The Main Molecular Mechanisms Underlying Methamphetamine- Induced Neurotoxicity and Implications for Pharmacological Treatment

1Methamphetamine (METH), also known as “ice” or “crystal,” is an addictive pharmacologic psychostimulant with strong neurotoxic effects on the central nervous system (CNS). It has been abused by >33 million people worldwide and seen a steady increase in use over the last few decades. Such use is associated with deleterious effects on families, loss of productivity, major public-health concerns, and a consumption of substantial resources for medical intervention. A recent editorial in the Lancet stated that the shift in public-health priorities to opioids in the last few years in the United States has enabled the METH market to flourish; as a result, this market is primed for resurgence. Accordingly, drug control may be more challenging than anticipated as a second “METH wave” begins.

2Methamphetamine belongs to a class of synthetic drugs known as amphetamine-type stimulants, which includes amphetamine, METH, methylenedioxy-methamphetamine, and other designer drugs. METH is similar to amphetamine with regard to pharmacodynamic effects; however, users are more likely to become addicted to METH because of its better penetration into the CNS and longer duration of action. Long-term abuse of METH causes serious physical and mental damage. Overall, METH abuse is associated with an increased risk of infection by the human immunodeficiency virus, hepatitis viruses, as well as dangerously high body temperature, periodontal disease, pulmonary hypertension, adrenergic storm, cerebrovascular events, stroke, circulatory collapse, and kidney failure. METH abusers are more likely to develop Parkinson’s disease, depression, schizophrenia, psychosis, and other neuropsychiatric and cognitive sequelae; these are mostly attributed to METH-induced neurotoxicity. The neurotoxic effects of METH are of strong concern, and exploration of the mechanisms underlying this neurotoxicity has become a research hotspot in recent years.

3In general, neurotoxicity is defined as physical damage to neurons. In a broader sense, neurotoxicity may refer to a permanent or reversible adverse effect of a substance on neuronal structure/function that induces disruption of neuronal components, collapse of entire neurons, histologic signs of neuronal injury, and/or behavioral abnormalities. METH-induced neurotoxic effects include damage to dopaminergic and serotonergic terminals, neuronal apoptosis, as well as activated astroglia and microglia that lead to a neuroinflammatory response within the brain.

4In clinical trials, psychological therapies have been shown to generate small-to-moderate reductions in METH use; however, these have not yet been translated into clinical practice. Poor outcomes of psychosocial interventions may be related to METH-produced neurotoxicity. Further, psychosocial treatments must be utilized clinically in conjunction with other strategies as pharmacotherapies.

5In this review, we discuss briefly some of the principal mechanisms underlying the neurotoxicity induced by METH and summarize targeted pharmacologic treatments. We anticipate that more efficacious intervention strategies that protect neural cells against METH-induced neurotoxic consequences may be implemented in the future.

6The neurotoxic mechanism of METH is complex and involves multiple pathways. Oxidative stress has been demonstrated to be a significant factor contributing to cellular toxicity. METH induces the considerable production of reactive oxygen species (ROS), such as hydroxyl radicals (OH-), hydrogen peroxide (H2O2), and the superoxide anion (O2-), by increasing the oxidation of dopamine (DA). METH passes through the blood–brain barrier and penetrates the brain readily due to its high lipid solubility. Then, it enters dopaminergic terminals via the dopamine transporter (DAT) because of its similarity to DA, as well as by passive diffusion. METH enhances DA concentration in the cytosol and synaptic cleft significantly by impairing vesicle monoamine transporter 2 (VMAT2) function and promoting DA release; this process may represent the main mechanism underlying the neurotoxic effect of METH in the brain. Within dopaminergic terminals and in synaptic clefts, excess DA is autoxidized to quinone or semi-quinone to generate large amounts of H2O2, OH-, and O2-. Further, a small proportion of DA metabolism mediated by monoamine oxidase (MAO) or catechol-O-methyltransferase (COMT) produces H2O2 as a byproduct. H2O2 reacts with transition-metal ions to produce highly toxic OH-. Eventually, abundant ROS promote a series of oxidative stress reactions, such as lipid peroxidation and activation of proteases, which trigger the cell-death cascade. Furthermore, highly toxic peroxynitrite ions (ONOO-) produced via O2- react with nitric oxide (NO) to damage proteins, nucleic acids, and phospholipids in cells by circumventing antioxidative enzymes (; Figure 1).

