DSIP

Delta Sleep-Inducing Peptide

DSIP (Delta Sleep-Inducing Peptide) is a sleep research compound. Neuromodulatory peptide that influences sleep architecture and the stress/cortisol axis.

How DSIP works

Neuromodulatory peptide that influences sleep architecture and the stress/cortisol axis.

Proposed benefits

Sleep-onset support and stress modulation.

Where to get DSIP

DSIP is sold in 2 forms. They are not interchangeable — the dose and the route differ, so pick the one this page describes unless you know why you want another.

The evidence for DSIP

Graded by what exists behind each claim.

✅ Clinically validated

• Human studies exist and are old and small — European work in the 1980s examined DSIP in insomnia, chronic pain and opioid withdrawal, with some reporting improvement. None was replicated and no development followed.

📊 Correlative data

• Used for sleep quality. Reported experience is notably inconsistent — some describe a clear effect on sleep depth, many describe nothing, which is one of the widest response spreads in the Vault.

🧪 Theoretical / extrapolated

• Delta Sleep-Inducing Peptide, isolated from rabbit cerebral venous blood during sleep. Proposed to promote delta-wave sleep and to modulate corticotropin release.
• Its own biology is contested: whether endogenous DSIP is a real sleep-regulating hormone was never settled, and it degrades in minutes in serum. A peptide whose native role is unproven and whose half-life is that short is a thin foundation, which the inconsistent reports probably reflect.

How to read these tiers: they say how much human evidence exists, not how well something works — and ✗ flags harm, never a disappointing trial. How the evidence tiers work →

What DSIP actually does

Start with the sequence, because it is nine residues long and it explains almost everything about this peptide. DSIP is Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu Schoenenberger 1978. Read it residue by residue and a picture emerges that no vendor page draws. Three of the nine are glycine and two more are alanine — five of nine positions carry either no side chain at all or a single methyl group. Glycine is the residue that makes a peptide backbone hinge; a nonapeptide that is one-third glycine has almost no capacity to hold a defined shape in solution.

The charge follows from the same reading. Aspartate and glutamate each carry a side-chain carboxyl, the C-terminus carries another, and the free N-terminal amine is the only positive charge. At blood pH that is a net charge of about −2 on a molecule of roughly 849 daltons. An 849 Da dianion does not cross a lipid membrane by diffusion, in either direction, which means any account of DSIP reaching the brain has to name a transporter — and no such transporter has ever been named for it.

Then the structural feature that is actually distinctive: the N-terminal tryptophan. Trp is the largest and most hydrophobic of the twenty, and it is sitting at the free amino end. On a peptide with no D-amino acids, no C-terminal amide and no cyclization, that N-terminus is an open invitation to an aminopeptidase — the enzymes that chew peptides one residue at a time from the amino end. The one interesting residue is in the position least likely to survive.

And here is the part that separates this page from every other compound in this cohort: there is no receptor. Kovalzon 2006, reviewing the field three decades after the discovery, concluded that the link between DSIP and sleep has never been further characterized and that the hypothesis regarding DSIP as a sleep factor is extremely poorly documented. No receptor was ever identified. No gene was ever identified. There is no known precursor protein from which this nonapeptide is cleaved. So the honest mechanistic statement is not “DSIP acts on X”; it is that a synthetic nine-residue anion with no known receptor, no known transporter and no identified biosynthetic origin was reported to change an EEG in rabbits in the 1970s, and the mechanism question was abandoned rather than answered.

Cell, rodent, human — and where it stops

Step one, rabbits, and the method is stranger than the summaries suggest. Schoenenberger 1977 obtained the material from an extracorporeal dialysate of cerebral venous blood in rabbits — blood drawn from the brain’s venous outflow, dialysed, and the low-molecular-weight fraction tested. Electroencephalogram leads from the neocortex and the archicortex were directly fast-Fourier transformed, across 58 rabbits. This is a 1977 experiment in which a substance was defined by an effect before anybody knew what it was.

Step two, the sequence and the synthetic peptide. Schoenenberger 1978 reported the amino acid analysis, the sequence Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu, the synthesis and the activity of the synthetic nonapeptide: a mean increase in EEG delta activity of 35% in the neocortex and limbic cortex. That 35% is the founding number of this entire field.

Step three, humans, six of them, intravenous. Schneider-Helmert 1981 gave six normal volunteers (four male, two female) 25 nmol/kg by slow intravenous infusion. The acute result was a 59% increase in the median of total sleep time within a 130-minute interval. The delayed results were shorter sleep onset, a reduced percentage of stage 1, and better sleep efficiency.

Step four, humans with the condition, also six. Schneider-Helmert 1981 gave six middle-aged chronic insomniacs the same 25 nmol/kg intravenously and reported longer sleep duration and higher sleep quality with fewer interruptions, and slightly more REM sleep, an effect described as lasting up to 6 hours of night sleep.

