CRF antagonist peptides

Corticotropin-releasing factor is the 41-amino-acid hypothalamic peptide that starts the stress response: it drives pituitary ACTH release and therefore cortisol, and independently produces anxiety, arousal, fragmented sleep and colonic motility changes through CRF receptors in the amygdala, locus coeruleus and gut. The peptide antagonists are truncated or modified CRF analogues. Alpha-helical CRF(9-41) was the first-generation, non-selective and relatively weak. Astressin and astressin-B are cyclised, far more potent, and astressin-B in particular became notorious for an accidental finding: it caused striking hair regrowth in chronically stressed mice. Astressin-2B is CRF2-selective. All of these are peptides with poor oral availability and poor brain penetration, which is why the clinical programmes moved to small molecules — and those small-molecule CRF1 antagonists then failed repeatedly in depression, anxiety and PTSD trials, which is the most important fact about this whole target.

Targets: CRF1 receptor (CRHR1), CRF2 receptor (CRHR2), Hypothalamic-pituitary-adrenal axis, Amygdala and locus coeruleus stress circuitry

• · These are laboratory reagents. The animal work uses microgram-per-kilogram to milligram-per-kilogram dosing by intravenous, subcutaneous or intracerebroventricular routes depending on the analogue, and none of it translates to a human protocol. No CRF peptide antagonist has an approved or even conventionally studied human dose.

Work out your exact syringe units →

• commonBlunted stress and cortisol response— The intended pharmacology, and also the main hazard - you need a cortisol response to handle illness, injury and surgery.
• commonUnknown human safety profile— No peptide CRF antagonist has been through human safety characterisation.
• uncommonHypotension— CRF receptor blockade affects cardiovascular stress responses in animals.
• uncommonHepatic enzyme elevation— Seen with several small-molecule CRF1 antagonists in clinical development; whether it extends to the peptides is unknown.

• Any human use - these compounds have never been given to a person in a published, controlled protocol.
• Adrenal insufficiency or anyone on corticosteroid replacement - suppressing CRF signalling on top of an already compromised axis is dangerous.
• Acute illness, injury or surgery, where an intact stress response is what keeps you alive.
• Pregnancy - CRF plays a defined role in the timing of parturition.

• synergydsip — Mechanistically adjacent - one of the proposed actions of DSIP is suppression of CRH and ACTH output, making this the same target approached far more crudely.
• redundantselank-anxiolytic — Same therapeutic intent, opposite ends of the evidence spectrum: Selank is a registered medicine somewhere, these are bench reagents.

• · Not applicable outside a laboratory. If anyone were to use these, morning cortisol and ACTH would be the obvious axis to watch.
• · The relevant clinical lesson is that pharmacological CRF1 blockade did not produce antidepressant or anxiolytic benefit in properly powered human trials.

Research reagents. Not approved for human use in any jurisdiction and not marketed for human consumption by legitimate suppliers.

• Rivier et al., design and characterisation of astressin CRF antagonists (preclinical)
• Wang et al. 2011, PLoS ONE - astressin-B reverses stress-induced alopecia in mice (preclinical)
• Coric et al. 2010, pexacerfont in generalised anxiety disorder (negative) (trial)
• Binneman et al. 2008, CP-316,311 in major depressive disorder (negative) (trial)
• Zorrilla & Koob, CRF antagonists in stress-related disorders review (review)

