DSIP (delta sleep-inducing peptide): sleep research, side effects and dosage

DSIP was isolated in 1977 from cerebral venous blood of rabbits whose brains had been electrically stimulated to induce sleep ^[1][2]. The Basel group led by Schoenenberger and Monnier determined its sequence as Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu and showed that the synthetic peptide, infused into rabbit brain ventricles, increased delta (slow-wave) EEG activity by about 35% compared with controls ^[2].

Despite its name, DSIP has never been firmly established as a natural sleep factor. A 2006 review noted that no gene, precursor protein or specific receptor for DSIP had been identified, and described the evidence that it regulates sleep as “extremely poorly documented” ^[3]. DSIP is now sold by research peptide vendors for sleep, a use not supported by modern trials.

What is DSIP peptide in the body? DSIP-like material has been measured in the plasma of several mammals and in human plasma, urine, cerebrospinal fluid and breast milk, where it was found bound to a larger carrier protein ^[24]. The Basel group also made a phosphorylated form, DSIP-P, which it described as many times more potent, and reported that both forms occur in human cerebrospinal fluid ^[24]. DSIP is also called emideltide, the name the US FDA uses ^[10].

How DSIP acts is not known. Researchers have proposed that it modulates the body’s opioid systems and influences stress hormones such as cortisol, describing a “programming” rather than a directly sedating action ^[7]. In the first human study, sophisticated EEG and behavioural analyses found no classic sedation even though participants slept more ^[4]. The 2006 review proposed that a still-unidentified DSIP-like peptide may explain the DSIP-like activity measured in tissues ^[3].

DSIP is broken down quickly in blood: when incubated in human or rat blood it released tryptophan-like fragments, and the authors concluded that the rapid disappearance of injected DSIP was due to degradation ^[8].

The delta sleep-inducing peptide mechanism of action, as far as it was studied, depends on DSIP reaching the brain. In dogs given 100 µg/kg intravenously, intact DSIP appeared in cerebrospinal fluid, at higher ratios to blood than insulin, and the free (unbound) peptide appeared to be the form that crossed ^[18]. The Basel group reported that its effects followed a bell-shaped (parabolic) dose-response curve, with too much or too little producing less effect, whether given into the brain, intravenously or under the skin ^[24].

Human hormone studies are mixed. In 11 healthy men given 25 nmol/kg intravenously, ACTH-like activity in plasma fell for at least 3 hours compared with saline, but cortisol followed its normal daily pattern ^[20]. A later German study found that DSIP infusions of 3 to 4 mg did not change ACTH or cortisol responses to a stimulating hormone or to a midday meal, and concluded the data did not support an inhibitory role for DSIP on these hormones in people ^[21]. The idea that DSIP lowers cortisol is therefore not established.

Measurements of the body’s own DSIP also cut against a simple sleep-hormone model. In 12 volunteers sampled every 30 minutes for 24 hours, plasma DSIP-like immunoreactivity peaked around 15:00 and was lowest around 01:00, tracked body temperature, and was lower during REM and slow-wave sleep than during wakefulness ^[22]. In people with sleep disorders, morning plasma DSIP did not differ between sleep apnoea, narcolepsy and healthy controls ^[28]. In Cushing syndrome, people with more delta sleep had lower, not higher, morning DSIP, which the authors said argued against DSIP causing delta sleep ^[29].

In rabbits, the synthetic peptide enhanced delta and spindle EEG activity ^[2]. Beyond sleep, researchers reported effects on stress responses and interactions with opioids and other drugs, mostly in animals ^[7]. These remain hypotheses rather than established effects, and the 2006 review concluded that DSIP’s biological role is still obscure ^[3].

The discoverers themselves described DSIP as having many actions beyond sleep, including interactions with stress, with drugs such as morphine, amphetamine and barbiturates, and with daily rhythms of locomotor activity, plasma proteins and cortisol, mostly from animal work ^[24]. A 1986 review listed insomnia, pain and withdrawal as the therapeutic uses then being explored ^[30].

The sections below cover the uses people search for, starting with sleep. Almost all of the human data are from the 1980s, and much of it comes from the Basel group that discovered the peptide.

