Sunifiram (DM-235)

updated

Sunifiram (DM-235) is a research compound sold for acute memory and learning enhancement through AMPA receptor potentiation. Its published work is rodent and brain-slice pharmacology, mostly from the group that discovered it between 2000 and 2006, with one independent group in 2013, and no controlled human study exists: a PubMed query on 10 October 2026 found no record of sunifiram given to a person. It is a different molecule from unifiram (DM-232), and the AMPA explanation is contested.

Acylpiperazine nootropic (mechanism contested: AMPA or NMDA glycine site)

  • Sunifiram is PubChem CID 4223812, C14H18N2O2, molecular weight 246.30, CAS 314728-85-3, an acylpiperazine; unifiram (DM-232) is a different molecule, CID 9861054.
  • In mice sunifiram prevented scopolamine-induced amnesia in passive avoidance at 0.001 to 0.1 mg/kg intraperitoneally or 0.01 to 0.1 mg/kg orally, with potency the authors put about 1,000 times that of the most active piracetam-like compounds.
  • Every efficacy result is rescue of a chemically induced or lesion-induced deficit in mice, rats or brain slices, and none shows enhancement in unimpaired animals.
  • A 2013 independent paper found sunifiram at 10 to 100 nM enhanced long-term potentiation in mouse hippocampal slices through the NMDA receptor glycine site, and no abstract shows direct AMPA receptor binding or potentiation.
  • Related compounds in the same series can be amnesic, including two compounds nearly equipotent to scopolamine in one structure-activity study.
  • A PubMed query on 10 October 2026 returned 17 records, and none reports sunifiram given to a person or any pharmacokinetic, toxicology or safety study.

What is sunifiram, and what is its regulatory status?

Sunifiram (DM-235) is an acylpiperazine research compound, PubChem CID 4223812, and a different molecule from unifiram (DM-232). It has no FDA label, no application and no registered study.

Sunifiram is PubChem CID 4223812, C14H18N2O2, 246.30 g/mol, CAS 314728-85-3, 1-(4-benzoylpiperazin-1-yl)propan-1-one, with the synonyms DM-235 and DM235. It is an acylpiperazine.

Unifiram (DM-232) is a different molecule: PubChem CID 9861054, C13H15FN2O3S, 298.34 g/mol, CAS 272786-64-8, a fluorophenylsulfonyl hexahydropyrrolo[1,2-a]pyrazin-6-one. The two are not the same compound and are not interchangeable by identity; see the unifiram page. Sapunifiram (MN19) is another related compound in the same literature.

Papers describe sunifiram's lineage inconsistently: as related in structure to piracetam (2002), related to ampakines (2003) and a pyrrolidone nootropic (2013), while the 2000 chemistry paper says the acylpiperazine group can mimic the 2-pyrrolidinone ring and the PubChem name shows no pyrrolidone ring. No public record ties any vendor material to the studied compound.

On 10 October 2026 ClinicalTrials.gov queries for sunifiram and for DM235, DM-235 and DM 235 returned no studies, DailyMed returned no label, and an openFDA Drugs@FDA query returned no match. Status outside the United States is not covered here.

PubChem: sunifiram identity, and how it differs from unifiram

Chemical identity record

PubChem, National Library of Medicine. Compound summary for CID 4223812, 1-(4-benzoylpiperazin-1-yl)propan-1-one (sunifiram, DM-235).

Sunifiram is PubChem CID 4223812, C14H18N2O2, 246.30 g/mol, CAS 314728-85-3, IUPAC name 1-(4-benzoylpiperazin-1-yl)propan-1-one, with the synonyms DM-235 and DM235. It is an acylpiperazine. Unifiram (DM-232) is a different molecule (PubChem CID 9861054, C13H15FN2O3S, 298.34 g/mol, CAS 272786-64-8, a fluorophenylsulfonyl hexahydropyrrolo[1,2-a]pyrazin-6-one), so the two are not the same compound and are not interchangeable by identity. Papers describe sunifiram's lineage inconsistently: as related in structure to piracetam (2002), related to ampakines (2003) and a pyrrolidone nootropic (2013), while the 2000 chemistry paper says the acylpiperazine group can mimic the 2-pyrrolidinone ring and the PubChem name shows no pyrrolidone ring. Sapunifiram (MN19) is another related compound in the same literature.

Participants / model
Chemical identity, not a clinical population
Treatment
Defined molecule, free form
Follow-up
Not applicable
Study design
Chemical database record; original reports
Funding
not stated in the source

Identity record only. No public record ties any vendor material to the studied compound.

Read the original source
Label and registry searches, 10 October 2026: no label, no match, no study

Regulatory record

U.S. Food and Drug Administration (DailyMed label search; openFDA Drugs@FDA query) and ClinicalTrials.gov. Searches for sunifiram and for DM235, DM-235 and DM 235, run 10 October 2026.

ClinicalTrials.gov queries for sunifiram and for DM235, DM-235 and DM 235 returned no studies, DailyMed returned no label, and the openFDA Drugs@FDA query returned no match.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Searches run on 10 October 2026
Study design
Dated label, application and registry search
Funding
not stated in the source

An absence finding as of 10 October 2026. It covers United States label, application and registry records only.

Read the original source

What did the mouse and rat studies find?

In mice and rats sunifiram prevented memory loss caused by scopolamine and several other amnesia-inducing drugs at very low doses, and improved memory in a surgical lesion model. Every result is rescue of an induced deficit, mostly from the group that discovered it, and no study shows enhancement in unimpaired animals or any effect in people.

In the 2000 piperazine paper, 1-acylpiperazines made by molecular simplification of the 1,4-diazabicyclo[4.3.0]nonan-9-one series kept high nootropic activity in the mouse passive avoidance test, with a pharmacological profile very similar to piracetam but much higher potency. Compound 13 (DM235) was active at 0.001 mg/kg subcutaneously, and DM232 and DM235 were the most potent of their classes and increased acetylcholine release in rat brain, although the molecular mechanism remains to be elucidated.