Corticotropin-releasing factor is the top of the stress cascade. Hypothalamic CRF neurons release it into the hypophyseal portal system, it acts on CRF1 receptors on pituitary corticotrophs to drive ACTH, and ACTH drives adrenal cortisol. That is the endocrine arm and it is only half the story. CRF also acts as a neurotransmitter in its own right in the central nucleus of the amygdala, the bed nucleus of the stria terminalis and the locus coeruleus, where it produces anxiety, hypervigilance, fragmented sleep and suppressed feeding independently of cortisol. Peripherally it drives colonic motility, which is why CRF antagonists have a substantial irritable bowel literature. CRF1 and CRF2 are both class B GPCRs, predominantly Gs-coupled through adenylate cyclase and cAMP, with CRF1 carrying most of the anxiogenic and ACTH-driving load and CRF2 having a more complex, often opposing role. The peptide antagonists were built as pharmacological tools and they are excellent ones. Gulyas, Rivier and colleagues at the Salk Institute published the structurally constrained agonists and antagonists in PNAS in 1995 - the design principle was to lock the peptide into its bioactive alpha-helical conformation with a lactam bridge, which produced dramatic gains in potency and duration over the floppy linear first-generation compounds. Koerber's 1998 medicinal chemistry paper details the i-(i+3) Glu-Xaa-DXbb-Lys bridge chemistry. Spina, Zorrilla, Koob and Rivier then characterised the behavioural effects of central astressin in rats in 2000. This is genuinely good chemistry and genuinely good pharmacology. The famous result is the one nobody was looking for. Wang, Million, Rivier, Taché and colleagues published in PLoS ONE in 2011 that astressin-B given peripherally at 5 micrograms per mouse for five consecutive days induced pigmentation and hair regrowth in CRF-overexpressing alopecic mice, retained for over four months, and prevented alopecia developing in younger animals. Histology showed hair follicles pushed from telogen back into anagen. Notably the drug did not lower the elevated circulating stress hormones, which points to a local peripheral CRF receptor mechanism in skin rather than a systemic HPA effect. This is the finding that put astressin-B into internet circulation, and it is worth being precise about it: it is a mouse genetically engineered to overexpress CRF, which is not a model of ordinary human hair loss, and the effect was on a stress-driven alopecia phenotype rather than androgenetic alopecia. Now the part that matters most, and the part that is systematically omitted wherever these compounds are sold. The CRF1 target has been tested properly in humans, repeatedly, with small molecules that actually reach the brain, and it failed. Binneman and colleagues at Pfizer ran CP-316,311 at 400 mg twice daily against sertraline and placebo in 123 patients with recurrent major depression and stopped the trial for futility at interim - no separation from placebo, while sertraline separated cleanly. Coric and colleagues at Bristol-Myers Squibb ran pexacerfont at 100 mg daily against escitalopram and placebo in 260 patients with generalised anxiety disorder for eight weeks; response rates were 42 percent for pexacerfont, 42 percent for placebo and 53 percent for escitalopram, with escitalopram separating and pexacerfont not. Both drugs were safe and well tolerated. They simply did not work. Similar programmes at other companies produced similar results in depression, anxiety and PTSD. That is the most informative evidence available about this target, and it is negative. The preclinical rationale was strong, the mechanism was clean, the compounds engaged the target, and the clinical benefit was absent. Anyone offering a peptide CRF antagonist as a stress or sleep intervention is asking you to bet on a hypothesis that has already been tested with better tools than theirs and lost.

The framing goes wrong before the pharmacology does. These compounds are sold on the idea that blocking the stress hormone will reduce stress, which sounds mechanistically obvious and has been tested properly and found false. Two well-designed phase 2 trials with brain-penetrant small molecules, working active comparators, and 383 patients between them, produced no benefit in either generalised anxiety disorder or major depression. Other programmes at other companies replicated that failure in PTSD and depression. The target engaged; the benefit did not appear. A peptide that reaches the brain worse than those molecules did is not going to succeed where they failed. The second thing that goes wrong is the one that can hurt you. Cortisol is not a waste product of modern life. It maintains blood pressure, mobilises glucose, and is the reason you survive infection, trauma and surgery. Suppressing the top of that axis pharmacologically produces exactly the state of secondary adrenal insufficiency, and the failure mode is not feeling calm - it is collapsing during a bout of gastroenteritis. People with steroid-induced adrenal suppression carry emergency cards and injectable hydrocortisone for this reason. Someone self-administering a CRF antagonist has the same physiology and none of the safety net. The third is the hair loss story. Astressin-B got its reputation from a genuinely striking mouse result, and the details do not survive translation. The mice overexpressed CRF - a genetic model of chronic stress-driven alopecia, not androgenetic hair loss. The effect appears to be local to skin CRF receptors rather than systemic. Nobody has given it to a human. Minoxidil, which is approved, cheap and available, showed partial efficacy in the same experiment. If hair is the goal, the evidence-based options are minoxidil and finasteride, and a research peptide with no human exposure is not a third option. The fourth is practical: these are cyclised peptides with poor aqueous solubility. Laboratory practice often requires dilute acetic acid or a small amount of DMSO to get them into solution at all. That is not a formality - a vial that appears to dissolve in bacteriostatic water may be a suspension of undissolved peptide, and what you draw up may bear little relation to the labelled dose. The fifth: the family is heterogeneous and vendors are careless. Alpha-helical CRF(9-41), astressin, astressin-B and astressin-2B differ in potency by orders of magnitude and in receptor selectivity entirely. Product labelled as one may be another. There is no way for a buyer to tell.

