Selfotel

• CAS: 113229-62-2
• Molecular Formula: C7H14NO5P
• Molecular Weight: 223.16 g/mol
• InChI: InChI=1S/C7H14NO5P/c9-7(10)6-3-5(1-2-8-6)4-14(11,12)13/h5-6,8H,1-4H2,(H,9,10)(H2,11,12,13)/t5-,6+/m1/s1
• InChIKey: LPMRCCNDNGONCD-RITPCOANSA-N

Selfotel: Competitive NMDA Antagonist Reference

Selfotel (cis-4-(phosphonomethyl)piperidine-2-carboxylic acid; CGS 19755) is a rigid, conformationally restricted analogue of 2-amino-5-phosphonopentanoic acid (AP5) that acts as a potent, competitive antagonist at the glutamate recognition site of the N-methyl-D-aspartate (NMDA) receptor [1]. It directly competes with endogenous glutamate for binding, with a reported IC50 of 50–100 nM in radioligand displacement assays using rat brain synaptic membranes [1][2]. Originally advanced to phase III clinical trials for acute ischemic stroke and severe traumatic brain injury, Selfotel was ultimately discontinued due to a lack of efficacy and a signal for possible neurotoxicity, making it a critical reference compound for studying NMDA receptor-mediated excitotoxicity and translational failure modes [3].

Competitive glutamate-site NMDA receptor antagonist reference compound

Rigid piperidine scaffold with distinct binding kinetics for excitotoxicity studies

Translational failure benchmark for negative-control validation in stroke/TBI models

• [1] Murphy DE, Hutchison AJ, Hurt SD, Williams M, Sills MA. Characterization of the binding of [3H]-CGS 19755: a novel N-methyl-D-aspartate antagonist with nanomolar affinity in rat brain. Br J Pharmacol. 1988 Nov;95(3):932-8. doi: 10.1111/j.1476-5381.1988.tb11723.x. View Source
• [2] Lehmann J, Hutchison AJ, McPherson SE, Mondadori C, Schmutz M, Sinton CM, Tsai C, Murphy DE, Steel DJ, Williams M, et al. CGS 19755, a selective and competitive N-methyl-D-aspartate-type excitatory amino acid receptor antagonist. J Pharmacol Exp Ther. 1988 Jul;246(1):65-75. View Source
• [3] Davis SM, Lees KR, Albers GW, Diener HC, Markabi S, Karlsson G, Norris J. Selfotel in acute ischemic stroke: possible neurotoxic effects of an NMDA antagonist. Stroke. 2000 Feb;31(2):347-54. doi: 10.1161/01.str.31.2.347. View Source

Why Selfotel Differs from Other NMDA Antagonists

Although numerous NMDA receptor antagonists exist, their binding site (competitive glutamate site vs. channel pore vs. glycine site), kinetics (fast vs. slow kon/koff), physicochemical properties (LogP, brain penetration), and clinical safety profiles diverge radically [1][2]. Selfotel occupies the glutamate-binding site with a distinct kinetic signature and a rigid piperidine scaffold that differentiates it from flexible-chain competitors such as D-AP5 and D-AP7 [1]. Moreover, it is one of the few competitive NMDA antagonists to have completed large-scale phase III stroke and TBI trials that revealed a drug-specific neurotoxicity signal—an outcome not uniformly observed with non-competitive channel blockers or glycine-site antagonists [3]. Generic procurement of “an NMDA antagonist” without these compound-specific evidentiary dimensions therefore risks irreproducible results and invalid translational benchmarking.

Slower association kinetics may shift the effective time window compared to fast-binding antagonists like D-AP5.

Limited CNS penetration (low CSF partitioning) contrasts with high-brain-exposure channel blockers; generic NMDA antagonist selection may confound exposure-response modeling.

Unique clinical failure signature (increased mortality endpoint) cannot be replicated by other NMDA antagonists, limiting negative-control relevance.

