(The FASEB Journal. 2003;17:1775-1788.)
© 2003 FASEB
Speed kills: cellular and molecular bases of methamphetamine-induced nerve terminal degeneration and neuronal apoptosis
JEAN LUD CADET1,
SUBRAMANIAM JAYANTHI and
XIAOLIN DENG
Molecular Neuropsychiatry Branch, NIH/NIDA, Intramural Research Program, Department of Health and Human Services, Baltimore, Maryland, USA
1Correspondence: Molecular Neuropsychiatry Branch, NIH, NIDA, Intramural Research Program, DHHS, 5500 Nathan Shock Dr., Baltimore, MD 21224, USA. E-mail: jcadet{at}intra.nida.nih.gov
 |
ABSTRACT
|
|---|
Methamphetamine (METH) is a drug of abuse that has long been known to damage monoaminergic systems in the mammalian brain. Recent reports have provided conclusive evidence that METH can cause neuropathological changes in the rodent brain via apoptotic mechanisms akin to those reported in various models of neuronal death. The purpose of this review is to provide an interim account for a role of oxygen-based radicals and the participation of transcription factors and the involvement of cell death genes in METH-induced neurodegeneration. We discuss data suggesting the participation of endoplasmic reticulum and mitochondria-mediated activation of caspase-dependent and -independent cascades in the manifestation of METH-induced apoptosis. Studies that use more comprehensive approaches to gene expression profiling should allow us to draw more instructive molecular portraits of the complex plastic and degenerative effects of this drug.—Cadet, J. L., Jayanthi, S., Deng, X. Speed kills: cellular and molecular bases of methamphetamine-induced nerve terminal degeneration and neuronal apoptosis.
Key Words: Bcl-2 family proteins • cDNA array • IEGs • mitochondria • endoplasmic reticulum • neurodegeneration • terminal apoptosis
|
INTRODUCTION
|
|---|
THE USE OF METHAMPHETAMINE (METH) has increased considerably during the past two decades. Although METH abuse was initially reported on the West Coast, recent data have shown significant spread to the rest of the country (1
, 2)
. The frequency of emergency room visits due to acute METH intoxication has increased dramatically in the past few years (3
, 4)
. There were twice as many coroner cases involving METH than cocaine between 1980 and 1988 (2)
. Deaths related to the amphetamines are associated with assaults, suicides, homicides, vehicular accidents, driving impairment, and maternal-fetal and infant exposures (5
6
7
8
9
10)
.
|
CLINICAL TOXICITY
|
|---|
In toxic doses, METH can cause agitation, anxiety, hallucinations, delirium, psychosis, cognitive and psychomotor impairment, seizures, and death (4
, 11
12
13
14
15
16)
. Patients who are METH abusers sometimes arrive in emergency rooms in an unconscious state due in part to the direct effects of intravenous METH use or to the secondary effects of amphetamine-induced seizures (17)
. Other brain disorders associated with amphetamine abuse include cerebral vasculitis, cerebrovascular accidents due to hemorrhage or vasospasm, and cerebral edema (18
, 19)
. Myocardial infarction and death due to heatstroke have also been reported (3)
. There has been recent documentation of neurodegenerative effects of the drug on the human brain. PET studies have revealed substantial loss of striatal dopamine transporters (DAT) in METH users who have been abstinent for an average of 
1–3 years (20
, 21)
. Decreases in DAT were reported to be associated with psychomotor slowing and memory impairments (22)
. The results of PET studies are consistent with observations reported from postmortem investigations that documented decreases in DAT concentration in the basal ganglia of METH abusers (12)
. A recent study using magnetic resonance spectroscopy (MRS) has also provided evidence of possible neuronal death and reactive gliosis in nondopaminergic regions of the brains of METH patients (23)
. These abnormalities might be related to the large doses of METH some intravenous abusers inject during binging episodes (14
, 24)
.
|
TOXIC EFFECTS OF METHAMPHETAMINE IN ANIMALS
|
|---|
Monoaminergic toxicity
Using drug administration patterns purported to reflect METH overdoses or binging by human abusers, several groups of investigators have administered either single large doses or multiple doses of the drug to animals in order to assess its effects on the central nervous system of various mammalian species. However, because there are significant differences in METH elimination half-lives between various mammalian species (25)
, it is likely that even the binging patterns used in those studies are only approximations since it has been suggested that the drug should be given every 30 min to rodents in order to better represent the close to 3 h intervals of self-administration that some human METH addicts use during binges (25)
. This caveat notwithstanding, METH-induced neurotoxicity in monoaminergic systems has been observed in rats given single large doses (in the range of 40–100 mg/kg) (26
27
28
29
30)
or multiple smaller doses such as 15 mg/kg, given four to six times, separated by intervals of 6–12 h (31
32
33
34)
, or 5–10 mg/kg, given four times at intervals of 2 h (35)
. Toxicity in dopaminergic systems was reported in mice given either a single dose of METH (25 mg/kg) (36)
or multiple doses, with the animals receiving four injections of METH varying from 2.5 to 10 mg/kg at 2 h intervals (37
38
39)
. Four injections of METH (0.5–2.0 mg/kg) can also cause damage to DA terminals of nonhuman primates (40
41
42)
.
Some neuronal markers used to assess the integrity of dopaminergic terminals are affected by toxic doses of METH. These include decreases in neostriatal DA levels (39
, 43
44
45)
, decreased activity of tyrosine hydroxylase (TH) (46)
, as well as decreases in the concentration of DA uptake sites (45
, 47
48
49)
and of vesicular monoamine transporters (VMAT-2) (50)
. Although the majority of the earlier papers were concerned with the long-term effects of the drug, some recent studies have provided evidence that METH can inhibit DAT activity in striatal synaptosomes almost immediately after its administration (51
52
53)
. This inhibition is followed by normalization of DAT activity by 24 h after injection, followed by sustained decreases in DAT activity several days later (51
52
53)
.