The core efficacy paper found that sunifiram prevented amnesia induced by scopolamine (1.5 mg/kg i.p.) in a mouse passive avoidance test after intraperitoneal (0.001 to 0.1 mg/kg) or oral (0.01 to 0.1 mg/kg) dosing, an effect comparable to piracetam (30 to 100 mg/kg i.p.), aniracetam (100 mg/kg p.o.) or rolipram (30 mg/kg p.o.). It also prevented amnesia induced by mecamylamine, baclofen and clonidine. In rats in the Morris water maze, 0.1 mg/kg i.p. 20 minutes before each daily training prevented scopolamine-induced impairment. At the highest effective doses it did not impair rota-rod coordination or change spontaneous motility, and the authors report a potency about 1,000 times that of the most active piracetam-like compounds.

In olfactory bulbectomized mice, sunifiram at 0.01 to 1.0 mg/kg orally once a day for 7 to 12 days improved spatial reference memory in the Y-maze and short-term memory in novel object recognition, and restored hippocampal long-term potentiation; the effects were blocked by the glycine-site NMDA antagonist gavestinel (10 mg/kg i.p.). Sunifiram did not improve depressive behavior in the tail suspension task. This is a surgical lesion model of Alzheimer-type deficits, not healthy animals.

Every endpoint is rescue of a chemically induced deficit or a lesion-model deficit, and none concerns people. No source says what sunifiram does to attention or memory in an unimpaired animal; the 2002 paper reports only the lack of effect on motility and coordination at effective doses. Most of the work is from one academic group between 2000 and 2006, with one independent group in 2013, and the newest sunifiram pharmacology paper in the records reviewed is from 2013.

2000 discovery paper: DM235 active at 0.001 mg/kg subcutaneously in mouse passive avoidance

Animal study

Manetti D et al. J Med Chem. 2000;43(23):4499-507. PMID 11087574.

In the 2000 piperazine paper, 1-acylpiperazines made by molecular simplification of the 1,4-diazabicyclo[4.3.0]nonan-9-one series kept high nootropic activity in the mouse passive avoidance test, with a pharmacological profile very similar to piracetam but much higher potency. Compound 13 (DM235) was active at 0.001 mg/kg subcutaneously, and DM232 and DM235 were the most potent of their classes and increased acetylcholine release in rat brain, although the molecular mechanism 'remains to be elucidated'.

Participants / model
Mice; rats
Treatment
Subcutaneous in mice
Follow-up
not stated in the source
Study design
Original preclinical study
Funding
not stated in the source

From the group that discovered the compound.

Read the original source
Core efficacy paper: rescue of scopolamine and other drug-induced amnesia, about 1,000 times piracetam-like potency

Animal study

Ghelardini C et al. Naunyn Schmiedebergs Arch Pharmacol. 2002;365(6):419-26. PMID 12070754.

Sunifiram prevented amnesia induced by scopolamine (1.5 mg/kg i.p.) in a mouse passive avoidance test after intraperitoneal (0.001 to 0.1 mg/kg) or oral (0.01 to 0.1 mg/kg) dosing, an effect comparable to piracetam (30 to 100 mg/kg i.p.), aniracetam (100 mg/kg p.o.) or rolipram (30 mg/kg p.o.). It also prevented amnesia induced by mecamylamine, baclofen and clonidine. In rats in the Morris water maze, 0.1 mg/kg i.p. 20 minutes before each daily training prevented scopolamine-induced impairment. At the highest effective doses it did not impair rota-rod coordination or change spontaneous motility, and the authors report a potency about 1,000 times that of the most active piracetam-like compounds.

Participants / model
Mice and rats with drug-induced amnesia
Treatment
Intraperitoneal (0.001 to 0.1 mg/kg) and oral (0.01 to 0.1 mg/kg) sunifiram
Follow-up
Acute dosing in mice; daily training in the rat water maze
Study design
Original preclinical study
Funding
not stated in the source

Every endpoint is rescue of a chemically induced deficit; none shows enhancement in unimpaired animals, and none concerns people.

Read the original source
Lesion model: olfactory bulbectomized mice, oral dosing for 7 to 12 days

Animal study

Moriguchi S et al. Behav Brain Res. 2013;242:150-7. PMID 23295391.

In olfactory bulbectomized mice, sunifiram at 0.01 to 1.0 mg/kg orally once a day for 7 to 12 days improved spatial reference memory in the Y-maze and short-term memory in novel object recognition, and restored hippocampal long-term potentiation; the effects were blocked by the glycine-site NMDA antagonist gavestinel (10 mg/kg i.p.). Sunifiram did not improve depressive behavior in the tail suspension task.

Participants / model
Olfactory bulbectomized mice
Treatment
Oral gavage, 0.01 to 1.0 mg/kg once daily
Follow-up
7 to 12 days
Study design
Original preclinical study
Funding
not stated in the source

This is a surgical lesion model of Alzheimer-type deficits, not healthy animals.

Read the original source
PubMed, 10 October 2026: 17 records, none gives sunifiram to a person

PubMed search record

PubMed search run 10 October 2026: sunifiram[tiab] OR DM-235[tiab] OR DM235[tiab] OR "DM 235"[tiab], with a follow-up query adding human, humans, patients, volunteers, clinical, trial, pharmacokinetic*, toxicity or safety.

The search returned 17 records and abstracts were read for all 17: 1 unrelated (a 1999 diabetic artery ultrasound paper), the retracted 2022 paper and its retraction notice, 8 chemistry or structure-activity papers, the 2006 and 2016 reviews, and 4 pharmacology papers. Adding human, humans, patients, volunteers, clinical, trial, pharmacokinetic*, toxicity or safety returned 6, none reporting sunifiram given to a person or any pharmacokinetic, toxicology or safety study; the human terms matched an Alzheimer's disease sentence, a review and the unrelated paper. The newest sunifiram pharmacology paper in the records is from 2013. No controlled human trial of acute memory and learning exists in the records reviewed.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Searches run on 10 October 2026
Study design
Dated literature search
Funding
not stated in the source

An absence finding as of 10 October 2026, not proof that no unpublished or unindexed work exists. No source says what sunifiram does to attention or memory in an unimpaired animal.