The DSIP sleep research is old and mixed, and the split falls largely along who ran the study. Studies from the Basel group and its collaborators were positive. After the first volunteer study ^[4], an intravenous dose of 25 nmol/kg in six middle-aged people with chronic insomnia gave longer and less interrupted sleep, slightly more REM sleep and no daytime sedation ^[11]. A 1983 summary of five double-blind studies reported good tolerability, a sleep-onset delay of about an hour, effects lasting up to 20 hours, and “complete normalization” of disturbed sleep after four consecutive injections in insomniacs ^[12]. A 1987 placebo-controlled study in 14 chronic insomniacs, given DSIP on 7 successive nights, reported better night sleep and daytime alertness, with benefits persisting into the first placebo night afterwards ^[13].

Independent studies were less positive. A 1987 crossover study in insomniacs found that sleep changes under DSIP did not differ significantly from placebo or baseline ^[6], and a 1992 double-blind study of 16 insomniacs found only weak effects and no improvement in how people rated their sleep ^[5]. An open study in seven patients with severe insomnia, given a series of 10 injections, reported normalised sleep in six at 3 to 7 months, but had no control group ^[14].

So does DSIP work for sleep? The controlled trials run outside the discovering group did not find a clinically useful effect on insomnia ^[5][6], and no trial has been published since the early 1990s. Claims that DSIP peptide benefits sleep rest mainly on small studies from one group.

Another group gave DSIP intravenously to people withdrawing from alcohol or opiates, on the theory that it acts on opioid systems. In 1983, 25 nmol/kg as the only treatment in 67 patients was followed by rapid relief of physical withdrawal signs in 48 of the 49 who could be evaluated, with anxiety settling more slowly ^[16]. A 1984 report of 107 inpatients described similar results and good tolerance apart from headaches in a few patients ^[17]. Neither study had a placebo group; outcomes were clinical judgements by staff.

Other reports were single cases or uncontrolled. In one man with narcolepsy, repeated DSIP injections reduced sleep attacks and increased daytime alertness ^[15]. In one woman with delayed sleep phase insomnia and benzodiazepine dependence, a week of inpatient DSIP treatment shifted her main sleep period 5 hours earlier and allowed her to stop flunitrazepam ^[27]. The chronic pain pilot is described under Human studies ^[7].

The first human study, in six healthy volunteers, gave DSIP by slow intravenous infusion in a double-blind crossover design; total sleep in the following 130 minutes rose by a median of 59% compared with placebo, with no side effects observed ^[4].

Later independent double-blind trials were less encouraging. In a study of 16 people with chronic insomnia, DSIP produced slightly higher sleep efficiency and shorter time to fall asleep, but the effects were weak, partly explained by changes in the placebo group, and subjective sleep quality did not improve; the authors concluded DSIP was unlikely to be of major benefit ^[5]. A crossover study in insomniacs found that sleep improvements under DSIP were not significantly different from placebo or baseline, and judged them of little clinical significance ^[6].

A 1984 pilot study in seven patients with chronic pain conditions reported reduced pain in six after intravenous DSIP, but had no control group ^[7]. In a 2009 randomised study in 24 women having anaesthesia, DSIP unexpectedly lightened rather than deepened anaesthesia and increased heart rate ^[9].

DSIP clinical trials, in short, were small, mostly from the 1980s, all used intravenous DSIP, and were often uncontrolled. We found no trials registered in recent years. The two FDA-listed concerns for compounded DSIP, immunogenicity and difficulty characterising the peptide and its impurities, have not been studied in people ^[10].

In the positive Basel studies, sleep promotion started in the second hour after injection, with a slight arousing effect in the first hour ^[11], and sleep onset was delayed by about an hour ^[12]. Effects were reported to last up to 20 hours after a dose ^[12] and to carry over into the night after a course ended ^[13]. The independent trials found little effect at any time point ^[5][6].

Does DSIP make you sleepy? The first human study found more sleep but no classic sedation on detailed testing ^[4], and the insomnia study reported no daytime sedation ^[11]. The Basel group described DSIP as raising alertness and performance while awake ^[12]. So it was not described as a sedative in the way sleeping pills are.

Every human study we found gave DSIP intravenously, and the Basel group said slow injection was essential ^[12]. In animals it was also given into the brain ventricles and under the skin ^[24]. We found no published human study of a DSIP nasal spray or of subcutaneous DSIP, so vendor nasal and subcutaneous products are untested routes.

