Critical Review Neuropsychotoxic and Neuroprotective Potentials of Dextromethorphan and Its Analogs

Home , Levomethorphan

J Pharmacol Sci 116, 137 – 148 (2011) Journal of Pharmacological Sciences © The Japanese Pharmacological Society Critical Review Neuropsychotoxic and Neuroprotective Potentials of Dextromethorphan and Its Analogs

Eun-Joo Shin1, Jae-Hyung Bach1, Sung Youl Lee1, Jeong Min Kim2, Jinhwa Lee2, Jau-Shyong Hong3, Toshitaka Nabeshima4, and Hyoung-Chun Kim1,* 1Neuropsychopharmacology and Toxicology Program, College of Pharmacy, Kangwon National University, Chunchon 200-701, Korea 2Central Research Center, Green Cross Corp, Yongin 446-799, Korea 3Neuropharmacology Section, Laboratory of Pharmacology and Chemistry, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, NC 27709, USA 4Department of Pharmacology, Graduate School of Pharmaceutical Sciences, Meijo University, Nagoya 468-8503, Japan

Received March 7, 2011; Accepted March 28, 2011

Abstract. Dextromethorphan (3-methoxy-17-methylmorphinan) has complex pharmacologic effects on the central nervous system. Although some of these effects are neuropsychotoxic, this review focuses on the neuroprotective effects of dextromethorphan and its analogs. Some of these analogs, particularly dimemorfan (3-methyl-17-methylmorphinan) and 3-hydroxymorphinan, have promising neuroprotective properties with negligible neuropsychotoxic effects. Their neuroprotective effects, the mechanisms underlying these effects, and their therapeutic potential for the treatment of diverse neurodegenerative disorders are discussed.

Keywords: dextromethorphan, dimemorfan, 3-hydroxymorphinan, neuroprotection, neuropsychotoxicity, neurodegenerative disorder

1. Introduction receptors and a neuroprotective agent. However, its neuroprotective effect is only seen at a dosage much Dextromethorphan (DM, 3-methoxy-17-methylmor- higher that that used for cough suppression. The psycho- phinan), a common ingredient in more than 125 com- tropic effects that can result from the clinical use of high mercial cough and cold remedies, is a dextrorotatory doses of DM (2, 3) and the fact that DM has been recog- optical isomer of levomethorphan, a typical morphine- nized as the object of drug-seeking behavior in several like opioid that is the codeine analog of the morphinan countries (4, 5) have hampered its pharmacologic devel- derivative levorphanol. Patented by Hoffmann–La Roche opment as a useful neuroprotective agent. In this review, in 1954 as an antitussive agent, DM has strong safety and we will discuss the complex behavioral effects of DM; efficacy profiles with no sedative or addictive properties demonstrate its neuroprotective effects; and discuss its at recommended antitussive doses (1). As an antitussive, analogs, such as dimemorfan (DF, 3-methyl-17-methyl- DM is superior to opioids used at antitussive doses in that morphinan) and 3-hydroxymorphinan (3-HM), which it lacks their gastrointestinal side effects, such as consti- have neuroprotective effects with improved safety pro- pation, and causes less depression of the central nervous files. The chemical structures of DM and its analogs are system. shown in Fig. 1. In the past decade, investigators have documented that DM is an antagonist of N-methyl D-aspartate (NMDA) 2. Metabolism of DM

DM is rapidly absorbed from the gastrointestinal tract *Corresponding author. [email protected] into the bloodstream. At oral therapeutic doses, it begins Published online in J-STAGE on May 21, 2011 (in advance) doi: 10.1254/jphs.11R02CR to act within 15 – 30 min of administration, reaching its peak serum level within 2 – 2.5 h (6, 7). It crosses the Invited article blood–brain barrier with a cerebral spinal fluid/plasma

137 138 E-J Shin et al

Codeine Dextromethorphan Dextrorphan (DM) (DX)

3-Methoxymorphinan 3-Allyloxy- 3-Cyclopropylmethoxy- (3-MM) 17-methylmorphinan 17-methylmorphinan (3-AM) (3-CM)

Dimemorfan 3-Hydroxymorphinan (DF) (3-HM) GCC1290K

Fig. 1. Chemical structures of codeine, dextromethorphan (DM), and DM analogs.

ratio of 33% – 80% (6) and a duration of action of 5 – 6 phencyclidine [PCP, 1-(1-phenylcyclohexyl)piperidine] h (7). binding site on the complex is higher than that of DM; DM is metabolized primarily through O-demethylation, DX binding to the NMDA-receptor complex produces which produces dextrorphan (DX), and also to some ex- PCP-like discriminative stimulus effects (10). The extent tent through N-demethylation, which yields 3-methoxy- of the metabolic conversion of DM to DX is highly de- morphinan. Both metabolites are further demethylated to pendent on the route of administration (8); intraperitoneal 3-hydroxymorphinan (3-HM) (Fig. 2). Urinary recovery (i.p.) administration of DM, whereby it is absorbed into studies in humans and rats have shown that that DX and the hepatic portal circulation, greatly favors production 3-HM are excreted largely (> 95%) as glucuronide con- of DX, while subcutaneous (s.c.) administration is less jugates (8). The pharmacology of 3-methoxymorphinan favorable for DX production (8). is not significant, and it does not have major behavioral side effects. However, as discussed later in this review, 3. DM-induced neuropsychotoxicity we have observed that 3-HM offers anti-parkinsonian effects with absolute safety. 3.1. Behavioral side effects DX, the major metabolite of DM, has neuroprotective Results of animal studies have suggested that most potential (9). Its affinity for the NMDA Ca2+-channel symptoms observed in abusers of DM are caused by DX, complex is lower than that of DM, but its affinity for the which binds to the same central nervous system receptors Neuromodulations by DM and Its Analogs 139

