| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,4
* Faculty of Pharmacy and
Department of Pharmacology, University of Toronto, Toronto, Ontario, Canada
4Correspondence: Faculty of Pharmacy, University of Toronto, 19 Russell St., Toronto, Ontario, Canada M5S 2S2. E-mail: pg.wells{at}utoronto.ca
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: MDMA • methamphetamine • brainstem • DNA oxidation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
), constitute the largest group of designer drugs, and cause numerous acute and reversible adverse effects, presumably due to receptor-mediated mechanisms, as discussed elsewhere (1)
. This study focused on irreversible changes leading to long-term neurodegeneration. Recently associated consequences include psychiatric and cognitive effects (2
, 3)
and psychomotor disorders (4
, 5)
, possibly resulting from the irreversible loss of monoaminergic nerve terminals, decreased tryptophan/tyrosine hydroxylase activity and long-lasting depletion of serotonin (5-HT) or dopamine (DA) uptake sites (6
, 7)
. The underlying molecular basis of amphetamine-mediated neurodegeneration, and the associated risk factors, are poorly understood and controversial, but reactive oxygen species (ROS) have been implicated (ref 6
and references therein). MDMA and MDA are postulated to be metabolized by cytochromes P450 (CYP) in brain target tissues to a catechol intermediate (8
9
10)
, which is converted to a reactive quinone that undergoes redox cycling and generates ROS. Although these amphetamines are good substrates for CYP isoforms that are abundant in liver, the contribution of this mechanism in brain may be limited by the relatively low CNS expression of CYP isoforms (11)
, and particularly the CYP2D6 isoform that catalyzes the demethylenation of MDMA and MDA to catechol metabolites (12)
. Even with chronic drug exposure, CYP2D6 expression is uniquely resistant to induction in the liver (13)
, and only limited CYP2D6 induction occurs in the brain (14)
. Rats deficient in CYP2D1, the homologue to human CYP2D6, are still susceptible to MDMA and MDA neurotoxicity (15)
. These discrepancies suggest that other mechanisms or enzymes may contribute to brain ROS generation by amphetamines like MDA and METH. In the absence of adequate antioxidative cytoprotection or macromolecular repair, ROS can cause irreversible damage to cellular macromolecules such as DNA, RNA, protein, and lipid membranes, which may result in altered cellular function and cellular death.
|
As an alternative hypothesis in another model with low CYP expression, the developing embryo, our laboratory has investigated the role of embryonic prostaglandin H synthase (PHS) in catalyzing the bioactivation of xenobiotics to teratogenic free radical intermediates (Fig. 2
). PHS exists as two isozymes, cyclooxygenase-1 (COX-1 or PHS-1) and cyclooxygenase-2 (COX-2 or PHS-2) (16
, 17)
. In adults, PHS-1 is constitutively expressed to varying degrees in all mammalian tissues, whereas PHS-2 is constitutive only in the brain and kidney, and is induced rapidly in response to many stimuli such as proinflammatory cytokines and growth factors (18
, 19)
. Numerous teratogens, including phenytoin, benzo[a]pyrene, and thalidomide, are bioactivated by embryonic PHS to free radical reactive intermediates that initiate the formation of ROS, and particularly hydroxyl radicals, which oxidatively damage embryonic DNA and other cellular macromolecules, with embryopathic and teratogenic consequences (Fig. 2)
(20
21
22
23
24
25
26
27)
.
|
Accordingly, we hypothesized that PHS in the brain may catalyze the bioactivation of amphetamines to free radical intermediates that initiate the formation of neurotoxic ROS. In an in vitro study, we investigated whether purified PHS can stereoselectively bioactivate MDA and METH to reactive free radical intermediates that enhance ROS formation and oxidatively damage DNA. In vivo, the time and dose dependency of amphetamine-initiated DNA oxidation, dopaminergic neuronal degeneration and motor function deficits were characterized in relation to brain regional levels of PHS protein and the protective potential of pretreatment with the PHS inhibitor acetylsalicylic acid (ASA). Our results provide the first evidence that amphetamines are excellent substrates for PHS-catalyzed free radical formation, resulting in neuronal DNA oxidation, nerve terminal degeneration and functional deficits.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
95%) and 8-hydroxy-2'-deoxyguanosine were obtained from Cayman Chemical Co. (Ann Arbor, MI, USA). Arachidonic acid, hematin, 
-phenyl-N-t-butylnitrone (PBN), 2'-deoxyguanosine, nuclease P1, Escherichia coli alkaline phosphatase, Ponceau S solution, and acetylsalicyclic acid were obtained from Sigma-Aldrich (Oakville, ON, Canada). Redistilled phenol was from Aldrich Chemical Co. (Oakville, ON, Canada). Chloroform:isoamyl alcohol:phenol (CIP, 24:1:25) was obtained from Life Technologies, Inc. (Burlington, ON, Canada). Broad range (premixed format) protein marker was obtained from New England Biolabs (Mississauga, ON, Canada). Proteinase K and Complete, Mini, EDTA-free protease inhibitor cocktail tablets were obtained from Roche Diagnostics (Laval, QC, Canada). All other reagents used were of analytical or HPLC grade.