Step five: nothing, for forty-five years. That is the complete human record — two studies, six subjects each, both in 1981, both from the group that discovered the peptide, both by intravenous infusion. Kovalzon 2006 is the field’s own verdict on what followed: the link was never further characterized and the sleep-factor hypothesis is extremely poorly documented.

The obstacle, stated exactly, and there are three of them stacked. First, n = 6 twice, unreplicated by any independent group — at that sample size a 59% median change is a hypothesis, not a result. Second, the route: every human number above came from a slow intravenous infusion, and every person using this compound today injects it subcutaneously, for which no pharmacokinetic study exists in any species. Third, and most fundamental, the mechanism was never found, so there is no receptor occupancy, no dose-response curve and no biomarker to anchor a translation to. A translation chain needs something at the far end to translate into.

DSIP pharmacokinetics — how much of it actually gets in

Route: the only human route with data is intravenous, and it is not the route anybody uses. Both human studies infused 25 nmol/kg slowly Schneider-Helmert 1981 Schneider-Helmert 1981. There is no published subcutaneous, intranasal or oral pharmacokinetic study of DSIP in a human or in any animal.

The arithmetic that converts that dose into something comparable, done out loud. DSIP’s nine residues give a molecular weight of about 849 Da. So 25 nmol/kg is about 21 µg/kg, which for a 70 kg adult is roughly 1.5 mg — infused intravenously. Set that against the 100–500 µg subcutaneous doses this site records and the everyday dose is three to fifteen times lower than the only dose ever shown to do anything in a person, delivered by a route with no data at all. That single comparison explains the reported response spread better than any pharmacology on this page.

What degrades it: peptidases, quickly, and the structure says why. This is a linear nonapeptide with a free N-terminus, a free C-terminal carboxyl, all L-amino acids, no amidation and no cyclization. Every one of those is a feature that peptide chemists remove when they want a peptide to survive in serum. Aminopeptidases attack the exposed N-terminal tryptophan; endopeptidases have a floppy, glycine-rich backbone with no secondary structure to hide the scissile bonds. Serum half-lives for unmodified linear peptides of this size and design are measured in minutes, and no DSIP-specific number has been published to replace that reasoning.

The oral barrier is absolute and worth stating. An 849 Da dianion faces brush-border peptidases first and a charged membrane second; there is no oral DSIP and there is no mechanism by which there could be one without a delivery technology nobody has applied to it.

And here is the pharmacokinetic–pharmacodynamic mismatch that is the most interesting unexamined thing about this peptide. The 1981 human study reported delayed effects — shorter sleep onset, less stage 1, better sleep efficiency — separate from the acute 130-minute window Schneider-Helmert 1981. A molecule cleared in minutes cannot be present to cause an effect days later. So either the peptide triggers something durable downstream, which would be a genuinely important finding and has never been chased, or the delayed effects are noise in six people. Nobody has done the experiment that would distinguish those, and both readings remain live.

What would have to be true, and how you would know it was not

Four predictions. The third is the one that cuts against the compound as it is actually used, and the fourth is the design that would settle it.

1. If anything is happening, it is in the delta band — so the read-out has to be an EEG, not a wrist. The founding result is a 35% increase in EEG delta activity Schoenenberger 1978, and the human results were changes in total sleep time, stage 1 percentage and sleep efficiency Schneider-Helmert 1981 — all polysomnograph variables. Actigraphy cannot see any of them; a wrist tracker infers sleep stages from movement and heart rate and would not detect a delta-power change if one occurred. This is the single most common way a person self-testing this peptide gets a meaningless answer.

2. The corticotropin claim is testable and has never been tested at these doses. DSIP has been proposed to modulate corticotropin release, and that is a claim with a blood test attached. Draw cortisol at a fixed morning hour, and ACTH alongside it, at baseline and after two weeks. The prediction this page will make is that neither moves — because the peptide has no identified receptor and is cleared in minutes Kovalzon 2006. A reproducible shift in either would be new information about a molecule that has produced almost none since 1981.

3. The prediction that cuts against it: at 100–500 µg subcutaneous, the published effect should not reproduce. The human effect came from about 1.5 mg intravenously Schneider-Helmert 1981. The usual dose is three to fifteen times smaller, by a route that must first survive subcutaneous tissue peptidases and then be absorbed. So the falsifiable claim is directional: a properly instrumented night should show no delta-power change at these doses. If someone runs a home EEG and finds one, that is a genuinely surprising result worth reporting; if they find nothing, the compound has behaved exactly as the arithmetic predicts and no further dose escalation is indicated by anything published.

4. And the design that would separate a real effect from expectation, which matters more here than anywhere else in this cohort. Because the original report includes delayed effects Schneider-Helmert 1981, a simple before-and-after comparison cannot distinguish carryover from belief. Run 14 nights baseline, 14 nights on, 14 nights off, with the same instrument throughout. The prediction that separates the hypotheses: a real peptide-triggered downstream effect should persist into the off block and then decay, while expectation should collapse immediately when dosing stops. Nobody has ever run that design on DSIP.