DM self-administration in rats but does not affect the response to a food reward (15). In addition, the co-administration of DM (20 mg/kg, i.p.) with methamphetamine (2 mg/kg, CYP3A4 CYP2D6 i.p.) attenuates methamphetamine-induced psychological dependence [conditioned place preference (CPP)] and behavioral sensitization (16). Pretreatment with DM (20 mg/kg, s.c.) potentiates the effects of acute morphine 3-MM DX exposure, while it attenuates the effects of chronic morphine exposure on dopamine release in the nucleus ac- cumbens (15). Pulvierenti et al. (17) demonstrated that DM (25 mg/ kg, i.p.) reduces the self-administration of cocaine at various doses (0.12, 0.25, and 0.5 mg/kg per 2-h infu- 3-HM sion) in rats, consistent with other findings (18). How- Conjugated ever, Kim et al. (18) reported that oral DM (40 mg/kg) CYP2D6 CYP3A4 DX increases the rate of reinforced responses to lower doses of cocaine (0.06 and 0.03 mg/kg per infusion). Therefore, DM shifts the dose–response curve for cocaine self-administration to the left, suggesting a sensitized response Conjugated to the reinforcing effects of cocaine. Consistent with this 3-HM interpretation, Jhoo et al. (19) reported that DM (i.p.) has a biphasic effect on cocaine-induced psychological de- Fig. 2. Metabolic pathway of DM. CYP=cytochrome P450, DM= pendence, as evaluated by CPP. DM decreases CPP in- dextromethorphan, 3-MM=3-methoxymorphinan, DX=dextrorphan, 3-HM=3-hydroxymorphinan. duced by high doses of cocaine and increases CPP induced by low doses (19). Furthermore, Kim et al. (20) found that DM exerts biphasic effects on cocaine-induced locomotor stimulation in mice and that the locomotor as PCP. In mice, long-term oral administration of a activities parallel Fos-related antigen immunoreactivity neuroprotective dose of DM produces behavioral sup- in the striatal complex. Therefore, the cocaine-induced pression followed by behavioral tolerance, and prenatal behavioral response is influenced by pre-exposure to exposure to DM produces hyperlocomotor activity in the DM, although the responses might also be affected by the offspring. When these offspring receive additional DM, route, period, or interval of administration or by the ani- they demonstrate marked behavioral depression (11). mal model. Mice exposed to long-term oral DM treatment exhibit impaired B-cell function and natural killer-cell cytotoxic- 3.2. Neurotoxicity ity, similar to the immunosuppressive effects of PCP A single exposure to a non-competitive NMDA- (12). receptor antagonist, such as MK-801, PCP, or ketamine, Ample evidence indicates that DM has effects of its has been reported to cause pathological damage in the own that are independent of its conversion to DX. posterior cingulate cortex and retrosplenial cortex of rat Holtzman (13) has reported that DM (30 mg/kg, s.c.) has brain (21). These chemicals can induce heat shock pro- PCP-like discriminative effects that are independent of tein HSP-70, which may play a role in cellular repair its conversion to DX; these effects are most easily ap- and/or protective mechanisms in the same regions (22). preciated in vitro, where metabolism is not an important Hashimoto et al. (23) found that the induction of HSP-70 issue. Interestingly, DM (13) and DX (10) are completely protein in the posterior cingulate and retrosplenial cortex interchangeable for PCP in rats trained to distinguish of the rat brain was maximal at a DM dose of 75 mg/kg PCP from saline, and DM also substitutes for DX in pi- (i.p.). In contrast, Carliss et al. (24) found that high-dose geons trained to discriminate DX from saline (14). oral DM did not produce the neuropathologic changes Therefore, the route of administration is an important caused by other NMDA-receptor antagonists, suggesting determinant in the deposition of DM and its metabolites, that the disposition of DM and its metabolites in rat brain as well as in its behavioral effects. depends on the route of administration. However, our DM alters the behavioral response to several drugs of preliminary findings (data not shown) indicate that an abuse, including morphine, methamphetamine, and co- understanding of the precise neuropathologic mecha- caine. DM (25 mg/kg, i.p.) reduces methamphetamine nisms mediated by high-dose DM will require more 140 E-J Shin et al study. this point. In addition to interacting with σ receptors, DM also 3.3. Cognitive dysfunction weakly inhibits NMDA receptor–mediated responses Substantial evidence indicates that NMDA receptors (48, 49), which seems to be mediated by binding to a play a significant role in the synaptic plasticity (25) be- non-competitive antagonist site of the NMDA-receptor lieved to form the basis for certain commonly studied cation channel (50). Compared with DM, DX is several learning and memory processes involving the hippocam- times more potent as an NMDA-receptor antagonist (51), pus. Some forms of long-term potentiation, a putative which is consistent with its higher anticonvulsant effects physiological process underlying learning and memory, in vivo. have been shown to be NMDA receptor–dependent (26). Noting that DM decreases seizure intensity in fully Furthermore, many studies indicate that NMDA-receptor kindled rats, Feeser et al. (38) have endorsed clinical antagonists that interfere with long-term potentiation testing of DM as an anticonvulsant drug, but Takazawa also disrupt performance on learning tasks, particularly et al. (39) have demonstrated that DM has a narrow thera- those involving spatial memory (27). Because NMDA peutic window as an anticonvulsant for kindled seizures. receptors are involved in both synaptic plasticity and Furthermore, the active metabolite DX has been reported neurodegenerative processes, potential anticonvulsant/ to be ineffective against amygdala-kindled seizures (52). neuroprotective drugs targeting this receptor must be Takazawa et al. (39) have also reported that treatment of evaluated in terms of their possible interference with animals with DM (30 – 50 mg/kg body weight) leads to normal learning and memory processes. hindlimb ataxia and sedation in a dose-dependent man- DM has been suggested to prevent the induction of ner, although without resting electroencephalographic long-term potentiation in vivo (28). In addition, investi- changes. The proconvulsant activity of DM may depend gators have demonstrated that DM (29) and DX (30) on the experimental model, as Echevarria et al. (53) have impair passive avoidance in rats. Similarly, DM has been shown in rats that DM has anticonvulsant effects in the reported to impair spatial learning in the Morris water maximal electroshock test and proconvulsant effects in maze in a dose-dependent manner (31), consistent with the flurothyl test. the water-maze effects of other NMDA-receptor antago- Leander (37) has argued that antitussives have an an- nists (e.g., MK-801 and AP-5) reported by other investi- ticonvulsant action separate from their PCP-like blockade gators (26). On the other hand, the memory impairment of NMDA receptors. In this respect, a previous report caused by DX is more pronounced than that of DM (32), that DM and DX, but not MK-801, can effectively block and the prolonged use of DM produces cognitive deterio- voltage-gated inward calcium and sodium currents in ration in humans (33). neurons (54) is of considerable interest, as these effects could explain why antitussives, but not more selective 4. Neuroprotective effects of DM NMDA-receptor antagonists such as MK-801, act as anticonvulsants in kindling models. Indeed, it has been 4.1. Anticonvulsant effects suggested that the primary mechanism of action of DM DM has been reported to suppress seizures induced is simply a blockade of certain ion channels (9). In addi- by maximal electroshock (34, 35), NMDA (36, 37), tion to its direct effects on voltage-gated cation channels, amygdala kindling (38, 39), kainic acid (40, 41), DM might also have anticonvulsant effects resulting BAY k-8644 (42, 43), and trimethyltin (44). DX (the from its binding to σ receptors (9, 34, 44, 47, 55), dextrorotatory form of levorphanol) has similarly been although the role of σ receptors in anticonvulsant drug reported to suppress seizures induced by maximal elec- actions remains to be determined. Both anticonvulsant troshock (34), NMDA (36), and BAY k-8644 (42, 43). and proconvulsant effects have been reported for σ-site Most studies have found that DX is several times more ligands (9). potent than DM as an anticonvulsant, suggesting that the Radiolabeled DM binds to high-affinity sites in the metabolite formed in vivo significantly contributes to the brain; these sites are strikingly similar in their binding anticonvulsant activity of the parent drug (45). However, characteristics and regional distribution to σ-binding sites some data suggest that the conversion of DM to DX is and are modulated by the antiepileptic drug phenytoin (9, not necessarily required for DM to produce its effects in 45). Indeed, at least one of the high-affinity DM-binding vivo (46). Both DM and DX show high binding affinities sites is identical to the σ1 site (56). Autoradiographic for σ1-type receptors, but DM binds 2 – 5 times more studies have shown that high-affinity DM-binding sites tightly than DX (47), despite the greater anticonvulsant are distinct from the high-affinity DX-binding site that is potency of DX. Thus, the mechanisms underlying the associated with the activated state of the NMDA-receptor anticonvulsant action of DM and DX remain unclear at cation channel complex (57). DX has a low affinity for Neuromodulations by DM and Its Analogs 141