• CRF1 receptor (CRHR1) — Astressin analogues bind with nanomolar affinity; specific Ki values vary by analogue and assay and were not resolved to a primary source in this session

  Competitive antagonism. Blocks the pituitary ACTH response to CRF and the anxiogenic and arousal effects of central CRF in the amygdala, BNST and locus coeruleus. This is the receptor every failed clinical programme targeted.

• CRF2 receptor (CRHR2) — Astressin and astressin-B are non-selective; astressin-2B was engineered for CRF2 selectivity

  Antagonism. CRF2 has a more ambiguous role - often opposing CRF1, involved in the recovery phase of the stress response, and important in cardiovascular and gastrointestinal function. Blocking it is not obviously desirable and non-selective blockade is one reason these compounds are blunt tools.

• Pituitary corticotrophs and the HPA axis — Downstream of CRF1 blockade

  Suppressed ACTH and therefore suppressed cortisol. This is simultaneously the intended pharmacology and the principal hazard - a suppressed stress axis is what kills you during intercurrent illness, injury or surgery.

• Peripheral CRF receptors in skin and hair follicle — Not characterised

  Peripheral astressin-B pushed hair follicles from telogen into anagen in CRF-overexpressing mice without lowering circulating stress hormones (Wang et al., PLoS ONE 2011), implying a local skin mechanism. The origin of the entire astressin-B hair-loss interest.

• Colonic CRF receptors — Not characterised

  CRF drives colonic motility and secretion; antagonists block stress-induced colonic responses in animals. The basis for the irritable bowel syndrome literature around this target, which is arguably its most plausible remaining application.

• Coric, Feldman, Oren, Shekhar et al. - multicentre, randomised, double-blind, active comparator and placebo-controlled trial of the CRF1 antagonist pexacerfont in generalised anxiety disorder (Depression and Anxiety 2010) Phase 2 · n=260 · 8 weeks · 2010

  Pexacerfont 100 mg/day versus placebo versus escitalopram 20 mg/day in generalised anxiety disorder. Pexacerfont did not separate from placebo on the primary outcome. Response rates were 42 percent (pexacerfont), 42 percent (placebo) and 53 percent (escitalopram); the active comparator separated cleanly, confirming assay sensitivity. Negative, and informative precisely because the trial was well designed enough that the negative result means something.

• Binneman, Feltner, Kolluri, Shi, Qiu and Stiger - a 6-week randomised placebo-controlled trial of CP-316,311, a selective CRH1 antagonist, in the treatment of major depression (American Journal of Psychiatry 2008) Phase 2 · n=123 · 6 weeks · 2008

  CP-316,311 at 400 mg twice daily versus sertraline 100 mg daily versus placebo in recurrent major depressive disorder. Stopped for futility at interim analysis with 89 evaluable patients: no separation from placebo on depression severity, while sertraline separated significantly. Safe and well tolerated, and ineffective.

• No human trial of any peptide CRF antagonist None

  Stated explicitly because the absence is the point. Astressin, astressin-B, astressin-2B and alpha-helical CRF(9-41) have never been given to humans in a published, controlled protocol. Every clinical trial at this target used small molecules. There is no human dose, no human safety data and no human pharmacokinetics for any peptide in this family.

In humans: unknown for every member of the family. No human has received a peptide CRF antagonist in a published protocol. In animals: effects on stress-induced behaviour appear within minutes to about an hour of central or systemic administration. Endocrine effects - suppression of the ACTH and corticosterone response to a stressor - are measurable within the same window. The cyclised analogues sustain effects over hours rather than minutes, which is what the conformational constraint was designed for. A useful anchor for how long the downstream consequences can run: in the astressin-B alopecia study, five daily peripheral doses produced hair regrowth that persisted for over four months. The peptide was long gone. What that illustrates is that a short course of a CRF antagonist can set off biological changes with a timescale entirely disconnected from the drug's own pharmacokinetics - which cuts both ways, and is a reason not to assume that stopping the compound promptly reverses whatever it started. For the HPA axis specifically: recovery from pharmacological suppression of the stress axis is not instantaneous. With corticosteroids, axis recovery after prolonged suppression takes weeks to months. Whether that applies to CRF-level blockade is unknown, but the assumption that you can stop and be fine the next day is not supported by anything.

