Reduced 3,4-methylenedioxymethamphetamine (MDMA, Ecstasy)-initiated oxidative DNA damage and neurodegeneration in prostaglandin H synthase-1 knockout mice

Abstract

The neurodegenerative potential of 3,4-methylenedioxymethamphetamine (MDMA, ecstasy) and underlying mechanisms are under debate. Here, we show that MDMA is a substrate for CNS prostaglandin H synthase (PHS)-catalyzed bioactivation to a free radical intermediate that causes reactive oxygen species (ROS) formation and neurodegenerative oxidative DNA damage. In vitro PHS-1-catalyzed bioactivation of MDMA stereoselectively produced free radical intermediate formation and oxidative DNA damage that was blocked by the PHS inhibitor eicosatetraynoic acid. In vivo, MDMA stereoselectively caused gender-independent DNA oxidation and dopaminergic nerve terminal degeneration in several brain regions, dependent on regional PHS-1 levels. Conversely, MDMA-initiated striatal DNA oxidation, nerve terminal degeneration, and motor coordination deficits were reduced in PHS-1 +/- and -/- knockout mice in a gene dose-dependent fashion. These results confirm the neurodegenerative potential of MDMA and provide the first direct evidence for a novel molecular mechanism involving PHS-catalyzed formation of a neurotoxic MDMA free radical intermediate.

Keywords: 3,4-methylenedioxymethamphetamine; Ecstasy; neurodegeneration; oxidative DNA damage; prostaglandin H synthase; reactive oxygen species.

Figure 1

Electron paramagnetic resonance (EPR) spectra for in vitro PHS-1-catalyzed bioactivation of amphetamines to free radical intermediates. Left panel: vehicle control (upper panel), 3,4-methylenedioxymethamphetamine (dl-MDMA) (middle panel), and phenytoin (lower panel). Each reaction contained 1000U/mL PHS-1, 1.0 μM hematin, and 0.5 mM phenol. After preincubation for 1 min at 37 °C, 67 μM arachidonic acid, 1 mM PBN, and 1 mM of dl-MDMA were incubated for 1 min. The positive control phenytoin (0.5 mM) was incubated with PHS for 30 min as previously described (29). The vehicle control incubation contained all components except the drug. Right panel: time course for in vitro PHS-1-dependent MDMA free radical formation. Shown are the EPR spectra for the incubation of dl-MDMA for 1, 5, and 10 min with PHS-1.

Figure 2

In vitro PHS-1-dependent oxidation of DNA initiated by MDMA. Incubations (the number is indicated in parentheses) included 2′-deoxyguanosine (2′-dG), dl-MDMA or vehicle, PHS-1, hematin, and arachidonic acid as described in Figure 1 (excluding PBN). Phenytoin (PHT) was the positive control. Oxidative DNA damage was quantified by the formation of 8-oxo-2′-deoxyguanosine (8-oxo-dG). Upper panel: effect of drug concentration at 1 min. ^ap < 0.02, ^bp < 0.006, and ^cp < 0.009 compared to their vehicle control. *p < 0.02 compared to 0.25 mM PHT. Lower panel: effect of incubation time. ^cp < 0.009 compared to the vehicle control at the same incubation time point.

Figure 3

In vitro PHS-1-dependent oxidation of MDMA is inhibited by the PHS inhibitor 5,8,11,14-eicosatetraynoic acid (ETYA). Preincubation of the PHS/lipoxygenase inhibitor ETYA (40 μM) with PHS-1 abolished 8-oxo-dG formation for 1 mM dl-MDMA. Incubations (the number is indicated in parentheses) included 2′-dG, dl-MDMA or its saline vehicle, PHS-1, hematin, and arachidonic acid as described in Figure 1 (excluding PBN). One control incubation included DMSO, the vehicle for ETYA, containing all of the components excluding ETYA. ^ap < 0.001 compared to the saline vehicle, ^bp < 0.001 compared to MDMA, ^cp < 0.008 compared to the DMSO vehicle for ETYA.

Figure 4

Stereoselectivity in the PHS-1-dependent oxidation of DNA by MDMA enantiomers. Upper panel: in vitro PHS-1 oxidation of DNA by MDMA. d-MDMA or l-MDMA (1 mM) was incubated for 1 min with the components of the in vitro system as detailed in Figure 1 (excluding PBN) and analyzed for DNA oxidation. The number of incubations is indicated in parentheses. ^ap < 0.001 compared to the vehicle control; *p < 0.04 and **p < 0.001 compared to the l-enantiomer at the same concentration. Lower panel: PHS-1 knockout mice are protected against brain region-dependent in vivo oxidation of DNA by MDMA. dl-, d-, or l-MDMA were dissolved in 0.9% saline and administered in 4 doses (10 mg/kg ip) with each dose given at 2 h intervals. Saline was used as the control. The mice were sacrificed 1 h after the last injection, and tissues from different brain regions were isolated by microdissection and analyzed for oxidative DNA damage. A minimum of 6 mice were used for each group. ^1,3p < 0.04, ²p < 0.002, ⁴p < 0.05, ⁵p < 0.03, and ⁶p < 0.03 compared to the corresponding saline-treated control for the same PHS-1 genotype. ^ap < 0.05 and ^bp < 0.02 compared to dl-MDMA-treated +/+ PHS-1-normal mice. ^αp < 0.009 and ^βp < 0.006 compared to d-MDMA-treated +/+ PHS-1-normal mice. Note the different scales for levels of DNA oxidation in different brain regions.

