IUPAC: (+)-3-methoxy-17-methylmorphinan · (4bS,8aR,9S)-3-methoxy-11-methyl-6,7,8,8a,9,10-hexahydro-5H-9,4b-(epiminoethano)phenanthrene · C18H25NO · MW 271.40 g/mol · CAS 125-71-3 · ChEMBL CHEMBL52440
Dextromethorphan (DXM). The dextrorotatory morphinan — a mirror-image relative of the opioid levorphanol that lost almost all of its opioid activity but kept a use-dependent NMDA open-channel block, a high-affinity sigma-1 agonism, and, at high dose, serotonin-reuptake inhibition. It is the "DM" in a thousand cough syrups, and its dissociative pharmacology is largely carried by its CYP2D6 metabolite dextrorphan. Context names: DXM, DM, robo, dex, "Triple C", robotripping, skittling.
Dextromethorphan is the d-isomer of a codeine analog that was deliberately engineered to keep the antitussive action of opioids while shedding the analgesia and dependence. In the process it became a genuinely polypharmacological molecule. Its recreationally and clinically relevant actions are three: non-competitive, use-dependent open-channel block of the NMDA-type glutamate receptor, sigma-1 (σ1) receptor agonism, and serotonin (and, weaker, norepinephrine) reuptake inhibition. Unlike morphine, DXM has negligible μ-opioid affinity — naloxone does not reverse its core effects.
The load-bearing subtlety: DXM itself is only a modest NMDA blocker (human PCP-site Ki ≈ 1.7 µM). The dissociative "trip" is driven far more by its O-demethylated metabolite dextrorphan (DXO), a roughly 8-fold more potent open-channel blocker (Ki ≈ 220 nM, same PCP site as ketamine and PCP). Because that conversion is done by the highly polymorphic enzyme CYP2D6, an individual's genotype rewrites their pharmacology — this is one of the cleanest real-world examples of pharmacogenetics shaping a subjective drug effect.
Protonated DXM/dextrorphan enter the open NMDA channel and lodge in the pore near the GluN1/GluN2 asparagine (N-site) ring — the same PCP/ketamine site. Use-dependent, partially trapped, and voltage-sensitive. Dextrorphan carries most of the potency.
DXM's highest-affinity target (Ki ≈ 348 nM, [³H](+)-pentazocine displacement). σ1 is an ER chaperone at mitochondria-associated membranes; agonism modulates Ca²⁺ signaling, BDNF, and — controversially — contributes to the dissociative, mood, and neuroprotective profile.
At the high plasma levels reached during recreational megadosing, DXM inhibits SERT (and NET) — the mechanistic basis of the very real serotonin-syndrome risk with MAOIs and SSRIs. This is not a footnote; it is the deadliest interaction on the page.
As with ketamine, NMDA block on tonically active GABAergic interneurons disinhibits pyramidal cells, releasing a glutamate surge onto AMPA receptors. The dissociative and dream-like phenomenology tracks the same corticothalamic uncoupling.
DXM also antagonizes α3β4 nicotinic receptors and weakly blocks neuronal Na⁺ channels ([³H]BTX-site Ki low-µM). These contribute to the antitussive action and to autonomic/cardiac effects at overdose more than to the "high".
Combining DXM with a CYP2D6 inhibitor (quinidine in Nuedexta; deuteration in deudextromethorphan/AVP-786) is used clinically to raise DXM levels for pseudobulbar affect — deliberately biasing the parent-vs-metabolite balance the opposite way from a recreational user chasing dextrorphan.
DXM is well absorbed orally but hit hard by first-pass metabolism. The committed step is CYP2D6-mediated O-demethylation to dextrorphan; a minor arm is CYP3A4/3A5 N-demethylation to 3-methoxymorphinan. Both eventually converge on 3-hydroxymorphinan, which is glucuronidated and renally cleared. Because CYP2D6 is genetically polymorphic, the same 1 mg/kg dose produces radically different exposure between people.
Metabolism cascade: the CYP2D6 arm both activates DXM (to the dissociative dextrorphan) and, in fast metabolizers, shortens its stay — the enzyme giveth and taketh away.
Extensive metabolizers (EMs, the majority) convert DXM to dextrorphan rapidly: strong dissociation but a short parent half-life. Poor metabolizers (PMs, ~5–10% of Europeans, higher in some populations) — plus anyone taking a CYP2D6 inhibitor — form little dextrorphan, accumulate the parent drug, and show markedly prolonged half-lives and higher σ1/serotonergic parent exposure. That combination (more DXM, less clearance, more SERT engagement) is exactly the pharmacology that raises serotonin-syndrome and toxicity risk. SSRIs (fluoxetine, paroxetine) and bupropion are potent CYP2D6 inhibitors — they turn an EM into a functional PM and add serotonergic load: a double hit.