7Mitochondria represent a major site of METH-induced ROS production within neural cells. Mitochondria, which are intracellular organelles composed of two bilayers, act as the energy generators of cells through oxidative phosphorylation and adenosine triphosphate (ATP) production. Defects in mitochondrial respiration have been implicated in neuronal death and several neurodegenerative diseases. Several studies have suggested that dysfunction of mitochondrial metabolism plays a critical role in METH-induced dopaminergic neurotoxicity through inhibition of the Krebs cycle and electron transport chain (ETC) as well as by promotion of oxidative stress; these effects result in imbalance between oxidation and antioxidation in neural cells. ROS and reactive nitrogen species (RNS) generated by DA oxidation inhibit several key enzymes directly as complexes I, II, III, and IV of the ETC, causing mitochondrial dysfunction and damage to DNA structure as well as loss of genetic information. In turn, inhibition of ETC components by METH enhances O2- production due to electron leakage. This positive-feedback loop aggravates disturbance of mitochondrial energy metabolism and neurotoxicity.

8The changes in mitochondrial enzymes in response to METH remain controversial. found significant inhibition of expression of striatal complex I following repeated METH treatment in mice. In accordance with this finding, and showed that intraperitoneal administration of METH decreased the activity of striatal complex I significantly but had no significant effect on the activity of complex IV in vivo or in vitro. Several researchers have found that acute exposure to METH induces a decrease in glutathione (GSH) levels and an increase in levels of oxidized glutathione (GSSG) in the striatum; this leads to a reduction in the GSH/GSSG ratio, which is essential for the inhibition of striatal complex I activity. In contrast, a regimen of rapid (<1 h) binge administration of METH decreased complex II activity in the striatal brain regions, but did not decrease the activity of complex I in vivo. also found that a binge regimen of METH every 2 h, via four successive intraperitoneal injections, did not alter the levels of mitochondrial complex I in striatal synaptosomes in vivo. An in vitro study showed no significant alterations in the protein content of mitochondrial respiratory complex I; however, METH treatment caused time-dependent reductions in the protein contents of complex IV. Although the reactions of the mitochondrial ETC differ according to the route of drug delivery and modes of METH administration, these studies suggest that METH inhibits the activity of the mitochondrial respiratory chain complex; this effect is considered to play a crucial role in the pathogenesis of several psychiatric disorders such as depression, bipolar disorder and schizophrenia.

9In recent years, dynamic disorders of mitochondria have been reported to result from mitochondrial dysfunction triggered by METH. In general, mitochondrial biogenesis coupled with dynamic fusion and fission maintain healthy mitochondria, whereas damaged mitochondria are degraded by mitophagy. The key molecules of mitochondrial biogenesis are peroxisome proliferator-activated receptor gamma coactivator-1α (PGC-1α), nuclear respiratory factors (NRFs), and mitochondrial transcription factor A (TFAM). PGC-1α regulates and coordinates the activity of NRF and TFAM to serve as a nutrient-sensing system that increases mitochondrial biogenesis. Using a repeated escalating METH regimen in rats, detected an increase in expression of PGC1α and TFAM in the substantia nigra, but not the striatum. However, revealed a significant decrease in expression of PGC-1α, NRF, and TFAM in the hippocampus of rats with a different repeated METH administration regimen. These inconsistent results suggest that the indicators of mitochondrial biogenesis appear to be expressed aberrantly in various brain regions or according to the route of drug administration.

10The regulation of mitochondrial fusion is primarily through mitofusin-1, mitofusin-2, and optic atrophy protein 1 (Opa1), whereas mitochondrial fission is regulated by mitochondrial fission 1 protein (Fis1) and dynamin-related protein 1 (Drp1). In in vitro studies,, ) revealed that METH decreased cell viability significantly and increased the levels of Fis1 and Drp1 in isolated mitochondria, whereas Drp1 expression in the cytosol of METH-treated cells showed no significant differences compared with the control group. Interestingly, reported a larger and more rod-shaped morphology and dysfunction of mitochondria in astrocytes during prolonged exposure to low levels of METH. This may have been mediated by inhibition of phosphorylation of Drp1 and an increase in mitofusin levels, implying an overall increase in the number of mitochondria in astrocytes. Considering the difference of dose and time points of METH used in these studies, the regimen of drug administration may be the main cause of the alteration in mitochondrial dynamics. Nonetheless, the mechanisms of mitochondrial impairment induced by METH have not been elucidated to date.