In addition to the dose of METH used, the age of the animals tested can strongly affect the neurotoxic effects of the drug. In gerbils, METH has been reported to cause an age-dependent degradation in synaptic components, accompanied by an accumulation of lysosomes in fibers and axon filaments in caudate putamen (54)
. In C57BL/6 mice, increasing age potentiates METH-induced striatal DA depletion and reactive gliosis (55)
. METH-induced DA depletion is potentiated in aged rats, with the depletion being correlated with greater production of reactive toxic intermediates in these animals (56)
. Other investigators have sought to pinpoint the age at which rodents become susceptible to METH toxicity (57
, 58)
. For example, only rats of postnatal day (PND) 90 showed long-term deficits in striatal dopaminergic systems whereas adolescent rats (PND 40) did not manifest any such changes (57
, 58)
. Major factors underlying the ontogeny of METH-induced dopaminergic neurotoxicity might include increases in the number of DAT sites, with possible influence on DA uptake (59)
or age-dependent interactions between glutamatergic and dopaminergic systems (60)
. Age-related differences in pharmacokinetics need to be taken into consideration because lower levels of METH were found in the brains of adolescent rats when compared with those of older rats treated with similar doses of METH (10 mg/kg, four injections at 2 h intervals) (58)
.
Damage to serotonin (5-HT) axon terminals in rats has been reported after METH treatment (61
, 62)
. The drug can cause reduction in the concentrations of 5-HT and its metabolite, 5-hydroxylindoleacetic acid (5-HIAA), decreases in 5-HT transporter binding sites (63)
, and reduced activity of tryptophan hydroxylase (TPH), the rate-limiting enzyme in 5-HT synthesis (34
, 64)
. Similar to its effects on dopaminergic systems, toxic doses of METH cause rapid decreases in striatal synaptosomal 5-HT uptake; these are followed by a more prolonged period of reduced activity with a short intervening period of recovery (65
66
67)
. Significant decreases in synaptosomal 5-HT uptake are observed many days after METH injections to rodents (68)
and nonhuman primates (41)
.
Cell death
METH was initially thought to damage only monoaminergic terminals such as those of the DA and 5-HT systems. However, recent in vitro and in vivo studies have provided evidence that this dogma may have been wrong (69)
. For example, Cadet et al. (70)
used an immortalized neural cell line and documented that METH could cause cell death via a process that resembles apoptosis. This report included observations of METH-induced DNA strand breaks, DNA ladder formation, chromatin condensation, as well as nuclear fragmentation. These findings were later supported by the report of Stumm et al. (71)
, who also found that METH killed neocortical cells via the induction of apoptotic pathways. They demonstrated that METH can cause increases in the expression of the proapoptotic Bcl-XS but decreases in the anti-apoptotic Bcl-XL members of the Bcl-2 family of proteins (71)
. Cell death is also observed in rodents after administration of various doses of METH, including 10 mg/kg of the drug given four times at 2 h intervals or after a single dose of 40 mg/kg (72
73
74
75
76
77)
. Administration of METH doses typically shown to cause damage to striatal DA terminals were shown to induce cell death in the rat parietal cortex (77
, 78)
. METH injections can also kill glutamatergic neurons in the rat somatosensory cortex (76)
. Reports by Deng et al. (73
, 75)
and Jayanthi et al. (74)
have provided further evidence that METH can induce neuronal death in rodent brains via a process that resembles apoptosis. Indeed, these effects appear to be more widespread than initially thought, involving not only the striatum and the cortex but the hippocampus, the indusium griseum, and the habenular nucleus (73
, 79)
. Thus, the accumulated data support the view that in addition to its deleterious effects on monoaminergic terminals, METH can cause neuronal apoptosis.
|
MECHANISMS INVOLVED IN METH-INDUCED TERMINAL DEGENERATION
|
|---|
Oxidative mechanisms
The mode of action of METH is thought to involve its rapid entry into the brain followed by influx into monoaminergic terminals, interaction with vesicular monoaminergic transporter, entry into monoaminergic vesicles and displacement of monoamines into the cytoplasm of the terminals, and subsequent release of the monoamines into the synaptic cleft (80
81
82
83)
. METH triggers DA release from the cytosol to the extracellular space by means of reverse transport through DAT (80
, 84)
. Some studies had suggested earlier that DA released in the synaptic cleft was important to the neurotoxic effects of the drug, mostly based on observations that uptake blockers can protect animals against the toxic effects of the drug in DA terminals (26
, 85)
. Although a recent paper reporting that DAT knockout mice are resistant to METH-induced degeneration of DA nerve terminals has provided support for the contention that DA transporters might be indispensable for the expression of METH toxicity on these nerve terminals (86)
, it is nevertheless possible that METH-induced DA release within the cytoplasm of DA terminals might be a more important trigger of its toxicity because VMAT2 knockout mice are more susceptible to the toxic manifestations of the drug (87)
. This idea is supported by the observation that METH toxicity is related to quinone formation, a process dependent on increased DA levels within nerve terminals (88)
. METH-induced quinone formation would be accompanied by production of superoxide radicals and hydrogen peroxide through the process of redox cycling (89)
. Indeed, some investigators have proposed a role for oxygen-based free radicals in the toxic actions of this drug (28
, 44
, 81
, 90
91
92)
. These ideas are supported by the evidence that administration of antioxidants, such as ascorbic acid or vitamin E can attenuate METH-induced damage on monoaminergic terminals (91
, 93)
whereas inhibition of superoxide dismutase (SOD) by diethyldithiocarbamate potentiates the nefarious effects of the drug (91)
. The specific participation of superoxide radicals in the neurotoxic effects of METH on DA nerve terminals was assessed by using transgenic (Tg) mice that overexpress the human CuZnSOD gene (94
95
96)
. These mice have been shown to have much higher CuZnSOD activity than wild-type animals from similar backgrounds (96
, 97)
. The differences in enzyme activity help to show that the protective effects of SOD occurred in a gene dosage-dependent fashion, with homozygous SOD Tg mice showing greater protection than heterozygous mice (95
, 96)
. Since CuZnSOD is involved in the dismutation of superoxide radicals (98)
, these results indicate that METH administration can cause increased production of superoxide radicals via a process that might be secondary to METH-induced DA release within the cytoplasm of DA terminals. This idea suggests that DA released within these locations might be accompanied by DA oxidation within these terminals with redox cycling of dopaquinone and consequent formation of oxygen-based radicals within these sites (88)
. The role of oxidative stress in the biochemical actions of this drug is also supported by the fact that METH can cause lipid peroxidation (96
, 99)
and formation of protein carbonyls (99)
in various regions of rodent brains.