Drugs
Pure racemic (d/l)-MDA, its d- and l-isomers, and d/l-METH were provided by the Research Technology Branch of the National Institute on Drug Abuse (Rockville, MD, USA) and by the Healthy Environments and Consumer Safety Branch of Health Canada (Ottawa, ON, Canada).
Mice
Outbred virgin CD-1 female mice, 8- to 10-wk-old (Charles River Canada, St. Constant, QC, Canada) were housed in plastic cages with ground corn cob bedding (Beta Chip; Northeastern Products, Warrensburg, NY, USA) and maintained in temperature-controlled rooms with a 12 h light-dark cycle. Food (Laboratory Rodent Chow 5001; Ralston Purina, Strathroy, ON, Canada) and tap water were provided ad libitum. All animal studies were approved by the University of Toronto Animal Care Committee in accordance with the standards of the Canadian Council on Animal Care.
In vitro bioactivation of METH and MDA to free radical reactive intermediate by PHS-1
PHS-1 (1000 units/mL) was incubated with hematin (1.0 µM) and phenol (0.5 mM) for 1 min at 37°C in 80 mM potassium phosphate buffer, pH 7.9. After the addition of MDA or METH (500 µM or 1 mM) or their vehicle and the free radical spin trap PBN (1 mM), arachidonic acid (67 µM) was added to initiate the reaction. After incubation for 1–10 min at 37°C, reactions were terminated and extracted with 2 mL of ethyl acetate. The ethyl acetate layer was then completely reduced under nitrogen, reconstituted with 300 µL ethyl acetate and analyzed by EPR spectrometry for free radical adducts as described previously (25)
.
Oxidation of 2'-deoxyguanosine (2'-dG)
2'-Deoxyguanosine (1 mg) was incubated with or without MDA or METH in the presence of PHS-1 using the conditions mentioned above with the following alterations: PBN was replaced with 2'-dG and a higher concentration of arachidonic acid (140 µM) was used to start the reaction. The resulting mixture was filtered (0.22 µm) and the filtrate was analyzed by high-performance liquid chromatography with electrochemical detection (HPLC-EC) as described previously (28)
. Samples were injected in duplicate.
Animal treatment and analysis for DNA oxidation
Mice were acclimatized for 1 wk before drug administration. Drugs were dissolved with sterilized 0.9% saline and the drug or its vehicle were injected intraperitoneally (i.p.) in a fixed volume of 0.1 mL/10 g body weight. Mice were administered 4 doses of MDA (10 or 20 mg/kg), METH (5 or 10 mg/kg), or 0.9% saline (vehicle), with each dose given at 2 h intervals. For inhibition of PHS, the same MDA dosing regimen was followed except ASA (100 mg/kg i.p.) was given 2 h prior to the first MDA injection. By irreversibly acetylating a serine residue at the catalytic site (29)
, this dose of ASA inhibits PHS cyclooxygenase activity in vivo, and blocks the embryopathic effects of ROS-initiating teratogens like phenytoin and thalidomide (20)
, as well as MPTP-initiated striatal cell loss, dopamine depletion, and associated alterations in locomotor activity in mice (30)
. The mice were anesthetized by isofluorane and killed at various time points. The brains were isolated, rinsed in ice cold 1.15% KCl solution and subsequently microdissected on ice into the cortex, hippocampus, striatum, substantia nigra, brainstem, and cerebellum. Dissected brain regions were snap-frozen in liquid nitrogen and stored at –80°C until sampling. For imunohistochemistry, mice were perfused with PBS, followed by 10% formalin. Brains were subsequently fixed in 10% formalin. DNA oxidation was quantified by HPLC-EC as described previously (28)
.