11Oxidative stress occurs primarily in mitochondria and leads to mitochondrial dysfunction by attacking mitochondria in the CNS. Accordingly, METH-induced mitochondrial damage may contribute to dopaminergic toxicity by enhancing susceptibility to oxidative stress and promoting the apoptosis of neural cells. This phenomenon is of clinical relevance as it eventually results in devastating neuropathological effects in the brain due to mitochondrial impairment, subsequent caspase activation, and apoptotic neuronal death following METH administration. For instance, METH exposure has been shown to increase expression of the pro-apoptotic proteins Bax, Bad, and Bid and decrease the expression of the anti-apoptotic proteins Bcl-2 and Bcl-xL. The decrease in the Bcl-2/Bax ratio in mitochondrial fractions has been shown to promote cytochrome c (Cyt c) release from mitochondria. Subsequently, Cyt c becomes part of the apoptosome with apoptotic peptidase activating factor-1 (Apaf-1) and induced sequential activation of the apoptosis executioners caspase-3, -6, and -7.

12Several studies have reported that multiple molecules are involved in the apoptotic death of neurons induced by METH. For instance, found that inhibition of protein kinase C delta (PKCδ) or overexpression of a cleavage-resistant PKCδ mutant protected against METH-elicited apoptosis in mesencephalic dopaminergic cell cultures in vitro. In follow-up studies, they showed that PKCδ inhibition may rescue METH-elicited mitochondrial burden, pro-apoptosis, and dopaminergic degeneration, implying that PKCδ is an important gene involved in METH-induced mitochondrial dysfunction and apoptosis in dopaminergic neuronal cells.

13Recently, reported that p53-upregulated modulator of apoptosis (PUMA) was involved in the mitochondrial apoptotic pathway induced by METH in PC12 cells and SH-SY5Y cells. They suggested that PUMA interacts with Bax and Bcl-2 in mitochondrial membranes to drive Cyt c relocation from mitochondria to the cytoplasm, causing activation of caspase-3, poly-ADP-ribose polymerase (PARP) and apoptosis. Furthermore, through microinjection of anti-micro (mi)R143 into the hippocampi of mice, revealed that miR143-dependent PUMA upregulation reversed the METH-induced decrease in microglial survival via regulation of apoptosis and autophagy. In addition to miRNAs, long non-coding RNAs (lncRNAs) appear to participate in METH-induced neuronal apoptosis by regulating the coding genes of neurons. In our recent investigation, we reported that several lncRNAs were expressed differentially in primary cultured prefrontal cortical neurons treated with METH. Further, using bioinformatics, we hypothesized that lncRNA GAS5 modulates downstream molecules involved in p53-mediated neuronal apoptosis, although more direct evidence from in vivo and in vitro studies is needed (Figure 2; ).

14Glutamate (Glu), which is the main excitatory neurotransmitter in the brain, has been reported to have an important role in METH-induced excitotoxicity. Glu accumulation over-activates various downstream signaling pathways, mostly involving a surge in Ca2+ influx, to trigger an increase in intracellular Ca2+ concentration. Specifically, excessive Glu activates N-methyl-D-aspartate receptors (NMDARs) and metabotropic glutamate receptors (mGluRs). Activation of mGluRs induces protein kinase C (PKC) phosphorylation and upregulates NMDAR function, leading to an increase in Ca2+ influx, which acts as a pervasive and pluripotent second messenger. Excess intracellular Ca2+ triggers a cascade of reactions within cells to activate protein kinases, phosphatases, and nitric oxide synthase (NOS). The latter subsequently promotes NO production, which causes endoplasmic reticulum (ER) stress, activation of apoptotic pathways and, eventually, METH-produced neurotoxic sequelae.

