Neuropeptide S (mood/arousal)

NPS is produced by a small cluster of neurons near the locus coeruleus and acts on a single known receptor, NPSR1, which signals through both Gq/phospholipase C and Gs/cAMP pathways. The pharmacological oddity that makes it interesting is the dissociation of arousal from anxiety: intracerebroventricular NPS in rodents increases locomotion and wakefulness and reduces REM and slow-wave sleep, yet produces anxiolysis across several standard behavioural models, apparently by driving inhibitory interneurons in the amygdala that suppress fear output. It also enhances fear extinction, which is why it drew attention as a possible adjunct to exposure therapy. Human relevance comes indirectly through the NPSR1 Asn107Ile functional polymorphism, which has been linked to panic disorder and to sleep and asthma phenotypes.

Targets: Neuropeptide S receptor 1 (NPSR1 / GPR154), Amygdala inhibitory interneuron circuits, Locus coeruleus / arousal network

• · There is no human dosing data for NPS. Every rodent study that generated the interesting findings used intracerebroventricular injection at nanomole doses, a route that has no peripheral or intranasal equivalent. Anyone quoting a specific human microgram dose for NPS is making it up.

Work out your exact syringe units →

• commonInsomnia and suppressed sleep— Consistently observed in rodents; NPS reduces both REM and slow-wave sleep. Human frequency is inferred, not measured.
• commonIncreased locomotor activity and restlessness— The core rodent phenotype.
• commonUnknown human safety profile— This is the honest entry: nobody has characterised what NPS does in a person.

• Any use in humans - there is no human safety, dosing or pharmacokinetic data whatsoever, and no route of administration validated to reach the receptor.
• Pregnancy and breastfeeding.
• Sleep disorders of any kind - the reliable rodent effect is suppressed sleep.

• redundantorexin-a-wake — Both push the same wake-promoting axis; NPS neurons sit close to the arousal network orexin also drives.
• redundantselank-anxiolytic — Overlapping anxiolytic intent, with Selank the only one of the two with any human data.
• conflictdsip — NPS suppresses slow-wave sleep, which is exactly what DSIP is taken to increase.

• · No monitoring protocol exists because no validated human use exists.
• · If anyone is going to use it regardless, sleep tracking is the only meaningful readout of the best-characterised effect.

Research reagent only. Not approved for human use anywhere and never formally studied in humans.

• Xu et al. 2004, Neuron - neuropeptide S identification, arousal and anxiolytic effects (preclinical)
• Reinscheid & Xu 2005, neuropeptide S and its receptor review (review)
• Jungling et al. 2008, NPS and fear extinction in the amygdala (preclinical)
• Domschke et al., NPSR1 Asn107Ile polymorphism and panic disorder (trial)