At therapeutic antitussive doses (10–30 mg) DXM suppresses the medullary cough reflex with no appreciable psychoactivity. Recreational use scales the dose by one to two orders of magnitude, and the subjective effects famously arrive in discrete steps that the user community (after William White's DXM FAQ) calls "plateaus" — a genuine dose–response staircase reflecting progressive NMDA blockade layered on σ1 and serotonergic tone.
Mild stimulation, euphoria, music enhancement, a slight "buzz" and restlessness — closer to an intoxicant than a dissociative. Body load and nausea already possible.
Stronger euphoria and intoxication, exaggerated "robo-walk" ataxia, slurred speech, closed-eye imagery, short-term memory disruption. The classic "robotripping" zone.
Frank dissociation: altered perception of time and body, dream-like visuals, impaired motor control, and a distinct disconnection from the environment. This is dextrorphan-dominated NMDA territory.
Full dissociative anesthesia — profound sensory disconnection, near-immobility, ego dissolution, and a state some users compare to the ketamine "K-hole". Also where cardiovascular, thermoregulatory, and airway risks climb steeply; overdose and combination-product toxicity dominate real-world harm here.
Clinically, the same molecule keeps finding new lives: Nuedexta (DXM + quinidine) for pseudobulbar affect, and Auvelity (DXM + bupropion) — FDA-approved in 2022 for major depressive disorder, riding the same NMDA-plus-σ1 logic that made ketamine a rapid antidepressant, with bupropion deployed as a CYP2D6 inhibitor to keep DXM levels up. The recreational and psychiatric stories are the same pharmacology read at different doses.
Evidence-based, non-moralistic. Pure DXM has a fairly wide margin alone, but the two things that actually kill people are (1) serotonergic drug interactions and (2) combination cough-and-cold products — the acetaminophen, antihistamine, and decongestant riding along in the same bottle. Megadosing "DM" syrup for dissociation means megadosing everything else in it.
| Target | Affinity | Rel. | Action |
|---|---|---|---|
|
σ1 · DXM
Sigma non-opioid receptor 1 · human
|
Ki = 348 nM
(+)-pentazocine displ. · pChEMBL 6.46
|
Agonist | |
|
NMDA-R · dextrorphan
Glutamate [NMDA] receptor · human (metabolite)
|
Ki = 220 nM
PCP-site displ. · pChEMBL 6.66
|
Open-channel block | |
|
SERT · DXM
Serotonin transporter (high-dose)
|
Ki ≈ 40 nM
5-HT uptake · literature (see note)
|
Reuptake inhibitor | |
|
NMDA-R · DXM
Glutamate [NMDA] receptor · human (parent)
|
Ki = 1680 nM
PCP-site displ. · pChEMBL 5.78
|
Open-channel block | |
|
NMDA-R · DXM (rat)
[³H]MK-801 / PCP site · rat brain
|
Ki ≈ 2.2 – 4.5 µM
|
Channel blocker | |
|
σ1 · DXM (rat)
Sigma-1 site · rat (selectivity assay)
|
Ki = 5.07 µM
|
Agonist | |
|
M2 mAChR · DXM
Muscarinic ACh receptor (secondary)
|
Ki = 10 µM
|
Antagonist |
DXM is a near-rigid morphinan: a fused tetracyclic cage with a single rotatable bond (the methyl ether). When it — or dextrorphan — plugs the pre-formed, water-filled NMDA pore, it surrenders almost no internal conformational entropy. As with ketamine, the dominant −TΔS term is desolvation: displacing and re-ordering the channel waters around the N-site ring. The Shannon-entropy collapse on binding lives in the solvent and pore side-chain microstates, not in the ligand.
The pharmacogenetic twist has an entropic reading too. The subjective effect is dominated by dextrorphan (Ki 220 nM) rather than DXM (1.7 µM) — a ~1 kcal/mol swing in ΔG concentrated almost entirely in the enthalpic/desolvation gain from removing the 3-methyl ether (adding an H-bonding phenol) while the rigid scaffold's ΔSconf stays essentially constant. A defensible ΔG decomposition here weights ΔSsolvent and pore-water release far above ligand ΔSconf — the same signature FlexAID∆S flags for rigid channel blockers versus floppy orthosteric binders.