Neuronal damage can occur through excitotoxic damage secondary to glutamate release and activation of glutamate receptors (100)
. Glutamate toxicity appears to depend on the production of nitric oxide in some models (101)
. Oxygen-based and NO-mediated pathways might act separately or in concert to cause degenerative changes observed in neuropathological states (102
, 103)
, including METH neurotoxicity (104)
. The idea of glutamate involvement in the neurochemical effects of METH effects is supported by the demonstration of METH-induced glutamate release in the brain (82
, 83)
. Sonsalla et al. (105)
provided support for this suggestion by showing that glutamate antagonists can attenuate METH-induced dopaminergic toxicity. However, because some of these drugs can suppress METH-induced hyperthermia, their protective effects via their influence on glutamatergic receptors have been called into question. Nevertheless, the observations that knockout mice that are deficient in either neuronal (nNOS) or inducible (iNOS) NOS are resistant to METH-induced toxic damage to monoaminergic terminals have solidified the argument for a role of the glutamate/NO pathway in METH neurotoxicity (106
, 107)
. This idea is further supported by observations that various nNOS inhibitors, which do not affect METH-induced hyperthermia, are also protective (108
, 109)
.
Even though most investigators have sought to determine the role of reactive species in cellular demise, these substances are known activators of various signal transduction pathways (110)
. It is thus not farfetched to suggest that NO might participate in the development of some stimulant-induced behavioral changes (111)
by activating specific transduction mechanisms (110)
. For example, the observations that nNOS knockout mice do not manifest METH-induced behavioral sensitization (111)
is consistent with a role of NO as a retrograde messenger that is involved in neuronal plasticity (112)
and with the proposed participation of glutamate in psychostimulant-induced behavioral sensitization (see ref 113
for further discussion).
Temperature homeostasis
Changes in temperature regulation can markedly influence the expression of METH neurotoxicity, with higher temperatures triggering greater toxicity and lower temperatures affording various degrees of protection (114
115
116
117)
. Many pharmacological compounds initially thought to be protective via interactions at specific receptors were subsequently reported to attenuate or prevent METH-induced hyperthermia (114
115
116
117)
. This does not appear to be the case for DA uptake blockers, which are reported to exert their protective effects independent of any influence on temperature regulation (118)
. It is thus possible that METH-induced increases in core temperature in the rodent might serve to exacerbate the effects of the drug on the production of free radicals in the brain. In contrast, lowering of body temperature might inhibit the formation of these radicals or might block secondary biochemical reactions that participate in the drug-induced toxic cascades.
Ion dysregulation
Independent of DAT function and temperature regulation, changes in ionic homeostasis have been reported to participate in the deleterious effects of METH on monoaminergic systems (118)
. It is thought that because DAT activity can influence ionic flux across membranes, with each molecule of substrate transported by the DAT being accompanied by the cotransport of at least two Na+ ions into DA terminals (119
120
121)
, METH might cause substantial changes in the balance of ions in various compartments within monoaminergic terminals with subsequent impairment of the function of Na+/H+ and Na+/Ca2+ pumps (118)
and associated perturbations of intra-terminal Ca2+ ions (122)
. These changes could play a role in DA-induced terminal degeneration through the activation of calcium-dependent enzymes such as calpain (123)
or activation of lysosomal enzymes (124)
, with subsequent proteolytic cleavage of structural elements essential to the normal functional architecture of dopaminergic terminals. Since mitochondria are present in nerve terminals, it is highly likely that oxidative stress and calcium dysregulation might cause activation of proteolytic cascades within these sites, with the monoaminergic terminals undergoing "falling off" in a fashion similar to what is observed in neuronal apoptosis, except for the absence of nuclear breakdown products. This idea remains to be tested.
|
PATHWAYS OF METH-INDUCED APOPTOSIS
|
|---|
Oxidative mechanisms
Although it is now clear that oxidative mechanisms are involved in the toxic effects of METH on monoaminergic terminals, few studies have tested the role of oxidative stress in METH-induced cell death. This is important because reactive species, including O·2-, OH-, H2O2, NOO·, have been shown to be involved in causing neuronal death in numerous models of brain damage including ischemia and Parkinson’s disease (125)
. To test the possibility that METH-induced reactive species might participate in causing apoptosis in the mouse striatum, CuZnSOD transgenic mice were injected with METH and various markers of apoptosis were measured (72)
. SOD transgenic mice were reported to show decreased number of TUNEL-positive cells, less activation of caspase-3, and decreased PARP cleavage in their striata when compared with wild-type mice treated with similar doses of the drug (72)
. Using cultures of rat fetal mesencephalic cells, Sheng et al. (126)
have provided evidence for the participation of the NO-dependent toxic cascade in METH-induced cell death. It remains to be established whether NO might serve as a trigger for METH-induced apoptosis in the rodent brain.
