DSIP: A Peptide Named After an Effect Nobody Could Pin to It

A Basel group dialysed cerebral venous blood from rabbits while stimulating a thalamic region that induces sleep. From that dialysate they isolated a nine-residue peptide, tested it by intraventricular infusion, saw slow-wave EEG enhancement, and named it after what they had seen [1][2].

That was 1977. The name has outlived the evidence. A 2006 review in Journal of Neurochemistry put it bluntly: the link to sleep has never been further characterised, and the hypothesis remains poorly documented [6].

No gene, no isolated protein, no receptor. Nearly fifty years on from the sequence, DSIP remains a defined molecule attached to an undefined biology. That makes it an unusual reference standard and an honest one to describe. Kimera supplies DSIP as 5 mg of lyophilised powder.

Everything below reports findings from animals, cells and small human studies. This material is for research use only, not for human or veterinary use.

What DSIP is

The molecule is a linear nonapeptide with no unusual residues and no modifications. Its sequence is the only thing that distinguishes it.

Identity and physical data

The sequence has no family

Two aspects stand out. Three of nine residues are glycine, which makes the backbone unusually flexible. Two carry carboxylates, aspartate and the C-terminal glutamate, so the peptide is acidic at physiological pH.

The 2006 review notes that the structure resembles no other known peptide family [6]. That is unusual for a putative signalling molecule, and it is part of why the biology has stayed hard to place.

The tryptophan matters practically

Position one is tryptophan, the only residue here with a strong ultraviolet chromophore. It gives the peptide usable absorbance near 280 nanometres, which makes concentration measurement straightforward. It also oxidises, which the handling section returns to.

How it was found

The isolation experiment

Rabbits received hypnogenic electrical stimulation of the intralaminar thalamic area while their cerebral venous blood was dialysed extracorporeally [2]. The peptide came out of that dialysate. Amino acid analysis and sequencing gave the nine residues [2], and the group published the characterisation in 1977 [1][3].

The synthetic comparison

The team then synthesised the nonapeptide along with five possible metabolic fragments, two analogues with residues exchanged, and a related tripeptide [1][2]. All nine went into rabbits by intraventricular infusion under double-blind conditions, with EEG analysed by Fourier transform [1]. A separate paper compared original and synthetic material directly [4], and another examined transport across the blood-brain barrier in rabbit [5].

What that design did well

Double-blind infusion, a fragment series and a synthetic-versus-natural comparison make this a careful piece of 1970s peptide chemistry. The isolation work is not the weak part of this story. What followed it is.

The name is a hypothesis

What the review found

The 2006 assessment is the reference point for anyone reading this literature [6]. It records three things: the sleep link never gained further characterisation, the natural occurrence and biological activity remain obscure, and the sleep-factor hypothesis stands weak on the published evidence (PubMed).

A 1994 paper had already asked the question in its title, weighing whether the compound is a sleep peptide or an unrecognised hypothalamic hormone [7].

Sleep biochemistry moved on without it

A 2003 review of biochemical regulation of non-rapid-eye-movement sleep surveys the molecules with established roles in that process [17]. Adenosine, cytokines and prostaglandins carry that literature. The peptide named for delta sleep sits at its margin rather than its centre.

Three things that were never found

No gene

Nobody has isolated a DSIP gene [6]. For an endogenous peptide, that absence is the central problem: without a gene there is no precursor, no expression pattern and no knockout.

No receptor

Nobody has identified a receptor either [6]. Binding studies exist in the older literature, and none produced a cloned target. A peptide with no receptor and no gene fits into no pathway. Mechanism sections in this literature therefore describe downstream observations rather than a route.

No confirmed endogenous form

A group raised monoclonal antibodies against the peptide in 1992 and applied them immunohistochemically [12]. Immunoreactivity is not identity, since an antibody reports an epitope rather than a molecule. The 2006 review proposes that a DSIP-like peptide, rather than this exact sequence, may be what those studies detected [6].

What a 1977 result can and cannot carry

The isolation work deserves reading on its own terms, and so do its limits.

The endpoint was EEG, not sleep

Delta EEG enhancement and sleep are related, and they are not the same measurement. The original papers measured spectral power after intraventricular infusion in rabbits [1][2]. Behavioural sleep architecture, latency and duration are separate endpoints, and the founding work did not establish them (PubMed).

Intraventricular delivery bypasses everything

Infusing a peptide directly into the ventricles answers what it does once inside the brain. It says nothing about whether the molecule reaches the brain from the blood under normal conditions, or whether the concentrations used resemble anything physiological. A route that bypasses the barrier also bypasses the question of endogenous relevance.