Neuropeptide S is the most interesting compound in this class and the one you have the least business taking. Both of those are true at once and the record should not soften either. The pharmacology is genuinely unusual. Xu, Reinscheid, Civelli and colleagues described it in Neuron in 2004: central NPS administration increased locomotor activity in mice, reduced both REM and slow-wave sleep in rats, and produced anxiolytic-like effects across four separate stress paradigms. Arousal up and anxiety down, in the same animal, from the same molecule. Almost nothing else does that. Stimulants raise arousal and raise anxiety. Anxiolytics lower anxiety and lower arousal. NPS dissociates the two axes, and that dissociation is why the compound has a following that far exceeds its evidence base. The receptor is NPSR1, also called GPR154, a class A GPCR that was orphaned until NPS deorphanised it. It couples to both Gq and Gs, so activation raises both intracellular calcium via phospholipase C and cAMP via adenylate cyclase. Dual coupling is not unusual in itself, but it does mean the functional consequence varies by which pathway dominates in a given cell type, and that is part of why the behavioural profile is not simply arousal. The anatomy explains the dissociation. NPS-expressing neurons sit in a small, previously uncharacterised cluster between the locus coeruleus and Barrington's nucleus - the Xu paper's identification of this cluster was one of its notable findings. From there NPS reaches the arousal network, which accounts for the wakefulness. But NPSR1 is also densely expressed in the amygdala, and Jüngling, Pape and colleagues showed in Neuron in 2008 exactly how the anxiolysis works: NPS increases excitatory drive onto the intercalated GABAergic cell masses of the amygdala via presynaptic NPS receptors on principal neurons. The intercalated cells are inhibitory gatekeepers. Driving them harder suppresses the output of the central amygdala, which is the amygdala's fear-expression relay. So NPS is not blunting fear the way a benzodiazepine does, by turning down everything. It is recruiting a specific inhibitory circuit that already exists to shut fear expression down. That same circuit is what fear extinction depends on, which is why NPS in the amygdala also enhanced extinction of conditioned fear and why an NPSR antagonist did the opposite. That finding is the reason NPS drew serious attention as a possible adjunct to exposure therapy rather than as a standalone anxiolytic. The only human thread is genetic and it points in a slightly awkward direction. Domschke and a large German consortium reported in Molecular Psychiatry in 2011 that the more active NPSR1 rs324981 T allele was associated with panic disorder, predominantly in women, along with higher anxiety sensitivity, elevated heart rate during behavioural testing, and reduced prefrontal and anterior cingulate activation to fearful stimuli in patients. Read that carefully: more NPSR1 signalling associated with more panic, not less. That is not a straightforward contradiction of the rodent work - the genetic variant acts across a lifetime of development while acute administration acts on a mature circuit, and the direction of an association study is not the direction of a drug effect - but it is a real complication, and anyone selling NPS as a human anxiolytic on the strength of the rodent data should be made to account for it. And then the barrier that ends the discussion: every one of these findings used intracerebroventricular delivery. Nobody has shown that peripherally or intranasally administered NPS reaches NPSR1 in the amygdala in any species at any dose. There is no human dose, no human route and no human data. Two translational gaps, stacked.

The specific thing that goes wrong with NPS is a double extrapolation that vendors present as a single small step. The first jump is across species. Every behavioural finding is in rats and mice. That is normal for early pharmacology and not by itself damning. The second jump is across route, and this is the one that is usually hidden. The findings came from intracerebroventricular injection - a cannula through the skull, delivering peptide directly into the ventricular system. There is no peripheral or intranasal equivalent for that. Nobody has demonstrated that intranasally administered NPS reaches NPSR1 in the amygdala in any species. A vendor quoting a microgram intranasal dose has taken a nanomole ICV dose and converted it into something that sounds like a protocol. That number is invented. There is no arithmetic that gets you from one to the other, because the missing variable - what fraction of a nasal dose reaches the target receptor - has never been measured. The reliable effect, if any of it translates, is suppressed sleep. NPS reduced both REM and slow-wave sleep in rats. Someone taking this for mood in the evening is taking an insomnia agent. The human genetic data point the wrong way for the marketing. The gain-of-function NPSR1 variant was associated with panic disorder, higher anxiety sensitivity and higher heart rate on behavioural testing. That is not proof that acute NPS causes panic in humans - developmental genetic effects and acute drug effects genuinely are different things - but it is the only human evidence that exists, and it is not the evidence a seller would choose. And the practical one: there is no way to tell whether what is in the vial is NPS. It is a 20-residue linear peptide with no distinguishing structural features, sold in a market with no meaningful quality enforcement, to buyers who have no way to detect an effect and therefore no way to detect a substitution.

• Neuropeptide S receptor 1 (NPSR1 / GPR154 / GPRA) — Low nanomolar potency in calcium mobilisation assays. A specific EC50 could not be resolved to a primary source in this session; figures in the region of 10 nM are commonly quoted from the original deorphanisation work.

  The only known receptor for NPS. Dual Gq and Gs coupling, raising both intracellular calcium via PLC and cAMP via adenylate cyclase. Distributed across amygdala, midline thalamus, cortex and hypothalamus. Everything NPS does runs through this one receptor, which makes it a clean target and also means there is no selectivity dial to turn.