There is no stack because there is no human protocol. What is worth saying is what makes the situation actively dangerous rather than merely unsupported. Do not combine with any corticosteroid, including inhaled or topical steroids at meaningful doses. Both suppress the HPA axis, from different points, and the combined suppression is worse than either alone. If you are on prednisolone, hydrocortisone replacement, or a high-dose inhaled steroid, a CRF antagonist is stacking suppression on suppression. Do not combine with anything else that blunts the stress axis - high-dose opioids, and in a more speculative way DSIP, whose one supported mechanism is reduction of CRF-induced adrenal output. The DSIP interaction is listed in the Core tier as synergy, and mechanistically that is correct, but synergy on this particular axis means synergy toward adrenal insufficiency. The circumstance that actually kills people with a suppressed HPA axis is intercurrent illness. Influenza, gastroenteritis with vomiting, a broken bone, an emergency operation. In all of those you need a cortisol surge, and if you have pharmacologically removed the ability to produce one and nobody knows you have, that is an adrenal crisis - hypotension, hyponatraemia, collapse. It is treatable with intravenous hydrocortisone if someone knows to give it. Nobody will know to give it if you bought the compound off a website. The only defensible use of a peptide CRF antagonist is in a laboratory, in an animal, as a pharmacological probe. That is what they were built for and they are very good at it.

Against small-molecule CRF1 antagonists: the small molecules are strictly better tools for the human question - orally available, brain-penetrant, and actually tested. They failed. The peptides are worse on every pharmacokinetic axis and untested. There is no argument for the peptide except availability, which is not an argument. Against an SSRI for anxiety or depression: escitalopram beat pexacerfont and placebo in the same trial. Sertraline beat CP-316,311 and placebo in the same trial. These are not indirect comparisons - the active comparators were in the same studies and they worked. That is about as clean an answer as clinical pharmacology produces. Against Selank: same therapeutic intent, wildly different positions. Selank is a registered medicine in Russia with a randomised active-comparator trial. Peptide CRF antagonists are bench reagents with none. The Core tier calls this pairing redundant, which is generous - it is not redundant, it is one compound with human data and one without. Against DSIP for the same axis: DSIP reduces CRF-induced corticosterone release in rats and has been given to humans in sleep-laboratory studies since 1981. It is a far cruder intervention at the same target, and it is also the one with human exposure. That is an unusual situation - the compound with less mechanistic elegance has more of the evidence that matters. Against minoxidil for hair: minoxidil is approved, costs almost nothing, has decades of human data, and showed partial efficacy in the same mouse experiment that made astressin-B famous. Where the target might still go: irritable bowel syndrome and stress-induced gastrointestinal dysfunction have arguably the most plausible remaining case for CRF antagonism, since the peripheral colonic CRF receptors are accessible without needing to cross the blood-brain barrier - which turns the peptides' main weakness into a non-issue. That is a real research direction. It is not a reason to inject one.

Deliberately null. These are laboratory reagents sold in microgram to low-milligram quantities for research use, and there is no human protocol against which a monthly cost could be calculated. Astressin-B and its relatives are expensive per milligram because they are long, cyclised, and difficult to synthesise. Any figure presented as a monthly cost for human use would be inventing a protocol that does not exist.