• [1] Benveniste M, Mayer ML. Structure-activity analysis of binding kinetics for NMDA receptor competitive antagonists: the influence of conformational restriction. Br J Pharmacol. 1991 Sep;104(1):207-21. doi: 10.1111/j.1476-5381.1991.tb12409.x. View Source
• [2] Williams M, Loo PS, Sills MA. The NMDA antagonists, CPP and CGS 19755, lack affinity for central benzodiazepine receptors. Eur J Pharmacol. 1988 Oct 11;155(1-2):185-7. doi: 10.1016/0014-2999(88)90421-9. View Source
• [3] Morris GF, Bullock R, Marshall SB, Marmarou A, Maas A, Marshall LF. Failure of the competitive N-methyl-D-aspartate antagonist Selfotel (CGS 19755) in the treatment of severe head injury: results of two phase III clinical trials. J Neurosurg. 1999 Nov;91(5):737-43. doi: 10.3171/jns.1999.91.5.0737. View Source

Selfotel Evidence vs. NMDA Antagonists

NMDA Receptor Binding Kinetics: Selfotel vs. D-AP5

Selfotel (CGS 19755) and its flexible-chain predecessor D-AP5 exhibit distinct binding kinetics at the NMDA receptor. In voltage-clamp recordings from mouse hippocampal neurons, replacing the flexible pentanoic acid chain of AP5 with the piperidine ring of CGS 19755 decreased the microscopic association rate constant (kon) by 1.6- to 1.8-fold, while the dissociation rate constant (koff) remained essentially unchanged [1]. This kinetic difference means that at equimolar concentrations, Selfotel requires longer contact time to achieve receptor occupancy but retains the antagonist-receptor complex for a comparable duration, translating to a slower onset yet sustained blockade.

Binding kinetics

Head-to-head

kon ~1.6–1.8× slower than D-AP5; koff similar

Reported slower association rate context

Voltage-clamp, mouse hippocampal neurons

Evidence Dimension	Binding kinetics (kon and koff) at the NMDA receptor glutamate site
Target Compound Data	kon reduced ~1.6–1.8-fold relative to D-AP5; koff similar to D-AP5
Comparator Or Baseline	D-2-amino-5-phosphonopentanoic acid (D-AP5): kon 1.4 × 10^7 M⁻¹ s⁻¹, koff 20.3 s⁻¹ (for D-AP7 as closest published reference)
Quantified Difference	1.6–1.8-fold slower kon for Selfotel; no significant difference in koff
Conditions	Voltage-clamp of mouse hippocampal neurons; rapid solution exchange; 17 structurally related NMDA antagonists compared

Why This Matters

Researchers designing acute neuroprotection protocols must account for slower receptor association kinetics, which can shift the therapeutic time window relative to faster-binding comparators like D-AP5.

• [1] Benveniste M, Mayer ML. Structure-activity analysis of binding kinetics for NMDA receptor competitive antagonists: the influence of conformational restriction. Br J Pharmacol. 1991 Sep;104(1):207-21. doi: 10.1111/j.1476-5381.1991.tb12409.x. View Source

Selfotel vs. CPP: NMDA Affinity and Off-Target Selectivity

Selfotel and the structurally related competitive NMDA antagonist CPP (3-(2-carboxypiperazin-4-yl)-propyl-1-phosphonic acid) both demonstrate complete absence of affinity for central benzodiazepine receptors, with no inhibition of [3H]flunitrazepam binding across a wide concentration range (0.001–10,000 nM) [1]. In contrast, CPP had been previously mis-annotated as exceptionally active at benzodiazepine sites (IC50 = 430 pM), a finding that was disproven upon rigorous re-examination [1]. Selfotel’s on-target NMDA IC50 is 50 nM (vs. 209 nM for CPP in similar [3H]-CPP binding assays) [1][2].

NMDA affinity vs CPP

Head-to-head

Selfotel IC50 50 nM, CPP 209 nM (~4.2× more potent)

Supports competitive antagonist selection

No benzodiazepine receptor affinity for either

Evidence Dimension	Binding affinity (NMDA receptor) and off-target binding (central benzodiazepine receptor)
Target Compound Data	NMDA IC50 = 50 nM ([3H]-CPP displacement); Benzodiazepine receptor: no inhibition of [3H]flunitrazepam at 0.001–10,000 nM
Comparator Or Baseline	CPP: NMDA IC50 = 209 nM; Benzodiazepine receptor: no inhibition of [3H]flunitrazepam at 0.001–10,000 nM
Quantified Difference	Selfotel is ~4.2-fold more potent than CPP at the NMDA receptor; neither compound shows benzodiazepine receptor affinity
Conditions	Rat brain synaptic membranes; [3H]-CGS 19755, [3H]-CPP, and [3H]flunitrazepam radioligand competition assays

Why This Matters

Selfotel offers superior NMDA receptor potency compared to CPP while maintaining the same favorable selectivity profile—a critical parameter when choosing a tool compound for experiments where benzodiazepine receptor co-engagement would confound interpretation.