15In particular, ER stress occurs under various toxic stimuli along with accumulation of misfolded proteins and activation of unfolded protein response (UPR), which removes unfolded and/or misfolded proteins in the ER, thereby recovering ER homeostasis. ER stress leads to the activation of three ER-resident transmembrane proteins: activating transcription factor-6 (ATF6), inositol requiring protein-1 (IRE1) and protein kinase RNA-like ER kinase (PERK). The roles of these three signaling pathways involve a reduction in protein synthesis and expression of specific genes to cope with proteotoxic stress. During prolonged ER stress, IRE1, PERK, and ATF6 may induce pro-apoptotic signaling through activation of C/EBP homologous protein (CHOP), which subsequently leads to the initiation of ER stress-mediated apoptosis through regulation of Bcl-2 family members. ER stress leads to apoptosis, including various mechanisms involving activation of death receptors and participation of the mitochondria-dependent cell death pathway. It has been shown that neurotoxic doses of METH induce expression of several ER stress genes, including those that encode the 78-kDa glucose-regulated protein (GRP-78), CHOP, and ATF4, in the rat striatum. The ER stress induced by METH appears to be associated with dopaminergic toxicity and activation of the DA D1 receptor. Recently, suggested that METH-induced apoptotic death is mediated (at least in part) through an ER-dependent mechanism involving CHOP, spliced X-box binding protein 1 (XBP1), caspase-12, and caspase-3 in vitro. In addition, exposure to relatively high-dose METH increases nuclear protein 1 (Nupr1) expression, which promotes dopaminergic neuronal apoptosis and autophagy through a Nupr1/CHOP pathway (Figure 3; ).

16Reactive neurogliocytes are considered to be sensitive markers of nerve damage, which is a common response to CNS injury. The neuroinflammation caused by METH shows a close correlation with microglial activity as it is activated rapidly after METH administration in DA-innervated areas. Microglial activation following METH exposure may result from the neuronal release of damage-associated molecular patterns (DAMPs). For instance, high-mobility group box-1 (HMGB1) expression was found up-regulated in response to METH treatment and shown to mediate the neuroin?ammatory response in the nucleus accumbens, ventral tegmental, and prefrontal cortex of the brain. Thomas and colleagues suggested that METH-induced microglial activation was regulated by DA-quinone (DAQ), a metabolite of DA, because it activated microglia dose- and time-dependently, and because inhibition of DAQ formation blocked (at least partially) microgliosis. The vital role of DA in neuroinflammation was also evidenced by the observation that excessive DA released into the synaptic cleft may stimulate regional microglia directly and trigger a neurotoxic signal cascade. The underlying mechanism of microglial activation induced by METH is related to the Toll-like receptor 4 (TLR4) located on microglia, which is involved in the immune surveillance of pathogens and exogenous small molecules. Activation of microglia may also be mediated through the sigma-1 receptor, which involves ROS generation and activation of the mitogen-activated protein kinase (MAPK) and PI3K/Akt pathways in the neurotoxicity of METH.

17Activated microglial cells secrete not only neurotrophic factors to prolong neuronal survival, but also cytotoxic mediators and cytokines that induce inflammation and neurotoxicity. Studies have suggested that METH activates nuclear factor-kappa B (NF-κB), inducing its transfer to the nucleus and promoting the transcription of pro-inflammatory cytokines in microglia. This results in the release of various pro-inflammatory factors such as interleukin 6 (IL-6), interleukin 1β (IL-1β), tumor necrosis factor-α (TNF-α), monocyte chemo-attractant protein 1 (MCP-1), and cellular adhesion molecule (ICAM-1), which are thought to play key roles in METH-induced neuroinflammation (Figure 4). Interestingly, astrocytes (which protect neurons and promote sprouting) are activated because glial fibrillary acidic protein (GFAP) immunoreactivity increases in the striatum and indusium griseum upon METH treatment, thereby implying complex mechanisms involving guidance molecules and cytotoxic mediators in the neurotoxic consequences of METH.

18The effects of long-term exposure to METH are different from those of acute injury. The former is characterized by impairment of expression of tyrosine, tyrosine hydroxylase (TH), DAT, and serotonin transporters (SERT), as well as DA depletion, decrease in the density of DA D2 receptors, hypo-dopaminergic status, and neuronal degeneration. In observed decreased expression of DA, TH, and DAT in the striatum of chronic METH abusers post mortem, which might explain the dysphoric effects of the drug and dose escalation observed in some METH users. The decrease in levels of dopaminergic markers in the striatum was shown by other studies, and could last for months to years after METH abstinence in human abusers of METH. Besides the striatum, a long-lasting decrease in DAT levels was also observed in the nucleus accumbens and prefrontal cortex of the brain. In accordance with this study, several previous reports have demonstrated that the density of DAT, VMAT2, SERT, and DA D2 receptors is significantly lower than that in healthy controls according to positron emission tomography (PET) scans of the brains of METH abusers. Similar to that observed in METH abusers, significant and sustained reductions in the levels of DA, TH, and DAT were shown following single, high-dose, or binge administration of METH in animal models (Figure 1;;; ). These results suggest persistent DA deficits as well as structural and metabolic dysfunctions in specific brain regions that correlate with several types of behavioral neurologic sequelae induced by METH.