Read the original source

How does sunifiram act, and is it AMPA potentiation?

The mechanism is contested. AMPA receptor potentiation is one proposal, resting on indirect findings in mice and slices, and a 2013 independent paper instead found that sunifiram enhanced long-term potentiation through the glycine site of NMDA receptors. No abstract shows direct binding to or potentiation of AMPA receptors by sunifiram.

A 2006 review says both sunifiram and unifiram show no affinity in binding studies for the most important central receptors or channels, and that an AMPA role is a hypothesis supported by hippocampal slice experiments. The 2000 paper reports increased acetylcholine release in rat brain and says the molecular mechanism remains to be elucidated.

In mice, sunifiram at 0.1 mg/kg i.p. reversed amnesia induced by the AMPA antagonist NBQX (30 mg/kg i.p.), and in rat hippocampal slices sunifiram reversed kynurenic acid's block of NMDA-mediated [3H]noradrenaline release, an effect abolished by NBQX. The increase in excitatory synaptic transmission in that paper was shown for unifiram, not for sunifiram.

A 2013 paper from a different group found that in mouse hippocampal slices sunifiram at 10 to 100 nM enhanced long-term potentiation in a bell-shaped dose-response peaking at 10 nM, that the enhancement was blocked by an antagonist of the NMDA receptor glycine-binding site but not by ifenprodil, and that AMPA receptor phosphorylation rose through CaMKII downstream. The authors conclude that sunifiram stimulates the glycine site of NMDA receptors. In olfactory bulbectomized mice the effects of sunifiram were blocked by a glycine-site antagonist, gavestinel.

Contested mechanism: NBQX amnesia reversal, no binding affinity, and a 2013 NMDA glycine-site finding

Animal study

Galeotti N et al. Naunyn Schmiedebergs Arch Pharmacol. 2003;368(6):538-45. PMID 14600801.

In mice, sunifiram at 0.1 mg/kg i.p. reversed amnesia induced by the AMPA antagonist NBQX (30 mg/kg i.p.), and in rat hippocampal slices sunifiram reversed kynurenic acid's block of NMDA-mediated [3H]noradrenaline release, an effect abolished by NBQX; the increase in excitatory synaptic transmission in that paper was shown for unifiram, not for sunifiram. A 2006 review says both compounds show no affinity in binding studies for the most important central receptors or channels and that an AMPA role is a hypothesis supported by hippocampal slice experiments (PMID 16834757). A 2013 paper from a different group found that in mouse hippocampal slices sunifiram at 10 to 100 nM enhanced long-term potentiation in a bell-shaped dose-response peaking at 10 nM, that the enhancement was blocked by an antagonist of the NMDA receptor glycine-binding site but not by ifenprodil, and that AMPA receptor phosphorylation rose through CaMKII downstream, concluding sunifiram stimulates the glycine site of NMDA receptors (PMID 23733502). No abstract shows direct binding to or potentiation of AMPA receptors by sunifiram.

Participants / model
Mice; rat hippocampal slices; mouse hippocampal slices
Treatment
Intraperitoneal in mice; tissue bath in slices
Follow-up
not stated in the source
Study design
Original preclinical studies; review
Funding
not stated in the source

AMPA potentiation is one proposal and is indirect. The 2013 independent paper attributes the effect to the NMDA glycine site.

Read the original source
2000 discovery paper: DM235 active at 0.001 mg/kg subcutaneously in mouse passive avoidance

Animal study

Manetti D et al. J Med Chem. 2000;43(23):4499-507. PMID 11087574.

In the 2000 piperazine paper, 1-acylpiperazines made by molecular simplification of the 1,4-diazabicyclo[4.3.0]nonan-9-one series kept high nootropic activity in the mouse passive avoidance test, with a pharmacological profile very similar to piracetam but much higher potency. Compound 13 (DM235) was active at 0.001 mg/kg subcutaneously, and DM232 and DM235 were the most potent of their classes and increased acetylcholine release in rat brain, although the molecular mechanism 'remains to be elucidated'.

Participants / model
Mice; rats
Treatment
Subcutaneous in mice
Follow-up
not stated in the source
Study design
Original preclinical study
Funding
not stated in the source

From the group that discovered the compound.

Read the original source
Lesion model: olfactory bulbectomized mice, oral dosing for 7 to 12 days

Animal study

Moriguchi S et al. Behav Brain Res. 2013;242:150-7. PMID 23295391.

In olfactory bulbectomized mice, sunifiram at 0.01 to 1.0 mg/kg orally once a day for 7 to 12 days improved spatial reference memory in the Y-maze and short-term memory in novel object recognition, and restored hippocampal long-term potentiation; the effects were blocked by the glycine-site NMDA antagonist gavestinel (10 mg/kg i.p.). Sunifiram did not improve depressive behavior in the tail suspension task.

Participants / model
Olfactory bulbectomized mice
Treatment
Oral gavage, 0.01 to 1.0 mg/kg once daily
Follow-up
7 to 12 days
Study design
Original preclinical study
Funding
not stated in the source

This is a surgical lesion model of Alzheimer-type deficits, not healthy animals.

Read the original source

Does sunifiram enhance memory through AMPA receptors?

Only as reversal of drug-induced amnesia and a lesion-model deficit in rodents. The AMPA explanation is indirect in one paper and replaced by an NMDA glycine-site explanation in another, and nothing shows either effect in a person.

Sellers describe acute memory and learning enhancement through AMPA receptor potentiation. The memory part is supported only as reversal of drug-induced amnesia in mice and rats and of a deficit in olfactory bulbectomized mice. The AMPA part rests on indirect results, including reversal of amnesia caused by an AMPA antagonist, and a 2013 independent paper points to the NMDA glycine site instead.

