Endomorphin-1 and -2

Endomorphin-1 (Tyr-Pro-Trp-Phe-NH2, about 610.7 Da) and endomorphin-2 (Tyr-Pro-Phe-Phe-NH2, about 571.7 Da) were described by Zadina's group in 1997 and remain the most mu-selective endogenous opioid ligands identified — selectivity for mu over delta and kappa in the thousands-fold range. Their significance is not as drugs in themselves; native endomorphins are degraded within minutes and cross the blood-brain barrier poorly. It is that engineered analogues, particularly cyclised and glycosylated versions, produce analgesia in rodents equal to or greater than morphine while showing markedly less respiratory depression, less tolerance, less motor impairment and less reward-related behaviour. That dissociation is thought to reflect biased signalling — favouring G-protein coupling over beta-arrestin recruitment — and is the same hypothesis that drove oliceridine to approval. Whether the separation survives translation to humans is genuinely unknown; nothing from this family has completed a human analgesia trial.

Targets: Mu opioid receptor (OPRM1), G-protein biased mu signalling versus beta-arrestin recruitment

• · There is no human dose. All published dosing is intrathecal or intracerebroventricular in rodents, which does not convert to a subcutaneous human protocol by any valid arithmetic. The native peptides are also poorly brain-penetrant, so peripheral self-administration would most likely produce peripheral mu effects — constipation and itch — and little analgesia.

Work out your exact syringe units →

• commonConstipation and GI slowing— Peripheral mu receptors are hit regardless of CNS penetration.
• commonPruritus
• uncommonRespiratory depression— Reduced relative to morphine in animal models, but 'reduced' is not 'absent' and this has never been tested in humans.
• uncommonTolerance— Markedly less than morphine in rodents; unquantified in humans.

• Any concurrent opioid or CNS depressant. The claimed respiratory safety margin is a rodent finding with an engineered analogue, not a property you can assume of research-grade native peptide.
• Reliance on the 'safer opioid' framing. Nothing in this family has been shown safe in a human being.

• redundantdermorphin — Same receptor, and dermorphin is far more potent; combining them is pure additive risk.
• redundantenkephalins — Both endogenous opioid peptides with the same degradation-limited profile.
• cautiondifelikefalin — Mu and kappa agonism together produce opposing effects on mood and reward and muddy any assessment of what is working.

• · No human monitoring standard exists. In research contexts: respiratory rate, oxygen saturation, analgesic threshold testing and signs of tolerance.

Not approved anywhere; sold as research peptides. As mu opioid agonists they raise controlled-substance analogue questions in several jurisdictions.

• Zadina et al. 1997, Nature — a potent and selective endogenous agonist for the mu-opiate receptor (preclinical)
• Zadina et al. 2016, endomorphin analogues with analgesia comparable to morphine and reduced side effects in rodents (preclinical)

Endomorphin-1 and -2 are the most mu-selective opioid ligands ever described from a natural source — selectivity for mu over delta and kappa reported in the thousands-fold range, far exceeding morphine, dermorphin or the enkephalins. As drugs in their native form they are hopeless: degraded in minutes, poorly brain-penetrant, no human dosing. Their importance is entirely as a scaffold, and the reason that scaffold has absorbed two decades of medicinal chemistry is the biased-signalling hypothesis. When a mu agonist binds, the receptor can couple to Gi/Go — producing analgesia — or recruit beta-arrestin-2, which is associated in the classic model with respiratory depression, constipation, tolerance and receptor internalisation. If a ligand could favour G-protein coupling over arrestin recruitment, you would in principle get the analgesia without the apnoea. Endomorphin analogues appeared to do exactly that. Zadina's group produced cyclised and modified endomorphin analogues that in rodents delivered analgesia equal to or greater than morphine while showing markedly reduced respiratory depression, less tolerance, less motor impairment, less glial activation and less reward-related behaviour. That is a striking result and it is real preclinical data. What has to be said alongside it is that the same hypothesis produced oliceridine, which reached FDA approval in 2020 and delivered a considerably more modest real-world advantage over morphine than the preclinical work predicted, and that the beta-arrestin model itself has been substantially challenged — the original knockout mouse findings have proved harder to replicate than the field assumed, and intrinsic efficacy may explain more of the separation than bias does. So the honest position is this: endomorphin analogues represent the best-developed version of a hypothesis that, when it finally reached humans in another molecule, underdelivered. That does not make the hypothesis wrong. It does mean that anyone treating 'less respiratory depression in rodents' as a safety property they can rely on in themselves is making an inference the field has already seen fail once. And critically, none of this applies to the native peptides that are actually sold as research chemicals — those have no demonstrated dissociation of any kind. There is also a real and unresolved oddity worth knowing: despite thirty years of searching, no gene or precursor protein encoding either endomorphin has been found. A systematic search of the human proteome came up empty. Their status as genuine endogenous ligands is therefore not fully settled.