DM-binding sites in the brain but is eight times more in the amplitude of anoxic depolarization (64). DM also potent than DM as a [3H]DX-binding inhibitor (57), attenuates the in vitro degeneration induced by acute consistent with its higher potency as an NMDA-receptor glucose deprivation (65). antagonist. In addition to σ1-site ligands, a variety of calcium channel antagonists compete for the σ1 sites that 4.3. Anti-parkinsonian effects presumably act as high-affinity DM-binding sites (9), NADPH oxidase, the primary producer of extracellular raising the possibility of an association between DM and superoxide in microglial cells, has been proposed to voltage-gated ion channels. contribute to the neurotoxicity of 1-methyl-4-phenyl- The fact that DM and DX exert biphasic effects on 1,2,3,6-tetrahydropyridine (MPTP) (66, 67). In a study of neural excitation, as exemplified by the observations of the mechanisms underlying the neuroprotective activity both anti- and proconvulsant effects in the same convul- of DM, DM was shown to have a significant protective sive animals, suggests that these drugs will have only a effect against the dopaminergic toxicity induced by narrow therapeutic window in epileptic patients, if they MPTP (67). Wild-type mice receiving daily MPTP injec- have any effect at all. Earlier studies have suggested that tions exhibited a significant loss of dopaminergic neurons epileptogenesis renders the brain more susceptible to the in the substantia nigra pars compacta, but DM treatment PCP-like adverse effects of NMDA-receptor antagonists significantly attenuated the neuronal loss elicited by (58). Many findings are not compatible with the idea that MPTP (67). Relative to the wild-type mice, NADPH DM is just a prodrug for DX. Rather, they strongly sug- oxidase–deficient mice exhibited a significant resistance gest that the parent drug is primarily responsible for the to MPTP-induced DA toxicity (67). In addition, the anticonvulsant effects observed in vivo, at least in kin- neuroprotective effect of DM was observed only in wild- dling and kainic acid models. type mice, but not in NADPH oxidase–deficient mice, Previous studies have demonstrated that DM prevents suggesting that NADPH oxidase is a critical target medi- seizures, mortality, and hippocampal cell loss in a dose- ating the neuroprotective activity of DM in the MPTP dependent manner (34, 40, 44, 55). One interpretation for mouse model. the neuroprotective effects of DM vis-à-vis its anticon- Observing that MPTP-induced dopaminergic neurode- vulsant effect is that it reduces the excitotoxic effect of generation is associated with reactive microgliosis, Gao glutamate on neurons by inhibiting NMDA receptors et al. (66) proposed that inhibition of reactive microglio- (59). DM has also been shown to attenuate kainic acid– sis is the basis for the neuroprotective effect of DM induced increases in activator protein-1 (AP-1) binding against MPTP toxicity. In wild-type mice, the initial activity and c-Jun/Fos–related antigen expression in the dopaminergic neuron-destroying effect of MPTP induces hippocampus, collectively suggesting that DM is an ef- reactive microgliosis, which in turn activates NADPH fective antagonist of kainic acid and a powerful protectant oxidase through the translocation of cytosolic subunits to for convulsants (40, 49). the cell membrane and the generation of superoxide. The neuroprotective effect of DM is attributed to its blockade 4.2. Effects on cerebral ischemia of reactive microgliosis induced by DA neuronal death DM treatment has been shown to protect the brain through the inhibition of both extracellular superoxide against infarction and the related pathophysiological and and intracellular reactive oxygen species (67). functional consequences of ischemic injury in several in In addition, DM blocks the MPTP toxicity–enhancing vivo ischemia models (60). In addition, several in vitro effect of the superoxide dismutase inhibitor diethyl- (61) and in vivo (62) studies have confirmed the neuro- dithiocarbamate in mice (68). A histological analysis protective actions of DM and have offered critical in- demonstrated that the depletion of dopaminergic neurons sights into its possible cellular mechanism of action. In induced by the co-administration of diethyldithiocarbam- particular, comprehensive studies undertaken with a ro- ate and MPTP is completely prevented by DM. DM also dent focal ischemia/reperfusion injury model have shown effectively prevents glutamate toxicity in mesencephalic that DM has a potent ability to decrease the volume of cell cultures, as evaluated by the uptake of [3H]-labeled cerebral infarction and to improve functional recovery as dopamine. DM shows the same pattern of protection as post-injury therapy (62). nicotine and MK801, both of which increase fibroblastic DM attenuates the loss of vulnerable neurons in the growth factor levels in the striatum, suggesting that DM CA1 region of the hippocampus in global ischemia might also work through a neurotrophic mechanism. models (63) and decreases cerebral infarct size in areas DM has also been found to reverse haloperidol-induced of severe neocortical damage after ischemia and reperfu- catalepsy in rats (69) and to potentiate the effect of sion (62). In in vitro hypoxia models, DM reduces neu- levodopa and D1 (but not D2) agonists in reserpine-treated ronal loss and dysfunction, which manifests as a decrease mice (70). However, conflicting results have been ob- 142 E-J Shin et al served in human studies (71, 72). release of peptides and excitatory amino acids and acti- vating NMDA receptors (82, 83). This hyperexcitability 4.4. Effects on pseudobulbar affect event can prolong and intensify the sensation of pain. Pseudobulbar affect (PBA), also known as emotional Although not widely used today as an analgesic, DM lability, is characterized by frequent and inappropriate was initially considered a pharmacologic alternative to episodes of crying and/or laughing and is associated with morphine for pain management (84). DM modulates the neurological disorders such as stroke, amyotrophic lateral sensation of pain by reducing the excitatory transmission sclerosis, Alzheimer’s disease, Parkinson’s disease, and of the primary afferent pathways along the spinothalamic traumatic brain injury. As PBA has been proposed to tract. This process occurs in the dorsal horn of the spinal arise (at least in part) from neurochemical dysregulation cord, where DM blocks NMDA receptors, reducing the of serotonergic, dopaminergic, and/or glutamatergic threshold for pain transmission via the Aδ- and C-sensory neurotransmission (73), agents shown to be useful thera- fibers (85). The activation of neuronal firing by NMDA peutically against PBA will probably be found to modu- receptors increases intracellular calcium levels (86). DM late this altered state of neurotransmission. has been shown to reduce and regulate the influx of in-