Oral DSIP is unlikely to be absorbed intact. In a model of the human gut lining, DSIP applied to the gut side was rapidly broken down, with only about 8% left after 2 hours, and none crossed to the other side even when the main enzymes were blocked ^[23]. Unweaned rats took up DSIP from the gut ^[24], but that does not show absorption in adults.

The insomnia and volunteer studies used 25 nmol/kg by intravenous infusion ^[4][5][6]. The anaesthesia study gave intravenous bolus doses of 25, 50 or 100 nmol/kg ^[9]. In rabbits, 6 nmol/kg was infused into the brain ventricles ^[2]. All human doses were given intravenously under supervision; these are study doses, not recommendations.

The same 25 nmol/kg intravenous dose was used in the Basel insomnia work ^[11] and the withdrawal studies ^[16]. Courses ranged from a single injection to 7 successive nights ^[13] or a series of 10 injections ^[14]. The hormone study used total intravenous doses of 3 to 4 mg ^[21]. With a molecular weight of 848.8 g/mol, 25 nmol/kg works out at about 21 µg per kg of body weight. That is a unit conversion of a study dose given intravenously in hospital, not a DSIP dosage for sleep by any other route.

Vendor DSIP dosing charts and “DSIP protocols” for nasal or subcutaneous use are not drawn from these studies. For vial arithmetic, such as DSIP 5mg reconstitution, the concentration is the peptide amount divided by the liquid volume; our peptide calculator shows the calculation without implying a dose.

The small early studies reported DSIP as well tolerated ^[4], but the total number of people studied is small and follow-up was short. In the anaesthesia study DSIP increased heart rate and reduced heart rate variability ^[9]. The US FDA’s compounding page (as of April 2026) lists emideltide (DSIP) among substances previously placed in its category of bulk substances that may present significant safety risks, citing possible immunogenicity and difficulties with peptide-related impurities and characterisation; the nomination was later withdrawn ^[10].

Other DSIP side effects reported were headaches in a few patients in the withdrawal studies ^[17], and the 1983 withdrawal study reported no major side effects ^[16]. Is DSIP safe? Nobody can say from these data: the published studies were small, gave DSIP intravenously, mostly once or over a few days, and were run in the 1980s. There are no data on repeated nasal or subcutaneous use of today’s research products, and Health Canada named DSIP among unauthorised peptide products it warned the public about in 2026 ^[25].

DSIP is cleared very quickly. After intravenous injection, its half-life in plasma was 4.0 minutes in dogs, 2.9 minutes in a monkey and 2.0 minutes in rats ^[19]. Brain tissue removed the first amino acid, tryptophan, with a half-life of about 15 minutes, while DSIP-like material bound to a carrier protein in plasma was protected from breakdown ^[24]. We found no human half-life figure for injected DSIP in the abstracts we read. The long effects claimed in the Basel studies therefore cannot be explained by DSIP staying in the blood.

DSIP and epitalon are often sold as a sleep or “longevity” pairing. We found no study that tested them together, so the combination has no evidence base.

DSIP is not approved as a medicine in the UK, US, Canada, Australia or New Zealand, and we found no approval elsewhere. See our legal status overview and the UK page.

Is DSIP FDA approved? No. The FDA moved emideltide in 2026 to a group of compounding nominations that were withdrawn, which is not an authorisation ^[10]. Its Pharmacy Compounding Advisory Committee considered emideltide on 24 July 2026; as of FDA’s August 2026 page, no outcome or FDA decision had been posted ^[26]. In Canada, Health Canada’s April 2026 advisory on unauthorised peptides named DSIP ^[25]; see the Canadian legal status page and the US page.

Research DSIP is usually sold as a freeze-dried powder. In blood, the phosphorylated analogue broke down more slowly than DSIP and formed complexes with other material, a reminder that small chemical changes alter how the peptide behaves ^[8]. Because DSIP degrades quickly in blood ^[8], its stability in solution should not be assumed; see our peptide storage guide and the bacteriostatic water guide.

You can compare prices, learn how to read a COA, and see our list of third-party tested suppliers. Other neuropeptides sold as research compounds include Selank and Semax.

Country pages list DSIP in the US and DSIP in the UK. A listing is not a statement that a product is legal or suitable for use. Because no quality standard exists for research DSIP, independent testing matters; see our guide to third-party peptide testing.