacologically, administration of dantrolene has been used, and this has sometimes led to a successful outcome (Henry, 1992; Tehan, 1993; Mallick & Bodenham, 1997). However, its use has been questioned by Barrett (1992) on the basis of the fact that it is a peripherally acting muscle relaxant with no central effects and is therefore not acting on the primary mechanisms involved in the production of the hyperthermia.
Surprisingly, there has been limited experimental investigation into the pharmacology of the MDMA-induced acute hyperthermic response, despite the fact that administration of MDMA to rats, as in humans, produces an acute dose-dependent hyperthermic response (Nash et al., 1988; Gordon et al., 1991; Colado et al., 1993; Dafters, 1994; O'shea et al., 1998). It has long been known that acutely increasing 5-HT synthesis and release in rat brain by administration of L-tyrptophan and a monoamine oxidase inhibitor results in hyperthermia (Grahame-Smith, 1971a; Green & Grahame-Smith, 1976). Since MDMA produces an acute and massive release of 5-HT from serotonergic nerve endings (Stone et al., 1986; 1987; Schmidt et al., 1987; Colado & Green, 1994; Mechan et al., 2000) it has sometimes been assumed (e.g. Shankaran & Gudelsky, 1999) that MDMA-induced hyperthermia is a consequence of 5-HT release and subsequent stimulation of the 5-HT receptors involved in thermoregulation. Such a view has been reinforced by the observations that p-chloroamphetamine, another 5-HT releasing drug, also produces hyperthermia (Colado et al., 1993) and that 5-HT2 receptor antagonists such as ketanserin and MDL 11,939 antagonize MDMA-induced hyperthermia (Nash et al., 1988; Schmidt et al., 1990).
One problem encountered in previous studies on MDMA-induced neurotoxicity has been the fact that some drugs produce marked hypothermia in rats when given alone. This effect therefore confounds any evidence that a drug has a selective action on the MDMA-induced hyperthermic response and makes interpretation difficult (see for example Colado et al., 1999). We have now investigated the effect of compounds known to alter either 5-HT or dopamine function on their ability to antagonize MDMA-induced hyperthermia by examining their effect on both saline and MDMA treated rats.
- Top of page
Any attempt to characterize the hyperthermic response that follows MDMA administration is fraught with difficulties in terms of interpretation of data. Following drug injection any effect seen may result from either an action on the key neurotransmitter or, secondarily, on another neurotransmitter which modulates the action of the initiating transmitter. Also, of course, since drugs are administered peripherally, the effect could also occur via a peripheral mechanism such as a change in blood flow. There is also the additional problem of elucidating the brain region, or regions, involved in the hyperthermic response. An observed neurotransmitter change in one region may not reflect a change in another more functionally relevant region. Nevertheless, even with all these provisos, such studies remain relevant since they give indications as to the best approaches that might be employed clinically to alleviate the life-threatening hyperthermia that can occur in persons overdosing on this commonly used recreational drug.
At the outset it should be emphasized that the dose of MDMA chosen in this investigation is clinically relevant. It has previously been shown that a dose of MDMA of 10 mg kg−1 (a 20% lower dose than in the current study) produced a plasma drug concentration of 6.3 nmol ml−1 45 min later (Colado et al., 1995). This value is in exactly the same range as the plasma concentrations reported in patients suffering acute adverse responses to ingestion of the drug (Henry et al., 1992; Dowling et al., 1987). Furthermore, as McCann & Ricaurte (2001) have recently pointed out, the principle of interspecies drug dose scaling (see Mordenti & Chappell, 1989) should always be used when extrapolating from doses in rat studies to those used by humans.
Using interspecies dose scaling, the dose used in this current study is the equivalent to a human (70 kg) taking a dose of MDMA of 560 mg. Given the most recent evidence that many tablets now being ingested contain between 200 – 300 mg of MDMA (see: www.dancesafe.org), this means that the current study is equivalent to a human ingesting 2 – 3 tablets, a not unusual recreational dose.