DM is well established as a σ1-receptor agonist that tracellular calcium through NMDA receptor–gated suppresses the release of excitatory neurotransmitters channels (87), thereby antagonizing the effects of excit- (74) and as an uncompetitive antagonist of the NMDA- atory amino acids and reducing the release of various sensitive ionotropic glutamate receptor (9). However, its peptides, such as glutamate and aspartate. Thus, it can therapeutic potential is limited by the fact that it is exten- ultimately lead to an overall reduction of pain sensation sively metabolized by cytochrome P450 2D6 to DX, (83). which is rapidly glucuronidated (6, 75) and unable to NMDA-receptor antagonists have been found to pre- cross the blood-brain barrier (76). Concomitant dosing vent the induction of central sensitization under experi- with quinidine, one of the most potent inhibitors of cyto- mental conditions (85). These preclinical observations chrome P450 2D6 activity (77), at 5% – 10% of the anti- have led to the hypothesis that NMDA antagonists might arrhythmic dose increases and sustains the concentration reduce postoperative pain (85). In fact, pretreatment with of DM in plasma, thereby enhancing its potential for DM effectively reduces postoperative pain after tonsil- therapeutic efficacy (78). lectomy, and the pain is reduced not only at rest but also Although the precise neuroanatomical basis of PBA is on swallowing, suggesting that DM prevents the develop- still in question, it is thought to involve the disruption of ment of central sensitization after tonsillectomy (88). inhibitory signals descending from cerebral cortex to motor regions of the brainstem implicated in the regula- 5. Neuroprotective effects of DM analogs with im- tion of emotional output (79) Recent evidence suggests proved safety profiles that the severity of cerebellar disconnection is directly related to PBA onset (80, 81). 5.1. Neuroprotective potential of DF Interestingly, neurons in the brainstem and cerebellum Dimemorfan (DF, 3-methyl-17-methylmorphinan), an are richly decorated with σ1 receptors, suggesting that the effective non-narcotic antitussive agent with a low inci- effect of DM on emotional control is mediated, at least in dence of adverse events (89), was discovered through part, through its interactions with these receptors. As a extensive screening of morphinan analogs and was first

σ1-receptor agonist, DM is expected to inhibit the release used in Japan in 1975 (89, 90). Its antitussive action ap- of glutamate, and as an uncompetitive NMDA-receptor pears to arise from a direct effect on the cough center in antagonist, it is expected to suppress the response to ex- the medulla. It does not induce any significant physical citatory neurotransmitters (80). However, the precise or psychological dependence (89), and its antitussive mechanism of action of DM/quinidine in the regulation action is not affected by the opioid receptor–blocker of PBA remains uncertain. levallorphan. Studies in animal models indicate that DF The above findings expand the recent clinical evidence is up to three times more potent as an antitussive agent that DM/quinidine co-administration markedly reduces than codeine and is equivalent in potency to DM. Its PBA frequency and severity with satisfactory safety and other possible beneficial effects have been evaluated high tolerability, increasing patient quality-of-life (81). using diverse models of neurodegeneration. Interestingly, the evidence indicates that the recognition sites for DM 4.5. Effects on pain sensation and DF are identical or overlapping (55, 91). NMDA can cause somatic and neuropathic pain. Upon Previous studies demonstrated that DF is a high-affinity tissue injury, pain transmission passes through Aδ- and ligand for the σ1 receptor (Ki = 151 nM) but not the PCP C-sensory fibers to the dorsal horn neurons, causing the site (Ki = 16798 nM) and has anticonvulsant properties Neuromodulations by DM and Its Analogs 143

(43, 55, 91). In addition, it has been reported to activate showing that the neurotrophic effect of 3-HM is glial

σ1 receptors to attenuate scopolamine- and β-amyloid cell–dependent and that 3-HM fails to show any protec- peptide–induced amnesia in mice (92), possibly through tive effect in neuron-enriched cultures, they subsequently its enhancement of hippocampal acetylcholine release demonstrated that it is the astroglial cells, and not the (93). Our group has reported that the anticonvulsant ef- microglial cells, that contribute to the neurotrophic effect fect of DF correlates with its σ1-receptor–activated modu- of 3-HM (98). This conclusion was based on reconstitu- lation of the AP-1 transcription factor (43, 55). Similarly, tion studies in which microglial or astroglial cells were DF exhibits a protective effect against cerebral ischemia– added back to neuron-enriched cultures at various final reperfusion injury in rats through a σ1-receptor–dependent percentages (10% – 20% and 40% – 50%, respectively); mechanism that blocks glutamate accumulation and the 3-HM was neurotrophic after the addition of astroglial downstream pathologic events that cause ischemic brain cells but not microglial cells (98). injury (94). In contrast, DF acts via σ1-receptor– Conditioned medium from 3-HM–treated astroglial independent mechanisms to modulate intracellular cal- cells exerts a significant neurotrophic effect on dopami- cium release, NADPH oxidase activity, and NF-κB sig- nergic neurons. 3-HM appears to induce the astroglia to naling, inhibiting the expression of inducible nitric oxide release certain neurotrophic factors (i.e., epidermal (NO) synthase and the production of NO and pro-inflam- growth factor, glial cell-derived neurotrophic factor, matory cytokines. These activities may contribute to its transforming growth factor-β1 and -α1, and activity- anti-inflammatory properties and its protective effects dependent neurotrophic factor) (98) that are responsible against endotoxin shock in mice (95). for its neurotrophic effects. As might be expected, in addition to the detrimental effect of Parkinson’s disease 5.2. Anticonvulsant effects of other DM analogs on DA neurons, a considerable glial reaction that is po- Some newly developed DM analogs have also shown tentially protective takes place in the substantia nigra promising anticonvulsant effects (20, 34, 96). In investi- pars compacta in Parkinson’s disease. The glial cells gating the effects of a series of synthetic DM analogs (particularly the astroglia) protect stressed dopaminergic (modified at positions 3 and 17 of the morphinan ring neurons by producing neurotrophic factors that counter- system: see Fig. 1) on maximal electroshock convulsions act oxidative stress. One potential therapeutic avenue for in mice, Kim et al. (34) found that DM, DX, 3-allyloxy- Parkinson’s disease would be the stimulation of astroglial 17-methylmorphinan (3-AM), and 3-cyclopropylmethox cells to produce these neurotrophic factors to rescue y-17-methylmorphinan (3-CM) had anticonvulsant ef- damaged or dying neurons. fects, whereas 3-methoxymorphinan and 3-HM did not. In addition to its neurotrophic effect, 3-HM has an According to these studies (20, 34, 96), DM, DX, 3-AM, anti-inflammatory mechanism that is also important for and 3-CM have high-affinity for σ1 receptors, but all have its neuroprotective activity; in the reconstitution experi- low affinity for σ2 receptors. In binding to PCP sites, DX ments of Zhang et al., LPS-induced neurotoxicity and has a higher affinity than DM, whereas 3-AM and 3-CM 3-HM-mediated neuroprotection increased as the per- have very low affinities, suggesting that PCP sites are not centage of microglial cells added back to the neuron- required for the anticonvulsant actions of 3-AM and enriched cultures increased (98). 3-HM provides potent 3-CM (34). These studies show that these new DM ana- neuroprotection by acting on two different cell targets: it logs are promising anticonvulsants that are devoid of has a neurotrophic effect mediated by astroglial cells and PCP-like behavioral side effects and that their anticon- an anti-inflammatory effect mediated by inhibition of vulsant actions may be, at least in part, mediated via σ1 microglial cell activation. The higher potency of 3-HM receptors (34, 55, 97). relative to DM is attributable to the neurotrophic effect that occurs in addition to the anti-inflammatory effect 5.3. Anti-parkinsonian effect of 3-HM shared by both substances (97, 98). In structure–activity studies designed to find more Zhang et al. (98) also observed a neuroprotective effect potent neuroprotective DM analogs, 3-HM, a DM me- for 3-HM in both in vivo and in vitro studies using the tabolite lacking methyl groups at the O and N sites, MPTP model. In their in vitro system using primary emerged as a novel candidate for the treatment of mixed mesencephalic neuron–glial cell cultures, 3-HM Parkinson’s disease (98). Zhang et al. showed that 3-HM (1 – 5 μM) significantly and dose-dependently attenuated is a more potent neuroprotective agent than its parent the reduction in dopamine uptake induced by MPTP or compound, DM, against lipopolysaccharide (LPS)- its active component, 1-methyl-4-phenylpyridinium induced neurotoxicity and that it is neurotrophic to do- (MPP+). At the same concentrations, 3-HM also signifi- paminergic neurons in primary mixed mesencephalic cantly increased the dopamine uptake capacity of the neuron–glial cell cultures (98, 99). Furthermore, after cells, confirming its neurotrophic effect. In their in vivo 144 E-J Shin et al studies, the administration of 3-HM significantly reduced Thus, in in vivo and in vitro MPTP models of Parkin- the MPTP-induced loss of nigral dopaminergic neurons, son’s disease, 3-HM is beneficial both in reducing do- as previously seen for DM. Significant depletion of dop- paminergic neuronal degeneration in the substantia nigra amine and its metabolites 3,4-dihydroxyphenylacetic pars compacta and in reversing the depletion of biogenic acid and homovanillic acid is observed in the striata of amines in the striatum through dual mechanisms —a MPTP-treated mice, and levels of all of these biogenic neurotrophic effect and the reduction of reactive micro- amines are attenuated in lesioned mice previously treated gliosis— consequently providing significant neuropro- with 3-HM. The advantage of 3-HM over DM is the tection. neurotrophic effect of the former, which may enhance the sprouting of dopaminergic terminal fibers in the stria- 5.4. Anti-parkinsonian effect of GCC1290K, a 3-HM tum, accelerate the formation of dopamine-containing prodrug vesicles, assist the re-synthesis of dopamine after lesions, To enhance the oral bioavailability of 3-HM, which is and eventually re-establish the return of the DA system approximately 18%, we synthesized the 3-HM prodrug to normal functioning. In addition, the anti-inflammatory GCC1290K (Fig. 1), which has a much higher oral bio- effect exerted by 3-HM resulting from the inhibition of availability of approximately 92%. The anti-parkinsonian microgliosis generated from the MPTP/MPP+-damaged effects produced by i.p. administration of 3-HM are dopaminergic neurons is another important mechanism comparable to those produced by oral administration of of 3-HM. 3-HM inhibits the reactive microgliosis in GCC1290K (data not shown). These findings are de- primary mesencephalic neuron–glial cells cultures after scribed in our PCT patent application WO/2008/111767A1 MPTP/MPP+ treatment (98). [“Orally bioavailable prodrug of (+)-3-hydroxymorph-