Immunoblotting
A 10–30% homogenate was prepared with microdissected brain tissue, in radioimmunoprecipitation (RIPA) buffer (1% NP-40, 0.5% Na+ deoxycholate, 0.1% SDS in PBS) supplemented with Complete, Mini, EDTA-free protease inhibitor cocktail tablets (Roche Diagnostics, Nutley, NJ, USA). The homogenate was allowed to lyse for 3 h at 4°C and subsequently centrifuged for 10 min at 16,000 x g. The supernatant was quantified for total cellular protein using the modified Lowry assay as described previously (31)
. An 80 µg sample of total cellular protein was separated using a 12% SDS-PAGE, under reducing and denaturing conditions and transferred onto a nitrocellulose membrane (Bio-Rad Laboratories, Hercules, CA, USA). Equal protein loading was confirmed by Ponceau S staining. PHS-1 was detected with rabbit polyclonal antiserum against PHS-1 (1:500, Caymen Chemicals, Ann Arbor, MI, USA) and goat anti-rabbit IgG-horseradish peroxidase (1:5000, Santa Cruz Biotechnologies Inc., Santa Cruz, CA, USA). Protein bands were visualized by an enhanced chemiluminescence (ECL) detection system (Amersham Biosciences, Piscataway, NJ, USA).
Immunohistochemistry
Brain sections (5 microns) were deparaffinized in xylene and ethanol, followed by a high temperature unmasking in 0.01 M sodium citrate buffer (pH 6.0) for 5 min. Tissue sections were blocked for at least 2 h with 3% BSA, 20 mM MgCl2, 0.3% Tween 20, and 5% goat serum in PBS followed by an overnight incubation with the goat anti-rabbit tyrosine hydroxylase primary antibody (1:600, Chemicon International Inc., Temecula, CA, USA). Sections were incubated with biotinylated goat anti-rabbit IgG reagent (1:200, Vector Laboratories, Burlington, ON, Canada) and detection was performed using the Vectastain Elite ABC Reagent kit and DAB kit (Vector Laboratories, Burlingame, CA, USA) according to the manufacturer’s specifications.
Quantitation of immunoblot and immunohistochemical data
Determination of protein level and immunohistochemical staining was performed using the NIH Imaging freeware (Scion v4.02). For Western blots, a representative value was taken from the means of the integrated density acquired from three independently obtained measurements from each sample; with a minimum of n = 4/treatment group. For immunohistochemistry, the density of surviving dopamine axon terminals per square millimeter of tissue area in tyrosine hydroxylase immunoreactive-stained (TH-IR) sections were used as an index of striatal density of TH innervation and was measured in coronal sections in the caudate putamen of the striatum. Quantification of TH-IR neurons within a unit area of tissue was conducted using a Leitz DMIL inverted microscope (Leica Microsystems Inc., Richmond Hill, ON, Canada) equipped with a Nikon COOLPIX 995 (Nikon Corporation, Mississauga, ON, Canada). Digital images were made in a standard frame sample area taken at 400x magnification. TH-IR was assessed using counts of TH-stained neuropil labeled above threshold. The threshold was set at the same level for each section sampled. To minimize variations attributed to morphometric analysis associated with immunohistochemical staining and in the quantification technique; three different coronal sections were used for measurements in each CD-1 mouse with 1) saline, 2) MDA, or 3) METH. For each section, 9 fields within the caudate putamen were analyzed bilaterally. Thus for one mouse, the analysis of 3 separate coronal sections gave rise to a representative value taken from the means of the integrated density acquired from 54 (3x9x2) independently obtained measurements. This value was then divided by the unit area of tissue and the result expressed as a percentage. Quantification of tissue staining was performed in a blinded fashion using experimental codes such that animal treatments were not known during measuring.
Behavioral studies
Treated and untreated mice were conditioned and trained on the Economex Rota-Rod (Columbus Instruments, Columbus, OH, USA) before performing the motor coordination test. Mice were given 60 min to adjust to their new surrounds followed by trial conditioning and training. Briefly for trial conditioning, mice were required to perch on the stationary rod for 30 s to accustom themselves before being allowed to run with a constant speed of 5 rpm for 90 s. Mice that succeeded 3 trials (2 h intervals) without falling were tested. Once conditioned, mice were tested by increasing the acceleration to 0.1 rpm/s or testing at a constant speed of 20–30 rpm for a maximum of 5 min. The performance time and speed at which the mice fell from the rod were recorded.