A 2016 review by Gualtieri, who led the group that discovered unifiram and sunifiram, says the molecules were not protected by a patent, and that around 12 years after their 2000 disclosure he found dozens of web sites offering them as drugs that improve cognition in healthy individuals even though only few preclinical studies were done and their long-term toxicity was unknown. A 2022 review of smart drugs in healthy people lists unifiram analogues among popular cognitive enhancers and says efficacy in healthy individuals is uncertain and reported improvement temporary; its harm statements are for the stimulant-type enhancers it covers and not specific to sunifiram.

A PubMed query on 10 October 2026 for sunifiram, DM-235, DM235 and DM 235 returned 17 records, and abstracts were read for all 17. Adding human, humans, patients, volunteers, clinical, trial, pharmacokinetic*, toxicity or safety returned 6, and none reports sunifiram given to a person; the human terms matched an Alzheimer's disease sentence, a review and an unrelated 1999 ultrasound paper. No controlled human trial of acute memory and learning exists in the records reviewed.

Core efficacy paper: rescue of scopolamine and other drug-induced amnesia, about 1,000 times piracetam-like potency

Animal study

Ghelardini C et al. Naunyn Schmiedebergs Arch Pharmacol. 2002;365(6):419-26. PMID 12070754.

Sunifiram prevented amnesia induced by scopolamine (1.5 mg/kg i.p.) in a mouse passive avoidance test after intraperitoneal (0.001 to 0.1 mg/kg) or oral (0.01 to 0.1 mg/kg) dosing, an effect comparable to piracetam (30 to 100 mg/kg i.p.), aniracetam (100 mg/kg p.o.) or rolipram (30 mg/kg p.o.). It also prevented amnesia induced by mecamylamine, baclofen and clonidine. In rats in the Morris water maze, 0.1 mg/kg i.p. 20 minutes before each daily training prevented scopolamine-induced impairment. At the highest effective doses it did not impair rota-rod coordination or change spontaneous motility, and the authors report a potency about 1,000 times that of the most active piracetam-like compounds.

Participants / model
Mice and rats with drug-induced amnesia
Treatment
Intraperitoneal (0.001 to 0.1 mg/kg) and oral (0.01 to 0.1 mg/kg) sunifiram
Follow-up
Acute dosing in mice; daily training in the rat water maze
Study design
Original preclinical study
Funding
not stated in the source

Every endpoint is rescue of a chemically induced deficit; none shows enhancement in unimpaired animals, and none concerns people.

Read the original source
Contested mechanism: NBQX amnesia reversal, no binding affinity, and a 2013 NMDA glycine-site finding

Animal study

Galeotti N et al. Naunyn Schmiedebergs Arch Pharmacol. 2003;368(6):538-45. PMID 14600801.

In mice, sunifiram at 0.1 mg/kg i.p. reversed amnesia induced by the AMPA antagonist NBQX (30 mg/kg i.p.), and in rat hippocampal slices sunifiram reversed kynurenic acid's block of NMDA-mediated [3H]noradrenaline release, an effect abolished by NBQX; the increase in excitatory synaptic transmission in that paper was shown for unifiram, not for sunifiram. A 2006 review says both compounds show no affinity in binding studies for the most important central receptors or channels and that an AMPA role is a hypothesis supported by hippocampal slice experiments (PMID 16834757). A 2013 paper from a different group found that in mouse hippocampal slices sunifiram at 10 to 100 nM enhanced long-term potentiation in a bell-shaped dose-response peaking at 10 nM, that the enhancement was blocked by an antagonist of the NMDA receptor glycine-binding site but not by ifenprodil, and that AMPA receptor phosphorylation rose through CaMKII downstream, concluding sunifiram stimulates the glycine site of NMDA receptors (PMID 23733502). No abstract shows direct binding to or potentiation of AMPA receptors by sunifiram.

Participants / model
Mice; rat hippocampal slices; mouse hippocampal slices
Treatment
Intraperitoneal in mice; tissue bath in slices
Follow-up
not stated in the source
Study design
Original preclinical studies; review
Funding
not stated in the source

AMPA potentiation is one proposal and is indirect. The 2013 independent paper attributes the effect to the NMDA glycine site.

Read the original source
Inventor's 2016 account: not patented, sold online for healthy users, long-term toxicity unknown

Narrative review (preclinical literature)

Gualtieri F. J Enzyme Inhib Med Chem. 2016;31(2):187-94. PMID 25831025.

A 2016 review by Gualtieri, who led the group that discovered unifiram and sunifiram, says the molecules were not protected by a patent, that around 12 years after their 2000 disclosure he found dozens of web sites offering them as drugs that improve cognition in healthy individuals 'even if only few preclinical studies were done and their long-term toxicity was unknown'. A 2022 review of smart drugs in healthy people lists unifiram analogues among popular cognitive enhancers and says efficacy in healthy individuals is uncertain, reported improvement temporary, and harms possible, mainly from the stimulant-type enhancers it covers (PMID 35366192; its harm statements are not specific to sunifiram).

Participants / model
Healthy-user market; review literature
Treatment
Not applicable
Follow-up
Not applicable
Study design
Review articles
Funding
not stated in the source

Reviews, not studies of sunifiram in people.

Read the original source
PubMed, 10 October 2026: 17 records, none gives sunifiram to a person

PubMed search record

PubMed search run 10 October 2026: sunifiram[tiab] OR DM-235[tiab] OR DM235[tiab] OR "DM 235"[tiab], with a follow-up query adding human, humans, patients, volunteers, clinical, trial, pharmacokinetic*, toxicity or safety.