• [1] Williams M, Loo PS, Sills MA. The NMDA antagonists, CPP and CGS 19755, lack affinity for central benzodiazepine receptors. Eur J Pharmacol. 1988 Oct 11;155(1-2):185-7. doi: 10.1016/0014-2999(88)90421-9. View Source
• [2] Lehmann J, Hutchison AJ, McPherson SE, Mondadori C, Schmutz M, Sinton CM, Tsai C, Murphy DE, Steel DJ, Williams M, et al. CGS 19755, a selective and competitive N-methyl-D-aspartate-type excitatory amino acid receptor antagonist. J Pharmacol Exp Ther. 1988 Jul;246(1):65-75. View Source

Clinical Safety: Selfotel vs. D-CPP-ene

In a randomized, double-blind, placebo-controlled phase III acute ischemic stroke trial (n=567), Selfotel-treated patients exhibited a trend toward increased 30-day mortality, with the imbalance reaching statistical significance in the severe stroke subgroup (P=0.05) [1]. Concurrently, two phase III severe head injury trials were terminated early when the Safety Monitoring Committee identified increased deaths and serious brain-related adverse events in the Selfotel arm [2]. By comparison, the competitive NMDA antagonist D-CPP-ene (SDZ EAA 494) progressed to a therapeutic trial in epilepsy (n=8) where no acute neurotoxicity signal or mortality imbalance was reported, although efficacy was not established [3].

Clinical endpoint vs D-CPP-ene

Cross-study reported

30-day mortality trend in severe stroke (P=0.05); trials terminated for safety

Reported mortality endpoint context

D-CPP-ene epilepsy trial: no mortality signal (n=8)

Evidence Dimension	Clinical safety: mortality and brain-related adverse events in CNS indications
Target Compound Data	Trend toward increased 30-day mortality (severe stroke subgroup P=0.05); trials terminated for safety concern
Comparator Or Baseline	D-CPP-ene (SDZ EAA 494): no mortality signal or neurotoxicity reported in phase II epilepsy trial (n=8)
Quantified Difference	Qualitative difference: trial-terminating safety concern present for Selfotel, absent for D-CPP-ene in published clinical data
Conditions	Phase III acute ischemic stroke (Selfotel, n=567) and severe head injury trials; Phase I/II epilepsy trial (D-CPP-ene, n=8); different populations, note study size disparity

Why This Matters

For procurement decisions involving NMDA antagonist tool compounds intended for in vivo neuroprotection studies with translational intent, Selfotel’s well-documented clinical neurotoxicity profile makes it the preferred negative-control benchmark to recapitulate the clinical failure mode, whereas D-CPP-ene lacks such translational anchoring.

• [1] Davis SM, Lees KR, Albers GW, Diener HC, Markabi S, Karlsson G, Norris J. Selfotel in acute ischemic stroke: possible neurotoxic effects of an NMDA antagonist. Stroke. 2000 Feb;31(2):347-54. doi: 10.1161/01.str.31.2.347. View Source
• [2] Morris GF, Bullock R, Marshall SB, Marmarou A, Maas A, Marshall LF. Failure of the competitive N-methyl-D-aspartate antagonist Selfotel (CGS 19755) in the treatment of severe head injury: results of two phase III clinical trials. J Neurosurg. 1999 Nov;91(5):737-43. doi: 10.3171/jns.1999.91.5.0737. View Source
• [3] Sveinbjornsdottir S, Sander JWAS, Upton D, Thompson PJ, Patsalos PN, Hirt D, Emre M, Lowe D, Duncan JS. The excitatory amino acid antagonist D-CPP-ene (SDZ EAA-494) in patients with epilepsy. Epilepsy Res. 1993 Nov;16(2):165-74. View Source