• ?No peptide CRF antagonist has ever been given to a human in a published controlled protocol. There is no human dose, route, safety data or pharmacokinetics for any member of the family.
• ?The sequence field is left empty and unverified because this is a family of related analogues rather than a single molecule, and I did not resolve the exact modified sequences of astressin, astressin-B or astressin-2B to a primary source in this session. The structural descriptions given come from the Gulyas 1995 and Koerber 1998 design papers.
• ?Specific Ki or IC50 values for astressin analogues at CRF1 and CRF2 were not resolved to a primary source in this session.
• ?Whether the hepatic enzyme elevation seen with several small-molecule CRF1 antagonists in development reflects the target or the chemotype is unresolved. Its relevance to peptide antagonists is unknown.
• ?How rapidly the HPA axis recovers after CRF-level blockade in humans is entirely uncharacterised. The comparison to corticosteroid-induced suppression is inference, not data.
• ?Clearance and half-life for astressin and astressin-B are not characterised in any species by a study I could resolve; the statement that cyclisation extends duration comes from the design rationale and animal behavioural duration rather than a measured pharmacokinetic curve.
• ?The astressin-B alopecia finding is in CRF-overexpressing transgenic mice. Its relevance to any human hair loss phenotype, including stress-related telogen effluvium, is completely unestablished.
• ?Whether peripheral administration of any astressin analogue produces meaningful central CRF receptor occupancy is doubtful and, as far as I can determine, unmeasured.
• ?Solubility requirements vary by analogue and the practical handling described here reflects general laboratory practice for cyclised peptides rather than a verified protocol for a specific product.
• ?The 383 total participants across the two negative trials is the sum of the randomised populations (260 plus 123); the Binneman trial stopped at interim with 89 evaluable.

• Multicenter, randomized, double-blind, active comparator and placebo-controlled trial of a corticotropin-releasing factor receptor-1 antagonist in generalized anxiety disorder Coric V, Feldman HH, Oren DA, Shekhar A, Pultz J, Dockens RC, et al., Depression and Anxiety, 2010 · PMID 20455246

  Depress Anxiety 27(5):417-25. The definitive negative trial at this target in anxiety, with a working active comparator. If you read one thing before buying a CRF antagonist, read this.

• A 6-week randomized, placebo-controlled trial of CP-316,311 (a selective CRH1 antagonist) in the treatment of major depression Binneman B, Feltner D, Kolluri S, Shi Y, Qiu R, Stiger T, American Journal of Psychiatry, 2008 · PMID 18413705

  Am J Psychiatry 165(5):617-20. The parallel negative trial in depression, stopped for futility. Two well-run negative phase 2 trials at the same target is not bad luck.

• CRF receptor antagonist astressin-B reverses and prevents alopecia in CRF over-expressing mice Wang L, Million M, Rivier J, Rivier C, Craft N, Stenzel-Poore MP, Tache Y, PLoS ONE, 2011 · PMID 21359208

  PLoS One 6(2):e16377. The famous hair regrowth result. 5 micrograms per mouse for five days, regrowth retained over four months, no change in elevated circulating stress hormones. Read the model description carefully before extrapolating - these are CRF-overexpressing mice, not men with male pattern baldness.

• Potent, structurally constrained agonists and competitive antagonists of corticotropin-releasing factor Gulyas J, Rivier C, Perrin M, Koerber SC, Sutton S, Corrigan A, Lahrichi SL, Craig AG, Vale W, Rivier J, Proceedings of the National Academy of Sciences USA, 1995 · PMID 7479843

  PNAS 92(23):10575-9. The original astressin design paper from the Salk Institute. The conformational constraint strategy that made these compounds potent enough to be useful tools.

• Constrained corticotropin-releasing factor (CRF) agonists and antagonists with i-(i+3) Glu-Xaa-DXbb-Lys bridges Koerber SC, Gulyas J, Lahrichi SL, Corrigan A, Craig AG, Rivier C, Vale W, Rivier J, Journal of Medicinal Chemistry, 1998 · PMID 9836618

  J Med Chem 41(25):5002-11. The lactam bridge chemistry underpinning the astressin family, and the source for the structural modifications listed in this record.

• Behavioral effects of central administration of the novel CRF antagonist astressin in rats Spina MG, Basso AM, Zorrilla EP, Heyser CJ, Rivier J, Vale W, Merlo-Pich E, Koob GF, Neuropsychopharmacology, 2000 · PMID 10693150

  Neuropsychopharmacology 22(3):230-9. The behavioural characterisation of central astressin, from the Koob lab. Note the route: central administration, because peripheral peptide does not reach these receptors.

• Corticoliberin (CRF/CRH) precursor, UniProtKB entry P06850 UniProt Consortium, UniProtKB

  Source for the parent CRF peptide length, PCSK2 processing and C-terminal amidation described in this record. Resolved directly in this session.