The search returned 17 records and abstracts were read for all 17: 1 unrelated (a 1999 diabetic artery ultrasound paper), the retracted 2022 paper and its retraction notice, 8 chemistry or structure-activity papers, the 2006 and 2016 reviews, and 4 pharmacology papers. Adding human, humans, patients, volunteers, clinical, trial, pharmacokinetic*, toxicity or safety returned 6, none reporting sunifiram given to a person or any pharmacokinetic, toxicology or safety study; the human terms matched an Alzheimer's disease sentence, a review and the unrelated paper. The newest sunifiram pharmacology paper in the records is from 2013. No controlled human trial of acute memory and learning exists in the records reviewed.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Searches run on 10 October 2026
Study design
Dated literature search
Funding
not stated in the source

An absence finding as of 10 October 2026, not proof that no unpublished or unindexed work exists. No source says what sunifiram does to attention or memory in an unimpaired animal.

Read the original source

What is known about the safety of sunifiram?

No human safety, pharmacokinetic or toxicology data exist for sunifiram, and its discoverer says long-term toxicity was unknown while it was sold online. In mice and rats it did not impair rota-rod coordination or change motility at the highest effective doses, and related compounds in the same series can be amnesic, so the identity of vendor material matters.

No record reports sunifiram given to a person, so no human adverse event, tolerability, pharmacokinetic or interaction information exists. No pharmacokinetic, toxicology or safety study of sunifiram appears among the 17 records reviewed.

A 2016 review by the discoverer says the long-term toxicity of the molecules was unknown while they were sold online to healthy users. A 2022 review of smart drugs says harms are possible, mainly from the stimulant-type enhancers it covers, and its harm statements are not specific to sunifiram.

In the core efficacy paper, sunifiram at the highest effective doses did not impair rota-rod coordination or change spontaneous motility in mice. That is a rodent finding from acute dosing and not a safety result in people.

Product identity is a separate concern. Related piperazines and amides in the same research series include amnesic compounds, so a product that contains a related compound would not be expected to behave as the studied molecule, and sunifiram and unifiram are different molecules.

PubMed, 10 October 2026: 17 records, none gives sunifiram to a person

PubMed search record

PubMed search run 10 October 2026: sunifiram[tiab] OR DM-235[tiab] OR DM235[tiab] OR "DM 235"[tiab], with a follow-up query adding human, humans, patients, volunteers, clinical, trial, pharmacokinetic*, toxicity or safety.

The search returned 17 records and abstracts were read for all 17: 1 unrelated (a 1999 diabetic artery ultrasound paper), the retracted 2022 paper and its retraction notice, 8 chemistry or structure-activity papers, the 2006 and 2016 reviews, and 4 pharmacology papers. Adding human, humans, patients, volunteers, clinical, trial, pharmacokinetic*, toxicity or safety returned 6, none reporting sunifiram given to a person or any pharmacokinetic, toxicology or safety study; the human terms matched an Alzheimer's disease sentence, a review and the unrelated paper. The newest sunifiram pharmacology paper in the records is from 2013. No controlled human trial of acute memory and learning exists in the records reviewed.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Searches run on 10 October 2026
Study design
Dated literature search
Funding
not stated in the source

An absence finding as of 10 October 2026, not proof that no unpublished or unindexed work exists. No source says what sunifiram does to attention or memory in an unimpaired animal.

Read the original source
Inventor's 2016 account: not patented, sold online for healthy users, long-term toxicity unknown

Narrative review (preclinical literature)

Gualtieri F. J Enzyme Inhib Med Chem. 2016;31(2):187-94. PMID 25831025.

A 2016 review by Gualtieri, who led the group that discovered unifiram and sunifiram, says the molecules were not protected by a patent, that around 12 years after their 2000 disclosure he found dozens of web sites offering them as drugs that improve cognition in healthy individuals 'even if only few preclinical studies were done and their long-term toxicity was unknown'. A 2022 review of smart drugs in healthy people lists unifiram analogues among popular cognitive enhancers and says efficacy in healthy individuals is uncertain, reported improvement temporary, and harms possible, mainly from the stimulant-type enhancers it covers (PMID 35366192; its harm statements are not specific to sunifiram).

Participants / model
Healthy-user market; review literature
Treatment
Not applicable
Follow-up
Not applicable
Study design
Review articles
Funding
not stated in the source

Reviews, not studies of sunifiram in people.

Read the original source
Core efficacy paper: rescue of scopolamine and other drug-induced amnesia, about 1,000 times piracetam-like potency

Animal study

Ghelardini C et al. Naunyn Schmiedebergs Arch Pharmacol. 2002;365(6):419-26. PMID 12070754.

Sunifiram prevented amnesia induced by scopolamine (1.5 mg/kg i.p.) in a mouse passive avoidance test after intraperitoneal (0.001 to 0.1 mg/kg) or oral (0.01 to 0.1 mg/kg) dosing, an effect comparable to piracetam (30 to 100 mg/kg i.p.), aniracetam (100 mg/kg p.o.) or rolipram (30 mg/kg p.o.). It also prevented amnesia induced by mecamylamine, baclofen and clonidine. In rats in the Morris water maze, 0.1 mg/kg i.p. 20 minutes before each daily training prevented scopolamine-induced impairment. At the highest effective doses it did not impair rota-rod coordination or change spontaneous motility, and the authors report a potency about 1,000 times that of the most active piracetam-like compounds.

Participants / model
Mice and rats with drug-induced amnesia
Treatment
Intraperitoneal (0.001 to 0.1 mg/kg) and oral (0.01 to 0.1 mg/kg) sunifiram
Follow-up
Acute dosing in mice; daily training in the rat water maze
Study design
Original preclinical study
Funding
not stated in the source

Every endpoint is rescue of a chemically induced deficit; none shows enhancement in unimpaired animals, and none concerns people.

Read the original source
Related compounds can be amnesic; a 2022 sunifiram hybrid paper was retracted in 2026

Animal study

Scapecchi S et al. Bioorg Med Chem. 2004;12(1):71-85. PMID 14697772.