Activation of the JNK/SAPK–c-Jun pathway
Because METH toxicity appears to be dependent on ROS generation (90)
and ROS can stimulate various signal transduction pathways (110)
, it is likely that the identification of intracellular signals influenced by METH-induced ROS will help to dissect the cellular and molecular mechanisms for the myriad actions of this drug. We thus sought to determine whether c-Jun activation was involved in METH toxicity by further elaborating on our initial cDNA array data that had provided evidence that METH caused very early induction of several transcription factors, including c-Src and c-Jun (127)
. Because of the close relationship between ROS, Src, Cas, and JNK activation (128)
, we wanted to know whether METH, which can promote ROS production (90
, 96)
, might activate the SAPK stress signal cascade. Expression of JNK1 and JNK3 but not that of JNK2 transcripts was indeed increased in METH-treated mice in comparison to saline-treated mice. In the frontal cortex, the increases in expression of the various members of JNK cascade peaked at 8 h. In contrast, METH-mediated up-regulation of these transcripts in the mouse striatum occurred earlier (2–4 h) (129)
.
MAP kinase cascades are believed to be among the most important intracellular signaling pathways for transmission of signals from the cell membrane to the nucleus (130)
. c-Jun activity is regulated via phosphorylation at serines 63 and 73 located in its activation domain. This is mediated by three members of the JNK/SAPKs family of kinases: JNK1, JNK2, and JNK3 (131)
. Phosphorylation of c-Jun potentiates its ability to activate the transcription of AP1 target genes (132)
. Therefore, we measured possible changes in these proteins and in their phosphorylation states by using antibodies specific for the phosphorylated forms of c-Jun and JNK. There was an almost immediate increase in striatal c-Jun protein after the METH injection. Phosphorylation of c-Jun at ser73 was very intense at 4–16 h whereas phosphorylation at ser63 was greatest at 2–4 h after METH treatment. There were METH-induced increases in JNK protein expression in the mouse brain. JNK protein expression increased with time, then showed a pattern that was similar to that observed for phosphorylated c-Jun (129)
. Because the activity of JNK is increased after their phosphorylation at threonine 183 and tyrosine 185 (133)
, we sought to determine whether there were any changes in JNK phosphorylation after METH treatment. This was indeed the case (129)
. These findings are consistent with those of other investigators who had reported increases in phospho-SAPK after injections of a toxic regimen of METH (134)
. We characterized some upstream members of the JNK pathway such as CrkII whose SH2 and SH3 domains have been shown to bind to JNK (128)
and to interact with the tyrosine-phosphorylated effector molecule, Cas. JNK activation can be accomplished by the upstream kinases MKK4 (135)
or MKK7 (136)
. c-Src, Cas, and Crk II proteins all showed increases after treatment with METH; these values normalized after 1 wk. MKK4, also called MEK4, showed transient increases whereas the increases in MKK7 were more prolonged.
Because its optimum action as a transcription factor depends on its phosphorylation at serine residues 63 and 73 (131)
, the METH-induced increases in phosphorylated ser63 and ser73 c-Jun are probably of significance to METH-induced cellular suicide. This idea is supported by the observations of increases in c-Jun phosphorylated at ser63 in models of neuronal death caused by trophic factor withdrawal (137)
and by a report of the involvement of c-Jun phosphorylated at ser73 in ischemia-induced neuronal damage (138)
. Thus, given the observations that JNK appears to be involved in several models of neuronal death (137
, 138)
, METH-induced increases in phosphorylated c-Jun and of its kinases might serve as triggers to METH-induced apoptotic events in the rodent brain (139)
. This suggestion is supported by our recent demonstration that c-jun knockout mice are protected against METH-induced apoptosis (139)
. Figure 1
provides a working model for the involvement of the JNK/SAPK pathway in METH-induced neuropathology.
View larger version (23K):
[in this window]
[in a new window]
|
Figure 1. Mechanisms of activation of the JNK/SAPK pathway after METH administration. METH activates JNK/SAPK pathway via the coordinate regulation of Src-Cas-Crk located upstream of the JNKs. METH-induced formation of free radicals may mediate these Src-dependent signaling events. Although these genes remain to be identified, activation of c-JUN via its phosphorylation might enhance the expression of gene products that are active players in the endgame of apoptosis.
|
|
In addition to their possible involvement in the appearance of METH-induced neuronal apoptosis, MAPK pathways appear to participate in the development of METH-mediated behavioral sensitization (140
, 141)
. These investigators have reported that two mitogen-activated protein phosphatases (MKP-1 and MKP-3) were activated by chronic administration of a dose of METH that has been shown to cause behavioral sensitization (140)
. Their observations of MKP-1 up-regulation are consistent with our findings of METH-induced changes in the SAPK/JNK pathway (129)
, a known MKP-1 activator (142)
. Therefore, when taken together with the observations that toxic doses of METH cause behavioral sensitization (143
, 144)
, these observations suggest that the MAPK pathways might be important triggers of the various neuroplastic and neurodegenerative effects generated by this psychostimulant.