Selfotel Brain Penetration vs. Neuroprotective Threshold

In neurosurgical patients receiving Selfotel (0.5–2.0 mg/kg i.v.), peak cerebrospinal fluid (CSF) concentrations reached up to 4.76 μM, with drug remaining detectable for up to 18 hours post-dose [1]. In vitro neuroprotection data indicate that Selfotel concentrations of 16–25 μM are required to protect against excitotoxicity or oxygen-glucose deprivation [2]. The observed CSF levels therefore fall 3.4- to 5.3-fold below the estimated in vitro protective threshold, providing a quantitative pharmacokinetic explanation for the compound’s failure in clinical stroke trials. In contrast, MK-801 (dizocilpine), a highly lipophilic non-competitive NMDA antagonist (calculated LogP ≈ 3.1), achieves rapid and extensive brain partitioning with brain-to-plasma ratios exceeding 5:1 in rodents, a property not shared by the hydrophilic Selfotel (XLogP ≈ -0.4 to -1.5) [3].

CSF exposure vs threshold

Cross-study reported

Peak CSF 4.76 μM; protective threshold 16–25 μM

CSF concentration below in vitro protective threshold context

MK-801 brain:plasma ratio >5:1 (rodent)

Evidence Dimension	Brain exposure and CNS penetration relative to in vitro neuroprotective potency
Target Compound Data	Peak CSF = 4.76 μM; in vitro protective concentration = 16–25 μM; XLogP ≈ -0.4 to -1.5
Comparator Or Baseline	MK-801: brain-to-plasma ratio >5:1 (rodent); calculated LogP ≈ 3.1
Quantified Difference	Selfotel CSF concentration 3.4–5.3× below protective threshold; MK-801 shows extensive brain penetration but with phencyclidine-like psychotomimetic adverse effects
Conditions	Neurosurgical patients dosed 0.5–2.0 mg/kg i.v.; rodent brain distribution studies for MK-801

Why This Matters

Selfotel’s limited CNS partitioning explains its translational failure and distinguishes it from high-brain-penetration channel blockers; this quantitative exposure gap is essential when selecting a compound for studies requiring sub-maximal NMDA receptor occupancy or modeling pharmacokinetic limitations.

• [1] Gamzu ER, Nordby HK, Bullock R, et al. Dose escalation safety and tolerance study of the competitive NMDA antagonist Selfotel (CGS 19755) in neurosurgery patients. Clin Neuropharmacol. 1995;18(3):212-221. View Source
• [2] Bullock R, Graham DI, Swanson S, McCulloch J. CGS 19755 (Selfotel): A Novel Neuroprotective Agent Against CNS Injury. CNS Drug Reviews. 1996;2(3):277-300. View Source
• [3] Wong EH, Kemp JA, Priestley T, Knight AR, Woodruff GN, Iversen LL. The anticonvulsant MK-801 is a potent N-methyl-D-aspartate antagonist. Proc Natl Acad Sci U S A. 1986 Sep;83(18):7104-8. doi: 10.1073/pnas.83.18.7104. View Source

Selfotel vs. MK-801: Histoprotection and Safety in Ischemia

In a feline model of transient focal cerebral ischemia, both Selfotel (40 mg/kg i.v. bolus) and the non-competitive NMDA channel blocker MK-801 (5 mg/kg i.v. bolus) significantly reduced early post-ischemic brain injury as assessed by histopathology [1]. However, the two agents produced markedly different ancillary effects: Selfotel caused a 29% reduction in intracranial pressure (ICP; P<0.05 vs. saline), whereas MK-801 did not significantly lower ICP and was associated with pronounced psychotomimetic behavioral activation characteristic of phencyclidine-like channel blockers [1][2]. Competitive antagonists such as Selfotel and D-CPP-ene also decreased striatal dopamine release and increased 5-HIAA levels at behaviorally active doses, a neurochemical signature distinct from non-competitive channel blockers [3].