Acutely, MDMA produces a major release of both 5-HT (Schmidt et al., 1986; Stone et al., 1986; 1987; Colado & Green, 1994; Mechan et al., 2000) and dopamine (Koch & Galloway, 1997; Sabol & Seiden, 1998; Colado et al., 1999). Methamphetamine also induces the release of both these neurotransmitters (Schmidt & Gibb, 1985a, 1985b; Baldwin et al., 1993) and hyperthermia follows administration to rats of both MDMA (Nash et al., 1988; Gordon et al., 1991; Colado et al., 1993; O'shea et al., 1998; Dafters, 1994) and methamphetamine (Clark & Lipton, 1986; Bronstein & Hong, 1995). Hyperthermia also results from administration of the 5-HT releasing drug p-chloroamphetamine (Colado et al., 1993; 1997), certain 5-HT agonists such as MK 212 (Yamawaki et al., 1983), 5-methoxy N,N-dimethyltryptamine (Grahame-Smith, 1971b) and quipazine (Yamawaki et al., 1983) and after injection of a monoamine oxidase inhibitor and L-tryptophan (Grahame-Smith, 1971a). Therefore it has often been assumed that MDMA-induced hyperthermia is also 5-HT mediated (see for example: Shankaran & Gudelsky, 1999). However, since studies with methamphetamine have implicated dopamine in the hyperthermia seen after administration of that compound (Bronstein & Hong, 1995), there is little reason not to assume that dopamine release could also be involved in the hyperthermic action of MDMA.
MDMA induces dopamine release, at least in the striatum, through several mechanisms. It has been reported to have a direct dopamine-releasing effect via both a calcium-dependent and -independent mechanism at the level of the nerve ending (Crespi et al., 1997; Koch & Galloway, 1997; Yamamoto & Spanos, 1988; Nash & Brodkin, 1991; Johnson et al., 1986; Schmidt et al., 1987). In addition, the increased 5-HT function resulting from the MDMA-induced 5-HT release has been suggested to stimulate 5-HT2 receptors thereby further enhancing dopamine release (Nash, 1990; Gudelsky et al., 1994; Schmidt et al., 1990). Finally, recent evidence also supports a role for MDMA in inhibiting the dopamine uptake carrier (Metzger et al., 1998).
MDMA (12.5 mg kg−1) produced a rise in rectal temperature of approximately 1.5°C in the current study. This is a similar rise to that seen in our other recent studies (O'shea et al., 1998; Mechan et al., 2001). We also briefly examined the tail skin temperature following this dose and saw no evidence for an increase occurring at the same time as the rise in rectal temperature. A similar lack of increase in tail temperature occurs following a hyperthermia-inducing dose of methamphetamine (Mohaghegh et al., 1997). These authors interpreted this to indicate that methamphetamine impaired heat dissipation rather than heat conservation, since tail temperature is a major heat gain/heat loss effector mechanism in rats (Grant, 1963). Thus MDMA may similarly impair heat dissipation mechanisms.
Regardless of whether this mechanistic interpretation is correct or not, our current investigation does not suggest that the acute MDMA-induced release of 5-HT plays a major role in the hyperthermic response. The non-selective 5-HT1/2 antagonist methysergide was without effect on the hyperthermic response, while the selective 5-HT2 antagonist MDL 11,939 blocked the MDMA-induced hyperthermia, in agreement with Schmidt et al. (1990). However, one has to question whether this latter result was due to a lack of receptor selectivitiy of this compound, or its metabolites, given our observation that another potent and selective 5-HT2 antagonist, ritanserin, was without an effect at a dose of 1 mg kg−1, a dose that is still approximately 20 times the ED50 for inhibiting 5-HT2 mediated behaviour (Goodwin & Green, 1985). A lack of selectivity might also explain the fact that ketanserin blocks the MDMA-induced hyperthermic effect (Schmidt et al., 1990) since this compound is also an effective α1-adrenoceptor antagonist (McCall & Schuette, 1984). The effect of MDL 11,939 does stand out as being unique. All the other 5-HT2 antagonists in our study failed to antagonize the MDMA-induced hyperthermia, including MDL 100,907 (5-HT2A antagonist) and SB 242084 (5-HT2C antagonist).