Statistical analysis
Statistical significance of differences between paired data was determined by the 2-tailed Student’s t test, while multiple comparisons among groups were analyzed by the 1-way ANOVA with a subsequent Tukey’s test (GraphPad InStat® 3.05, GraphPad Software, Inc., San Rafael, CA, USA). For data with two independent variables, 2-way ANOVA was performed followed by a Bonferroni post test when F ratios were significant. The level of significance was determined to be at P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The EPR signal for d/l-MDA after a 1 min incubation indicated a carbon-centered free radical, seen as a triplet of doublets known for this radical adduct (20)
. The triplet of doublets observed for this radical adduct of d/l-MDA had hyperfine splitting constants (HFSCs) of aN = 15.08 G and aßH = 2.60 G. The positive control, phenytoin, also gave rise to a carbon-centered free radical with similar HFSCs, aN = 15.46 G and aßH = 2.73 G, which are consistent with published observations (25)
.
|
d/l-METH showed an overlapping, superimposed EPR spectrum consisting of both carbon- and nitrogen-centered free radicals (Fig. 3)
. No radical was detected in the negative control incubation containing only the vehicle solution.
Time-dependent incubations of d/l-MDA with PHS-1 produced a carbon-centered free radical with a maximal at 1 min (Fig. 4
).
|
In vitro DNA oxidation by amphetamines: time and concentration and enantiomeric dependency
d/l-MDA and d/l-METH were bioactivated by PHS-1 in vitro to a free radical reactive intermediate that initiated the oxidation of 2'-dG to 8-oxo-dG. Compared with vehicle controls, at the lowest amphetamine concentration (500 µM), d/l-METH caused a 3.2-fold (P<0.04) increase in DNA oxidation (Fig. 5
, upper panel). d/l-MDA produced a 2.0-fold increase, although this difference was not significant (P<0.09). The positive control, phenytoin (250 and 500 µM), caused a dose-dependent 3.4- to 5.1-fold increase in 2'-dG oxidation (P<0.02 and P<0.006, respectively), consistent with published observations (25)
.
|
A maximal concentration-dependent increase in 2'-dG oxidation was observed within 1 min with 1 mM of d/l-MDA and d/l-METH, resulting in respective 3.0-fold (P<0.0005) and 2.9-fold (P<0.0003) increases in 8-oxo-dG formation compared with vehicle controls (Fig. 5
, upper and lower panel). DNA oxidation was not increased further with a 10 min incubation (Fig. 5
, lower panel).
d-MDA is more potent than its l-isomer in causing acute neurotransmitter depletion (32
, 33)
and induction of acute stereotypic behavior and locomotion (34
, 35)
. At a low dose, d-MDA is more potent in causing long-term neurotransmitter depletion, but this stereoselectivity is lost at a higher dose (36)
. No publications have evaluated stereoselectivity in long-term behavioral deficits. Here, we determined whether stereoselective differences in PHS-dependent DNA oxidation could contribute to the stereoselective MDA neurotoxicity seen in vivo. d-MDA (P<0.005) appeared to cause more oxidative DNA damage than its l-isomer (P<0.01) at 1 mM, with respective 2.9-fold and 1.9-fold increases over the vehicle control. Although d-MDA caused 1.5-fold more DNA oxidation than its l-isomer, this difference was not statistically significant (P=0.058) (Fig. 6
).
|
In vivo DNA oxidation: time- and dose-dependent DNA oxidation by amphetamines in selective brain regions, and abolishment by PHS inhibitors
Both d/l-MDA and d/l-METH produced dose-, time-, and regionally dependent increases in brain DNA oxidation compared with vehicle controls, with sampling beginning 1 h after the last of 4 drug treatments (Fig. 7
, Fig. 8
). Within 1 h, a maximal effect was achieved with the lower dose (10 mg/kg) of d/l-MDA, and both doses of 10 and 20 mg/kg caused a 2- to 3-fold elevation in DNA oxidation in the striatum and substantia nigra, with delayed maximal elevations in the hippocampus at 3 h for the high dose and 6 h for the low dose (Fig. 7)
. Maximal DNA oxidation was seen at 6 h in the cortex for both doses compared with saline controls. DNA oxidation in the brainstem was elevated at all time points compared with their corresponding saline controls.
|
|
A similar pattern of increased DNA oxidation was seen for 5 and 10 mg/kg d/l-METH, with a 2- to 3.5-fold elevation in the striatum and substantia nigra (Fig. 8)
. Delayed maximal elevations were evident in the hippocampus at 3 h and in the cortex at 6 h for both low and high doses. In the cortex, the temporal trend suggested that the level of oxidative DNA damage may be increasing beyond 6 h, at least for the higher dose of METH. DNA oxidation in the brainstem was elevated by only the higher dose of METH at all time points. d/l-MDA and d/l-METH are not known to cause neurotoxic effects in the cerebellum, consistent with our results that showed no increase in DNA oxidation in that region for either drug.
