DSIP Peptide (Delta Sleep-Inducing Peptide): What the Research Shows

DSIP (delta sleep-inducing peptide) is a nine-amino-acid peptide, Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu, isolated in 1977 by Swiss researchers from the blood of rabbits in which sleep had been induced by brain stimulation.¹ Despite its name, a sleep-regulating role for the DSIP peptide has never been firmly established, and no gene encoding it, protein precursor or receptor has been isolated.² It is not an FDA-approved drug³ and is not named in Australia's Poisons Standard.⁴

This article traces what DSIP is, why its biology is still an open question almost fifty years after its discovery, and what the animal and human studies actually reported. It summarises research; it is not guidance on use.

What is the DSIP peptide?

DSIP came out of the Basel laboratory of Schoenenberger and Monnier.² The group induced sleep in rabbits by electrically stimulating the thalamus and characterised the peptide from the animals' cerebral venous blood plasma.⁵ In 1977 they reported the sequence of the active nonapeptide. They synthesised it along with five possible breakdown fragments and three related peptides, infused each into the brain ventricles of rabbits under double-blind conditions (58 animals including controls), and reported that only synthetic DSIP significantly and specifically enhanced delta and spindle EEG patterns.¹

"Delta" refers to delta waves, the slow, high-amplitude EEG rhythm characteristic of deep non-REM sleep. The name therefore describes what the peptide appeared to do in the original rabbit experiments, not an established physiological role. Its international non-proprietary name is emideltide.⁶

DSIP at a glance

Chemical identifiers below are as listed by PubChem.⁶

The unsolved question: where does DSIP come from?

Most signalling peptides are cut from larger precursor proteins, each encoded by a known gene. DSIP is an exception. A 2006 review in the Journal of Neurochemistry titled it "a still unresolved riddle": the link between DSIP and sleep had never been characterised further, partly because no DSIP gene, precursor protein or receptor had been isolated, and the authors described the hypothesis of DSIP as a sleep factor as "extremely poorly documented and still weak". They also noted that DSIP's structure differs from every other known peptide family.²

Two lines of work have tried to fill the gap:

• DSIP-like immunoreactivity. Antibodies raised against DSIP detect "DSIP-like" material in the brain, in peripheral organs and in the plasma of several mammals.⁷ But when a team at the Karolinska Institutet in Stockholm purified a large immunoreactive substance from pig brain in 1993, it turned out to be a 77-residue peptide whose sequence was unrelated to DSIP.⁵ That is a reminder that antibody signals labelled "DSIP-like" may not be DSIP itself.
• A putative precursor. In 2011, a Moscow group searched gene and protein databases for DSIP-like sequences and found one in a family of histone demethylases encoded by the JMJD1B gene: WKGGNASGE, which differs from DSIP at positions 2 and 5. They proposed these enzymes as possible precursors of endogenous peptides with DSIP-like activity.⁸ That is a hypothesis about a related peptide, not evidence that DSIP itself is made in the body.

So, as far as the published literature shows, no gene encoding DSIP and no confirmed precursor protein or receptor has been identified. Any claim that DSIP is "the body's natural sleep peptide" goes beyond the evidence.

What has DSIP been studied for?

Most DSIP research is old: of roughly 400 PubMed records with the peptide's full name in the title or abstract, more than three-quarters were published before 2000. A 1984 review summarised reports that DSIP mainly promoted delta sleep in rabbits, rats, mice and humans, but affected REM sleep more in cats, and that its activity followed a U-shaped curve: effects appeared at some doses and infusion times and faded at higher or lower ones. The same review listed reported effects on electrophysiology, brain neurotransmitter levels, circadian and locomotor patterns, hormone levels and drug withdrawal.⁷ Such a broad list of reported effects from one short peptide is itself a reason for caution.