Activation of the mitochondrial cell death pathway
Recent observations have documented a role for members of the Bcl-2 family of genes in METH-induced neurodegeneration. We have shown that overexpression of Bcl-2 can cause significant protection against METH-induced apoptosis of immortalized neural cells (70)
while other investigators (71)
have demonstrated that METH-treated primary cortical neurons showed differential regulation of the Bcl-X splice variants, with induction of the proapoptotic Bcl-XS but inhibition of the anti-apoptotic Bcl-XL variant. Bcl-2 family proteins are functionally divided into death-inhibiting or death-inducing members. Bcl-2, Bclw, and Bcl-XL are known to enhance cell survival whereas BAX, BAK, BAD, and BID are inducers of death (145
, 146)
. Some of their effects might depend on the release of certain apoptogenic factors from mitochondria such as AIF and cytochrome c (147
148
149)
. They might depend on changes in the mitochondrial membrane permeability transition pore (150
, 151)
or on the recruitment of executors of apoptosis such as caspases and DNases (152
153
154)
. Thus, the ultimate cellular responses to pro- and anti-apoptotic signals might be dependent not only on transcriptional and translational changes, but on interactions between Bcl-2 homologs and structurally unrelated proteins (155)
as well as on their proteolytic cleavage, with cellular suicide being the product of adverse ratios of death promoters to death inhibitors (146
, 156)
. These ideas were tested by assessing the effects of METH on the expression of genes of the Bcl-2 family in the mouse brain by using cDNA array analyses (127)
. We observed significant up-regulation of proapoptotic genes of the Bcl-2 family several hours after METH injection. More detailed time course experiments using RT-PCR and Western blot confirmed that METH injections induced the prodeath genes BAX, BAK, BAD, and BID (74)
. Up-regulation of Bcl-2 prodeath genes, which peaked at 8 h postdrug treatment preceded apoptotic cell death, which was maximal at 3 days postdrug (75)
. In contrast, mRNA and protein expression of anti-death genes of the Bcl-2 family showed significant decreases. The most prominent decreases were observed with Bcl-2 and Bcl-XL (74)
. We found that METH caused significant increases in the prodeath/anti-death ratios, for BAX/Bcl-2, for BAD/Bcl-XL, and for BAD/Bcl-2 (74)
. These data indicate that injections of toxic doses of METH can cause a shift in the intrinsic ratios of death promoters to death repressors that might lead to neuronal death in a fashion similar to that reported in other models of apoptosis (157
158
159
160)
.
In addition to the possible involvement of pro- and anti-death genes of the Bcl-2 family (70
, 71)
, METH toxicity is associated with increases in caspase-3 activity and evidence of PARP cleavage in the brain (72)
. These are important regulatory events through which death promoters and caspase-dependent executioners, activated by substances released from the mitochondria, can promote apoptosis (152
, 154
, 155)
. When taken together with the recent in vitro demonstration that METH can cause release of cytochrome c from mitochondria, activation of caspases 9 and 3, as well as activation of DFF40 and its transit to the nucleus (160)
, these in vivo data implicate a formal role of mitochondria in METH-induced neuronal degeneration (Fig. 2
). Other factors released from mitochondria including Smac/DIABLO, endonuclease G, and AIF participate in dismantling cells during apoptosis (161)
. Although the molecular mechanisms involved in their release are not completely elucidated, it is clear that the anti-apoptotic properties of some members of the Bcl-2 family depend on their blocking of the release of these substances from mitochondria (147
148
149)
. In contrast, proapoptotic members of that family are reported to promote their release (162
163
164)
. The possible participation of AIF and Smac/DIABLO in METH-induced apoptosis was recently confirmed in the striata of METH-treated mice (unpublished observations).
View larger version (31K):
[in this window]
[in a new window]
|
Figure 2. Possible participation of ER- and mitochondria-dependent events in METH-induced apoptosis. METH activates caspase-12, a marker for ER-induced stress; this activation might be secondary to ROS-induced Ca2+ dysregulation within the ER and subsequent generation of unfolded proteins within that site. METH can cause release of cytochrome c from mitochondria. This is followed by activation of the caspase-dependent cascade which results in neuronal apoptosis.
|
|
Possible involvement of the endoplasmic reticulum (ER)-dependent stress pathway
In addition to its effects on mitochondria, oxidative stress has been reported to cause ER perturbations (165)
. The ER is a very important organelle that participates in the regulation of cellular homeostasis by regulating calcium signaling and protein folding (166)
. Dysregulation of intracellular calcium homeostasis can cause ER stress and ER-induced apoptosis (167)
. Calcium-mediated cell death is associated with the activation of various proteases (168)
that can cleave substrates, some of which include actin and fodrin, which are essential for maintenance of cellular homeostasis (169
170
171)
. ER stress and calcium dysregulation have recently been implicated in METH-induced cellular demise by our recent observations that apoptotic doses of the drug can cause activation of calpain, a Ca2+-responsive cytosolic cysteine protease (172)
that is an important early mediator of ER-dependent cell death (168)
. ER stress might be secondary to direct effects of METH, a very lipophilic drug (173)
; to METH-mediated oxidative stress (44
, 90
, 96)
; to shifts in the balance of BAX/Bcl-2 ratio induced by the drug (74)
; or to functional impairments of Na+/H+ and Na+/Ca2+ antiporters (118)
. We found, in addition, that apoptotic doses of METH influence the pattern of expression of proteins that participate in ER-induced apoptosis and in the ER-mediated unfolded protein response (UPR) (174)
. These are caspase-12, GRP78/BiP (glucose-regulated protein/immunoglobulin heavy chain binding protein), and CHOP/GADD153 (C/EBP homology protein/growth arrest and DNA damage 153) (unpublished observations). CHOP is a transcription factor that can form functionally negative heterodimers with members of the C/EBP family (175)
, and it has been implicated in causing apoptosis (176)
. A possible role for CHOP in the model of METH-induced cell death is supported by observations that cells that express high levels of CHOP protein show increased expression of active caspase-3 (unpublished observations). Because the link between CHOP and apoptosis is thought to occur in part via down-regulation of Bcl-2 expression and exaggerated production of reactive oxygen species (165)
, it is intriguing to suggest that METH-induced increases in CHOP expression might have contributed to our previous observations of METH-induced decreases in Bcl-2 expression in the mouse striatum (74)
. It is possible that METH-induced increased expression of proapoptotic Bcl-2 family of proteins such as BAX and BAK (74)
might play important roles in causing ER stress. This suggestion is consistent with the report that increased expression of either BAX or BAK resulted in their accumulation in the endoplasmic reticulum (ER) and mitochondria, followed by early caspase-independent Bcl-2-sensitive release of Ca2+ from the ER and subsequent Ca2+ accumulation in mitochondria (177)
. Elevated Ca2+ influx into mitochondria could lead to disruption of cellular metabolism and subsequent cell death via release of cytochrome c and associated activation of the caspase-dependent cell death pathway (Fig. 2)
. This discussion thus implicates ER- and mitochondria-mediated events as important interactive culprits in the manifestations of METH-induced neuropathology.