Altered receptor modulation in NMDA-/ 1-/ PCP-receptors Production of psychotoxicity including cognitive dysfunction DM PCP Psychotoxic DX binding site effects (major metabolite 3-AM & of DM) 3-HM (or GCC1290K) 3-CM DF

1 NMDA L-type Anti- receptor receptor calcium inflammation/ channel Antioxidant/ Neurotrophic effect

Anti-ischemic effect and Anti-convulsive alleviation of effect pseudobulbar affect Anti-parkinsonian effect Fig. 3. Neuropharmacological actions mediated Analgesic effect by DM and DM analogs. DM=dextromethorphan, DX=dextrorphan, DF=dimemorfan, 3-HM=3-hy- 3-AM, 3-CM, DF, and 3-HM (or GCC1290K) exert their droxymorphinan, 3-AM=3-allyloxy-17-methyl- neuroprotective effects with satisfactory behavioral morphinan, 3-CM=3-cyclopropylmethoxy-17- safety. methylmorphinan, PCP=phencyclidine, NMDA= N-methyl-D-aspartate. Neuromodulations by DM and Its Analogs 145

Table 1. Neuropsychopharmacological profiles of DM and its analogs 3-HM DM DX DF 3-MM 3-AM 3-CM (GCC1290K) Neuropharmacological properties DM binding site +++ 57 + 57 ND ND ND ND ND DX binding site + 97 +++ 97 – 97 – 97 – 97 – 97 – 97 PCP binding site + 34, 44 +++ 34 – 55, 91 ––– 34 – 34

34, 44, 47, 55, 91 34, 44, 47, 55, 91 55, 91 34 34 σ1 binding site +++ +/++ +++ ND ND +++ +++ Antagonism of + 9, 59 +/++ 9, 59 ND ND ND ND ND NMDA receptor Blockade of L-type +++ 42 + 42 +++ 43 ND ND ND ND calcium channel Neuropsychotoxicities (with high dose) PCP-like behavior ++ 10, 13, 34, 98 +++ 14, 34, 98 – 98 – 34, 98 – 34, 98 – 34, 98 – 34, 98 Cognitive dysfunction ++ 29, 31, 33 +++ 30, 32 ND ND ND ND ND Neuroprotections Anticonvulsant effect +++ 34-44 +++ 34, 36, 42 ++ 43, 55 – 34 – 34 ++ 34 ++ 34 Anti-ischemic effect +++ 60, 62, 63 +++ 100 ++ 94 ND ND ND ND Anti-parkinsonian effect +++ 67, 98 ND – 98 +++ 98, 99 ND – 98 + 98 Palliation of PBA syndrome ++ 81 ND ND ND ND ND ND Analgesic effect ++ 84, 88 ND ND ND ND ND ND +++: high, ++: moderate, +: low, –: negligible, ND: non-determined. Each superscript indicates cited reference number.

inan for Parkinson’s disease prevention or treatment”]. has a potential for abuse, although it also has diverse An Investigational New Drug Application for the prodrug positive effects. In addition, the evidence suggests that (IND 107,477) has been approved by the United States DM and its analogs offer strong neuroprotective benefits Food and Drug Administration. GCC1290K is totally in multiple neurodegenerative disorders through their free from safety issues, as shown by studies using 3-HM antioxidant, anti-inflammatory, and neurotrophic effects (data not shown) and is now undergoing clinical trials. and through their modulation of the interactions between

NMDA, PCP, and σ1 receptors (refer to Fig. 3 and Table 5.5. Behavioral effects of DM analogs 1). Thus, they offer a promising new direction for the The repeated administration of DM or its major me- development of therapeutic compounds for the treatment tabolite, PCP-like DX, significantly increases locomotor of excitotoxic and inflammatory neurodegenerative activity and circling behavior, although these behavioral disorders. responses are less pronounced than those caused by PCP. These behavioral responses are enhanced much less by Acknowledgments DF and 3-HM (or GCC1290K) than by DM and DX. In This study was supported by a grant from the Brain Research terms of behavioral effects, DM, DX, and PCP produce Center from 21st Century Frontier Research Program funded by the significant behavioral side effects (i.e., CPP and locomo- Ministry of Science and Technology, Republic of Korea; by a grant tor stimulation), but DF and 3-HM (or GCC1290K) (#A101069) of the Korea Healthcare Technology R&D project, produce very few behavioral side effects (as seen in DM- Ministry of Health & Welfare, Republic of Korea; and by a grant or DX-treated animals), suggesting that DF and 3-HM (#E00025) of the Korea-Japan Joint Research Program, National Re- have negligible psychotropic effects (34, 97, 98). search Foundation of Korea, Republic of Korea. Jae-Hyung Bach was supported by the BK 21 program. This work was, in part, supported 6. Conclusions by grants of the Research on Risk of Chemical Substances, Ministry of Health Labour, and Welfare and the Academic Frontier Project for Private Universities, Ministry of Education, Culture, Sports, Science, The research findings discussed in this review demon- and Technology, Japan. strate that DM produces neuropsychotoxic effects and 146 E-J Shin et al