19When using a regimen that involves gradual increases in METH administration to rats to mimic progressively larger doses of the drug used by some human abusers of METH, METH preconditioning protects against DA depletion caused by binge METH challenge in the brain. Further, chronic injections of METH activate programs that prevent DA toxicity without influencing drug-induced pathologic changes in serotoninergic systems. The plethora of exciting applications of preconditioning in neuroprotection reflect differences in the mechanisms involved in METH toxicity on monoaminergic systems between the various routes of METH administration. For instance, reported differential METH-induced gene expression in the striatum, including brain-derived neurotrophic factor (BDNF), heme oxygenase-1 (HO-1) and heat shock protein 27 (Hsp27), in the absence and presence of METH preconditioning and between various brain regions. This result implies that the brain cannot be considered as a homogeneous structure when assessing the molecular effects of METH preconditioning (; Table 1).

20Preclinical and clinical investigations have been applied to seek effective and efficacious pharmacologic strategies for treatment of METH neurotoxicity. As stated above, METH interferes with DA reuptake and leads to DA oxidation; this leads to the production of ROS and RNS, which trigger degeneration of dopaminergic terminals and neuronal apoptosis. This pathway demonstrates that oxidative stress is one of the main mechanisms through which METH injures the CNS. Accordingly, antioxidant pharmacotherapies have been applied to explore efficacious strategies for the protection of neural cells suffering from oxidative stress generated by METH.

21For example, vitamin C (Vit.C) reduces the production of free radicals, maintains GSH homeostasis, and induces the expression of HO-1, which is critical in limiting cellular damage by maintaining redox homeostasis within the brain. found that pretreatment with Vit.C enhanced METH-elicited HO-1 expression and attenuated METH-induced ROS production in neuronal/glial co-cultures. Conversely, pharmacologic inhibition of HO-1 activity abolished the suppressive effects of Vit.C. Recently, found that treating cells with Vit.C before METH exposure attenuated production of ROS and Beclin 1 time-dependently, suggesting that the protective effect of Vit.C against METH toxicity is achieved via the attenuation of ROS production and apoptosis, and that HO-1 induction by Vit.C serves as a strategy for alleviation of METH neurotoxicity. However, an in vivo study to test the neuroprotective action of Vit.C in terms of METH exposure has not been conducted.

22Selenium is an essential mineral found naturally in water, soil, and food. It is often used as an antioxidant and dietary supplement. In an in vivo study, found that selenium supplementation for 13 weeks significantly blocked a METH-induced increase in the lipid-peroxidation marker malondialdehyde (MDA), decreased the ratio of GSH/GSSG, and appeared to attenuate the loss of DA in the striatum and substantia nigra. Consistently, demonstrated that selenium pretreatment in drinking water for 1 week prevented (at least in part) the depletion of striatal DA induced by METH exposure in rodent brains. Further in vitro research supported their findings by showing that the increased oxidative stress induced by METH is reversed upon the treatment of SH-SY5Y neuronal cells with selenium, possibly through a reduction in glutathione peroxidase (GPx) levels. This effects is attributed to the incorporation of selenium into the amino acid selenocysteine by GPx1 and GPx4, which have antioxidant functions. However, in view of the narrow range between the therapeutic and toxic doses of selenium, as well as the dependence of the effect on the applied formulation, dose, and method of treatment, supplementation should be undertaken with appropriate precautions and avoidance of the side-effects of selenium.