Fragments were the control

Testing five metabolic fragments and two exchanged-residue analogues alongside the parent was good practice. A modern replication should copy that part of the design [2].

Where activity has been reported instead

The literature that exists scatters across effects unrelated to the name. Each result belongs to its model.

Seizure models

A 1998 study reported effects on NMDA-induced convulsive activity in rats [9]. A 2008 review grouped the peptide with neuropeptide Y as a candidate endogenous anticonvulsant [10]. That framing has more supporting work behind it than the sleep framing does.

Stress responses

A 2012 study examined c-Fos expression in the hypothalamic paraventricular nucleus of rats differing in resistance to emotional stress. Dizocilpine and cycloheximide both prevented the peptide’s inhibition of that expression [11]. The design implicates NMDA receptor signalling and protein synthesis without identifying a direct target.

Ischaemia models

A 2021 study reported recovery of motor function in rats after focal stroke [14]. Its companion paper tested a DSIP-like peptide and reported reduced brain infarction in mice, plus reduced myocardial infarction in rats given the peptide during reperfusion [15].

High altitude and memory

A 2018 study used a phosphorylated form and reported restored spatial memory and p-CREB expression, attributing the effect to improved sleep architecture at altitude [13].

The analogue problem

Read the last two sections again and a pattern appears. Several of the stronger recent results used modified peptides rather than the parent sequence.

Which molecule produced the result

The 2018 study used phosphorylated DSIP [13]. The 2021 infarction study used a DSIP-like peptide designated KND [15]. A 2024 study used a fusion construct designed to cross the blood-brain barrier, tested in a PCPA-induced insomnia mouse model [16].

Why that matters

Any modification that improves stability or delivery also changes the molecule under test. Results from a phosphorylated or fused analogue describe that analogue. Read the methods for the exact species before attributing anything to the parent nonapeptide. Treat analogue results as evidence about the scaffold.

The one human study

A 1993 study administered the peptide intravenously to healthy women and measured growth hormone and prolactin [8]. Basal levels sat in the normal range, the infusion changed nothing, and the circadian rhythm of both hormones held steady [8].

That is a negative result in a small sample, and it is worth knowing for two reasons. It shows the compound reached human study in the era when interest was highest. And it shows what that work returned.

Designing an experiment with DSIP

Decide which claim you are testing

Sleep, seizure threshold and ischaemic protection are three separate literatures with three separate endpoints. Pick one. A study that measures everything at once and finds one positive has run a fishing expedition, and this compound already has enough of those.

Match the route to the question

Intraventricular delivery reproduces the founding condition. Peripheral delivery tests something the founding work never tested. Say which one applies, because the two answer different questions and the older papers used the first [1][5].

Set the concentration from the literature you are testing

The founding work infused nanomole quantities per kilogram directly into rabbit ventricles [1][2]. Modern cell work runs micromolar. Those two numbers describe different experiments, and neither justifies the other.

Pick the range from the specific paper being followed, and say in the write-up which paper set it. Concentration inherited from a review rather than from a primary source is how ranges drift across a literature, and this one has had fifty years to drift.

Include a scrambled control

Nine residues in a shuffled order test whether an effect depends on sequence or on composition. For a peptide with no receptor, that control carries more weight than usual. Without a binding assay it is the only cheap route to showing sequence specificity.

Verify the peptide after the experiment

Run an aliquot of the working solution at the end of the run, not only at the start. Tryptophan oxidation happens on the bench timescale, and a post-experiment check tells you whether the compound survived the experiment it was meant to drive.

Reading the Russian-language branch

Several entries in this record are Russian-language papers with English abstracts, including the 1994 framing paper and the seizure work [7][9]. Methods detail stays behind the abstract, so treat those as pointers rather than as papers you have examined.

The pattern repeats across this catalogue’s older compounds. Where an English-language source covers the same ground, cite that one and keep the translation-limited entry as context.

Physicochemical properties and handling

Tryptophan is the residue to protect

Indole side chains oxidise on exposure to light, air and peroxides, and oxidation products differ in mass from the parent. A stored solution that has yellowed has almost certainly oxidised, and the resulting species are no longer the compound the experiment intends.

Amber vials, minimal headspace and prompt use handle it. Adding a reducing agent introduces its own confound, so prefer short storage over chemical protection.

Reconstitution discipline

Lyophilised peptide forms a loose cake that lifts easily. Add diluent slowly against the vial wall, let it wet the cake, then swirl. Vortexing shears peptide and foams the solution, and foam carries material out of the working volume.

Record the exact diluent volume rather than the nominal one. A 5 mg vial reconstituted to a slightly different volume shifts every concentration downstream by the same fraction.