Some findings cut against a simple sleep story. The authors of the 2006 review reported that their own early studies found significant slow-wave sleep promotion by certain synthetic DSIP analogues, but not by DSIP itself.²

Reaching the brain, and surviving in blood

Whether a peptide injected into the bloodstream reaches the brain matters for interpreting any effect. A 1984 rat and dog study reported that a radioiodinated analogue, [125I]N-Tyr-DSIP, crossed the blood-brain barrier and the blood-cerebrospinal fluid barrier by a non-competitive mechanism.⁹ Note that this was a labelled analogue, not DSIP itself. Stability is a separate problem: a 1987 study found that DSIP incubated in human or rat blood was rapidly degraded, releasing products with the chromatographic behaviour of free tryptophan, whereas two analogues persisted longer.¹⁰

Analogues and modified forms

Because DSIP itself breaks down quickly, researchers also made analogues, including a phosphorylated form (P-DSIP). In the 1987 blood-stability study, two radiolabelled analogues, one of them phosphorylated, degraded more slowly than DSIP and, unlike DSIP, formed complexes in serum. The authors left open whether this explains why analogues sometimes gave stronger or more consistent effects than the parent peptide.¹⁰

DSIP and melatonin are unrelated

DSIP is often mentioned alongside melatonin, but the two have nothing structurally in common. Melatonin is a small indoleamine hormone made by the pineal gland from tryptophan. DSIP is a nine-residue peptide with no confirmed site of synthesis, gene or receptor.² The two should not be treated as interchangeable.

Recent animal work

Interest has not disappeared entirely. In a 2021 study of rats with experimental stroke, animals given intranasal DSIP before the occlusion and for seven days afterwards had smaller infarcts than controls, though the difference was not statistically significant; their performance on the rotarod motor test recovered significantly better.¹¹

DSIP in human studies

A handful of small human studies were published, most in the early 1980s and several from the group that discovered the peptide.

The 1981 study was the first in humans, enrolled six volunteers and included the peptide's discoverers among its authors.¹² The 1992 study from Amsterdam kept 16 patients in a sleep laboratory for five consecutive nights, with an adaptation night, a baseline night and three nights after DSIP or a glucose placebo. It concluded that short-term DSIP was "not likely to be of major therapeutic benefit" in chronic insomnia.¹³ The 2009 Manchester study, which expected DSIP to deepen anaesthesia, found the opposite on its main EEG measures.¹⁴ A search of ClinicalTrials.gov in September 2026 found no registered trials of DSIP or emideltide.

Why the DSIP evidence is weak

• No biological anchor. Without a known gene, precursor or receptor, there is no confirmed physiological system for DSIP to act in.²
• Old, small studies. Most of the literature predates current standards for trial registration and reporting, and the human studies enrolled between 6 and 24 people.
• Inconsistent results. Reported effects vary with dose and timing in a U-shaped way⁷, analogues sometimes outperformed DSIP itself², and the better-controlled human studies found weak or paradoxical effects.^13,14
• Rapid breakdown. DSIP is quickly degraded in blood, which complicates any study that gives it systemically.¹⁰
• Concentrated authorship. Several of the early positive human reports came from the discovering group¹²; the independent studies that followed were less encouraging.^13,14

DSIP regulatory status

• United States. DSIP is not an FDA-approved drug; the Drugs@FDA database holds no product containing emideltide or DSIP.³ The FDA's page on bulk drug substances that may present significant safety risks (content current as of 22 April 2026) lists "Emideltide (DSIP)" among substances previously in category 2 whose nominations were withdrawn. It notes a potential immunogenicity risk and states that the FDA had not identified safety-related information for the proposed route, so it lacked sufficient information to know whether the drug would cause harm in humans.¹⁵
• Australia. DSIP and emideltide are not named in the Poisons Standard (the June 2026 instrument, SUSMP No. 48).⁴ For how research peptides are regulated more broadly, see are research peptides legal in Australia?

DSIP for laboratory research

DSIP is supplied for in vitro and animal-model research as a lyophilised (freeze-dried) powder. Its N-terminal tryptophan is susceptible to oxidation, and the 1987 blood-stability data show how quickly the peptide can be broken down enzymatically, so handling and storage conditions matter; see how to store lyophilised and reconstituted peptides and our primer on what research peptides are. The 1993 finding that a DSIP antiserum recognised an unrelated brain peptide is also a practical lesson: immunoassays can mistake other molecules for DSIP, so identity is better confirmed by mass spectrometry, as explained in how HPLC and mass spectrometry test peptide purity. Titan Peptides supplies DSIP 10 mg for laboratory research only.

Frequently asked questions

References