23Considering that mitochondrial dysfunction plays an important part in METH-induced neurotoxic insult, improvement of mitochondrial function might ameliorate the oxidative stress and neural damage associated with this disorder. The mood stabilizers lithium and valproate attenuate a series of METH-induced changes, such as reduction of mitochondrial Cyt c, the mitochondrial anti-apoptotic Bcl-2/Bax ratio, activity of mitochondrial Cyt c oxidase, and the expression of several proteins related to the apoptotic pathway. These phenomena illustrate that lithium and valproate enhance mitochondrial function and protect against the mitochondria-mediated toxicity of METH. found that lithium and valproate attenuated the effects of METH on the activity of enzymes in the Krebs cycle, thereby alleviating the impairment of respiratory chain complex activity (complexes I, II, III, and IV). In their study, the effects of lithium and valproate on some enzymes and in some brain areas were not always identical. For instance, interventions using lithium and valproate reversed the decrease in complex II activity in the hippocampus and striatum of rats; however, in the amygdala and prefrontal cortex, valproate (but not lithium) increased complex II activity, notably in the METH group. This region-specific effect on oxidative stress is in accordance with the previously shown heterogeneity of oxidative-stress parameters across brain regions and treatment regimens.

24Nicotinamide, which is a co-factor in the ETC, has been reported to be an efficacious treatment for mitochondrial encephalopathies, possibly through energy repletion. revealed that nicotinamide treatment before METH injection attenuated reductions in striatal DA and ATP content 5 days later in vivo. These findings suggest a close relationship between METH-induced perturbations of energy metabolism and dopaminergic neurotoxicity, and that potential therapeutic strategies might bypass bioenergetic defects if defective mitochondrial energy metabolism has a role in neurotoxicity. Inconsistently, found that nicotinamide perfusion during METH administration had no effect on the long-term toxicity induced by METH. Given the localized effect of nicotinamide on nerve terminals (i.e., striatum), it is likely to produce consequences vastly different from those produced by systemic administration of METH. The distinct routes of administration in the two studies, as well as altered pharmacokinetics and bioavailability, might account for this discrepancy.

25It has been reported that METH users experience a higher risk of developing Parkinson’s disease despite the different brain areas affected by METH and Parkinson’s disease. Talipexole is used as an antiparkinsonian agent in Japan. It was shown to react strongly with OH-, with similar reaction kinetics against METH neurotoxicity in vitro and in vivo, indicating the potency of the neuroprotective action of talipexole due to its scavenging of OH-. In addition, suggested that, in principle, the striatal DA deficiency caused by METH may be corrected by DA substitution if safety and patient selection could be resolved. We hope that these findings will prompt researchers to investigate the potential of antiparkinsonian drugs and therapies targeted at specific brain regions for amelioration of the effects of METH on the brain.

26As discussed above, the excitatory toxicity of METH is closely related to the high release of Glu, which activates NMDARs and GluRs, leading to the influx of excess Ca2+ and triggering a series of intracellular cascade reactions. Therefore, several drugs targeting these receptors have been explored in recent years.

27For example, melatonin is a potent protector against oxidative damage and regulates the movement of free Ca2+ intracellularly in the CNS. Using a melatonin preparation, demonstrated that the suppression of neuronal nitrogen-activated protein kinase and alteration of the NMDAR subunits NR2A and NR2B induced by METH were attenuated in the hippocampus of mice. suggested that melatonin reduced the METH-elicited inflammation, autophagy, and death of hippocampal progenitor cells in vitro. These findings are of interest to clinicians as melatonin protects against mitochondrial dysfunction, apoptosis, and dopaminergic degeneration, which are considered to contribute to several psychosomatic manifestations in METH abusers.

28N-acetylcysteine (NAC) is a pro-glutamatergic compound. It is of considerable interest because NAC attenuates the excitatory toxicity of Glu by standardizing extracellular Glu levels in the nucleus accumbens via stimulation of the cystine–glutamate antiporter. Other actions of NAC include antioxidant activity, modulation of DA release, improvement of mitochondrial dysfunction, and reductions in levels of pro-inflammatory cytokines. Several investigators have examined the effects of NAC in METH-exposure models owing to its broad efficacy against neuropsychiatric disorders involving schizophrenia, bipolar disorder, depression, memory impairments, and cognitive sequelae. They found that pre-treatment with NAC significantly improved the reduction in density of DA transporters induced by high-dose METH treatment in mice, rats, and monkeys using PET. These results suggest that NAC reduces neurotoxic damage and possibly alleviates associated neuropsychiatric symptoms in METH abusers; these data were supported by clinical findings. Nonetheless, full-scale clinical trials to establish definitively if NAC has a therapeutic benefit (and the nature of this benefit) are needed.

