IUPAC: 1-(1-phenylcyclohexyl)piperidine · C17H25N · MW 243.39 g/mol · CAS 77-10-1 · ChEMBL CHEMBL275528 · PDB ligand 1PC
Phencyclidine. The arylcyclohexylamine that named a receptor site. PCP is the archetypal open-channel blocker of the NMDA-type glutamate receptor — the eponymous "PCP site" deep in the pore — binding with roughly an order of magnitude higher affinity and far longer residence than its younger, gentler cousin ketamine. Originally the surgical anesthetic Sernyl (Parke-Davis, 1950s), abandoned over emergence delirium, reborn as a 1970s street drug. Street/context names: angel dust, sherm, wet, embalming fluid, PeaCePill, Sernyl(an), CI-395.
Like ketamine, PCP is not an agonist and not a transporter substrate: it is a physical plug wedged inside the ion-conducting pore of the NMDA-type ionotropic glutamate receptor — a heterotetramer of two obligatory GluN1 subunits plus two GluN2 subunits. Its binding pocket, in the central channel vestibule near the GluN1/GluN2 asparagine (N-site) ring that also coordinates the Mg2+ block, is literally called the "PCP site" — historically defined by [3H]TCP and [3H]MK-801 radioligand binding. PCP, ketamine, MK-801 (dizocilpine), dextromethorphan and memantine all occupy this same site. PCP just holds it harder and longer.
The block is use-dependent: the channel must first open — glutamate + the co-agonist glycine/D-serine bound and the membrane depolarized enough to relieve resting Mg2+ — before protonated PCP can reach its site. It then becomes deeply trapped: the gate can close around the bound molecule and unblock is slow. The distinguishing feature versus ketamine is kinetics and affinity. In matched human/pig cortex assays PCP displaces channel-blocker radioligands with Ki ≈ 15–60 nM — roughly 15–20× tighter than ketamine's ~420 nM PCP-site Ki — and its slower off-rate (longer residence) is what turns a brief ketamine "K-hole" into hours of unpredictable, fluctuating PCP intoxication.
Protonated PCP enters the open channel and lodges near the GluN1/GluN2 asparagine ring deep in the vestibule, occluding Na+/Ca2+ flux. The cryo-EM structure PDB 7SAB resolves PCP (ligand 1PC) in exactly this central cavity of a GluN1a/GluN2B receptor.
Same site as ketamine, but PCP's larger phenyl/cyclohexyl/piperidine scaffold buries more of the hydrophobic vestibule and unblocks more slowly. Human PCP-site Ki ~23 nM vs ketamine ~420 nM — the potency and duration gap is the whole story.
PCP is a strongly trapping blocker: the gate shuts on it and it escapes only when the channel re-opens. Slow, voltage-dependent unblock means effects outlast free plasma drug and preferentially silence the most active synapses — a built-in activity filter turned up to eleven.
By blocking NMDA receptors on tonically active GABAergic interneurons, PCP disinhibits pyramidal cells, driving a glutamate surge onto AMPA receptors — "excitation via inhibition of inhibition". This is the substrate for the agitation, thought disorganization, and psychotomimetic profile.
PCP is also a σ1 receptor ligand (human Ki ~1.1 µM) and a sub-µM–µM inhibitor of dopamine (DAT), norepinephrine and serotonin reuptake. This monoaminergic arm — stronger than ketamine's — adds a stimulant, hyperthermic, and euphoric/dysphoric colouring on top of the dissociation.
PCP intoxication reproduces positive, negative and cognitive features of schizophrenia more faithfully than any amphetamine psychosis. PCP/ketamine NMDA block is the empirical foundation of the glutamatergic (NMDA-hypofunction) hypothesis of schizophrenia.