The specific failure mode here is a reasoning error rather than a dosing error, and it is worth naming precisely because it is so seductive. The literature says endomorphin analogues produce analgesia with less respiratory depression, less tolerance and less reward than morphine. The compound on sale is not that. It is native endomorphin-1 or -2, which has no demonstrated dissociation of any kind, is degraded within minutes, and crosses the blood-brain barrier poorly — so the realistic outcome of a peripheral injection is peripheral mu effects (constipation, itch, flushing) with little analgesia. Taking more to compensate does not fix a blood-brain barrier problem, it just loads the peripheral receptors harder. And on the small fraction that does cross, the safety-margin assumption borrowed from the analogue papers does not hold. The second error is dose translation: all published endomorphin analgesia dosing is intrathecal or intracerebroventricular in rodents, and there is no valid arithmetic converting an ICV rodent dose into a human subcutaneous dose. Anyone who has produced such a number has invented it. The third is handling — endomorphin-1 contains tryptophan and degrades noticeably with light and oxygen in solution, so material stored casually is likely to be partly inactive, which introduces exactly the kind of unpredictable potency variation that is dangerous with an opioid. Fourth, the same legal caution as with dermorphin and DALDA: mu agonist activity raises controlled-substance analogue questions in multiple jurisdictions, and the 'research peptide' framing is not the protection people assume.

• Mu opioid receptor (OPRM1) — Sub-nanomolar with mu-over-delta and mu-over-kappa selectivity reported in the thousands-fold range — the highest selectivity of any natural opioid peptide. Specific Ki values appear throughout the literature but I did not resolve a primary source in this session, so none is asserted.

  High-efficacy agonism with standard Gi/Go coupling: adenylyl cyclase inhibition, GIRK opening, presynaptic calcium channel closure, activation of descending inhibition.

• Beta-arrestin-2 recruitment at the mu receptor — The signalling axis rather than a binding site.

  Engineered endomorphin analogues show reduced arrestin recruitment relative to G-protein coupling, which is the proposed explanation for the reduced respiratory depression, tolerance and reward seen in rodents. This is a hypothesis with real preclinical support and a mixed translational record — the same reasoning produced oliceridine, whose clinical advantage over morphine turned out to be modest.

• Peripheral mu receptors in gut and skin — Fully accessible; unlike the CNS, these do not require crossing a barrier.

  Constipation, slowed GI transit and pruritus. For the native peptides given peripherally, these are the effects you would actually get, because peripheral receptors are reached and central ones largely are not.

• Delta and kappa opioid receptors — Negligible at physiological or analgesic exposures.

  None of consequence. The selectivity is the scientifically interesting property of these peptides and the reason they remain standard laboratory tools for probing mu-specific effects.

Rapid after central or intravenous administration in animal models, with duration on the order of minutes for the native peptides. In humans, unknown in every respect — there is no published onset, peak or duration, because nobody has completed a human study. The engineered analogues have substantially longer durations in rodents, but those are different molecules and their human timeline is equally unknown.

There is no human stack, and the pharmacologically meaningful notes are all subtractive. Any second CNS depressant is additive on respiratory drive, and the 'reduced respiratory depression' finding is a rodent result with engineered analogues that does not transfer to native peptide bought as a research chemical. Stacking with dermorphin is redundant and dangerous — same receptor, and dermorphin is far more potent. Stacking with the enkephalins is redundant in the other direction: both are degradation-limited endogenous peptides with the same fatal flaw. Combining with difelikefalin is pharmacologically incoherent, since kappa agonism opposes mu-mediated reward and produces dysphoria, so you would get the respiratory risk of one and the mood cost of the other. The one thing that genuinely belongs alongside any mu agonist is naloxone, unexpired, with a second person present who knows how to use it.