Therefore, with the exception of MDL 11,939, the 5-HT antagonist studies reported here suggest that 5-HT2 receptors are not primarily involved in the MDMA-induced hyperthermic response. Supporting this interpretation are the data obtained with 5-HT uptake inhibitors. The 5-HT uptake inhibitor fluoxetine failed to alter the MDMA-induced hyperthermic response. Since both fluoxetine and MDMA are metabolized by the same hepatic enzyme (Crewe et al., 1992), it was conceivable that altered kinetics of one of these two drugs might be suggested as an explanation for this lack of effect. However, a recent study has demonstrated that fluoxetine administration does not alter the concentration of MDMA in the brain following its peripheral administration (Sanchez et al., 2001). Furthermore, administration of the 5-HT uptake inhibitor zimeldine, which is metabolized by an entirely different pathway to fluoxetine (Cashman et al., 1988), also had the same lack of effect on MDMA-induced hyperthermia.
Our studies also demonstrated that while fluoxetine treatment did not alter the hyperthermic response, it had nevertheless esssentially abolished the MDMA-induced 5-HT release in the hippocampus. This region was selected as one in which it was relatively easy to implant a microdialysis probe (in contrast to the hypothalamus, a more rational region for examining the relationship between neurotransmitter function and temperature related changes). While inhibition by fluoxetine of MDMA-induced 5-HT release has been reported by others (Berger et al., 1992; Schmidt et al., 1990; Malberg et al., 1996), we think this is the first study to also examine temperature in the same animals. As mentioned above, it is unlikely that the hippocampus is the locus of any 5-HT mediated thermoregulatory response, the fact remains that MDMA has similar actions in all major forebrain regions, including the hypothalamus (Sabol et al., 1996). It is therefore likely that the changes produced by MDMA and fluoxetine in the hippocampus reflect changes also occurring in other brain regions. It is also noteworthy that the rise in the extracellular concentration of 5-HT in the hippocampus following MDMA is of a considerably shorter duration than the rise in rectal temperature (Figure 5). This, in addition to the fact that inhibition of the MDMA-induced acute 5-HT release does not alter the hyperthermia, suggests that these events may be unrelated.
The data with zimeldine and fluoxetine reported here may also have clinical relevance. Liechti & Vollenweider (2000) recently reported that administration of a low dose of MDMA (1.5 mg kg−1) to healthy volunteers resulted in a modest increase in body temperature which was not attenuated by the 5-HT uptake inhibitor citalopram, even though other MDMA-induced physiological changes were inhibited by this compound.
Although our data fail to provide a clear role for 5-HT in the hyperthermic response, the current study does indicate a role for dopamine. Administration of SCH 23390, a D1 receptor selective antagonist, produced a dose-dependent inhibition of the hyperthermic response. This indicates that MDMA-induced hyperthermia may primarily result from stimulation of D1 receptors and is consistent with the observation that SCH 23390 also antagonizes methamphetamine-induced hyperthermia (Bronstein & Hong, 1995).
While the D2 selective compound remoxipride was without effect on the hyperthermia, D2 receptors may also be involved in some of the temperature changes seen following MDMA administration under specific conditions. For example, administration of MDMA produces hypothermia in rats kept in low ambient temperature conditions (Gordon et al., 1991; Dafters, 1994) while apomorphine, the prototypic D2 agonist produces hypothermia in rats at normal and low ambient temperature, but hyperthermia at high ambient temperature (Faunt & Crocker, 1987). It is, therefore, possible that D2 stimulation predominates in animals adminstered MDMA at ambient temperatures below approximately 18°C, which is why hypothermia is then seen in these animals (Gordon et al.; 1991Dafters, 1994). Similarly, it seems reasonable to propose that the hypothermia observed in the MDMA+SCH 23390 treated rats given the highest dose of SCH 23390 (Figure 7) is due to the total blockade of D1 sites allowing expression of D2 receptor stimulation by the dopamine released by MDMA. Thus one sees an ‘apomorphine-like’ effect and hypothermic response which is in agreement with Faunt & Crocker (1987) who reported that apomorphine-induced hypothermia could be potentiated by SCH 23390.