Secondary pharmacology is real and, unlike ketamine, contributes materially: PCP inhibits monoamine transporters at sub-µM–low-µM concentrations (dopamine reuptake block feeds the euphoria and the agitation/hyperthermia), engages σ1 receptors (~1.1 µM), the muscarinic-flavoured butyrylcholinesterase (Ki ~0.5 µM), and non-competitive nicotinic-receptor sites. None of these displaces NMDA channel block as the core mechanism, but the broader monoaminergic reach helps explain why PCP is more agitating, more sympathomimetic, and more dangerous in overdose than ketamine.
PCP's clinical menace is as much pharmacokinetic as pharmacodynamic. It is a small, very lipophilic weak base (logP ≈ 4.3, pKa ≈ 8.5–9.5), so it distributes into a large volume of distribution and sequesters in brain and fat. Two features make its time-course notoriously unpredictable: a long, variable elimination half-life (reported ~7 to >50 h; often cited around 17–24 h, longer in overdose) and gastric ion-trapping — PCP secreted into acidic stomach fluid is protonated, re-absorbed distally, and effectively recycled. The result is a fluctuating, relapsing intoxication that can re-intensify hours after the last dose.
Metabolism cascade: hepatic oxidation (CYP3A4 and CYP2B6 among others) hydroxylates the cyclohexyl/piperidine rings to PCHP and PPC, which are then glucuronidated and cleared renally. The parent drug carries essentially all the activity; the hydroxy metabolites are pharmacologically minor.
Because clearance of the ~10% excreted unchanged is pH-dependent, urinary acidification was historically proposed to speed elimination — but it is now generally avoided: acidosis worsens the rhabdomyolysis-driven risk of myoglobinuric renal failure that is itself a leading cause of PCP death. Detection windows are long precisely because of the lipophilic depot: urine immunoassays can stay positive for days to a week or more after heavy use.
A single activity-gated molecular action fans out along a steep, dose-dependent curve. Unlike ketamine, where sub-anesthetic dosing is fairly forgiving, PCP's higher potency, longer residence, and stronger monoaminergic reach make the low-dose-to-toxic window narrower and the trajectory harder to predict — especially when it is smoked on plant material (parsley, cannabis, tobacco), where the dose per hit is wildly uneven.
NMDA-dependent corticothalamic and cortico-cortical integration is disrupted, uncoupling sensory input from self-referential processing: depersonalization/derealization, distorted body schema, profound analgesia and numbness, euphoria or blank detachment, and the characteristic nystagmus (vertical and horizontal — a near-pathognomonic exam sign). Users describe feeling powerful, weightless, or "elsewhere".
As disinhibition and monoamine reuptake block deepen, the picture becomes unpredictable: ataxia, slurred speech, rigidity or waxy catatonia alternating with sudden agitation, sensory storms, and a striking insensitivity to pain. It is the analgesia plus disinhibition — not any intrinsic "PCP rage" — that produces the classic scenario of unpredictable, self-injurious behaviour with no felt injury. The reputation for superhuman strength is largely myth built on anesthesia + adrenergic drive.
The overdose is a dissociative-adrenergic emergency: hypertension, tachycardia, hyperthermia, diaphoresis, muscular rigidity and hyperactivity driving rhabdomyolysis → myoglobinuric acute kidney injury, plus seizures, hypertensive crisis, and fluctuating coma. Death is typically from hyperthermia/rhabdomyolysis, trauma sustained while anesthetized and agitated, aspiration, or the consequences of extreme hypertension — rarely from respiratory arrest by PCP alone.
Because PCP is potent, long-residence and lipophilically stored, a fraction of users develop a prolonged psychotic state — days to weeks of paranoia, thought disorder, and cognitive disruption outlasting measurable drug — that can be clinically indistinguishable from acute schizophrenia. This is the observation that anchored the NMDA-hypofunction model of psychosis and that separates PCP, experientially and psychiatrically, from the shorter, cleaner ketamine experience.
Emergency management is supportive and pharmacologic, not moralistic: minimize sensory stimulation, and use benzodiazepines first-line for agitation, hypertension, hyperthermia and seizures, with aggressive cooling and IV fluids to protect the kidneys against rhabdomyolysis. There is no antidote and no reversal agent.