Baseline levels of DNA oxidation in saline controls varied 1.5- to 2.0-fold among the brain regions, with the lowest values in the brainstem and cerebellum, and the highest in the hippocampus, striatum and substantia nigra. A greater, 1.3- to 3.0-fold variation in DNA oxidation at peak time intervals among brain regions was observed after amphetamine treatment, with the highest elevations in the hippocampus, striatum and substantia nigra.
Oxidative DNA damage in different brain regions correlates with the regional level of PHS
Western blotting revealed substantially different constitutive PHS-1 levels in different regions of the brain (Fig. 9
). Highest baseline (constitutive) levels were seen in the substantia nigra and lowest levels in the cortex, differing by 2.6-fold (P<0.001). There was some evidence for the induction of PHS-1 expression by amphetamines, in that PHS-1 protein levels in the striatum and substantia nigra of d/l-MDA- and d/l-METH-treated mice and in the cortex of d/l-METH-treated mice were 
1.5-fold higher in amphetamine-treated animals compared with saline controls (P<0.001).
|
Remarkably, the pattern of brain regional differences in baseline DNA oxidation, as well as DNA oxidation enhanced by both d/l-MDA- and d/l-METH, positively correlated with the level of PHS-1 in the different regions (Fig. 10
). This correlation was highest for MDA (r=0.986, P=0.014), while that for METH was relatively high but marginally significant (r=0.859, P=0.079). PHS-1 levels > 200 densitometric units in any brain region resulted in maximal oxidative DNA damage for both saline controls and amphetamine treatments, although the level of maximal DNA oxidation was 3-fold higher for amphetamine-treated animals compared with saline controls.
|
Ultrastructural damage to dopaminergic nerve terminals
Reduced TH staining was observed in the neuropil of the striatum within 18 h of dl-MDA and dl-METH administration, with progressive decreased density after 1 wk, indicative of degeneration of dopaminergic neuronal terminals (Fig. 11
, upper panel). After 1 wk, the density of dopaminergic terminals was reduced by 70% and 58% in dl-MDA- and dl-METH-treated mice, respectively (Fig. 11
, lower panel). No differences in tyrosine hydroxylase staining were evident in any brain region 6 h after drug treatment, or in the substantia nigra at any time tested.
|
Functional deficits in motor coordination
Structural damage to the dopaminergic nerve terminals in the striatum was confirmed functionally, with a significant impairment in motor coordination observed in the rotarod test (Fig. 12
). Motor deficits were evident beginning 2 wk after the low dose treatment with d/l-MDA (10 mg/kg), but less so for the low dose treatment with d/l-METH (5 mg/kg) (P>0.05). This correlated with a 2-fold greater enhancement of striatal DNA oxidation by d/l-MDA compared with d/l-METH (P<0.009). However, impairment of motor activity was seen for d/l-METH after 3 wk. Deficits in motor coordination were permanent and were observed even up to 6 months after drug administration (P<0.05).
|
DNA oxidation and functional deficits are abolished by a PHS inhibitor
Pretreatment with the PHS inhibitor ASA blocked the enhancement in DNA oxidation caused by d/l-MDA (10 mg/kg) in the striatum, substantia nigra and brainstem compared with control mice treated with only d/l-MDA (Fig. 13
). There was significantly less DNA oxidation in the cerebellum of mice treated with both ASA and d/l-MDA compared with the controls treated with saline or d/l-MDA alone. Pretreatment with ASA also abolished the development of d/l-MDA-initiated functional deficits tested at least up to 6 wk after drug treatment (P<0.02) (Fig. 14
). Treatment with ASA alone did not affect DNA oxidation levels or motor function.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 37
, 38)
. The further production of 8-oxo-dG under these in vitro conditions shows these amphetamine free radicals can initiate the formation of ROS that cause oxidative DNA damage, consistent with a neurodegenerative potential. The apparent stereoselective nature of this amphetamine-initiated DNA oxidation observed herein is consistent with published in vivo evidence of stereoselective neurotoxicity residing with the d-isomers (39)
. Evidence for a similar PHS-dependent bioactivation of amphetamines with resultant CNS DNA oxidation was observed in vivo, and this oxidative DNA damage correlated with permanent neurodegenerative cellular changes and functional motor deficits reflected in the rotarod test. Although motor coordination deficits were not apparent for several wk, in humans the clinical manifestations of Parkinson’s disease do not occur until 80% of dopaminergic neurons are lost from the substantia nigra (40)
, and a similar threshold likely occurs in our mouse model. The ability of a single pretreatment with the PHS inhibitor aspirin to block not only the initial PHS-catalyzed amphetamine bioactivation, but also the ensuing oxidative DNA damage, histological neurodegeneration, and functional deficits are consistent with a common mechanism.