29Other chemical compounds targeting excitatory receptors have also shown therapeutic potential against METH. found that dizocilpine, a non-competitive NMDAR antagonist, acted as a potent anti-convulsant and protected dopaminergic neurons in the striatum against METH-induced neurotoxicity in vivo. reported that topiramate showed considerable potential in reducing the excitatory toxicity of METH. They hypothesized that topiramate has a complex mechanism of action that includes antagonism of several Glu receptors, blockade of voltage-dependent Na+ channels, and inhibition of carbonic anhydrase. In addition, suggested that neuropeptide Y, with receptors Y1, Y2, and Y5, reduce Glu release to protect neurons from METH excitotoxicity. Activation of Y1 or Y2 receptors prevents METH-induced cell death, and the Y1 subtype is responsible for blocking the decrease in neuronal differentiation induced by METH.

30The excitability of METH has been shown to be closely related to ONOO- generation, which can be protected selectively by nNOS inhibitors, ONOO- scavengers, or decomposition catalysts. nNOS inhibitors, such as 7-nitroindazole (7-NI), AR-R17477AR, S-Methylthiocitrulline, 3-bromo-7-nitroindazole, and the new nitrone-based radical scavenger S34176, have shown significant protective effects and been used in investigations of METH-evoked neurotoxicity. found that 7-NI generated hypothermic effects and afforded total protection against the DA depletions elicited by METH in the striatum of mice at 20°C. However, at 28°C, 7-NI produced minimal effects on body temperature and failed to alleviate METH-induced DA reductions. This findings suggest that the neuroprotective action of 7-NI was likely related to its ability to induce hypothermia. revealed that pretreatment with 7-NI before METH injection afforded protection against the depletion of dopaminergic markers induced by METH, but did not affect persistent hyperthermia at a low dose. These data suggest that diminished production of NO by nNOS inhibitors, rather than thermoregulation, might prevent METH-produced neurotoxicity. However, the detailed mechanism of action, such as the possible role of NO regulation by 7-NI, has not been elucidated.

31The inflammatory response induced by activated microglia plays a crucial role in the neurotoxicity of METH. Hence, blockade of microglial activation seems to be a promising method for the suppression of METH-induced neurotoxic effects. The antibiotic minocycline has anti-inflammatory and neuroprotective effects in the CNS that are thought to be mediated by the inhibition of microglial activation. observed that the reduction of DA and DAT immunoreactivity after repeated administration of METH was attenuated dose-dependently if minocycline was used in the striatum of mice. They suggested that minocycline may (at least in part) protect against METH-induced neurotoxicity via inhibition of microglial activation in vivo. The phosphodiesterase-4 inhibitor ibudilast increases brain levels of glial-derived neurotrophic factor and reduces microglial activation and production of pro-inflammatory cytokines. Clinical trials have shown that ibudilast can improve METH-induced acute injury. Further, ibudilast reverses the decrease in levels of synaptic signaling protein produced by chronic intake of METH. Although further clinical trials and animal experiments are needed, minocycline and ibudilast may have therapeutic potential for METH neurotoxicity through modulation of neuroimmune signaling. In addition, revealed that modafinil (a cognitive enhancer with weak stimulant-like behavior used for the treatment of sleep apnea, narcolepsy, and shift work-induced sleep disorders) prevented METH-induced microglial and astroglial activation in the human brain, thus avoiding the induction of inflammatory processes.

32Immunotherapies reduce the amount of drug in the CNS by triggering the production of antibodies binding the drug molecule after systemic absorption of METH. found that, in rats receiving a KLH-conjugated METH-like hapten vaccine (MH6-KLH) and the vaccine succinyl MA (SMA–KLH), higher antibody titer-dependent METH serum concentrations, yet lower METH concentrations, in the CNS were observed, which suggested reduced METH concentrations in the brain. In addition, it was shown that anti-METH/AMP mAb4G9 therapy protected maternal and fetal rat brains from METH-induced damage. Another study reported that a human–mouse monoclonal antibody binding to methylphenidate, named anti-METH antibody (mAB7F9), elicited a significant reduction in the METH concentration. In particular, a combined approach using monoclonal and polyclonal antibodies was more effective in reducing the METH concentration in the brain. However, the limitations of a vaccine remain; these include incomplete blockade of drug effects, prolonged delay in the production of sufficient circulating antibodies, and considerable variation in the antibody titer. Furthermore, given that anti-METH antibodies cannot cross the blood–brain barrier, this therapy may be too expensive because antibodies must be administered repeatedly to maintain an effective level.