Evidence-based, non-moralistic. PCP is more potent, longer-acting and less predictable than ketamine — the dominant harms are unpredictable agitation and self-injury under anesthesia, hyperthermia/rhabdomyolysis, uneven street dosing, and combination with other depressants.
PCP was first synthesized in 1926 and developed by Parke-Davis in the 1950s as the intravenous surgical anesthetic Sernyl (CI-395). It delivered the dissociative dream — analgesia and anesthesia with preserved airway and cardiovascular tone — but human trials were derailed by severe emergence reactions: agitation, delirium, hallucinations and prolonged psychosis on waking. Human use was withdrawn around 1965; a veterinary formulation, Sernylan, persisted for a while as a large-animal immobilizer. (Ketamine, synthesized in 1962 as a shorter, milder analog, was the pharmaceutical industry's answer to exactly this problem.)
PCP re-emerged on the street in 1967 San Francisco as the "PeaCePill," and through the 1970s became "angel dust" — sprinkled or sprayed onto parsley, mint, tobacco or cannabis and smoked ("sherm", "wet", and the "embalming fluid" folklore of joints dipped in PCP ± formaldehyde). Its cheapness, potency and the lurid press coverage of agitated, pain-insensitive users made it a defining drug-panic of the era. The US placed it in Schedule II in 1978, and precursor controls followed. Recreational use has waxed and waned regionally ever since, but PCP's lasting scientific legacy is larger than its street footprint: it named the PCP site, and it handed neuroscience the NMDA-hypofunction model of schizophrenia.
| Target | Affinity | Rel. | Mechanism |
|---|---|---|---|
|
NMDA-R · PCP site
Glutamate [NMDA] receptor · human
|
Ki = 23 nM
human cortex · range 15–60 nM · pChEMBL 7.6
|
Open-channel block | |
|
NMDA-R (rat)
[3H]TCP / [3H]MK-801 site
|
Ki ≈ 38–65 nM
IC50 9.8–47 nM
|
Channel blocker | |
|
GluN1/GluN2A
NMDA-R GRIN1/GRIN2A · human
|
IC50 = 0.40 µM
functional Ca²⁺ (FLIPR)
|
Channel blocker | |
|
GluN1/GluN2B
NMDA-R GRIN1/GRIN2B · human
|
IC50 = 1.26 µM
subtype in PDB 7SAB
|
Channel blocker | |
|
σ1R
Sigma-1 receptor · human
|
Ki = 1,090 nM
|
Ligand | |
|
DAT / monoamines
DA · NE · 5-HT reuptake
|
sub-µM – µM
uptake inhibition (see note)
|
Reuptake inhibitor | |
|
BuChE
Butyrylcholinesterase (secondary)
|
Ki = 520 nM
|
Inhibitor |
PCP and ketamine bind the same pre-formed, water-filled vestibule, so both are gated by the receptor's own conformational cycle rather than by folding a flexible site around the ligand — the dominant −TΔS term is desolvation and channel-water release, not ligand conformational entropy (PCP, like ketamine, has only ~2 rotatable bonds and surrenders little internal entropy). What distinguishes PCP is buried nonpolar surface: the phenyl + cyclohexyl + piperidine scaffold displaces and orders more vestibule water than ketamine's chlorophenyl-cyclohexanone, so the favourable hydrophobic (solvent-entropy) driving force is larger — a first-principles read on why the PCP-site Ki is ~15–20× tighter.
In FlexAID∆S terms, a rigorous ΔG decomposition here should weight ΔSsolvent and pore-water release far above ligand ΔSconf. "Trapping" is a second, slower entropic gate: closing the channel around the bound blocker collapses the configurational entropy of the bound state and raises the escape barrier — and PCP's deeper, longer-residence occupancy sits further down that well than ketamine's. The Shannon-entropy collapse on binding is concentrated in solvent and side-chain microstates lining the pore, the opposite weighting from a floppy-ligand orthosteric binder.