Remarkably, regional differences in DNA oxidation were observed in selective brain regions associated with motor function, specifically the striatum and substantia nigra, and the level of MDA-initiated oxidative DNA damage among these brain regions correlated with the constitutive level of PHS protein in each region, consistent with PHS-catalyzed bioactivation of amphetamines to a neurotoxic free radical intermediate. A similar but less significant relationship was observed with METH. Furthermore, pretreatment with a single dose of the irreversible PHS inhibitor ASA inhibited not only amphetamine-initiated DNA oxidation but also the resulting long-term cellular and functional changes reflecting site-specific neurodegeneration. The short half-life of ASA and its PHS-inhibiting salicylate metabolite (29)
indicates that the protection by ASA was due to inhibition of PHS-catalyzed amphetamine bioactivation. This is consistent with preliminary results in PHS-1 knockout mice that were protected from the neurodegenerative effects of the amphetamine MDMA (41)
. Similarly, PHS-catalyzed bioactivation has been observed to contribute to the mechanism of chemical teratogenesis in the developing embryo (20
, 21
, 42)
, which like the brain has relatively low levels of cytochromes P450 but high expression of PHS-1 and PHS-2 protein (23
, 26)
.
Bioactivation of the amphetamine analogs MDA and METH by purified PHS-1 in our in vitro system generated putative carbon-centered radicals for MDA and METH and a putative nitrogen-centered radical for METH, as reported for a number of proteratogens (25)
. Although the molecular mechanism has yet to be determined, patients chronically taking ASA or other nonsteroidal anti-inflammatory drugs (NSAIDs) appear to be partially protected from some cancers and neurodegenerative diseases (43
, 44)
. Either basal or enhanced PHS activity may constitute a risk factor for teratogenesis, neurodegeneration, cancer, and aging. While our studies focused on PHS-catalyzed bioactivation of substrates to neurodegenerative free radical intermediates, we cannot exclude the possibility of contributions via other PHS-dependent mechanisms, such as alterations in prostaglandin biosynthesis, involvement of prostaglandins in cell signaling, or other PHS-mediated effects (45)
. However, the mechanism underlying the protective effect of a single pretreatment with ASA observed in the current study would not be expected to involve long-term changes in prostaglandin homeostasis.
Based on the time course of the EPR signal for d/l-MDA, the postulated metabolism of MDA by PHS-1 is summarized in Fig. 15
and the chemical basis is provided elsewhere (25)
. These amphetamine hydroperoxides could subsequently react to produce DNA-damaging hydroxyl radicals, as postulated for a number of ROS-initiating teratogens (25
, 46)
.
|
Both PHS-1 and PHS-2 are constitutive in the brain and spinal cord, although there are specific regions where either isoform is predominately expressed (47
48
49)
. The subcellular localization of PHS-1 and -2 includes the lumenal surfaces of the endoplasmic reticulum and nuclear envelope (49)
, the latter location being particularly conducive for catalyzing the bioactivation of amphetamines and other neurodegenerative chemicals to free radical intermediates that enhance DNA oxidation. The particular cellular target (dopaminergic or serotonergic neurons) has been reported to be species-dependent (6
, 50
, 51)
.