Expect fast clearance

Short linear peptides with no protecting modifications clear quickly in plasma. The blood-brain barrier transport study from 1977 is the closest thing to a distribution dataset in this literature [5], and it predates modern pharmacokinetic practice. Design in vivo work around a short exposure window rather than assuming sustained levels.

Analytical characterization

Mass and ultraviolet detection

Electrospray gives a protonated ion at m/z 849.8 for the C35H48N10O15 formula. The single tryptophan provides absorbance near 280 nanometres, which allows direct concentration determination by ultraviolet spectrophotometry rather than by mass alone.

Confirming the sequence

Nine residues permute into an enormous number of orders at one mass. Three glycines make several of those rearrangements isobaric. Tandem mass spectrometry resolves the order through the fragment series. Composition analysis and mass together still leave sequence unproven, which matters for a compound whose identity is only its sequence. Batch documentation for catalogue material sits in the certificate of analysis database.

Two checks that cost nothing

Read the ultraviolet absorbance of the working solution and compare it against the expected value for a single tryptophan. A reading well below expectation means the vial holds less peptide than the label implies, or that oxidation has already run.

Then inject the same solution and look at the peak shape. A shouldered or split main peak on a fresh preparation points at oxidation products, which elute close to the parent and are easy to miss on a fast gradient.

What a peptide certificate should carry

Purity by reversed-phase HPLC with the gradient stated, net peptide content, water content and counterion. Lyophilised peptides carry acetate or trifluoroacetate from purification, and that mass is not peptide.

Storage in practice

Aliquot on the day of reconstitution. Each freeze-thaw cycle costs peptide, and a vial opened repeatedly picks up moisture that no label records.

Write the reconstitution date on the vial rather than trusting a notebook. Two weeks later, a solution with no date on it is a solution nobody can defend in a methods section. Keep the aliquots small enough that one thaw serves one experiment.

A search trap worth knowing

The acronym is not unique. A PubMed search returns papers using DSIP for deuterium stable isotope probing in microbiology, and for a sanitation infrastructure project in Bangladesh.

Why the collision matters here

An acronym search that returns unrelated fields is an inconvenience for most compounds. For this one it distorts the apparent size of the literature, and a reader skimming counts rather than titles will overestimate how much work exists on the peptide.

That mistake compounds the one the name already invites. A large-looking record and a confident name together suggest a well-established molecule, and neither impression survives contact with the primary papers.

How to filter

Search the full name alongside the acronym, and read the titles before harvesting identifiers. Four hundred and sixty results shrink to a manageable set once the unrelated fields drop out. This is the same class of error that pulled materials-science papers into a racetam search earlier in this programme.

Where DSIP sits among peptide research compounds

All four are short synthetic peptides from mid-century or later Russian and European research programmes. Further reading sits in the peptides research library.

What this literature does not establish

That the compound induces sleep

The founding observation was delta EEG enhancement after intraventricular infusion in rabbits [1]. Fifty years later that remains the strongest sleep-related evidence for the parent peptide, and the 2006 review calls the sleep hypothesis weak [6].

That it occurs endogenously in this form

Antibody studies report immunoreactivity [12]. Isolation of a gene or a precursor has not [6].

A mechanism of any kind

Reported effects on seizures, stress markers and ischaemia have no identified receptor behind them [9][11][14]. The stress study reached NMDA signalling by pharmacological inference rather than by binding [11].

Which effects would replicate today

Nobody has repeated the founding EEG experiment with modern polysomnography and modern peptide quality control. The seizure and ischaemia work sits in small studies from a few groups [9][14][15].

A replication programme would start with the original route and endpoint, then add behavioural sleep measures the 1977 apparatus could not deliver. Until somebody runs it, the honest summary of DSIP is a well-characterised molecule with an unsettled biology, which is exactly how a reference standard should be described.

That analogue results transfer

Phosphorylated, fused and DSIP-like peptides carry several of the recent positive findings [13][15][16]. Each of those is a different molecule, and none of them establishes anything about the nonapeptide on its own.

Frequently asked questions

What is the sequence?

Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu, or WAGGDASGE. Nine residues, 848.8 g/mol.

Does DSIP actually induce sleep?

The 1977 rabbit work reported delta EEG enhancement after intraventricular infusion [1]. Later work did not build that into a characterised mechanism, and a 2006 review describes the sleep hypothesis as poorly documented [6].

Is there a receptor?

Nobody has found one [6]. Nor has a gene or an isolated endogenous protein.

Why do recent papers report positive results?