What nobody has tested yet

Nobody has identified a receptor, and it is fifty years on. Kovalzon 2006 states it plainly: no receptor, no gene, and a sleep-factor hypothesis that is extremely poorly documented. The methods that would settle it now — a labeled DSIP binding screen against a comprehensive receptor panel, or an affinity pull-down from brain lysate with mass spectrometry — did not exist when the field was active and have never been applied since. This is the experiment that would either revive the molecule or close it.

Nobody has measured a plasma DSIP concentration after a subcutaneous dose in anything. Every human number comes from an intravenous infusion Schneider-Helmert 1981 Schneider-Helmert 1981; every current use is subcutaneous. A single-timepoint LC-MS/MS assay in a handful of people would establish whether the compound reaches the circulation at all by that route, and it has never been published.

Nobody has replicated the 1981 studies, by any group, ever. Two trials of six people each, from one laboratory, constitute the entire human evidence base for a peptide that has been sold continuously for decades. A replication would need twelve people, a sleep laboratory and one night each, and it would be the first independent human test of this molecule in forty-five years.

And the version of that experiment anybody can now run. Consumer sleep-EEG headbands measure delta power directly — the exact variable that produced the founding 35% result Schoenenberger 1978. Ten people running the 14/14/14 on-off design with the same headband would generate more human delta-power data on this peptide than exists in the published literature. That is an unusual situation and it is the honest reason this page exists.

DSIP — its own safety story, not its class's

The entire human safety dataset for DSIP is twelve people, in 1981, by intravenous infusion. Six normal volunteers Schneider-Helmert 1981 and six chronic insomniacs Schneider-Helmert 1981, all given 25 nmol/kg, all in a laboratory, all in one year. Neither report describes a toxicity signal. That is not the same as a safety record, and this page will not present it as one: twelve single exposures cannot detect anything uncommon, anything delayed, or anything that requires repeated dosing to appear.

The route mismatch is this compound’s own specific risk. The published exposure is intravenous and slow; the actual exposure is a subcutaneous bolus. Subcutaneous tissue has its own peptidases and its own local reaction to a bolus of a synthetic peptide, and no study in any species has characterized what happens at a DSIP injection site. Nothing in the intravenous record covers it.

Identity is the risk that is easy to overlook and easy to state. There is no approved DSIP product anywhere and no pharmacopoeial monograph, so identity rests entirely on a vendor certificate. And the usual certificate does not settle the question: a mass spectrometer reading 849 Da confirms the composition Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu, and cannot distinguish it from any rearrangement of the same nine residues, because sequence isomers have identical mass. On a peptide that is one-third glycine, plenty of rearrangements are chemically easy to make and pharmacologically meaningless.

And the honest summary of the risk position, which is unusual. No receptor is known Kovalzon 2006, so there is no antagonist, no antidote, and no basis for predicting an interaction with anything else — but equally no known target through which harm would be expected. A microgram dose of a rapidly cleared endogenous-looking nonapeptide carries a low prior for acute toxicity, and decades of small-scale use without a reported catastrophe is weak evidence rather than none. The accurate sentence is that the risk is probably low and definitely uncharacterized, and those are different claims.

Sources read for this page

• Schoenenberger GA, Monnier M. Characterization of a delta-electroencephalogram (-sleep)-inducing peptide.. Proc Natl Acad Sci U S A 1977 · PMID 265572
• Schoenenberger GA, Maier PF, Tobler HJ, Wilson K, Monnier M. The delta EEG (sleep)-inducing peptide (DSIP). XI. Amino-acid analysis, sequence, synthesis and activity of the nonapeptide.. Pflugers Arch 1978 · PMID 568769
• Schneider-Helmert D, Gnirss F, Monnier M, Schenker J, Schoenenberger GA. Acute and delayed effects of DSIP (delta sleep-inducing peptide) on human sleep behavior.. Int J Clin Pharmacol Ther Toxicol 1981 · PMID 6895513
• Schneider-Helmert D, Schoenenberger GA. The influence of synthetic DSIP (delta-sleep-inducing-peptide) on disturbed human sleep.. Experientia 1981 · PMID 7028502
• Kovalzon VM, Strekalova TV. Delta sleep-inducing peptide (DSIP): a still unresolved riddle.. J Neurochem 2006 · PMID 16539679

DSIP — safety, predicted from mechanism

DSIP — interference & stacking

Bloodwork to run alongside DSIP

Baseline first, then again at 8–12 weeks.

The Sleep Quality & Recovery panel covers these in one order — 11 markers, $172.35 with the discount applied.

Check results you already have → · All 103 markers A–Z

DSIP — frequently asked questions

DSIP inside a finished plan

One arm of 2 Protocol Blueprints, free to read in full.

What DSIP is used for

DSIP appears under 2 goals in the goal router.

Related Sleep compounds

← Explore the full Protocol Vault

↑ Back to on this page