Since ASA inhibits both PHS-1 and -2 isoforms, the PHS-2 isoform, which like PHS-1 is constitutively expressed in brain, may contribute to METH bioactivation and its neurodegenerative effects. However, our results do not support an obligatory role for PHS-2 in the long-term neurodegenerative effects of METH, as was recently suggested for METH "neurotoxicity" (52)
. These authors defined neurotoxicity as a decrease in striatal dopamine concentration at 48 h after METH treatment, with no assessment of functional consequences, so it is not clear how this relatively early outcome in a single brain region relates to the longer term effects in multiple brain regions measured in our study, including the loss of dopaminergic nerve terminals measured by tyrosine hydroxylase staining remaining for at least 1 wk after METH administration, and the deficits in motor coordination persisting beyond 5 wk. In our study, the consistent protection by a single pretreatment with the irreversible PHS inhibitor ASA against METH-initiated oxidative DNA damage, nerve terminal degeneration, and motor coordination deficits also is contrary to the results of Thomas and Kuhn (52)
reporting an absence of protection against dopamine loss by multiple treatments with various reversible PHS inhibitors and the similar lack of protection in PHS-1 knockout mice, perhaps due in part to the early timing of their dopamine endpoint. Our results, including a demonstration of the direct bioactivation of both METH and the MDMA metabolite MDA to free radical intermediates by purified PHS-1, together with other studies from our laboratory showing the resistance of PHS-1 knockout mice to the long-term (>4 wk) neurodegenerative effects of MDMA (53)
, suggest that this PHS isoform plays a major role in the long-term neurodegenerative effects of amphetamines.
METH is reported to increase PHS-2 protein expression and lipid peroxidation and decrease DA levels in the striatum of BALB/c mice 72 h after treatment, which the authors postulated was due to PHS induction (54)
. However, neurodegeneration via induction of PHS is not consistent with several observations, including 1) not all PHS-2 induction or PHS inducing agents cause neurodegeneration (55
, 56)
; and 2) in the current study, both early DNA damage and long-term neurodegeneration were blocked by a single dose of the PHS inhibitor ASA. Therefore, 72 h may not constitute an adequate time to assess irreversible neurodegeneration as distinct from a short-term deficit resulting from a receptor-mediated reversible effect. Nevertheless, PHS induction and the subsequent generation of prostaglandins and oxygen free radicals is a commonly invoked mechanism by which PHS is suggested to contribute to various diseases, including cancer, Alzheimer’s disease, Down’s syndrome, and prion diseases (45
, 57
58
59
60)
. The protection by ASA via inhibition of PHS-catalyzed bioactivation of amphetamines to a reactive free radical intermediate is also consistent with a similar mechanism in chemical teratogenesis, wherein embryonic DNA oxidation and the embryopathic effects of several teratogens are reduced in embryo culture and/or in vivo by PHS inhibitors like eicosatetraynoic acid or ASA, or in vivo in PHS knockout mice (20
, 26)
.
Furthermore, our results provide the first evidence that amphetamine-mediated DNA oxidation is an early event, preceding the onset of both lipid peroxidation and protein oxidation reported in other studies (61
, 62)
. These results, together with the observed stereoselectivity of MDA bioactivation, suggest that the free radical intermediates characterized for the amphetamines are relevant to their molecular mechanism of neuronal damage, which may involve oxidative damage to neuronal nuclear and/or mitochondrial DNA.
In normal untreated animals, there is measurable oxidative damage to all cellular constituents in the brain, including nucleic acids, proteins, lipids, and carbohydrates, which significantly accumulate with increased age in most animal models (63)
. The age-related increased oxidation of nuclear and mitochondrial DNA and resultant gene modifications are associated with various oxidative DNA lesions (64
, 65)
that may be further enhanced in brains with neurodegenerative diseases (66
, 67)
. ROS-initiating xenobiotics can substantially enhance this baseline oxidative DNA damage, which if not repaired can cause a disruption of transcription, translation and DNA replication that ultimately may result in mutations, altered cellular function and/or cell death (68
, 69)
. This was evident herein with MDA- and METH-initiated ROS-mediated DNA oxidation in the brains of CD-1 mice, leading to degeneration of dopaminergic nerve terminals in the striatum, the latter of which correlates with published studies reporting a decrease in tyrosine hydroxylase staining, replicated in the current study along with associated functional motor deficits. Although there also was a high level of PHS-1 protein and DNA oxidation in the substantia nigra, we did not observe damage to dopaminergic cell bodies in this region in amphetamine-treated animals, although the possibility of other ultrastructural damage and related functional deficits cannot be excluded. The resistance of substantia nigral cell bodies to amphetamine-initiated neurodegeneration has been reported previously (39)
