PACAP-38: brain stress & pain signaling neuropeptide
A natural brain hormone that controls stress responses, pain, and nerve-cell survival; the lizard version is used as a lab research tool to study PTSD-like fear and migraine.
A researcher, an agent, or an algorithm wrote down the sequence and picked a target to hit.
An AI model like OpenFold3 or AlphaFold built a 3D structure and scored how well it fits the binding site.
A second contributor repeated the computation on their own hardware and the scores matched.
A chemistry service or a researcher ordered the sequence, it was manufactured, and mass spectrometry confirmed the right molecule was produced.
A binding or activity measurement confirmed that it actually does what the computer predicted — or didn't.
What this is
PACAP-38 (Pituitary Adenylate Cyclase-Activating Polypeptide, 38-residue form) is a naturally occurring neuropeptide found across all vertebrates, from fish to mammals. It acts on a specific brain receptor called PAC1 to trigger a cascade of signals that influence stress responses, memory, pain processing, and nerve-cell survival. The 38-residue sequence stored on this card was characterized from the Italian wall lizard (Podarcis sicula) by Valiante and colleagues (Brain Research, 2007); the lizard sequence is highly conserved relative to mammalian PACAP-38, differing only at a small number of positions, making it a useful cross-species tool for studying class B G-protein coupled receptor (GPCR) function. PACAP exists in two biologically active forms produced from the same precursor protein: PACAP-38 (the predominant form in the brain) and the shorter PACAP-27, which spans the N-terminal 27 residues of PACAP-38.
History
PACAP-38 was first isolated in 1989 by Miyata, Arimura, and colleagues at the Tulane University Health Sciences Center, who screened fractions of ovine hypothalamic tissue for their ability to stimulate adenylyl cyclase activity in cultured rat anterior pituitary cells (Miyata et al., Biochemical and Biophysical Research Communications, 1989). The peptide showed approximately 68% sequence homology with vasoactive intestinal peptide (VIP), placing it in the VIP/secretin/glucagon neuropeptide family. Within a year, the same group isolated the shorter PACAP-27 isoform from the same fractionation side-streams (Miyata et al., Biochemical and Biophysical Research Communications, 1990). The PAC1 receptor — the highly PACAP-selective receptor — was cloned in 1993 by Ogi and colleagues from a human pituitary cDNA library (Ogi et al., Biochemical and Biophysical Research Communications, 1993), with Pisegna and colleagues subsequently characterizing four functional splice variants of the human PAC1 receptor that fine-tune downstream signaling (Pisegna et al., Journal of Biological Chemistry, 1996). In the lizard model, Valiante and colleagues (2007) confirmed that PACAP and its PAC1 receptor are expressed in the brain of Podarcis sicula, with the sequence being highly conserved relative to mammalian orthologs, establishing the lizard as a useful non-mammalian model for studying PACAP biology.
What it does
PACAP-38 functions like a master regulator for nerve cells under stress. When released in the brain or peripheral nervous system, it binds primarily to PAC1 receptors on neurons and engages multiple survival and signaling programs simultaneously. In simple terms: it tells stressed neurons to stay alive, heightens the brain's sensitivity to threat-related cues, and in blood vessels, triggers dilation and mast-cell activation. In the hippocampus it promotes synaptic strengthening underlying fear memory consolidation; in the spinal cord and trigeminal system it sensitizes pain-processing circuits (Johnson et al., Frontiers in Cellular Neuroscience, 2020). Elevated PACAP-38 in the bloodstream has been linked to exaggerated startle responses and difficulties distinguishing danger from safety, particularly in women (Ressler and colleagues, Nature, 2011). Intravenous infusion of PACAP-38 in humans reliably provokes flushing, palpitations, and — in people prone to migraine — delayed migraine-like attacks several hours after infusion.
Evidence
- Human: Ressler and colleagues (2011, Nature, N=1,237) found a sex-specific association between elevated blood PACAP-38 levels and PTSD diagnosis and symptoms in women, with a SNP (rs2267735) in the PAC1 receptor gene (ADCYAP1R1) predicting PTSD symptom severity specifically in female subjects. Human provocation studies showed that intravenous PACAP-38 infusion induced migraine-like attacks in susceptible patients with a mean delay of approximately 6 hours; one double-blind crossover study reported headache in 11 of 12 migraine patients after PACAP-38 infusion versus 3 of 12 after placebo. A Phase II trial of AMG 301, a monoclonal antibody targeting the PAC1 receptor, did not demonstrate efficacy over placebo for reducing monthly migraine days, while the anti-PACAP-38 antibody Lu AG09222 is in Phase II development for migraine prevention.