• Amygdala intercalated GABAergic cell masses — Circuit-level, via presynaptic NPSR on principal neurons

  NPS increases excitatory transmission onto the intercalated inhibitory neurons, which suppresses central amygdala fear output (Jungling et al., Neuron 2008). This is the mechanism of both the anxiolysis and the enhanced fear extinction, and it is the most mechanistically satisfying part of the NPS story.

• Arousal network (NPS neurons between locus coeruleus and Barrington's nucleus) — Anatomical origin, not a binding site

  Projections from this small previously uncharacterised cluster to arousal circuitry drive the wakefulness and locomotor effects, and suppress both REM and slow-wave sleep in rats.

• NPSR1 Asn107Ile functional polymorphism (rs324981) — The Ile107 variant is a gain-of-function variant, reported to be roughly an order of magnitude more responsive to NPS than the Asn107 form

  The only human-relevant NPS finding. The more active T allele was associated with panic disorder in women, higher anxiety sensitivity, higher heart rate on behavioural testing and reduced prefrontal and cingulate activation to fear (Domschke et al., Mol Psychiatry 2011). Complicates the simple 'NPS reduces anxiety' story considerably.

• Domschke, Reif, Weber, Deckert et al. - neuropeptide S receptor gene: converging evidence for a role in panic disorder (Molecular Psychiatry 2011) Genetic association study with imaging and psychophysiological components, not an interventional trial · 2011

  Association of the functional NPSR1 rs324981 variant with panic disorder across two independent samples, plus anxiety sensitivity, heart rate during behavioural testing, and fMRI response to fearful faces. The more active T allele was associated with panic disorder, predominantly in women. This is the only human NPS-related study of any kind, and it is genetic rather than interventional.

• No interventional human trial of neuropeptide S exists None

  Stated explicitly because the absence is the most important fact about this compound. No human has been given neuropeptide S in a published, controlled protocol by any route. There is no phase 1, no first-in-human safety study, and no dose-finding work of any kind.

In humans: unknown, in every respect. Nobody has been dosed, so there is no onset, no duration, no time to effect and no time to offset. In rodents, after intracerebroventricular administration: locomotor and anxiolytic effects appear within minutes and the sleep suppression is evident in the same recording session. Reported effects are acute and short-lived, consistent with a linear peptide being degraded rapidly. The reason this section is short is the reason the compound should not be used. A timeline requires somebody to have measured something in a person, and nobody has.

There is no stack. Stacking implies a base protocol to add to, and NPS has no established protocol, no human dose, no validated route and no safety data. Anything written here about combinations would be fiction dressed as advice. What can be said honestly is mechanistic. NPS suppresses both REM and slow-wave sleep in rats, so pairing it with anything taken to deepen sleep - DSIP, MK-677, glycine - is self-cancelling by design. It sits functionally alongside orexin-A rather than complementing it: both push the same arousal axis, and the NPS neurons are anatomically adjacent to the arousal network orexin drives, so combining them is redundancy with double the unknowns. And its anxiolytic intent overlaps entirely with Selank, which is the one compound in that pair with actual human trials, which tells you which one to reach for. The genuinely interesting hypothetical combination - NPS as an adjunct to exposure therapy, exploiting the fear extinction enhancement - is the one thing in this space worth a real trial and the one thing nobody can do at home, because it needs a dose, a route and a therapist working from the same protocol. That is a study waiting to be run, not a stack.

Against Selank: same therapeutic intent, opposite ends of the evidence spectrum. Selank has a randomised active-comparator trial against a benzodiazepine in 62 patients, a mechanistic chain running from in vitro enzyme inhibition to a biomarker change in treated patients, and registered medicine status in Russia. NPS has rodent studies and a genetic association that points the wrong way. If anxiolysis is the goal there is no version of this comparison where NPS wins. Against orexin-A: both are wake-promoting neuropeptides with a delivery problem. Orexin-A at least has two controlled human studies, a clear anatomical rationale for intranasal delivery, and primate data comparing routes head to head. NPS has none of that. If you were going to try one uncharacterised arousal peptide, orexin-A is the one with a floor under it. Against a benzodiazepine or an SSRI: not a comparison, because one side has no human data at all. As a research target: NPS deserves more attention than it gets. The dissociation of arousal from anxiety is a genuinely rare pharmacological property, and the intercalated amygdala mechanism is one of the more elegant circuit stories in affective neuroscience. Small-molecule NPSR1 agonists with real pharmacokinetics would be an interesting class of drug and it is somewhat surprising nobody has pushed one into humans. The right conclusion from the NPS literature is that someone should run a trial, not that you should buy a vial.