|
PROTECTIVE MECHANISMS AGAINST METH NEUROTOXICITY
|
|---|
c-Fos activation and DNA repair mechanisms
Although the initial focus on the mechanisms of METH neurotoxicity had been on superoxide radicals, hydrogen peroxide, hydroxyl radicals (44
, 96
, 178)
, and nitric oxide (44
, 106
, 126)
, immediate early genes (IEGs) were also thought to be culpable because injections of toxic doses of the drug can cause sustained increases in c-fos (178)
and of AP-1 DNA binding activity in the rodent brain (179)
. These increases were partially dependent on the production of superoxide radicals because they were attenuated in SOD Tg mice (178
, 179)
.
To test the role of c-fos in METH neurotoxicity, we treated either homozygous (c-fos -/-), heterozygous (c-fos +/-) c-fos knockout or wild-type (WT) mice with toxic doses of the drug (75)
. These knockout mice showed exacerbation of the effects of METH on monoaminergic terminals in the striatum. There was a significant greater number of apoptotic cells in both the frontal cortex and striatum of the knockout mice compared with WT animals (75)
. These findings indicated that the c-fos increases observed after administration of METH might serve protective functions in the METH toxicity model. Although the mechanisms involved in this c-fos-mediated protection remain to be clarified, our data are consistent with those of other authors who have shown that cells obtained from c-fos null mice are more sensitive to the cytotoxic effects of a number of substances (180
, 181)
. Because of the pleiotrophic effects of c-fos (182
, 183)
, we chose to further dissect the mechanisms of protection by making use of cDNA array analysis techniques in order to assess possible differential molecular responses of WT and c-fos +/- mice to toxic doses of METH. We characterized gene expression programs in these mice by measuring their responses to toxic doses of METH that have been shown to cause greater apoptotic responses in the c-fos knockout mice (75)
. We found that 195 genes were up-regulated or down-regulated in either of the two genotypes. Hierarchical clustering of these genes revealed distinct molecular responses of the two genotypes to a toxic regimen of METH (184)
. Hierarchical clustering ( known as Gene Tree) allows one to visualize a set of genes by organizing them into a mock phylogenetic tree. In these trees, genes having similar effects on the gene expression patterns are clustered together. Many of the genes that were up-regulated in the WT but not in the c-fos +/- mice belong to classes of growth factors, cytokines, chemokines, and DNA repair genes. These genes include EGF, FGF 7 and 10, NGFß, GDF5, small inducible cytokines, interleukins 4 and 7, and many others (184)
. These genes may be involved in trophic and/or inflammatory responses induced by the drug since c-fos +/- mice showed less METH-induced reactive astrocytosis (75)
. The fact that the expression of the affected DNA repair genes, which include ATM, XPG, polB, MSH3, lig1, and APE, was observed mostly in the WT suggests that these responses might be related to the differential toxic effects of METH in the two genotypes. The lack of response in the c-fos +/- mice suggests that these mice were not able to mount adequate protective responses to METH-induced toxic insults and provide a partial explanation for the exacerbation of METH-induced apoptosis in the c-fos knockout mice (75)
.
The suggested DNA damage might be secondary to the toxic effects of METH-induced generation of quinone by-products of catecholamine metabolism (88
, 90)
. For example, because catechol-o-quinones can generate of O2- and H2O2, there would be secondary production of hydroxyl radicals due to interactions of hydrogen peroxide with iron to form. Because METH toxicity has been shown to involve the production of nitric oxide (NO) (106
, 126)
, NO could interact with superoxide radicals to form peroxynitrite (185)
. All of these reactive species are known to cause significant damage to DNA (70
, 186)
. To combat DNA damage, organisms have developed complex DNA repair networks (187
, 188)
. DNA repair can occur via base excision repair (BER), nucleotide excision repair (NER), and mismatch repair (MMR). Our observations that METH treatment caused up-regulation of APEX, PolB, and LIG1 suggest that these changes are compensatory increases aimed at counteracting ROS-induced DNA damage using the BER pathway. This idea is consonant with a previous report that oxidative stress can induce the expression of APE/Ref-1 and that such an increase afforded protection against toxic levels of hydrogen peroxide (189)
. In addition, the expression of PolB is stimulated by DNA damage (190
, 191)
whereas DNA ligase I expression is induced has been reported after oxidative stress (192)
. Thus, the exacerbation in METH-induced apoptosis observed in c-fos +/- mice (75)
may in part be related to the lack of compensatory increases in the expression of DNA repair genes after administration of toxic doses of the drug.