A structure-activity study on unifiram and sunifiram reports that none of the compounds made reached the parent potency and that two compounds (numbered 13 and 14) had amnesic activity 'nearly equipotent to scopolamine'. Later series report that some cyclohexane derivatives induce amnesia (PMID 18954993), and the isopropylsulfonyl analog of unifiram was amnesic. A product sold as sunifiram that contains a related piperazine or amide would not be expected to behave as the studied molecule. The 2022 paper on sunifiram-carbamate hybrids (PMID 35484855) was retracted in 2026 (retraction notice PMID 42565552) and is not used.

Participants / model
Mice
Treatment
Not stated in the abstracts
Follow-up
not stated in the source
Study design
Original chemistry and pharmacology studies; retraction notice
Funding
not stated in the source

The retracted 2022 paper is not cited as evidence.

Read the original source

Why does sunifiram sit in D tier?

Sunifiram has plausible rodent pharmacology, but every result is rescue of an induced deficit in animals or slices, mostly from one academic group, with no human exposure, pharmacokinetics or toxicology and a contested mechanism, while it is sold online to healthy users on few preclinical studies.

Sunifiram has a long-standing, plausible pharmacology in rodents: about 1,000-fold the potency of piracetam-like drugs in drug-induced amnesia models, effects on hippocampal long-term potentiation, and an oral route that worked in mice.

But every result is a rescue of an induced deficit in mice, rats or brain slices, mostly from one academic group between 2000 and 2006, with one independent group in 2013. There is no human exposure, no pharmacokinetics and no toxicology, the molecule's own discoverer says it is sold online for healthy users with few preclinical studies and unknown long-term toxicity, and the mechanism cited by sellers, AMPA potentiation, is contested by a 2013 paper pointing to the NMDA glycine site.

Core efficacy paper: rescue of scopolamine and other drug-induced amnesia, about 1,000 times piracetam-like potency

Animal study

Ghelardini C et al. Naunyn Schmiedebergs Arch Pharmacol. 2002;365(6):419-26. PMID 12070754.

Sunifiram prevented amnesia induced by scopolamine (1.5 mg/kg i.p.) in a mouse passive avoidance test after intraperitoneal (0.001 to 0.1 mg/kg) or oral (0.01 to 0.1 mg/kg) dosing, an effect comparable to piracetam (30 to 100 mg/kg i.p.), aniracetam (100 mg/kg p.o.) or rolipram (30 mg/kg p.o.). It also prevented amnesia induced by mecamylamine, baclofen and clonidine. In rats in the Morris water maze, 0.1 mg/kg i.p. 20 minutes before each daily training prevented scopolamine-induced impairment. At the highest effective doses it did not impair rota-rod coordination or change spontaneous motility, and the authors report a potency about 1,000 times that of the most active piracetam-like compounds.

Participants / model
Mice and rats with drug-induced amnesia
Treatment
Intraperitoneal (0.001 to 0.1 mg/kg) and oral (0.01 to 0.1 mg/kg) sunifiram
Follow-up
Acute dosing in mice; daily training in the rat water maze
Study design
Original preclinical study
Funding
not stated in the source

Every endpoint is rescue of a chemically induced deficit; none shows enhancement in unimpaired animals, and none concerns people.

Read the original source
Contested mechanism: NBQX amnesia reversal, no binding affinity, and a 2013 NMDA glycine-site finding

Animal study

Galeotti N et al. Naunyn Schmiedebergs Arch Pharmacol. 2003;368(6):538-45. PMID 14600801.

In mice, sunifiram at 0.1 mg/kg i.p. reversed amnesia induced by the AMPA antagonist NBQX (30 mg/kg i.p.), and in rat hippocampal slices sunifiram reversed kynurenic acid's block of NMDA-mediated [3H]noradrenaline release, an effect abolished by NBQX; the increase in excitatory synaptic transmission in that paper was shown for unifiram, not for sunifiram. A 2006 review says both compounds show no affinity in binding studies for the most important central receptors or channels and that an AMPA role is a hypothesis supported by hippocampal slice experiments (PMID 16834757). A 2013 paper from a different group found that in mouse hippocampal slices sunifiram at 10 to 100 nM enhanced long-term potentiation in a bell-shaped dose-response peaking at 10 nM, that the enhancement was blocked by an antagonist of the NMDA receptor glycine-binding site but not by ifenprodil, and that AMPA receptor phosphorylation rose through CaMKII downstream, concluding sunifiram stimulates the glycine site of NMDA receptors (PMID 23733502). No abstract shows direct binding to or potentiation of AMPA receptors by sunifiram.

Participants / model
Mice; rat hippocampal slices; mouse hippocampal slices
Treatment
Intraperitoneal in mice; tissue bath in slices
Follow-up
not stated in the source
Study design
Original preclinical studies; review
Funding
not stated in the source

AMPA potentiation is one proposal and is indirect. The 2013 independent paper attributes the effect to the NMDA glycine site.

Read the original source
Inventor's 2016 account: not patented, sold online for healthy users, long-term toxicity unknown

Narrative review (preclinical literature)

Gualtieri F. J Enzyme Inhib Med Chem. 2016;31(2):187-94. PMID 25831025.

A 2016 review by Gualtieri, who led the group that discovered unifiram and sunifiram, says the molecules were not protected by a patent, that around 12 years after their 2000 disclosure he found dozens of web sites offering them as drugs that improve cognition in healthy individuals 'even if only few preclinical studies were done and their long-term toxicity was unknown'. A 2022 review of smart drugs in healthy people lists unifiram analogues among popular cognitive enhancers and says efficacy in healthy individuals is uncertain, reported improvement temporary, and harms possible, mainly from the stimulant-type enhancers it covers (PMID 35366192; its harm statements are not specific to sunifiram).

Participants / model
Healthy-user market; review literature
Treatment
Not applicable
Follow-up
Not applicable
Study design
Review articles
Funding
not stated in the source

Reviews, not studies of sunifiram in people.

Read the original source
PubMed, 10 October 2026: 17 records, none gives sunifiram to a person

PubMed search record

PubMed search run 10 October 2026: sunifiram[tiab] OR DM-235[tiab] OR DM235[tiab] OR "DM 235"[tiab], with a follow-up query adding human, humans, patients, volunteers, clinical, trial, pharmacokinetic*, toxicity or safety.