33Long-term exposure to METH leads to DA depletion, impairment of dopaminergic markers, and neuronal degeneration. Therefore, extensive investigations, including those of several drugs in clinical trials, have been performed to explore efficacious pharmaceutical interventions for recovery from neural injury resulting from METH use. For instance, reported that 9 cis-retinoic acid (9cRA), a biologically active derivative of Vit.A, reversed the METH-induced decrease in immunoreactivity and apoptosis of T-helper cells via inhibition of the export of nuclear receptor 77 from the nucleus to the cytosol in primary cultured dopaminergic neurons. Their in vivo experiments showed that 9cRA (delivered via the intracerebroventricular route) also antagonized the immunoreactivity of T-helper cells and locomotor activity in the striatum of rats indirectly through a signaling mechanism involving bone morphogenetic protein. This result is similar to that of a study by showing that 9cRA produced neuroprotective effects in a rodent model of Parkinson’s disease involving reduced rotational behavior and loss of T-helper cells in the substantia nigra, but increased DA release in the striatum. In the report by, 9cRA was repeatedly administered through the intra-nasal route as intracerebral delivery is not feasible for repeated drug administration and may require chronic cannulation, whereas small molecules bypass the blood–brain barrier and reach brain parenchyma non-invasively if delivered via the intranasal route.

34Cognition-enhancing as well as antidepressant and neuroprotective effects are conferred by 7, 8-Dihydroxyflavone (7, 8-DHF), a high-affinity tropomyosin receptor kinase B (TrkB) agonist that activates downstream signaling. observed that 7, 8-DHF significantly prevented the reduction of DAT and microglial activation in the striatum of mice after repeated METH administration. In their study, it was investigated whether 7, 8-DHF prevents neurotoxicity through a signaling pathway upstream of DA terminals, and if reduction of microglial activation occurs as a consequence (rather than a cause) of METH-induced neurotoxicity. In view of the vital role of microglial activation in neurotoxicity, this should be explored in further detail. Endogenous hormones have shown therapeutic potential in METH treatment. For instance, Cholecystokinin-8 (CCK-8) pretreatment has been shown to attenuate the decrease in expression of TH and DAT in the striatum. Overall, these drugs have shown the potential for amelioration of hypo-dopaminergic status and neuron degeneration; however, their precise mechanism of action is not known (Table 2).

35The complex mechanisms underlying METH-evoked neurotoxicity affect sub-regions of the CNS in various ways. Important advances have been made regarding the basic neurobiology of METH-produced neurotoxicity; however, these findings have not resulted in the development of efficacious drug therapy. Nevertheless, in view of the great harm elicited by METH to the CNS, exploration of toxic mechanisms and pharmacotherapies are anticipated in preclinical/clinical investigations. Natural compounds such as 1-methyl-l, 2, 3, 4-tetrahydroisoquinoline (1MeTIQ), resveratrol, curcumin, and gingko biloba have garnered considerable inhibition owing to their ability to scavenge excess free radicals to protect neurons. Accordingly, some researchers have shifted focus to natural compounds to make major breakthroughs. suggested that 1MeTIQ can apply a brake to several processes related to neurotoxicity (e.g., elimination of free radicals, inhibition of MAO, and Glu-dependent excitotoxicity) without side-effects. Therefore, the use of natural compounds may represent a promising strategy for preventing METH neurotoxicity. In a recent study, the therapeutic effect of carbon nanotubes, which enabled oxidation of METH-enhanced extracellular DA in the striatum of mice, was demonstrated. Considering the favorable properties (e.g., large surface areas, superior bundle strength, and highly electrostatic attraction to neurotransmitters such as DA) of nanotubes, their use merits further research as a new approach for treatment of METH abuse-induced neurotoxicity.

36XY and JY wrote the manuscript. YW, QL, YZ, LC, and YD completed the figures and the tables. LL, KX, and C-xY provided advice. All authors reviewed the manuscript.

37The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.