DSIP

Origin and structure

From the 1960s the Basel researcher Marcel Monnier searched for a substance in the blood that could carry sleep from one animal to another. His group stimulated the thalamus of rabbits electrically to induce sleep and looked in the venous blood leaving their brains for a factor that, given to other rabbits, increased the slow, high-amplitude delta waves of deep sleep in the EEG. In 1977 Guido Schoenenberger and Monnier purified the active factor, read its sequence and made it synthetically. Of nine synthetic peptides infused into the brains of rabbits under double-blind conditions (58 animals, controls included), only this one clearly increased delta and spindle activity, and they named it delta sleep-inducing peptide.

DSIP is a chain of nine amino acids: Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu (C35H48N10O15, 848.8 g/mol, CAS 62568-57-4). Its international nonproprietary name is emideltide, the name the FDA uses. Research-grade DSIP is sold either as the free peptide or as an acetate salt. The FDA treats these as two different substances and in 2026 judged both poorly characterised, partly because names and submitted documents mixed the two up and data on impurities, aggregates and bacterial endotoxins were missing.

How it is thought to work

Nobody knows for certain. By 2006, when a review called DSIP 'a still unresolved riddle', no gene, precursor protein or receptor for it had been found, and its sequence resembles no known peptide family; the FDA's 2026 review still describes its mechanism as unknown. Antibodies against DSIP do pick up DSIP-like material in the brain and in blood, and the review's authors suspect that at least part of that signal, and of DSIP's activity, comes from a related peptide that has not yet been identified.

In animals the sleep effect was inconsistent. Several groups found more deep (non-REM) sleep in rabbits, rats and cats after injecting DSIP into the brain or a vein, but another group found no rise in delta activity in rats, and the authors of the 2006 review saw clear deep-sleep effects in their own rabbit and rat studies only with certain synthetic analogues, not with DSIP itself. Why the results differ has never been explained. Part of the effect seems to run through the body's own opioid system: in rats the opioid blocker naloxone prevented it, although DSIP does not bind opioid receptors itself and appears instead to release enkephalins in the brain. Rodent studies have also reported lower body temperature, damped stress responses, pain relief and protection against seizures.

DSIP crosses from the blood into the brain in animal experiments but survives only minutes in blood: in human serum in the test tube its half-life was 5 to 10 minutes, as enzymes break it down from the front end. The natural DSIP-like signal in human plasma follows a daily rhythm that tracks body temperature, highest around 3 p.m. and lowest around 1 a.m., and it is lower during REM and deep sleep than during waking. The researchers who measured this concluded that higher levels may go with less, not more, deep and REM sleep, which does not fit the idea of a simple sleep hormone.

Is DSIP approved?

Not in the EU, the Netherlands or the US. No medicine containing DSIP has been authorised by the EMA or the FDA, and the FDA found no monograph for it in the European, US, Japanese or International Pharmacopoeia. In the Netherlands it is not a registered medicine, so it may not be sold or promoted for use in people. Research-grade DSIP is laboratory material, and that is the basis on which our shop sells it.

The one country where we found DSIP registered as a medicine is Russia. Deltaran is a freeze-dried powder of 0.3 mg DSIP with 3 mg glycine per ampoule that is dissolved in water and given into the nose. Its Russian label lists it, as part of combined treatment, for stress-related conditions and for alcohol withdrawal and craving, and it is sold without a prescription. We could not find the clinical studies behind that registration in international databases, and the FDA's 2026 review reported that it had found no registration in the countries it searched.

In the United States the question has been pharmacy compounding. The FDA had placed emideltide in Category 2 of its compounding list, the group of substances that may pose significant safety risks, citing possible immune reactions, peptide impurities and a lack of safety information for the proposed route; the nominations were later withdrawn. The FDA reviewed it anyway. Its staff evaluation, dated 11 May 2026, looked at injection under the skin for opioid withdrawal, chronic insomnia and narcolepsy and recommended against adding it: the substance was poorly characterised, there were no human data at all for that route, the intravenous studies were inconclusive, and its action through the opioid system could carry a risk of addiction that has never been studied. On 24 July 2026 the FDA's Pharmacy Compounding Advisory Committee agreed, voting 7 to 6 against with one abstention; it was the only one of the seven peptides reviewed that week that the committee turned down. No final FDA decision had been published by October 2026, and DSIP is not on the list of substances US pharmacies may compound from.

In sport DSIP is not named on the 2026 WADA Prohibited List. Category S0 bans at all times any substance without a current approval by a government health authority for human therapeutic use, which would normally cover a research peptide, but the Russian registration of Deltaran makes it uncertain whether S0 applies, and we found no WADA ruling on DSIP. Athletes in the Netherlands can put the question to the Dopingautoriteit, which applies the WADA list.

What the research shows

The human research is small and old, and much of it comes from the Swiss group that discovered DSIP. By the FDA's 2026 count, 209 people received it intravenously, at 25 to 150 nmol/kg for 1 to 15 days; no published study has given it under the skin, and ClinicalTrials.gov lists no DSIP trial. The FDA concluded that the insomnia studies were 'inconclusive and at best preliminary', with conflicting findings between research groups, small samples and weak control groups, and that the narcolepsy and withdrawal data were too limited to judge. The main studies:

• First human study (Schneider-Helmert et al. 1981): 6 healthy volunteers, double-blind crossover against placebo. After a slow morning infusion, total sleep in the next 130 minutes was 59% higher (median) than after placebo, without signs of sedation; the following night, 13 to 22 hours later, they fell asleep faster and slept more efficiently.
• Insomnia, single dose (Schneider-Helmert and Schoenenberger 1981): 6 middle-aged insomniacs, double-blind crossover against placebo. Sleep was longer and less interrupted from the second hour after the injection, after a slightly arousing first hour. The FDA noted that time to fall asleep and wake-up time did not change and that results were reported mainly as p values.
• Insomnia, one week (Schneider-Helmert 1986 and 1987): 18 and 14 insomniacs, injected each evening for about a week. In the 1987 study average time to fall asleep fell from 58 to 28 minutes and total sleep rose from 313 to 385 minutes, but there was no placebo group: patients were compared with their own baseline and with healthy volunteers studied separately.
• Insomnia, placebo-controlled by other groups: Monti et al. (1987), 6 severe insomniacs in a double-blind crossover over four nights, found no differences from placebo that were not already present before treatment and called the improvement of little clinical significance. Bes et al. (1992, University of Amsterdam), 16 insomniacs split between DSIP and placebo for three nights, found weak effects that may partly reflect a chance change in the placebo group and no change in subjective sleep quality, and concluded that short-term DSIP is unlikely to be of major benefit.
• Narcolepsy: a single case (Schneider-Helmert 1984). In a 35-year-old man, sleep attacks fell from six in the baseline week to three in each week of injections.
• Withdrawal (Dick et al. 1984, Geneva): open study without a control group in 107 inpatients withdrawing from alcohol (47) or opiates (60). Of those who could be evaluated, symptoms disappeared or improved markedly in 97% of the opiate and 87% of the alcohol patients, but in the opiate group anxiety and insomnia often returned within one to three days of stopping. A later open study in 7 opioid-dependent patients (Backmund et al. 1998) saw withdrawal scores fall after the first injection, but symptoms returned and only 2 completed the course.
• Pain (Larbig et al. 1984): pilot study without a control group in 7 patients with migraine attacks, other headaches, tinnitus or psychogenic pain; compared with their own baseline period, pain levels fell significantly in 6 of 7.
• Stress hormones (Bjartell et al. 1989): in 11 healthy men in a randomised double-blind crossover, a single dose lowered the stress hormone ACTH in the blood, measured by immunoassay, for at least 3 hours compared with saline, while cortisol did not change.
• Animals: later work, much of it Russian, has studied stress, seizures and ageing in rodents. In one lifelong study, female mice given Deltaran five days a month developed fewer mammary cancers and leukaemias, but the FDA noted that the glycine in the product may explain the effect and that the study was not designed to test cancer risk.

Side effects and safety

In the insomnia studies intravenous DSIP was reported as well tolerated, without daytime drowsiness or a hangover. The largest safety record comes from the Geneva withdrawal study: according to the FDA's reading of it, 9 of 107 patients had brief minor effects such as sweating, headache, nausea or dizziness, and 3 had reactions counted as serious, two of them a drop in blood pressure at the first injection, which in one case got worse after a second injection. Withdrawal itself causes similar symptoms, which makes these reports hard to interpret.

Beyond that, little is known. There are no repeat-dose toxicity, genotoxicity or reproductive toxicity studies and no proper carcinogenicity study, no human data for injection under the skin, and no study of whether DSIP can be addictive, a question the FDA raised because part of its effect runs through the opioid system. The FDA also flagged a risk of immune reactions to injected peptide and its impurities. Its adverse-event database held no reports for DSIP up to March 2024. The Russian Deltaran label lists allergic reactions and irritation of the nasal lining, rules out pregnancy, breastfeeding and under-18s, and advises against it in people with a severely slowed heart rate.

Frequently asked questions

Does DSIP help you sleep?

That has not been shown. The early studies by the Swiss group that discovered DSIP reported longer, less interrupted sleep, but they were tiny and the larger ones had no placebo group. The two placebo-controlled insomnia trials run by other groups, with 6 and 16 patients, found no meaningful benefit, and the FDA called the evidence inconclusive in 2026. All of these studies gave DSIP into a vein; it has never been tested as an injection under the skin.

Is DSIP legal in the Netherlands?

DSIP is not a registered medicine in the Netherlands or elsewhere in the EU, so it may not be sold or promoted for use in people. It may be bought and sold as research material, which is the basis our shop works on. The Russian nasal medicine Deltaran is not authorised in the Netherlands.

Is DSIP banned in sport?

It is not named on the WADA Prohibited List. Unnamed substances without any government approval for human use fall under category S0 and are banned at all times, but DSIP is registered as a medicine in Russia, so it is not clear that S0 applies, and we found no WADA ruling on it. Athletes in the Netherlands can check with the Dopingautoriteit.

Is DSIP FDA-approved?

No. The only US question has been pharmacy compounding: FDA staff advised against adding DSIP (emideltide) to the list of substances pharmacies may compound from, and on 24 July 2026 the FDA's advisory committee agreed, voting 7 to 6 against with one abstention. It was the only one of seven peptides the committee rejected that week.

Does the human body make DSIP?

Possibly, but it has never been proven. Antibodies against DSIP detect DSIP-like material in human blood, with a daily rhythm that peaks in the afternoon, but no gene or precursor protein for DSIP has been found, and part of that signal may come from other peptides that the antibodies also recognise.

How long does DSIP last in the body?

Only minutes in blood: in human serum in the test tube its half-life was 5 to 10 minutes, and in dogs, rats and a monkey it was 2 to 6 minutes. Yet the first human study reported effects on sleep the following night, 13 to 22 hours after the infusion, which the short half-life does not explain. There are no human pharmacokinetic data for injection under the skin, and we found none for nasal use.

Storage and handling

Store according to product-label instructions and batch handling notes. Lyophilized research peptides are commonly kept dry, protected from light, and handled with clean laboratory technique.

References

Checked against these sources on 28 October 2026.

Related peptides