The search returned 17 records and abstracts were read for all 17: 1 unrelated (a 1999 diabetic artery ultrasound paper), the retracted 2022 paper and its retraction notice, 8 chemistry or structure-activity papers, the 2006 and 2016 reviews, and 4 pharmacology papers. Adding human, humans, patients, volunteers, clinical, trial, pharmacokinetic*, toxicity or safety returned 6, none reporting sunifiram given to a person or any pharmacokinetic, toxicology or safety study; the human terms matched an Alzheimer's disease sentence, a review and the unrelated paper. The newest sunifiram pharmacology paper in the records is from 2013. No controlled human trial of acute memory and learning exists in the records reviewed.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Searches run on 10 October 2026
Study design
Dated literature search
Funding
not stated in the source

An absence finding as of 10 October 2026, not proof that no unpublished or unindexed work exists. No source says what sunifiram does to attention or memory in an unimpaired animal.

Read the original source

Studies and sources

PubChem: sunifiram identity, and how it differs from unifiram

Chemical identity record

PubChem, National Library of Medicine. Compound summary for CID 4223812, 1-(4-benzoylpiperazin-1-yl)propan-1-one (sunifiram, DM-235).

Sunifiram is PubChem CID 4223812, C14H18N2O2, 246.30 g/mol, CAS 314728-85-3, IUPAC name 1-(4-benzoylpiperazin-1-yl)propan-1-one, with the synonyms DM-235 and DM235. It is an acylpiperazine. Unifiram (DM-232) is a different molecule (PubChem CID 9861054, C13H15FN2O3S, 298.34 g/mol, CAS 272786-64-8, a fluorophenylsulfonyl hexahydropyrrolo[1,2-a]pyrazin-6-one), so the two are not the same compound and are not interchangeable by identity. Papers describe sunifiram's lineage inconsistently: as related in structure to piracetam (2002), related to ampakines (2003) and a pyrrolidone nootropic (2013), while the 2000 chemistry paper says the acylpiperazine group can mimic the 2-pyrrolidinone ring and the PubChem name shows no pyrrolidone ring. Sapunifiram (MN19) is another related compound in the same literature.

Participants / model
Chemical identity, not a clinical population
Treatment
Defined molecule, free form
Follow-up
Not applicable
Study design
Chemical database record; original reports
Funding
not stated in the source

Identity record only. No public record ties any vendor material to the studied compound.

Read the original source
2000 discovery paper: DM235 active at 0.001 mg/kg subcutaneously in mouse passive avoidance

Animal study

Manetti D et al. J Med Chem. 2000;43(23):4499-507. PMID 11087574.

In the 2000 piperazine paper, 1-acylpiperazines made by molecular simplification of the 1,4-diazabicyclo[4.3.0]nonan-9-one series kept high nootropic activity in the mouse passive avoidance test, with a pharmacological profile very similar to piracetam but much higher potency. Compound 13 (DM235) was active at 0.001 mg/kg subcutaneously, and DM232 and DM235 were the most potent of their classes and increased acetylcholine release in rat brain, although the molecular mechanism 'remains to be elucidated'.

Participants / model
Mice; rats
Treatment
Subcutaneous in mice
Follow-up
not stated in the source
Study design
Original preclinical study
Funding
not stated in the source

From the group that discovered the compound.

Read the original source
Core efficacy paper: rescue of scopolamine and other drug-induced amnesia, about 1,000 times piracetam-like potency

Animal study

Ghelardini C et al. Naunyn Schmiedebergs Arch Pharmacol. 2002;365(6):419-26. PMID 12070754.

Sunifiram prevented amnesia induced by scopolamine (1.5 mg/kg i.p.) in a mouse passive avoidance test after intraperitoneal (0.001 to 0.1 mg/kg) or oral (0.01 to 0.1 mg/kg) dosing, an effect comparable to piracetam (30 to 100 mg/kg i.p.), aniracetam (100 mg/kg p.o.) or rolipram (30 mg/kg p.o.). It also prevented amnesia induced by mecamylamine, baclofen and clonidine. In rats in the Morris water maze, 0.1 mg/kg i.p. 20 minutes before each daily training prevented scopolamine-induced impairment. At the highest effective doses it did not impair rota-rod coordination or change spontaneous motility, and the authors report a potency about 1,000 times that of the most active piracetam-like compounds.

Participants / model
Mice and rats with drug-induced amnesia
Treatment
Intraperitoneal (0.001 to 0.1 mg/kg) and oral (0.01 to 0.1 mg/kg) sunifiram
Follow-up
Acute dosing in mice; daily training in the rat water maze
Study design
Original preclinical study
Funding
not stated in the source

Every endpoint is rescue of a chemically induced deficit; none shows enhancement in unimpaired animals, and none concerns people.

Read the original source
Contested mechanism: NBQX amnesia reversal, no binding affinity, and a 2013 NMDA glycine-site finding

Animal study

Galeotti N et al. Naunyn Schmiedebergs Arch Pharmacol. 2003;368(6):538-45. PMID 14600801.

In mice, sunifiram at 0.1 mg/kg i.p. reversed amnesia induced by the AMPA antagonist NBQX (30 mg/kg i.p.), and in rat hippocampal slices sunifiram reversed kynurenic acid's block of NMDA-mediated [3H]noradrenaline release, an effect abolished by NBQX; the increase in excitatory synaptic transmission in that paper was shown for unifiram, not for sunifiram. A 2006 review says both compounds show no affinity in binding studies for the most important central receptors or channels and that an AMPA role is a hypothesis supported by hippocampal slice experiments (PMID 16834757). A 2013 paper from a different group found that in mouse hippocampal slices sunifiram at 10 to 100 nM enhanced long-term potentiation in a bell-shaped dose-response peaking at 10 nM, that the enhancement was blocked by an antagonist of the NMDA receptor glycine-binding site but not by ifenprodil, and that AMPA receptor phosphorylation rose through CaMKII downstream, concluding sunifiram stimulates the glycine site of NMDA receptors (PMID 23733502). No abstract shows direct binding to or potentiation of AMPA receptors by sunifiram.