References 20 Kim HC, Bing G, Shin EJ, Jhoo HS, Cheon MA, Lee SH, et al. Dextromethorphan affects cocaine-mediated behavioral pattern 1 Bem JL, Peck R. Dextromethorphan. An overview of safety is- in parallel with a long-lasting Fos-related antigen-immunoreac- sues. Drug Saf. 1992;7:190–199. tivity. Life Sci. 2001;69:615–624. 2 Desai S, Aldea D, Daneels E, Soliman M, Braksmajer AS, Kopes- 21 Olney JW, Labruyere J, Price MT. Pathological changes induced Kerr CP. Chronic addiction to dextromethorphan cough syrup: a in cerebrocortical neurons by phencyclidine and related drugs. case report. J Am Board Fam Med. 2006;19:320–323. Science. 1989;244:1360–1362. 3 Wolfe TR, Caravati EM. Massive dextromethorphan ingestion 22 Sharp FR, Butman M, Aardalen K, Nickolenko J, Nakki R, and abuse. Am J Emerg Med. 1995;13:174–176. Massa SM, et al. Neuronal injury produced by NMDA antago- 4 Chung H, Park M, Hahn E, Choi H, Lim M. Recent trends of nists can be detected using heat shock proteins and can be drug abuse and drug-associated deaths in Korea. Ann N Y Acad blocked with antipsychotics. Psychopharmacol Bull. 1994;30: Sci. 2004;1025:458–464. 555–560. 5 Cranston JW, Yoast R. Abuse of dextromethorphan. Arch Fam 23 Hashimoto K, Tomitaka S, Narita N, Minabe Y, Iyo M, Fukui S. Med. 1999;8:99–100. Induction of heat shock protein (HSP)-70 in posterior cingulate 6 Hollander D, Pradas J, Kaplan R, McLeod HL, Evans WE, and retrosplenial cortex of rat brain by dizocilpine and phency- Munsat TL. High-dose dextromethorphan in amyotrophic lateral clidine: lack of protective effects of sigma receptor ligands. Ad- sclerosis: phase I safety and pharmacokinetic studies. Ann Neu- dict Biol. 1996;1:61–70. rol. 1994;36:920–924. 24 Carliss RD, Radovsky A, Chengelis CP, O’Neill TP, Shuey DL. 7 Pender ES, Parks BR. Toxicity with dextromethorphan-containing Oral administration of dextromethorphan does not produce neu- preparations: a literature review and report of two additional ronal vacuolation in the rat brain. NeuroToxicology. 2007;28: cases. Pediatr Emerg Care. 1991;7:163–165. 813–818. 8 Wu D, Otton SV, Kalow W, Sellers EM. Effects of route of 25 Cotman CW, Iversen LL. Excitatory amino acids in the brain – administration on dextromethorphan pharmacokinetics and be- focus on NMDA receptors. Trends Neurosci. 1987;10:263–265. havioral response in the rat. J Pharmacol Exp Ther. 1995;274: 26 Morris RG, Anderson E, Lynch GS, Baudry M. Selective impair- 1431–1437. ment of learning and blockade of long-term potentiation by an 9 Tortella FC, Pellicano M, Bowery NG. Dextromethorphan and N-methyl-D-aspartate receptor antagonist, AP5. Nature. 1986; neuromodulation: old drug coughs up new activities. Trends 319:774–776. Pharmacol Sci. 1989;10:501–507. 27 Malenfant SA, O’Hearn S, Fleming AS. MK801, an NMDA an- 10 Holtzman SG. Phencyclidine-like discriminative effects of opi- tagonist, blocks acquisition of a spatial task but does not block oids in the rat. J Pharmacol Exp Ther. 1980;214:614–619. maternal experience effects. Physiol Behav. 1991;49:1129– 11 Kim HC, Jhoo WK. Alterations in motor activity induced by high 1137. dose oral administration of dextromethorphan throughout two 28 Krug M, Matthies R, Wagner M, Brodemann R. Non-opioid anti- consecutive generations in mice. Arch Pharm Res. 1995;18: tussives and methadone differentially influence hippocampal 146–152. long-term potentiation in freely moving rats. Eur J Pharmacol. 12 Kim HC, Jhoo WK, Kwon MS, Hong JS. Effects of chronic 1993;231:355–361. dextromethorphan administration on the cellular immune re- 29 Murata S, Kawasaki K. Common and uncommon behavioural sponses in mice. Arch Pharm Res. 1995;18:267–270. effects of antagonists for different modulatory sites in the NMDA 13 Holtzman SG. Discriminative stimulus effects of dextrometh- receptor/channel complex. Eur J Pharmacol. 1993;239:9–15. orphan in the rat. Psychopharmacology (Berl). 1994;116:249– 30 Sierocinska J, Nikolaev E, Danysz W, Kaczmarek L. Dextror- 254. phan blocks long- but not short-term memory in a passive avoid- 14 Herling S, Solomon RE, Woods JH. Discriminative stimulus ance task in rats. Eur J Pharmacol. 1991;205:109–111. effects of dextrorphan in pigeons. J Pharmacol Exp Ther. 1983; 31 Bane A, Rojas D, Indermaur K, Bennett T, Avery D. Adverse 227:723–731. effects of dextromethorphan on the spatial learning of rats in the 15 Glick SD, Maisonneuve IM, Dickinson HA, Kitchen BA. Com- Morris water maze. Eur J Pharmacol. 1996;302:7–12. parative effects of dextromethorphan and dextrorphan on mor- 32 Dematteis M, Lallement G, Mallaret M. Dextromethorphan and phine, methamphetamine, and nicotine self-administration in dextrorphan in rats: common antitussives – different behavioural rats. Eur J Pharmacol. 2001;422:87–90. profiles. Fundam Clin Pharmacol. 1998;12:526–537. 16 Yang PP, Huang EY, Yeh GC, Tao PL. Co-administration of dex- 33 Hinsberger A, Sharma V, Mazmanian D. Cognitive deterioration tromethorphan with methamphetamine attenuates methamphet- from long-term abuse of dextromethorphan: a case report. J Psy- amine-induced rewarding and behavioral sensitization. J Biomed chiatry Neurosci. 1994;19:375–377. Sci. 2006;13:695–702. 34 Kim HC, Shin CY, Seo DO, Jhoo JH, Jhoo WK, Kim WK, et al. 17 Pulvirenti L, Balducci C, Koob GF. Dextromethorphan reduces New morphinan derivatives with negligible psychotropic effects intravenous cocaine self-administration in the rat. Eur J Pharma- attenuate convulsions induced by maximal electroshock in mice. col. 1997;321:279–283. Life Sci. 2003;72:1883–1895. 18 Kim HC, Park BK, Hong SY, Jhoo WK. Dextromethorphan alters 35 Tortella FC, Musacchio JM. Dextromethorphan and carbetapen- the reinforcing effect of cocaine in the rat. Methods Find Exp tane: centrally acting non-opioid antitussive agents with novel Clin Pharmacol. 1997;19:627–631. anticonvulsant properties. Brain Res. 1986;383:314–318. 19 Jhoo WK, Shin EJ, Lee YH, Cheon MA, Oh KW, Kang SY, et al. 36 Ferkany JW, Borosky SA, Clissold DB, Pontecorvo MJ. Dex- Dual effects of dextromethorphan on cocaine-induced condi- tromethorphan inhibits NMDA-induced convulsions. Eur J tioned place preference in mice. Neurosci Lett. 2000;288:76–80. Pharmacol. 1988;151:151–154. Neuromodulations by DM and Its Analogs 147