- Animal: Mice lacking the PACAP gene failed to develop inflammatory pain (carrageenan model) or neuropathic pain (L5 spinal nerve transection), while retaining normal acute nociceptive responses, demonstrating PACAP's requirement for pain sensitization. PAC1 receptor knockouts showed impaired contextual fear conditioning. Exogenous PACAP-38 reduced stroke infarct volume and promoted functional recovery in rodent ischemia models (Cherait et al., Journal of Molecular Neuroscience, 2025).
- In vitro: PACAP-38 activates adenylyl cyclase, phospholipase C, MEK/ERK, and Akt pathways via PAC1 receptor coupling to Gαs and Gαq/11 (Liao et al., Current Topics in Medicinal Chemistry, 2019). In rat trigeminal ganglion cells, PACAP-38 induces transcriptomic changes linked to calcium signaling, TRPM8 ion channel upregulation, and mitochondrial Complex I suppression — changes also produced by the truncated form PACAP(6–38), suggesting involvement of a non-canonical receptor mechanism (Takács-Lovász et al., International Journal of Molecular Sciences, 2022).
Known effects
- Fear memory consolidation — Preclinical (rodent knockouts, exogenous peptide infusion); sex-differentiated effects in females vs. males
- PTSD-associated stress physiology — Human associative data (Ressler et al. 2011, Nature); sex-specific
- Migraine provocation — Human provocation studies (intravenous infusion); mechanism involves meningeal vasodilation and mast-cell activation
- Pain sensitization (neuropathic and inflammatory) — Preclinical (PACAP-knockout mice, spinal sensitization models)
- Neuroprotection — Preclinical (stroke, TBI, Parkinson's, diabetic neuropathy rodent models); no validated human translation yet
- Vasodilation and cardiovascular effects — Human infusion studies; flushing and transient blood pressure reduction observed
Safety signals
Human infusion studies conducted for provocation research report consistent adverse effects: flushing and warm sensations in nearly all participants, palpitations in a substantial proportion, and transient reduction in diastolic blood pressure. Long-lasting flushing and heat sensation following PACAP-38 infusion can be attenuated by oral antihistamine treatment, consistent with mast-cell activation. No serious adverse events were reported in ascending-dose human studies. Cherait and colleagues (2025) noted that PACAP-38's very short plasma half-life — under 5 minutes in vivo due to rapid peptidase degradation — limits systemic exposure during brief infusion protocols, and that peripheral VPAC1/VPAC2 receptor activation contributes to the cardiovascular side-effect profile. The peptide is not approved for therapeutic use in any jurisdiction; all human exposure data comes from controlled research infusion protocols.
Regulatory status
- US/EU: Not approved for any therapeutic indication. PACAP-38 is an investigational research tool peptide; no IND or EMA marketing authorization has been granted for this molecule itself.
- WADA: Pharmacological substances without governmental regulatory approval for human therapeutic use are prohibited under WADA's S0 category (non-approved substances). PACAP-38 falls within this category.
- Research use: Widely used as a research tool in provocation studies (migraine, PTSD) and as a pharmacological probe for class B GPCR signaling.
Mechanism
PACAP-38 binds with high affinity and selectivity to the PAC1 receptor (encoded by ADCYAP1R1), a class B GPCR, with at least 1000-fold greater potency than VIP at this receptor. PAC1 couples to both Gαs and Gαq/11, triggering two major downstream branches: (1) Gαs → adenylyl cyclase → cAMP → PKA → CREB phosphorylation, activating transcription of neuroprotective and synaptic plasticity genes; and (2) Gαq/11 → phospholipase C → IP3/DAG → intracellular calcium release and PKC activation (Liao et al. 2019). Additional signaling through MEK/ERK and PI3K/Akt pathways supports neuronal survival. The PAC1 receptor undergoes alternative splicing, producing at least four human variants (Pisegna et al. 1996; Blechman et al., Frontiers in Endocrinology, 2013) that differ in the third intracellular loop region (the "hip" and "hop" cassettes), altering the balance between adenylyl cyclase and phospholipase C coupling. PACAP-38 also binds VPAC1 and VPAC2 receptors — the shared VIP/PACAP receptors — with lower selectivity, which accounts for peripheral vasodilatory and mast-cell-activating effects. The lizard sequence characterized by Valiante and colleagues (2007) is highly conserved in its N-terminal receptor-binding domain, which is also the region shared with PACAP-27; the C-terminal extension unique to PACAP-38 (residues 28–38, KRYKQRVRNK in this card's sequence) modulates receptor selectivity and signaling bias relative to the shorter isoform.
