DSIP

DSIP is short for delta sleep-inducing peptide. It is a nine-residue peptide with the sequence Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu. It was isolated in Basel in the 1970s from the cerebral venous blood of rabbits. The blood was drawn during electrical stimulation of the thalamus. The isolated fraction was then given to recipient rabbits. In them it produced a delta electroencephalogram (EEG) pattern. The peptide is named for that pattern (Schoenenberger and Monnier, 1977; Monnier et al., 1977).

So the name reflects the assay in which the peptide was found, not a proven function. Nearly five decades later, no DSIP gene, precursor protein or receptor has been found. A review in the Journal of Neurochemistry set out this open question in detail (Kovalzon and Strekalova, 2006).

Sequence and aliases come from PubChem.

Published work on DSIP falls into six areas. Each summary below states the model used. Readers should know up front that the evidence base is unusual. Most of the early work came from the one Basel group that isolated the peptide. Outside attempts to repeat that work were mixed or negative. A large share of the later papers come from a few Russian laboratories, many in Russian-language physiology and pharmacology journals.

Isolation and sequence

The first study split dialysate from rabbit cerebral venous blood into fractions. The team then made synthetic DSIP. They also made five possible breakdown products, two analogs with substituted residues and a related tripeptide, Trp-Ser-Glu. All nine synthetic peptides were infused into the cerebral ventricles of 58 rabbits under double-blind conditions. Fast Fourier analysis of neocortical and archicortical EEG recordings showed stronger delta and spindle patterns.

This was seen only with the intact nonapeptide (Schoenenberger and Monnier, 1977). Two early reviews from the same period collect the follow-up work on reported effects beyond the EEG. These include brain neurotransmitter levels, circadian and locomotor patterns, hormone levels and drug withdrawal (Graf and Kastin, 1984; Graf and Kastin, 1986).

Replication attempts in rodents

An outside group in Zurich gave DSIP to rats, both systemically and into the lateral and third ventricles. The group recorded motor activity and EEG. Neither sleep nor delta-band power rose significantly in the two hours after the peptide was given. The authors concluded that the peptide did not qualify as a specific sleep-promoting substance. They did note smaller changes in dark-time motor activity (Tobler and Borbely, 1980). This split between the first lab and outside groups is the central feature of DSIP research.

Human studies

Two small controlled studies looked at the peptide in people with chronic insomnia. One was a double-blind matched-pairs study of 16 patients, with peptide or placebo given on three nights. Polysomnography showed higher sleep efficiency and shorter latency than placebo. But the authors judged the differences weak and partly due to a change in the placebo group. They concluded that short-term use in chronic insomnia was unlikely to matter clinically (Bes et al., 1992). A double-blind crossover study gave the peptide or placebo on four nights.

It found increases in total and non-REM sleep time, but these were already present at baseline. The authors judged the change of little clinical significance (Monti et al., 1987). A later randomized study in 24 female patients used the peptide with volatile anesthesia. It found the opposite of the expected direction. Heart rate rose and heart-rate variability fell. Delta rhythm dropped and bispectral index rose. The authors read this as a lightening of anesthetic depth (Pomfrett et al., 2009).

Stability, transport and clearance

An enzyme immunoassay study in dogs, a monkey and rats found rapid clearance from plasma. Mean half-life was about four minutes in dogs and about two minutes in rats (Kato et al., 1984). Incubation in human and rat blood released material that eluted like tryptophan. This points to cleavage of the N-terminal residue.

Labeled analogs broke down more slowly and formed complexes (Graf et al., 1987). Transport studies in dogs found measurable entry into cerebrospinal fluid. Entry was linked to plasma level, plasma half-life and lipophilicity (Banks et al., 1986; Banks et al., 1982). A cultured brain microvessel endothelial model found non-saturable, temperature-insensitive flux. This fits simple diffusion (Raeissi and Audus, 1989).

What DSIP immunoassays actually detect

Much of the claimed tissue distribution of DSIP rests on antibody tests for DSIP-like immunoreactivity. One of those larger immunoreactive forms was isolated directly from porcine brain. It proved to be a 77-residue peptide with a sequence unrelated to DSIP. Yet antiserum raised against DSIP still recognizes it (Sillard et al., 1993). This is a standing caution for anyone reading older distribution data.

Stress, hypoxia and injury models

A Russian group studied isolated rat brain mitochondria. It found that the peptide raised phosphorylating respiration and the respiratory control ratio. Giving the peptide first also prevented the drop in mitochondrial respiratory activity after experimental hypoxia (Khvatova et al., 2003). Another study used a rat middle cerebral artery occlusion model.

The peptide was given through the nose before and after the occlusion. It did not significantly change infarct volume. But rotarod performance recovered more in the peptide group (Tukhovskaya et al., 2021). One recent mouse study tested a fusion construct rather than free DSIP. The construct is meant to increase brain entry (Mu et al., 2024).