37 Leander JD. Evaluation of dextromethorphan and carbetapentane methyl-D-aspartate-induced currents and voltage-operated in- as anticonvulsants and N-methyl-D-aspartic acid antagonists in ward currents in cultured cortical neurons. Eur J Pharmacol. mice. Epilepsy Res. 1989;4:28–33. 1993;238:209–216. 38 Feeser HR, Kadis JL, Prince DA. Dextromethorphan, a common 55 Shin EJ, Nah SY, Kim WK, Ko KH, Jhoo WK, Lim YK, et al. antitussive, reduces kindled amygdala seizures in the rat. Neuro- The dextromethorphan analog dimemorfan attenuates kainate- sci Lett. 1988;86:340–345. induced seizures via sigma1 receptor activation: comparison with 39 Takazawa A, Anderson P, Abraham WC. Effects of dextrometh- the effects of dextromethorphan. Br J Pharmacol. 2005;144: orphan, a nonopioid antitussive, on development and expression 908–918. of amygdaloid kindled seizures. Epilepsia. 1990;31:496–502. 56 Zhou GZ, Musacchio JM. Computer-assisted modeling of multi- 40 Kim HC, Pennypacker KR, Bing G, Bronstein D, McMillian ple dextromethorphan and sigma binding sites in guinea pig MK, Hong JS. The effects of dextromethorphan on kainic acid- brain. Eur J Pharmacol. 1991;206:261–269. induced seizures in the rat. NeuroToxicology. 1996;17:375–385. 57 Franklin PH, Murray TF. Identification and initial characteriza- 41 Kim HC, Suh HW, Bronstein D, Bing G, Wilson B, Hong JS. tion of high-affinity [3H]dextrorphan binding sites in rat brain. Dextromethorphan blocks opioid peptide gene expression in the Eur J Pharmacol. 1990;189:89–93. rat hippocampus induced by kainic acid. Neuropeptides. 1997; 58 Loscher W, Honack D. Responses to NMDA receptor antagonists 31:105–112. altered by epileptogenesis. Trends Pharmacol Sci. 1991;12:52. 42 Kim HC, Ko KH, Kim WK, Shin EJ, Kang KS, Shin CY, et al. 59 Choi DW. Dextrorphan and dextromethorphan attenuate gluta- Effects of dextromethorphan on the seizures induced by kainate mate neurotoxicity. Brain Res. 1987;403:333–336. and the calcium channel agonist BAY k-8644: comparison with 60 George CP, Goldberg MP, Choi DW, Steinberg GK. Dextro- the effects of dextrorphan. Behav Brain Res. 2001;120:169–175. methorphan reduces neocortical ischemic neuronal damage in 43 Shin EJ, Nabeshima T, Lee PH, Kim WK, Ko KH, Jhoo JH, et al. vivo. Brain Res. 1988;440:375–379. Dimemorfan prevents seizures induced by the L-type calcium 61 Klette KL, DeCoster MA, Moreton JE, Tortella FC. Role of cal- channel activator BAY k-8644 in mice. Behav Brain Res. 2004; cium in sigma-mediated neuroprotection in rat primary cortical 151:267–276. neurons. Brain Res. 1995;704:31–41. 44 Shin EJ, Nah SY, Chae JS, Bing G, Shin SW, Yen TP, et al. Dex- 62 Britton P, Lu XC, Laskosky MS, Tortella FC. Dextromethorphan tromethorphan attenuates trimethyltin-induced neurotoxicity via protects against cerebral injury following transient, but not per- sigma1 receptor activation in rats. Neurochem Int. 2007;50: manent, focal ischemia in rats. Life Sci. 1997;60:1729–1740. 791–799. 63 Bokesch PM, Marchand JE, Connelly CS, Wurm WH, Kream 45 Rogawski MA, Porter RJ. Antiepileptic drugs: pharmacological RM. Dextromethorphan inhibits ischemia-induced c-fos expres- mechanisms and clinical efficacy with consideration of promising sion and delayed neuronal death in hippocampal neurons. Anes- developmental stage compounds. Pharmacol Rev. 1990;42: thesiology. 1994;81:470–477. 223–286. 64 Luhmann HJ, Scheid M. Dextromethorphan attenuates hypoxia- 46 Kim HC, Bing G, Jhoo WK, Kim WK, Shin EJ, Im DH, et al. induced neuronal dysfunction in rat neocortical slices. Neurosci Metabolism to dextrorphan is not essential for dextromethor- Lett. 1994;178:171–174. phan’s anticonvulsant activity against kainate in mice. Life Sci. 65 Monyer H, Choi DW. Morphinans attenuate cortical neuronal 2003;72:769–783. injury induced by glucose deprivation in vitro. Brain Res. 1988; 47 Quirion R, Bowen WD, Itzhak Y, Junien JL, Musacchio JM, 446:144–148. Rothman RB, et al. A proposal for the classification of sigma 66 Gao HM, Liu B, Zhang W, Hong JS. Critical role of microglial binding sites. Trends Pharmacol Sci. 1992;13:85–86. NADPH oxidase-derived free radicals in the in vitro MPTP 48 Aram JA, Martin D, Tomczyk M, Zeman S, Millar J, Pohler G, model of Parkinson’s disease. FASEB J. 2003;17:1954–1956. et al. Neocortical epileptogenesis in vitro: studies with N-methyl- 67 Zhang W, Wang T, Qin L, Gao HM, Wilson B, Ali SF, et al. D-aspartate, phencyclidine, sigma and dextromethorphan recep- Neuroprotective effect of dextromethorphan in the MPTP tor ligands. J Pharmacol Exp Ther. 1989;248:320–328. Parkinson’s disease model: role of NADPH oxidase. FASEB J. 49 Kim HC, Bing G, Jhoo WK, Ko KH, Kim WK, Lee DC, et al. 2004;18:589–591. Dextromethorphan modulates the AP-1 DNA-binding activity 68 Vaglini F, Pardini C, Bonuccelli U, Maggio R, Corsini GU. induced by kainic acid. Brain Res. 1999;824:125–132. Dextromethorphan prevents the diethyldithiocarbamate enhance- 50 Mendelsohn LG, Kerchner GA, Kalra V, Zimmerman DM, ment of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine toxicity in Leander JD. Phencyclidine receptors in rat brain cortex. Biochem mice. Brain Res. 2003;973:298–302. Pharmacol. 1984;33:3529–3535. 69 Scotti de Carolis A, Popoli P, Pezzola A, Sagratella S. Differential 51 Cole AE, Eccles CU, Aryanpur JJ, Fisher RS. Selective depres- effects of morphinan drugs on haloperidol-induced catalepsy in sion of N-methyl-D-aspartate-mediated responses by dextrorphan rats: a comparative study with an N-methyl-D-aspartate antago- in the hippocampal slice in rat. Neuropharmacology. 1989;28: nist. Arch Int Pharmacodyn Ther. 1991;310:132–141. 249–254. 70 Kaur S, Starr MS. Antiparkinsonian action of dextromethorphan 52 Le Gal La Salle G, Calvino B, Ben-Ari Y. Morphine enhances in the reserpine-treated mouse. Eur J Pharmacol. 1995;280: amygdaloid seizures and increases inter-ictal spike frequency in 159–166. kindled rats. Neurosci Lett. 1977;6:255–260. 71 Bonuccelli U, Del Dotto P, Piccini P, Behge F, Corsini GU, 53 Echevarria E, Robles L, Tortella FC. Differential profiles of puta- Muratorio A. Dextromethorphan and parkinsonism. Lancet. tive dextromrthorphan sigma ligands in experimental seizure 1992;340:53. models. FASEB J. 1990;4:A331. 72 Montastruc JL, Fabre N, Rascol O, Senard JM, Blin O. N-methyl- 54 Netzer R, Pflimlin P, Trube G. Dextromethorphan blocks N- D-aspartate (NMDA) antagonist and Parkinson’s disease: a pilot 148 E-J Shin et al