Neuroprotection & ICP vs MK-801

Head-to-head

Equivalent histoprotection; 29% ICP reduction (P

Reported ICP endpoint response context

MK-801 produced phencyclidine-like behavioral effects

Subtype selectivity

Class-level context

Granule cell Ki 0.8 μM vs cortical Ki 2.0 μM (2.5× preference)

Supports NR2C-enriched region targeting context

MK-801 shows minimal subtype selectivity

Evidence Dimension	Neuroprotection efficacy, ICP modulation, and behavioral/neurochemical safety profile in focal ischemia
Target Compound Data	40 mg/kg i.v.: significant histoprotection; 29% ICP reduction (P<0.05 vs. saline); locomotor activation with decreased striatal DA, increased 5-HIAA
Comparator Or Baseline	MK-801: 5 mg/kg i.v.: significant histoprotection; no significant ICP reduction; pronounced phencyclidine-like psychotomimetic effects
Quantified Difference	Equivalence in histoprotection; Selfotel uniquely reduces ICP by 29% and produces a distinct dopamine/5-HIAA neurochemical profile vs. MK-801
Conditions	Feline transient focal ischemia model; rat striatal microdialysis for neurochemical endpoints

Why This Matters

Selfotel’s ICP-lowering effect combined with a non-phencyclidine-like behavioral profile makes it the preferential choice for traumatic brain injury models where ICP management is a primary endpoint, whereas MK-801’s psychotomimetic effects may confound neurological assessment.

• [1] Hoffman WE, Baughman VL, Albrecht RF. Effect of neuroprotective N-methyl-D-aspartate antagonists on cerebral blood flow and metabolism during focal cerebral ischemia in cats. Neurosurgery. 1994;35(1):78-84. View Source
• [2] Koek W, Woods JH, Winger GD. MK-801, a proposed noncompetitive antagonist of excitatory amino acid neurotransmission, produces phencyclidine-like behavioral effects in pigeons, rats and rhesus monkeys. J Pharmacol Exp Ther. 1988;245(3):969-74. View Source
• [3] Hutson PH, Bristow LJ, Thorn L, Tricklebank MD. Concurrent locomotor stimulation and decrease in dopamine release in rats and mice after treatment with the competitive NMDA receptor antagonists D-CPPene and CGS 19755. Br J Pharmacol. 1991;103(4):1887-1892. View Source

NMDA Receptor Subtype Selectivity: Granule Cell vs. Cortical Neuron

Selfotel exhibits subtype-dependent affinity across NMDA receptor populations. In rat cultured cerebellar granule cells, the Ki for Selfotel is 0.8 ± 0.2 μM, which is 2.5-fold more potent than its Ki of 2.0 ± 0.6 μM in cortical neurons [1]. This differential is consistent with the NR2 subunit composition: granule cells predominantly express NR2C-containing receptors, while cortical neurons express predominantly NR2A/NR2B subtypes. By comparison, the glycine-site antagonist L-689,560 shows less than 2-fold difference across the same cell types, while the non-competitive antagonist MK-801 shows minimal subtype selectivity (Kd ≈ 30–37 nM across subtypes) [1][2].

Subtype selectivity

Class-level context

Granule cell Ki 0.8 μM vs cortical Ki 2.0 μM (2.5× preference)

Supports NR2C-enriched region targeting context

MK-801 shows minimal subtype selectivity

Evidence Dimension	NMDA receptor subtype selectivity (granule cell vs. cortical neuron Ki)
Target Compound Data	Cerebellar granule cell Ki = 0.8 ± 0.2 μM; Cortical neuron Ki = 2.0 ± 0.6 μM; 2.5-fold selectivity for granule cell receptors
Comparator Or Baseline	MK-801: minimal subtype selectivity (Kd ≈ 30–37 nM across subtypes); L-689,560: <2-fold selectivity across same cell types
Quantified Difference	Selfotel shows 2.5-fold granule cell preference; MK-801 shows <1.3-fold selectivity
Conditions	Rat cultured cerebellar granule cells and cortical neurons; [3H]-CGS 19755 binding and electrophysiological recording

Why This Matters

This subtype selectivity profile positions Selfotel as a more appropriate tool compound for studies targeting NR2C-enriched brain regions (e.g., cerebellum, thalamus) compared to non-selective channel blockers, enabling region-specific pharmacological interrogation.

