DSIP: why the ‘sleep peptide’ never became a sleep drug

DSIP was isolated in 1977 as a possible natural sleep signal, and a series of small Swiss studies in the 1980s reported that it normalised sleep in insomniacs. When two other groups ran placebo-controlled trials, they found effects too weak or inconsistent to matter clinically. Nearly 50 years on, nobody has found the gene that makes it or the receptor it acts on, and we found no human trial of it published after 1995.

What DSIP is

Delta sleep-inducing peptide is nine amino acids long: Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu [1]. Guido Schoenenberger and Marcel Monnier in Basel isolated it from rabbits, and a later review describes the source as blood draining from the brain [1,2]. When they infused the synthetic version into the brain ventricles of rabbits, under double-blind conditions, it increased the slow “delta” waves and spindles on the EEG that mark deep non-REM sleep. Eight related peptides tested alongside it didn’t [1].

The name needs a word of explanation. “Delta” refers to delta waves, the slow, high-amplitude brain rhythm that dominates the deepest stage of non-REM sleep, sometimes called slow-wave sleep. So the original claim was specific: this peptide didn’t just make animals drowsy, it pushed the brain into deep sleep. That specificity is what made it exciting, and it’s also what later human trials struggled to reproduce.

That was a promising start. The trouble began afterwards. A 2006 review by two Russian sleep researchers, Vladimir Kovalzon and Tatyana Strekalova, laid out the problems plainly: no gene for DSIP had been found, no precursor protein and no receptor, so it wasn’t clear DSIP was a natural signalling molecule at all [2]. Their own experiments in rabbits and rats had found that some synthetic DSIP analogues promoted slow-wave sleep but DSIP itself did not [2]. They suggested the antibodies used to “detect” DSIP in tissues might be picking up a different, related peptide.

There is also a stability problem. When Schoenenberger’s group incubated DSIP in human or rat blood, it broke down quickly, releasing free tryptophan; two modified analogues lasted longer [8]. That makes it hard to know how much intact DSIP reaches the brain after an injection.

The human evidence

Almost all the positive human data come from one researcher, Dieter Schneider-Helmert, working with Schoenenberger. All the studies used intravenous injections, usually 25 to 30 nanomoles per kg of body weight.

• 1981, six insomniacs. A single injection gave longer, less interrupted sleep, slightly more REM sleep and no daytime sedation. The effect started in the second hour; in the first hour there were hints of mild arousal [3].
• 1983, a summary of five studies. The team reported that DSIP not only improved sleep but also raised daytime alertness and, on psychological testing, improved “stress tolerance and coping ability” [12]. These broader claims were never tested by anyone else.
• 1986, 18 insomniacs. A week of injections (six doses of 30 nmol/kg) followed by a week off. Middle-aged patients’ sleep improved to normal values; older patients, whose sleep was worse at the start, took longer to normalise [4]. The abstract doesn’t describe a placebo group.
• 1987, 14 insomniacs. Seven nights of DSIP under placebo-controlled, double-blind conditions. The author reports that night-time sleep efficiency reached the level of normal sleepers and daytime alertness and performance improved [5].

Two other groups tested DSIP in insomnia, and neither was impressed. Jaime Monti’s team gave 25 nmol/kg or placebo over four nights in a double-blind crossover [6]. Total sleep time went up, driven by light stage 2 sleep rather than deep slow-wave sleep, and the apparent differences from placebo were already present at baseline. They concluded the improvement was “of little clinical significance” [6].

The University of Amsterdam trial by Bes and colleagues, summarised above, reached a similar verdict. The objective gains were weak, the authors thought they were partly explained by a chance shift in the placebo group, and the patients didn’t feel they slept better [7]. Their conclusion was that short-term DSIP “is not likely to be of major therapeutic benefit” [7].

Reading the two sets of results fairly

It would be easy to say the independent trials simply cancel out the Swiss ones, but the picture is messier. The Swiss team’s own 1983 summary describes five double-blind human studies, a delay of about an hour before sleepiness began, and effects lasting up to 20 hours; it also says slow injection was essential [12]. Those are claims the independent groups weren’t set up to test in the same way. Bes and colleagues injected in the afternoon, hours before bedtime [7], which may or may not matter for a peptide said to act after an hour and break down fast in blood [8].

What both sides share is tiny samples: six, 14, 16 and 18 people. With numbers that small, a couple of unusually good or bad nights in one group can swing the result, and that is exactly what the Amsterdam team suspected had happened [7]. The honest summary is that DSIP never showed a benefit that was both reproducible and large enough to matter, and nobody went back to settle it with a bigger trial.

It is worth noticing that the one sleep variable that should rise if DSIP did what its name promises, deep slow-wave sleep, didn’t change in Monti’s trial [6].

Hormones and withdrawal

DSIP was also studied as a brake on the stress-hormone system. In a randomised, double-blind crossover in 11 healthy men, a single 25 nmol/kg injection lowered blood ACTH for at least 3 hours, but cortisol, the hormone ACTH controls, didn’t change [9]. A German team then tested it more directly: infusions of 3 to 4 mg had no effect on ACTH or cortisol responses to a hormone challenge or to a midday meal [10]. So even the hormone story doesn’t hold together.

In the early 1980s, DSIP was given intravenously to 107 inpatients in alcohol or opiate withdrawal. The doctors reported that symptoms cleared or improved markedly in 97% of opiate and 87% of alcohol patients who could be assessed [11]. There was no control group, the assessment was clinical impression by staff who knew what patients were given, and withdrawal symptoms improve with time anyway. It is best read as a hypothesis that was never followed up with a proper trial.

What the animal work suggests

Apart from the original rabbit experiments [1], animal research on DSIP is scattered and inconsistent, which is exactly the point of the 2006 review [2]. Sleep effects in rabbits and rats have been hard to reproduce with DSIP itself, while several analogues seemed more active [2].

Safety and side effects reported

The Basel studies describe DSIP as well tolerated, with no daytime sedation [3]. In the withdrawal study, a few patients reported headaches [11]. That is the entire safety record, all from small, short studies of intravenous injections four decades ago. There is no safety data on repeated long-term use, on under-the-skin injection (the way it is used today by people buying it online), or on the quality of products sold as research chemicals.

Legal and regulatory status

DSIP has never been approved as a medicine by the UK’s MHRA, the European Medicines Agency or the US FDA, or, as far as we can find, by any other regulator. It is sold online only as a research chemical, which means it isn’t manufactured or tested to medicine standards and isn’t meant for human use.

What we still don’t know

• Whether DSIP exists naturally as a signalling molecule, since no gene, precursor or receptor has been identified.
• Why Schneider-Helmert’s results were so much stronger than the independent trials.
• How much intact DSIP reaches the brain, given how quickly it breaks down in blood.
• Whether it does anything at all when injected under the skin rather than into a vein.
• Anything about long-term safety.

Educational content only — not medical advice. Many peptides discussed on HPR are not approved for human use. Talk to a qualified clinician before making any decision about your health.

Free live webinar · Thursday 1 October · 7pm UK time

Join the next webinar: Introduction to Peptides

with Mitch O’Callaghan, ISSCA Peptide Therapist · Founder, Proper Protocols

Save my free placeHPR×Proper Protocols