Figure 5

Relationship of baseline (constitutive or endogenous) and MDMA-enhanced DNA oxidation in brain regions of PHS-1 mice. Treatment and analysis of DNA oxidation is as described in Figure 4, lower panel. The constitutive levels represent the combined values for PHS-1 +/+, +/−, and −/− mice, which were identical for a given region. dl-MDMA-enhanced DNA oxidation was analyzed only in the PHS-1 +/+ mice, which was the only PHS-1 genotype exhibiting a significant increase in oxidative DNA damage. Upper panel: constitutive and dl-MDMA-enhanced DNA oxidation in different brain regions in PHS-1 mice. Lower panel: linear regression curve for constitutive and maximal dl-MDMA-initiated DNA oxidation (R = 0.7639, p < 0.0022). ^ap < 0.001 and ^cp < 0.001 compared to the brainstem and ^bp < 0.001 and ^dp < 0.001 compared to the cerebellum of constitutive brain regions. ^αp < 0.05 and ^βp < 0.001 compared to the cerebellum, ^γp < 0.01 compared to the cortex, and ^δp < 0.01 compared to the hippocampus of dl-MDMA-treated brain regions.

Figure 6

PHS-1 knockout mice are protected against d-MDMA-initiated persistent degeneration of dopaminergic nerve terminals in the striatum. Upper panel: d- (d) or l-MDMA (l) (10 mg/kg ip) was administered in 4 doses, each given at 2 h intervals. Saline (Sal) vehicle was used as the control. The mice were sacrificed 1 week after the last injection and perfused with 10% formalin. Brain sections were stained for tyrosine hydroxylase indicative of dopaminergic nerve terminals. Immunohistochemical staining is representative of n = 3/gender/treatment group; scale bar = 50 μm. Lower panel: quantification of immunohistochemical data reported in the upper panel. The number of animals is given in parentheses. ^ap < 0.001 and ^bp < 0.01 compared to the corresponding saline control for the same PHS-1 genotype; ^cp < 0.001, ^dp < 0.05 and ^ep < 0.01 compared to the corresponding d-MDMA-treated PHS-1 genotype; *p < 0.001 and **p < 0.05 compared to d-MDMA-treated wild-type PHS-1-normal mice; ^αp < 0.001 compared to d-MDMA-treated heterozygous PHS-1 knockout mice.

Figure 7

Slight long-term recovery of striatal dopaminergic nerve terminals in wild-type d-MDMA-treated PHS-1-normal mice. d-MDMA (10 mg/kg ip) was administered in 4 doses, each given at 2 h intervals. The mice were sacrificed 2 months after the last injection and perfused with 10% formalin. Brain sections were stained and quantified for tyrosine hydroxylase (TH), indicative of dopaminergic nerve terminals. The data were plotted as a percentage of TH staining in vehicle-treated wild-type mice (negative control). *p < 0.04 and **p < 0.01 compared to d-MDMA-treated wild-type PHS-1-normal mice sacrificed 1 week after the last injection (10 mg/kg ip). The number of mice is indicated in parentheses.

Figure 8

Female and male PHS-1 knockout mice are protected against d-MDMA-initiated persistent functional deficits. Motor coordination impairment was assessed by the rotarod test at 25 rpm for mice treated with d- or l-MDMA or their saline vehicle. The time at which the mice fell from the rod was recorded as the latency. A minimum of 4 mice were used per gender for each group. Left upper panel: d-MDMA compared to saline. ¹p < 0.001 compared to its genotypically identical saline control. ^ap < 0.003, ^bp < 0.001, and ^cp < 0.04 compared to d-MDMA-treated +/− PHS-1 knockout mice. ^αp < 0.001 and ^βp < 0.002 compared to d-MDMA-treated −/− PHS-1 knockout mice. Left middle panel: l-MDMA compared to saline. No significant differences. Left lower panel: d-MDMA compared to its l-enantiomer. *p < 0.001, **p < 0.03, and ***p < 0.02 compared to l-MDMA-treated +/+ PHS-1-normal mice. Right upper panel: d-MDMA compared to saline. ¹p < 0.001 and ²p < 0.004 compared to its genotypically matched saline control. ^bp < 0.03 compared to d-MDMA-treated +/− PHS-1 knockout mice. ^αp < 0.001, ^βp < 0.02, and ^γp < 0.03 compared to d-MDMA-treated −/− PHS-1 knockout mice. Right middle panel: l-MDMA compared to saline. No significant differences. Right lower panel: d-MDMA compared to its l-enantiomer. *p < 0.001 and **p < 0.006 compared to l-MDMA-treated +/+ PHS-1-normal mice.