There is no cost per month because there is no month of use to cost. NPS is sold as a research reagent by the milligram, typically in 1 mg or 5 mg quantities, at prices in line with other 20-residue custom peptides. Since there is no established human dose, no protocol and no validated route, any monthly figure would be invented. Deliberately left null.

• ?No human has been given neuropeptide S in any published protocol. There is no human dose, no human route, no human safety data and no human pharmacokinetics.
• ?No pharmacokinetic data exists in any species, including the rodents in which all the behavioural work was done.
• ?Whether intranasally or peripherally administered NPS reaches NPSR1 in the amygdala at any dose in any species is unknown and, as far as I can determine, has never been tested.
• ?The NPSR1 EC50 for NPS could not be resolved to a primary source in this session. The low-nanomolar figure given is the commonly quoted value from the deorphanisation literature.
• ?The reported ten-fold gain of function for the Ile107 variant is widely cited but I did not resolve it to a primary measurement in this session.
• ?Participant numbers for the Domschke 2011 study were not extracted; it used two independent case-control samples plus imaging and psychophysiological subsamples.
• ?How to reconcile the rodent anxiolysis with the human genetic association between more active NPSR1 signalling and panic disorder is genuinely unresolved in the literature, not just here.
• ?Whether the arousal-plus-anxiolysis dissociation survives translation to humans is unknown, and it is the single property the compound is sold on.
• ?The 2187.5 Da molecular weight in the Core record was not independently verified against an analytical source, though the sequence was confirmed via UniProt P0C0P6.
• ?Product identity and purity for commercially sold NPS cannot be assessed by any means available to a buyer, and there is no detectable effect to fall back on as a proxy.

• Neuropeptide S: a neuropeptide promoting arousal and anxiolytic-like effects Xu YL, Reinscheid RK, Huitron-Resendiz S, Clark SD, Wang Z, Lin SH, Brucher FA, Zeng J, Ly NK, Henriksen SJ, de Lecea L, Civelli O, Neuron, 2004 · PMID 15312648

  Neuron 43(4):487-97. The founding paper. Identified NPSR distribution, showed increased locomotion in mice, reduced REM and slow-wave sleep in rats, anxiolytic-like effects across four stress paradigms, and located the NPS neurons in a previously uncharacterised cluster near the locus coeruleus. Everything interesting about this compound is in here, and all of it is intracerebroventricular.

• Neuropeptide S-mediated control of fear expression and extinction: role of intercalated GABAergic neurons in the amygdala Jungling K, Seidenbecher T, Sosulina L, Lesting J, Sangha S, Clark SD, Okamura N, Duangdao DM, Xu YL, Reinscheid RK, Pape HC, Neuron, 2008 · PMID 18667157

  Neuron 59(2):298-310. The circuit mechanism for the anxiolysis and the fear extinction enhancement. The best-argued paper in the NPS literature and the reason the compound was taken seriously as a possible exposure-therapy adjunct.

• Neuropeptide S receptor gene - converging evidence for a role in panic disorder Domschke K, Reif A, Weber H, Richter J, Hohoff C, Ohrmann P, et al. (Deckert J, senior author), Molecular Psychiatry, 2011 · PMID 20603625

  Mol Psychiatry 16(9):938-48. The only human evidence touching NPS biology, and it complicates the rodent story rather than confirming it - the more active receptor variant went with more panic, not less.

• Neuropeptide S precursor, UniProtKB entry P0C0P6 UniProt Consortium, UniProtKB

  Source for the 20-residue mature human NPS sequence, its position in the 89-residue precursor, the rs4751440 variant and the structure-activity notes on Ser1 and Phe2. Resolved directly in this session.