study with dextromethorphan. Mov Disord. 1994;9:242–243. tion. Neurosci Lett. 1991;124:232–234. 73 Arciniegas DB, Topkoff J. The neuropsychiatry of pathologic 88 Kawamata T, Omote K, Kawamata M, Namiki A. Premedication affect: an approach to evaluation and treatment. Semin Clin with oral dextromethorphan reduces postoperative pain after Neuropsychiatry. 2000;5:290–306. tonsillectomy. Anesth Analg. 1998;86:594–597. 74 Musacchio JM, Klein M, Paturzo JJ. Effects of dextromethorphan 89 Ida H. The nonnarcotic antitussive drug dimemorfan: a review. site ligands and allosteric modifiers on the binding of (+)-[3H]3- Clin Ther. 1997;19:215–231. (-3-hydroxyphenyl)-N-(1-propyl)piperidine. Mol Pharmacol. 90 Kase Y, Kito G, Miyata T, Uno T, Takahama K, Ida H. Antitus- 1989;35:1–5. sive activity and other related pharmacological properties of d-3- 75 Palmer GC. Neuroprotection by NMDA receptor antagonists in a methyl-N-methylmorphinan (AT-17). Arzneimittelforschung. variety of neuropathologies. Curr Drug Targets. 2001;2:241– 1976;26:353–360. 271. 91 Chou YC, Liao JF, Chang WY, Lin MF, Chen CF. Binding of 76 Steinberg GK, Bell TE, Yenari MA. Dose escalation safety and dimemorfan to sigma-1 receptor and its anticonvulsant and loco- tolerance study of the N-methyl-D-aspartate antagonist dex- motor effects in mice, compared with dextromethorphan and tromethorphan in neurosurgery patients. J Neurosurg. 1996;84: dextrorphan. Brain Res. 1999;821:516–519. 860–866. 92 Wang HH, Chien JW, Chou YC, Liao JF, Chen CF. Anti-amnesic 77 Inaba T, Tyndale RE, Mahon WA. Quinidine: potent inhibition of effect of dimemorfan in mice. Br J Pharmacol. 2003;138: sparteine and debrisoquine oxidation in vivo. Br J Clin Pharma- 941–949. col. 1986;22:199–200. 93 Wang HH, Chou YC, Liao JF, Chen CF. Dimemorfan enhances 78 Pope LE, Khalil MH, Berg JE, Stiles M, Yakatan GJ, Sellers EM. acetylcholine release from rat hippocampal slices. Brain Res. Pharmacokinetics of dextromethorphan after single or multiple 2004;1008:113–115. dosing in combination with quinidine in extensive and poor me- 94 Shen YC, Wang YH, Chou YC, Liou KT, Yen JC, Wang WY, tabolizers. J Clin Pharmacol. 2004;44:1132–1142. et al. Dimemorfan protects rats against ischemic stroke through 79 Wilson SAK. Some problems in neurology. J Neurol Psycho- activation of sigma-1 receptor-mediated mechanisms by decreas- pathol. 1924;4:299–333. ing glutamate accumulation. J Neurochem. 2008;104:558–572. 80 Miller A, Panitch H. Therapeutic use of dextromethorphan: key 95 Wang YH, Shen YC, Liao JF, Lee CH, Chou CY, Liou KT, et al. learnings from treatment of pseudobulbar affect. J Neurol Sci. Anti-inflammatory effects of dimemorfan on inflammatory cells 2007;259:67–73. and LPS-induced endotoxin shock in mice. Br J Pharmacol. 81 Pioro EP, Brooks BR, Cummings J, Schiffer R, Thisted RA, 2008;154:1327–1338. Wynn D, et al. Dextromethorphan plus ultra low-dose quinidine 96 Zapata A, Gasior M, Geter-Douglass B, Tortella FC, Newman reduces pseudobulbar affect. Ann Neurol. 2010;68:693–702. AH, Witkin JM. Attenuation of the stimulant and convulsant 82 Aanonsen LM, Wilcox GL. Nociceptive action of excitatory effects of cocaine by 17-substituted-3-hydroxy and 3-alkoxy amino acids in the mouse: effects of spinally administered opi- derivatives of dextromethorphan. Pharmacol Biochem Behav. oids, phencyclidine and sigma agonists. J Pharmacol Exp Ther. 2003;74:313–323. 1987;243:9–19. 97 Shin EJ, Lee PH, Kim HJ, Nabeshima T, Kim HC. Neuropsycho- 83 Battaglia G, Rustioni A. Coexistence of glutamate and substance toxicity of abused drugs: potential of dextromethorphan and P in dorsal root ganglion neurons of the rat and monkey. J Comp novel neuroprotective analogs of dextromethorphan with im- Neurol. 1988;277:302–312. proved safety profiles in terms of abuse and neuroprotective 84 Weinbroum AA, Rudick V, Paret G, Ben-Abraham R. The role of effects. J Pharmacol Sci. 2008;106:22–27. dextromethorphan in pain control. Can J Anaesth. 2000;47: 98 Zhang W, Shin EJ, Wang T, Lee PH, Pang H, Wie MB, et al. 585–596. 3-Hydroxymorphinan, a metabolite of dextromethorphan, pro- 85 Woolf CJ, Chong MS. Preemptive analgesia – treating postopera- tects nigrostriatal pathway against MPTP-elicited damage both tive pain by preventing the establishment of central sensitization. in vivo and in vitro. FASEB J. 2006;20:2496–2511. Anesth Analg. 1993;77:362–379. 99 Zhang W, Qin L, Wang T, Wei SJ, Gao HM, Liu J, et al. 86 Church J, Lodge D, Berry SC. Differential effects of dextrorphan 3- hydroxymorphinan is neurotrophic to dopaminergic neurons and levorphanol on the excitation of rat spinal neurons by amino and is also neuroprotective against LPS-induced neurotoxicity. acids. Eur J Pharmacol. 1985;111:185–190. FASEB J. 2005;19:395–397. 87 Church J, Shacklock JA, Baimbridge KG. Dextromethorphan 100 Du C, Hu R, Hsu CY, Choi DW. Dextromethorphan reduces in- and phencyclidine receptor ligands: differential effects on K(+)- farct volume, vascular injury, and brain edema after ischemic and NMDA-evoked increases in cytosolic free Ca2+ concentra- brain injury. J Neurotrauma. 1996;13:215–222.