Trophic factors, protection, and regeneration
Although the above discussion has focused mainly on DNA repair mechanisms, it is to be pointed out that several trophic factors are up-regulated in the mouse striatum after METH injections (184)
. Several of these trophic factors, including BDNF, NGF (193)
, EGF and FGF (194)
are known to exert trophic or protective effects in various models of neurodegeneration both in vitro and in vivo (see review by Salehi et al., ref 195
). Of special interest is GDF5 (growth/differentiation factor 5), which has been shown to protect nigral dopaminergic cell bodies and striatal nerve terminals of adult rats injected with the catecholaminergic neurotoxin 6-hydroxydopamine (6-OHDA) (196)
. Our cDNA array findings (184)
are consistent with the documented role for trophic factor-induced protection in the METH model because GDNF administration protects against METH-induced toxicity on monoaminergic systems (197
198
199)
. Specifically, Cass (197)
showed that intra-cerebral injection of GDNF given 1 day before METH treatment can prevent METH-induced reductions in striatal DA concentration. They showed in addition that GDNF can restore striatal DA content in rats treated previously with neurotoxic doses of METH (198)
. Melega et al. (199)
reported that GDNF can cause recovery in markers of the integrity of DA systems in vervet monkeys. Thus, it is possible that METH-induced changes in these trophic factors (184)
might be responsible in part for the spontaneous, if delayed, recovery of some DA markers observed not only in the striatum of animals (41
, 200
201
202
203
204)
but also in the human basal ganglia (22)
. The pattern of recovery appears to vary across species and ages of the animals (41
, 200
, 201)
and to depend on doses of METH administered (202)
as well as on the intervals since the last drug use (203)
. Complete recovery has been reported in studies done in rats 12 months after being injected with METH (5 mg/kg), given four times at intervals of 2 h during 1 day (205)
. In contrast, rats treated with higher toxic doses of METH (12.5 mg/kg), given four times at intervals of 2 h, only showed partial recovery (200)
. In nonhuman primates, partial recovery of dopaminergic markers was evident at 12–15 months (201)
or at 4 years (41)
after METH administration. Nevertheless, the proposed scenarios that METH-induced trophic factors might be involved in the observed protracted recovery will need to be tested in mice models in which specific trophic genes have been mutated.
|
CONCLUSIONS AND PERSPECTIVES
|
|---|
Although earlier studies had suggested that METH neurotoxicity was essentially characterized by the destruction of monoaminergic terminals, recent observations have clearly documented that the deleterious effects of the drug are indeed more diverse, consisting of cell death in various brain regions including the cortex, striatum, and hippocampus. In addition, the accumulated evidence indicates that METH-induced cell death is accompanied by activation of several pathways whose roles have been well documented in other models of neuronal apoptosis. These events include increased AP-1 binding activity, activation of the JNK/SAPK pathway, respective up-regulation and down-regulation of prodeath and anti-death genes of the Bcl-2 family, as well as activation of ER- and mitochondria-dependent caspase cascades. These observations suggest that the neuropathological effects of this drug might be dependent on the dysregulation of various cellular and molecular protective transduction signals that attempt to maintain neuronal homeostasis after METH insults (Fig. 3
). Although the discussion in this review has focused on the neurodegenerative effects of the drug, it must be kept in mind that many of the METH-induced transcriptional changes (for example, changes in AP1 transcription factors) have been reported in models of drug-induced behavioral alterations used to mimic drug addiction (206)
. Thus, the possibility exists that these molecular events might participate in the substrates of METH-induced sensitization since such behavioral changes are observed after administration of toxic doses of this drug (143
, 144)
.
View larger version (27K):
[in this window]
[in a new window]
|
Figure 3. Hypothetical schema of possible interactions between AP-1 and Bcl-2 family genes in METH-induced cell death. METH neurotoxicity is exacerbated in c-fos knockout mice but partially prevented in c-jun heterozygous knockout mice. In addition, c-fos heterozygous mice failed to trigger protective molecular responses against METH effects. Thus, these observations suggest that METH-induced increases in c-fos and c-jun expression might constitute attempts by members of a conglomerate that seeks to maintain tissue homeostasis, with warden c-Jun calling in or triggering the executioners but chaperone c-Fos standing on protective suicide watches.
|
|
Because many interesting questions remain to be answered, the schemes summarized in Figs. 1
2
3
must, of necessity, represent only sketches of the complex networks that appear to be involved in causing METH neurotoxicity in monoaminergic terminals and in striatal neurons. For example, it will be important to determine which specific striatal cells undergo METH-induced apoptosis and why some cells are resistant to the toxic effects of the drug. Identification of the molecules that provide protection is of interest because some striatal cells survive excitotoxic damage in models of Huntington’s disease (207)
. Thus, it will be of interest to determine whether the resistant cells in these various models are identical. More information is also needed on the specific role that temperature regulation, aging, and their interactions with oxidative mechanisms might play in the development of METH neurotoxicity. The role that receptor-dependent apoptotic mechanisms such as the Fas pathway (208)
might play in METH neurotoxic processes also remains to be determined.
In conclusion, our approach to the studies of METH effects on the brain, using techniques such as microarray analyses to identify genes of interest followed by dissection of the impact of these transcriptional changes on various brain regions, should help to develop more sophisticated models of the neurodegenerative and neuroplastic effects of this drug. Lastly, these models should be helpful to investigators studying the molecular bases of cell death that occurs in other models of brain damage by identifying common mechanisms that might negatively affect the mammalian nervous system.
|
ACKNOWLEDGMENTS
|
|---|
This work is supported by the DHHS/NIH/NIDA Intramural Research Program.
Received for publication January 23, 2003.
Accepted for publication June 6, 2003.