Since Koch & Galloway (1997) showed that GBR 12909 prevented MDMA-induced dopamine release in vitro in brain slices and our results suggested that dopamine is the neurotransmitter primarily involved in MDMA-induced hyperthermia, it seemed surprising that pretreatment with the dopamine uptake inhibitor GBR 12909 had no effect on MDMA-induced hyperthermia. However, the current data on the role of the dopamine uptake site in MDMA-induced dopamine release is conflicting. The data of Koch & Galloway (1997) using an in vitro technique are hard to reconcile with the fact that, in vivo, in uptake inhibitor mazindol failed to block the acute dopamine depeletion which follows administration of the MDMA-related compound methamphetamine (Marek et al., 1990a).
We therefore decided to examine further the effect of GBR 12909 on MDMA-induced dopamine release in vivo. Nash & Brodkin (1991) had previously studied the effect of GBR 12909 on MDMA-induced dopamine release in vivo but they had infused MDMA directly into the brain which we postulated might have resulted in a different effect to that seen when MDMA was given peripherally. We obtained clear evidence that GBR 12909 has no effect on MDMA-induced dopamine release in the rat when both drugs are given peripherally. This is consistent with a recent study in mice which also demonstrated that GBR 12909 had no effect on MDMA-induced dopamine release in vivo (O'shea et al., 2001). In addition, our experiment explained why (if we assume from our other results that the hyperthermia is dopamine mediated) MDMA-induced hyperthermia was unaffected by GBR 12909.
One explanation for the failure of GBR 12909 to block MDMA-induced dopamine release in the current study could be that we gave an insufficient dose. While Rothman et al. (1991) suggested that the ED50 dose for inhibiting the uptake site was 10 mg kg−1 (the dose used in the current study) and Stephens & Yamamoto (1994) previously found this dose to be sufficient to block the dopamine releasing effect of methamphetamine, other studies have indicated that a much higher dose of GBR 12909 (45 mg kg−1) is required to completely block the dopamine uptake carrier (Nakachi et al., 1995). Further studies on the role of the dopamine uptake site in MDMA-induced dopamine release are, therefore, clearly required to finally determine whether MDMA is or is not transported by the dopamine uptake carrier. However, the fact that a high dose of mazindol (40 mg kg−1) failed to block the dopamine-releasing action of methamphetamine (Marek et al., 1990b) does cast some doubt on the role of the dopamine uptake site in the dopamine releasing action of amphetamine compounds.
There are several studies that demonstrate that MDMA, by increasing 5-HT release indirectly increases dopamine release, via an action at 5-HT2 receptors. For example, it has been shown that co-administration of fluoxetine and MDMA enhances dopamine release (Koch & Galloway, 1997), 5-HT2 receptor antagonists block the MDMA-induced increase in dopamine (Nash, 1990; Schmidt et al., 1992) and 5-HT agonists potentiate dopamine release (Gudelsky et al., 1994). The question therefore arises as to why MDMA-induced hyperthermia was not altered by administration of either fluoxetine or 5-HT2 antagonists if indeed the hyperthermia results from increased dopamine function as we are proposing. At present the most plausible explanation is that this association between 5-HT and dopamine function has only been demonstrated to occur in the striatum. It is probable that the dopamine release involved with hyperthermia is an event associated with a direct effect of MDMA on release in a different brain area, probably the hypothalamus.
In conclusion, we have demonstrated a clear dissociation between MDMA-induced hyperthermia and a change in 5-HT function, particularly that mediated by 5-HT2 receptors. However, a probable association between an MDMA-induced increase in dopamine release, D1 receptor activity and the hyperthermia has been demonstrated. This result suggests that administration of a D1 receptor antagonist could be a logical way to treat patients presenting with an acute and potentially life-threatening MDMA-induced hyperthermia.