Several used modified peptides rather than the parent: a phosphorylated form, a DSIP-like sequence and a fusion construct [13][15][16]. Check which molecule each study used.

What degrades in storage?

Tryptophan, by oxidation. Store dark and cold, and discard solutions that have changed colour.

Was the original work poorly done?

No. The 1977 isolation used double-blind infusion, a fragment series and a synthetic-versus-natural comparison [1][2][4]. The gap is in what followed, not in what they did.

How is concentration best measured?

Ultraviolet absorbance near 280 nanometres, using the tryptophan. Most short peptides lack a chromophore and cannot be measured this way.

References

1. Schoenenberger GA, Monnier M. Characterization of a delta-electroencephalogram (-sleep)-inducing peptide. Proc Natl Acad Sci U S A. 1977;74(3):1282-6. PubMed DOI
2. Schoenenberger GA, Maier PF, Tobler HJ, et al. The delta EEG (sleep)-inducing peptide (DSIP). XI. Amino-acid analysis, sequence, synthesis and activity of the nonapeptide. Pflugers Arch. 1978;376(2):119-29. PubMed DOI
3. Schoenenberger GA, Maier PF, Tobler JH, et al. A naturally occurring delta-EEG enhancing nonapeptide in rabbits. X. Final isolation, characterization and activity test. Pflugers Arch. 1977;369(2):99-109. PubMed DOI
4. Monnier M, Dudler L, Gachter R, et al. The delta sleep inducing peptide (DSIP). Comparative properties of the original and synthetic nonapeptide. Experientia. 1977;33(4):548-52. PubMed DOI
5. Monnier M, Dudler L, Gachter R, et al. Transport of the synthetic peptide DSIP through the blood-brain barrier in rabbit. Experientia. 1977;33(12):1609-10. PubMed DOI
6. Kovalzon VM, Strekalova TV. Delta sleep-inducing peptide (DSIP): a still unresolved riddle. J Neurochem. 2006;97(2):303-9. PubMed DOI
7. Kovalzon VM. [DSIP: the sleep peptide or an unknown hypothalamic hormone?]. Zh Evol Biokhim Fiziol. 1994;30(2):310-9. PubMed
8. Giusti M, Carraro A, Porcella E, et al. Delta sleep-inducing peptide administration does not influence growth hormone and prolactin secretion in normal women. Psychoneuroendocrinology. 1993;18(1):79-84. PubMed DOI
9. Shandra AA, Godlevskii LS, Brusentsov AI, et al. Effects of delta-sleep-inducing peptide on NMDA-induced convulsive activity in rats. Neurosci Behav Physiol. 1998;28(6):694-7. PubMed DOI
10. Stanojlovic O, Hrncic D, Radosavljevic T. [Endogenous anticonvulsants: neuropeptide Y and delta sleep inducing peptide]. Med Pregl. 2008;61(5-6):252-5. PubMed DOI
11. Umriukhin PE, Koplik EV, Sudakov KV. Dizocilpine and cycloheximide prevent inhibition of c-Fos gene expression by delta sleep-inducing peptide in the paraventricular nucleus of the hypothalamus in rats with different resistance to emotional stress. Neurosci Lett. 2012;506(2):184-7. PubMed DOI
12. Charnay Y, Golaz J, Vallet PG, et al. Production and immunohistochemical application of monoclonal antibodies against delta sleep-inducing peptide. J Chem Neuroanat. 1992;5(6):503-9. PubMed DOI
13. Roy K, Chauhan G, Kumari P, et al. Phosphorylated delta sleep inducing peptide restores spatial memory and p-CREB expression by improving sleep architecture at high altitude. Life Sci. 2018;209:282-290. PubMed DOI
14. Tukhovskaya EA, Ismailova AM, Shaykhutdinova ER, et al. Delta sleep-inducing peptide recovers motor function in SD rats after focal stroke. Molecules. 2021;26(17):5173. PubMed DOI
15. Tukhovskaya EA, Shaykhutdinova ER, Ismailova AM, et al. DSIP-like KND peptide reduces brain infarction in C57Bl/6 and reduces myocardial infarction in SD rats when administered during reperfusion. Biomedicines. 2021;9(4):407. PubMed DOI
16. Mu X, Qu L, Yin L, et al. Pichia pastoris secreted peptides crossing the blood-brain barrier and DSIP fusion peptide efficacy in PCPA-induced insomnia mouse models. Front Pharmacol. 2024;15:1439536. PubMed DOI
17. Obal F Jr, Krueger JM. Biochemical regulation of non-rapid-eye-movement sleep. Front Biosci. 2003;8:d520-50. PubMed DOI