|
REFERENCES
|
|---|
- Derlet, R. W., Rice, P., Horowitz, B. Z., Lord, R. V. (1989) Amphetamine toxicity: experience with 127 cases. J. Emerg. Med. 7,157-161[CrossRef][Medline]
- Puder, K. S., Kagan, D. V., Morgan, J. P. (1988) Illicit methamphetamine: analysis, synthesis, and availability. Am. J. Drug Alcohol Abuse 14,463-473[Medline]
- Perez, J. A., Jr, Arsura, E. L., Strategos, S. (1999) Methamphetamine-related stroke: four cases. J. Emerg. Med. 17,469-471[CrossRef][Medline]
- Lan, K. C., Lin, Y. F., Yu, F. C., Lin, C. S., Chu, P. (1998) Clinical manifestations and prognostic features of acute methamphetamine intoxication. J. Formos. Med. Assoc. 97,528-533[Medline]
- Stewart, J. L., Meeker, J. E. (1997) Fetal and infant deaths associated with maternal methamphetamine abuse. J. Anal. Toxicol. 21,515-517[Medline]
- Ellinwood, E. H., Cohen, S. (1971) Amphetamine abuse. Science 171,420-421[Free Full Text]
- Kalant, H., Kalant, O. J. (1975) Death in amphetamine users: causes and rates. Can. Med. Assoc. J. 112,299-304[Abstract]
- Logan, B. K. (2001) Amphetamines: an update on forensic issues. J. Anal. Toxicol. 25,400-404[Medline]
- Schepens, P. J., Pauwels, A., Van Damme, P., Musuku, A., Beaucourt, L., Selala, M. I. (1998) Drugs of abuse and alcohol in weekend drivers involved in car crashes in Belgium. Ann. Emerg. Med. 31,633-637[CrossRef][Medline]
- Vega, W. A., Gil, A., Warheit, G., Apospori, E., Zimmerman, R. (1993) The relationship of drug use to suicide ideation and attempts among African American, Hispanic, and white non-Hispanic male adolescents. Suicide Life Threat. Behav. 23,110-119[Medline]
- Simon, S. L., Domier, C., Carnell, J., Brethen, P., Rawson, R., Ling, W. (2000) Cognitive impairment in individuals currently using methamphetamine. Am. J. Addict. 9,222-231[CrossRef][Medline]
- Wilson, J. M., Kalasinsky, K. S., Levey, A. I., Bergeron, C., Reiber, G., Anthony, R. M., Schmunk, G. A., Shannak, K., Haycock, J. W., Kish, S. J. (1996) Striatal dopamine nerve terminal markers in human, chronic methamphetamine users. Nat. Med. 2,699-703[CrossRef][Medline]
- Yui, K., Goto, K., Ikemoto, S., Ishiguro, T., Angrist, B., Duncan, G. E., Sheitman, B. B., Lieberman, J. A., Bracha, S. H., Ali, S. F. (1999) Neurobiological basis of relapse prediction in stimulant-induced psychosis and schizophrenia: the role of sensitization. Mol. Psychiatry 4,512-523[CrossRef][Medline]
- Kramer, J. C., Fischman, V. S., Littlefield, D. C. (1967) Amphetamine abuse. Pattern and effects of high doses taken intravenously. JAMA 201,305-309[CrossRef][Medline]
- Volkow, N. D., Chang, L., Wang, G. J., Fowler, J. S., Franceschi, D., Sedler, M. J., Gatley, S. J., Hitzemann, R., Ding, Y. S., Wong, C., et al (2001) Higher cortical and lower subcortical metabolism in detoxified methamphetamine abusers. Am. J. Psychiatry 158,383-389[Abstract/Free Full Text]
- Conci, F., D’Angelo, V., Tampieri, D., Vecchi, G. (1988) Intracerebral hemorrhage and angiographic beading following amphetamine abuse. Ital. J. Neurol. Sci. 9,77-81[CrossRef][Medline]
- Hall, W., Hando, J., Darke, S., Ross, J. (1996) Psychological morbidity and route of administration among amphetamine users in Sydney, Australia. Addiction 91,81-87[CrossRef][Medline]
- Chynn, K. Y. (1968) Technique for experimental embolization of cerebral arteries and repetitive selective internal carotid arteriography in dogs. Invest. Radiol. 3,275-279[Medline]
- Salanova, V., Taubner, R. (1984) Intracerebral haemorrhage and vasculitis secondary to amphetamine use. Postgrad. Med. J. 60,429-430[Abstract]
- Volkow, N. D., Chang, L., Wang, G. J., Fowler, J. S., Franceschi, D., Sedler, M., Gatley, S. J., Miller, E., Hitzemann, R., Ding, Y. S., et al (2001) Loss of dopamine transporters in methamphetamine abusers recovers with protracted abstinence. J. Neurosci. 21,9414-9418[Abstract/Free Full Text]
- McCann, U. D., Wong, D. F., Yokoi, F., Villemagne, V., Dannals, R. F., Ricaurte, G. A. (1998) Reduced striatal dopamine transporter density in abstinent methamphetamine and methcathinone users: evidence from positron emission tomography studies with [11C]WIN-35,428. J. Neurosci. 18,8417-8422[Abstract/Free Full Text]
- Volkow, N. D., Chang, L., Wang, G. J., Fowler, J. S., Leonido-Yee, M., Franceschi, D., Sedler, M. J., Gatley, S. J., Hitzemann, R., Ding, Y. S., et al (2001) Association of dopamine transporter reduction with psychomotor impairment in methamphetamine abusers. Am. J. Psychiatry 158,377-382[Abstract/Free Full Text]
- Ernst, T., Chang, L., Leonido-Yee, M., Speck, O. (2000) Evidence for long-term neurotoxicity associated with methamphetamine abuse: a 1H MRS study. Neurology 54,1344-1349[Abstract/Free Full Text]
- Connell, P. H. (1968) Amphetamine dependence. Proc. R. Soc. Med. 61,178-181[Medline]
- Cho, A. K., Melega, W. P., Kuczenski, R., Segal, D. S. (2001) Relevance of pharmacokinetic parameters in animal models of methamphetamine abuse. Synapse 39,161-166[CrossRef][Medline]
- Marek, G. J., Vosmer, G., Seiden, L. S. (1990) The effects of monoamine uptake inhibitors and methamphetamine on neostriatal 6-hydroxydopamine (6-OHDA) formation, short-term monoamine depletions and locomotor activity in the rat. Brain Res. 516,1-7[CrossRef][Medline]
- Marek, G. J., Vosmer, G., Seiden, L. S. (1990) Pargyline increases 6-hydroxydopamine levels in the neostriatum of methamphetamine-treated rats. Pharmacol. Biochem. Behav. 36,187-190[CrossRef][Medline]
- Seiden, L. S., Vosmer, G. (1984) Formation of 6-hydroxydopamine in caudate nucleus of the rat brain after a single large dose of methylamphetamine. Pharmacol. Biochem. Behav. 21,29-31
TOP
ABSTRACT
INTRODUCTION