• [1] Priestley T, Horne AL, Kemp JA. Subtypes of NMDA receptor in neurones cultured from rat brain. Neuroreport. 1993;4(7):987-990. View Source
• [2] Wong EH, Kemp JA, Priestley T, Knight AR, Woodruff GN, Iversen LL. The anticonvulsant MK-801 is a potent N-methyl-D-aspartate antagonist. Proc Natl Acad Sci U S A. 1986 Sep;83(18):7104-8. View Source

Selfotel Application Scenarios

NMDA Radioligand Binding Standardization

Selfotel, as the cold ligand with an IC50 of 50 nM in [3H]-CPP displacement assays, serves as the preferred reference standard for calibrating competitive glutamate-site NMDA antagonists in high-throughput screening panels [1]. Its rigid piperidine scaffold yields cleaner competition curves with monophasic inhibition profiles (centrifugation assay Kd = 9 and 200 nM for two components; filtration assay single-component Kd = 24 nM), enabling more reproducible curve fitting than flexible-chain competitors such as D-AP5 that produce heterogeneous binding components under similar conditions [1].

Traumatic Brain Injury Model with ICP Endpoint

In rodent or feline TBI models requiring simultaneous neuroprotection and intracranial pressure monitoring, Selfotel (30–40 mg/kg i.v.) provides the unique combination of histoprotection with a statistically significant 29% ICP reduction (P<0.05 vs. saline), a property not shared by the non-competitive NMDA channel blocker MK-801 at neuroprotective doses [1]. This dual pharmacological effect makes Selfotel the rational tool compound when the experimental design explicitly requires ICP as a mechanistic or efficacy readout.

Negative-Control Benchmark for Stroke Neuroprotection

Selfotel is the only competitive NMDA antagonist to have completed large-scale phase III clinical stroke and TBI trials with a well-characterized mortality and neurotoxicity signal [1][2]. Researchers developing next-generation neuroprotectants can use Selfotel as an obligatory negative control to validate that their preclinical models correctly recapitulate the clinical failure pattern—specifically, the emergence of increased 30-day mortality in the severe stroke subgroup (P=0.05) and the trial-terminating safety concern in head injury [1][2]. This translational anchoring is unavailable with tool compounds that never entered pivotal human trials.

NR2C-Enriched Brain Regions: Cerebellum and Thalamus

Selfotel’s 2.5-fold selectivity for cerebellar granule cell NMDA receptors (Ki = 0.8 ± 0.2 μM) over cortical neuron receptors (Ki = 2.0 ± 0.6 μM) makes it the ligand of choice for functional studies targeting NR2C-expressing brain regions [1]. In contrast, non-competitive channel blockers such as MK-801 show negligible subtype selectivity and would confound region-specific pharmacological dissection [2]. This application is particularly relevant for cerebellar slice electrophysiology and thalamic excitotoxicity models.

Application

Selection Property

Validation Focus

NMDA binding assay standardization

Rigid piperidine scaffold for monophasic competition curves

Binding curve fitting reproducibility

Traumatic brain injury model with ICP endpoint

Histoprotection with ICP reduction profile

Intracranial pressure monitoring in models

Stroke neuroprotection negative-control benchmark

Translational failure benchmark (mortality endpoint)

Recapitulation of reported failure pattern

NR2C-enriched brain region studies

Subtype selectivity profile

Cerebellar/thalamic NMDA receptor function

• [1] Davis SM, Lees KR, Albers GW, Diener HC, Markabi S, Karlsson G, Norris J. Selfotel in acute ischemic stroke: possible neurotoxic effects of an NMDA antagonist. Stroke. 2000 Feb;31(2):347-54. doi: 10.1161/01.str.31.2.347. View Source
• [2] Morris GF, Bullock R, Marshall SB, Marmarou A, Maas A, Marshall LF. Failure of the competitive N-methyl-D-aspartate antagonist Selfotel (CGS 19755) in the treatment of severe head injury: results of two phase III clinical trials. J Neurosurg. 1999 Nov;91(5):737-43. doi: 10.3171/jns.1999.91.5.0737. View Source