. Similarly, MDMA causes 5-HT axonal terminal degeneration in rats without any effect on the cell bodies of the dorsal or median raphe nuclei (70)
. The apparent resistance of cells in the substantia nigra suggests that oxidative DNA damage may be necessary but not sufficient for the initiation and/or promotion of neurodegeneration, as with carcinogens that cause oxidative DNA damage in many cell types and tissues, but the cellular development of tumors is highly selective. Also, similar to carcinogenesis and teratogenesis (20
, 21
, 42)
, neurodegenerative oxidative DNA damage may involve changes in gene expression, as distinct from mutational changes. Ongoing rapid PHS-catalyzed bioactivation and ROS formation results in the accumulation of oxidative DNA damage up to a maximum at 1–6 h after drug administration or even later, depending on the brain region, where an equilibrium presumably is reached with the activity of constitutive or induced DNA repair proteins. The decline in DNA oxidation may be further accentuated by the loss of cells with a critical level of oxidative DNA damage, followed by remodeling anomalies. This duration of oxidative DNA damage can be pathogenic, as observed in the developing embryo, where it is sufficient to initiate in utero embryonic death and ultrastructural and functional postnatal anomalies caused by both endogenous and exogenously enhanced oxidative stress, presumably via nonmutational mechanisms involving altered gene expression, and p53, atm, and ogg1 knockout mice deficient in DNA repair are at increased risk (20
, 42
, 71
72
73)
. Differences in enzymatic or nonenzymatic antioxidant levels, and/or DNA damage recognition or repair capacity, may further modulate site-specific susceptibility to amphetamine-initiated neurodegeneration in the brain. We are the first to show motor coordination impairment in mice relating to the decreased DA levels resulting from the dopaminergic neurodegeneration in the striatum. Our results may be clinically relevant since there is evidence for slowed fine and gross motor skills that were directly proportional to decreased striatal dopamine transporters in abstinent METH abusers (4)
. In addition to decreased motor skills, METH abuse has been linked to permanently impaired cognitive skills associated with the striatum (4)
, so it would not be surprising if other brain regions exhibit irreversible changes in macromolecular function as a result of proximal PHS-catalyzed amphetamine bioactivation. The damage to monoaminergic nerve terminals observed herein may similarly provide a molecular basis for the speculation that amphetamines contribute to a variety of neurodegenerative diseases such as Parkinson’s disease, depression, and other psychiatric abnormalities (74
75
76)
.
In summary, our in vitro studies provide the first direct evidence that the amphetamines MDA and METH are bioactivated by PHS-1 and possibly by PHS-2 to free radical intermediates that initiate ROS-mediated DNA oxidation. The stereoselective nature of PHS-dependent oxidative DNA damage provides a novel mechanism for the published observations of stereoselective amphetamine neurotoxicity. These results, together with the correlation of higher PHS-1 regional brain levels with increased severity of amphetamine-initiated DNA oxidation, and the protective effects of a single dose of the PHS inhibitor ASA against amphetamine-initiated DNA damage and ultrastructural neurodegeneration and functional deficits, suggest that PHS-catalyzed bioactivation within the brain may constitute a novel and common molecular mechanism for neurodegeneration caused by some ROS-initiating xenobiotics and endobiotics.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Current address: Covance Laboratories Inc., Vienna, VA, USA.
3 Current address: SRI International, Menlo Park, CA, USA.
Received for publication October 9, 2005. Accepted for publication November 22, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. G. Wells, G. P. McCallum, C. S. Chen, J. T. Henderson, C. J. J. Lee, J. Perstin, T. J. Preston, M. J. Wiley, and A. W. Wong Oxidative Stress in Developmental Origins of Disease: Teratogenesis, Neurodevelopmental Deficits, and Cancer Toxicol. Sci., March 1, 2009; 108(1): 4 - 18. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. W. Wong, G. P. McCallum, W. Jeng, and P. G. Wells Oxoguanine Glycosylase 1 Protects Against Methamphetamine-Enhanced Fetal Brain Oxidative DNA Damage and Neurodevelopmental Deficits J. Neurosci., September 3, 2008; 28(36): 9047 - 9054. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Reese, J. R. Bunzow, S. Arttamangkul, M. S. Sonders, and D. K. Grandy Trace Amine-Associated Receptor 1 Displays Species-Dependent Stereoselectivity for Isomers of Methamphetamine, Amphetamine, and Para-Hydroxyamphetamine J. Pharmacol. Exp. Ther., April 1, 2007; 321(1): 178 - 186. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