Participants / model
Mice; rat hippocampal slices; mouse hippocampal slices
Treatment
Intraperitoneal in mice; tissue bath in slices
Follow-up
not stated in the source
Study design
Original preclinical studies; review
Funding
not stated in the source

AMPA potentiation is one proposal and is indirect. The 2013 independent paper attributes the effect to the NMDA glycine site.

Read the original source
Lesion model: olfactory bulbectomized mice, oral dosing for 7 to 12 days

Animal study

Moriguchi S et al. Behav Brain Res. 2013;242:150-7. PMID 23295391.

In olfactory bulbectomized mice, sunifiram at 0.01 to 1.0 mg/kg orally once a day for 7 to 12 days improved spatial reference memory in the Y-maze and short-term memory in novel object recognition, and restored hippocampal long-term potentiation; the effects were blocked by the glycine-site NMDA antagonist gavestinel (10 mg/kg i.p.). Sunifiram did not improve depressive behavior in the tail suspension task.

Participants / model
Olfactory bulbectomized mice
Treatment
Oral gavage, 0.01 to 1.0 mg/kg once daily
Follow-up
7 to 12 days
Study design
Original preclinical study
Funding
not stated in the source

This is a surgical lesion model of Alzheimer-type deficits, not healthy animals.

Read the original source
Related compounds can be amnesic; a 2022 sunifiram hybrid paper was retracted in 2026

Animal study

Scapecchi S et al. Bioorg Med Chem. 2004;12(1):71-85. PMID 14697772.

A structure-activity study on unifiram and sunifiram reports that none of the compounds made reached the parent potency and that two compounds (numbered 13 and 14) had amnesic activity 'nearly equipotent to scopolamine'. Later series report that some cyclohexane derivatives induce amnesia (PMID 18954993), and the isopropylsulfonyl analog of unifiram was amnesic. A product sold as sunifiram that contains a related piperazine or amide would not be expected to behave as the studied molecule. The 2022 paper on sunifiram-carbamate hybrids (PMID 35484855) was retracted in 2026 (retraction notice PMID 42565552) and is not used.

Participants / model
Mice
Treatment
Not stated in the abstracts
Follow-up
not stated in the source
Study design
Original chemistry and pharmacology studies; retraction notice
Funding
not stated in the source

The retracted 2022 paper is not cited as evidence.

Read the original source
Inventor's 2016 account: not patented, sold online for healthy users, long-term toxicity unknown

Narrative review (preclinical literature)

Gualtieri F. J Enzyme Inhib Med Chem. 2016;31(2):187-94. PMID 25831025.

A 2016 review by Gualtieri, who led the group that discovered unifiram and sunifiram, says the molecules were not protected by a patent, that around 12 years after their 2000 disclosure he found dozens of web sites offering them as drugs that improve cognition in healthy individuals 'even if only few preclinical studies were done and their long-term toxicity was unknown'. A 2022 review of smart drugs in healthy people lists unifiram analogues among popular cognitive enhancers and says efficacy in healthy individuals is uncertain, reported improvement temporary, and harms possible, mainly from the stimulant-type enhancers it covers (PMID 35366192; its harm statements are not specific to sunifiram).

Participants / model
Healthy-user market; review literature
Treatment
Not applicable
Follow-up
Not applicable
Study design
Review articles
Funding
not stated in the source

Reviews, not studies of sunifiram in people.

Read the original source
PubMed, 10 October 2026: 17 records, none gives sunifiram to a person

PubMed search record

PubMed search run 10 October 2026: sunifiram[tiab] OR DM-235[tiab] OR DM235[tiab] OR "DM 235"[tiab], with a follow-up query adding human, humans, patients, volunteers, clinical, trial, pharmacokinetic*, toxicity or safety.

The search returned 17 records and abstracts were read for all 17: 1 unrelated (a 1999 diabetic artery ultrasound paper), the retracted 2022 paper and its retraction notice, 8 chemistry or structure-activity papers, the 2006 and 2016 reviews, and 4 pharmacology papers. Adding human, humans, patients, volunteers, clinical, trial, pharmacokinetic*, toxicity or safety returned 6, none reporting sunifiram given to a person or any pharmacokinetic, toxicology or safety study; the human terms matched an Alzheimer's disease sentence, a review and the unrelated paper. The newest sunifiram pharmacology paper in the records is from 2013. No controlled human trial of acute memory and learning exists in the records reviewed.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Searches run on 10 October 2026
Study design
Dated literature search
Funding
not stated in the source

An absence finding as of 10 October 2026, not proof that no unpublished or unindexed work exists. No source says what sunifiram does to attention or memory in an unimpaired animal.

Read the original source
Label and registry searches, 10 October 2026: no label, no match, no study

Regulatory record

U.S. Food and Drug Administration (DailyMed label search; openFDA Drugs@FDA query) and ClinicalTrials.gov. Searches for sunifiram and for DM235, DM-235 and DM 235, run 10 October 2026.

ClinicalTrials.gov queries for sunifiram and for DM235, DM-235 and DM 235 returned no studies, DailyMed returned no label, and the openFDA Drugs@FDA query returned no match.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Searches run on 10 October 2026
Study design
Dated label, application and registry search
Funding
not stated in the source

An absence finding as of 10 October 2026. It covers United States label, application and registry records only.

Read the original source

Why is Sunifiram (DM-235) in D tier?

D fits a compound with plausible rodent pharmacology but no human exposure, pharmacokinetics or toxicology. Every result is rescue of an induced deficit in mice, rats or slices, mostly from one academic group, the AMPA mechanism that sellers cite is contested, and the molecule's own discoverer says it is sold online to healthy users on few preclinical studies.