Endomorphins are the best version of a story that this whole class keeps telling and never finishing: a mu agonist that relieves pain without stopping your breathing. The preclinical case is genuinely the strongest anyone has assembled — Zadina's 2016 analogues showed the separation across five separate endpoints simultaneously, which is more than a fluke. But the same hypothesis, tested in humans with a different molecule, gave oliceridine, whose real-world advantage over morphine has been modest enough that the biased-signalling model itself is now contested. Compare that with the two compounds in this class that actually solved their problem: ziconotide abandoned the opioid receptor entirely and blocked a calcium channel instead, and difelikefalin changed receptor and then engineered the molecule out of the brain. Both got approved. The endomorphin programme kept the mu receptor and tried to fix the signalling, and after nearly thirty years has not put a molecule into a completed human analgesia trial. That is not evidence the approach is wrong; it is evidence that it is hard. What it should tell a reader is that the native endomorphin in a vendor's vial carries the reputation of the analogue programme without any of its properties, and that gap is where people get hurt.

Deliberately not priced. These are research reagents sold in milligram quantities with no human protocol, and a monthly cost figure would imply a use pattern that does not exist.

• ?No human pharmacokinetics, dosing or safety data exist for either peptide or for any analogue. Everything here is animal work.
• ?No Ki values or selectivity ratios are asserted; the thousands-fold mu selectivity figure comes from the secondary literature and I did not resolve a primary binding source in this session.
• ?The biosynthetic origin of both peptides remains unexplained — no precursor gene has been identified, which leaves their status as true endogenous ligands genuinely unsettled.
• ?The extent of blood-brain barrier penetration is described as poor-to-partial from the secondary literature; there is no quantitative human figure.
• ?Whether the reduced respiratory depression seen with engineered analogues in rodents reflects beta-arrestin bias, lower intrinsic efficacy, or something else is actively contested in the field.
• ?The Core record leaves molecularWeightDa null; PubChem gives 610.7 Da for endomorphin-1 (CID 5311080) and 571.7 Da for endomorphin-2 (CID 5311081), verified in this session.
• ?Half-life figures for the native peptides are given qualitatively as 'minutes' because I did not resolve a primary pharmacokinetic source.

• A potent and selective endogenous agonist for the mu-opiate receptor Zadina JE, Hackler L, Ge LJ, Kastin AJ, Nature, 1997 · PMID 9087409

  The discovery paper. Endomorphin-1 described as the most mu-selective endogenous opioid ligand identified, which is the claim the entire field rests on.

• Endomorphin analog analgesics with reduced abuse liability, respiratory depression, motor impairment, tolerance, and glial activation relative to morphine Zadina JE, Nilges MR, Morgenweck J, et al., Neuropharmacology, 2016 · PMID 26748051

  The central paper for the 'safer opioid' claim. Engineered endomorphin analogues in rodents: analgesia comparable to morphine with markedly reduced respiratory depression, tolerance, motor impairment and reward behaviour. Read it, and then note that these are analogues in rats, not the native peptides sold as research chemicals.

• Search of the human proteome for endomorphin-1 and endomorphin-2 precursor proteins Terskiy A, Wannemacher KM, Yadav PN, et al., Life Sciences, 2007 · PMID 17964607

  The uncomfortable paper. A systematic search of the human proteome found no plausible precursor protein containing either endomorphin sequence. Thirty years on, the biosynthetic origin of these peptides remains unexplained, which is a genuine open question about whether they are true endogenous ligands.

• Novel mu opioid antagonists derived from the mu opioid agonists endomorphin and [Dmt(1)]DALDA Shi S, Xu J, Feng L, et al., Chemical Biology & Drug Design, 2020 · PMID 32526055

  Useful for seeing how closely the endomorphin and DALDA scaffolds are related in practice — the same medicinal chemistry programme works on both, and small changes flip agonists into antagonists.