Open questions
- Why are PACAP-38/PAC1 stress and PTSD associations female-specific, and how does estrogen regulate ADCYAP1R1 expression?
- Can the neuroprotective effects demonstrated across 30+ years of animal models (stroke, TBI, neurodegeneration) be translated to humans, and what delivery strategy can overcome the <5-minute plasma half-life and poor BBB penetration?
- What is the identity and pharmacology of the alternative PACAP-38 receptor implicated by the PACAP(6–38) findings in trigeminal ganglion cells?
- Does the anti-PACAP mAb approach (Lu AG09222) outperform anti-CGRP strategies for migraine prevention in head-to-head trials?
- How do the pharmacological profiles of the lizard PACAP-38 sequence differ, if at all, from human PACAP-38 at PAC1 and VPAC receptors?
Related peptides
- PACAP-27 — the N-terminal 27-residue form of PACAP, produced by alternative processing of the same precursor; comparable potency at PAC1R, slightly different receptor selectivity profile.
- VIP (Vasoactive Intestinal Peptide) — the closest structural relative (~68% homology); binds VPAC1 and VPAC2 with high affinity but has very low affinity for PAC1R, making it the functional contrast peptide for dissecting PAC1-specific effects.
- Secretin — a more distant member of the same VIP/secretin/glucagon family; acts on the secretin receptor (SCTR), another class B GPCR, primarily in the gastrointestinal tract.
Research directions for this peptide, selected from the current sources — hypotheses you can explore and model. None of it is proven yet; tap any one to see the full thinking.
Could giving extra PACAP-38 to someone with PTSD actually calm the overactive fear circuitry by making the receptor tune itself down, the way an alarm silences after being triggered too many times?
People with PTSD have very few medication options and most existing drugs carry side effects or risks. If PACAP-38 can quiet an overactive stress receptor by saturating it into silence, it might offer a new, biology-based treatment that addresses the root of the problem.
Could a slightly altered version of PACAP-38 block the same receptor it normally activates, offering a separate route to prevent migraines in people who do not respond to existing treatments?
About half of migraine sufferers do not fully benefit from the newest anti-CGRP medicines. Because the PACAP-PAC1R pathway runs alongside the CGRP pathway, a PACAP-38-derived blocker might reach those patients through a different mechanism.
Does the flexible, positively charged tail at the end of PACAP-38 fold only when it meets its target receptor, and could that help explain why it favors PAC1R over its close relatives?
If this holds up, drug designers could try to build PACAP-like compounds that lean toward the PAC1R pain and stress pathway with less activity at related receptors, which might help reduce side effects in treatments for PTSD or chronic pain.
If the floppy, positively charged end of PACAP-38 is replaced with a fat-like anchor, would the resulting peptide last longer in the body and get into the brain more easily, while still doing its protective job?
PACAP-38 breaks down too quickly in the body to use as a medicine right now. A version with a stabilizing tail swap could become a practical drug for protecting the brain after injury or in early neurodegeneration, benefiting patients who currently have very few options.
Could the extra positively charged segment on PACAP-38, which is absent from its shorter cousin PACAP-27, stick to sugar chains on neuron surfaces and concentrate the peptide where it acts?
If this surface-anchoring effect is real, it could help explain why PACAP-38 appears to protect neurons better than the shorter form, and might inform the design of neuroprotective drugs aimed at conditions like Alzheimer's disease or stroke.
▸full evidence table2 metrics
| metric | value | tool |
|---|---|---|
| ipTM | 0.7597019672393799 | openfold3-mlx |
| ranking score | 0.842347264289856 | openfold3-mlx |
▸structural qualityopenfold3
| metric | value | note |
|---|---|---|
| gpde | 0.869 | global PDE — lower = better |
| disorder | 0.194 | fraction disordered |
| chain pair ipTM (A, B) | 0.760 | interface quality |
▸3-letter notation
▸recipeopenfold3-mlx 0.3.1
| parameter | value |
|---|---|
| model | openfold3-mlx 0.3.1 |
| weights | aedd8f3eb814e392… |
| hardware | apple_m4_base_16gb |
| mlx version | 0.31.1 |
| python | 3.14.3 |
| random seed | 42 |
| msa strategy | colabfold |
| diffusion samples | 1 |
| runtime | 449s |
| predicted by | mlx@peptide |
| predicted at | 2026-04-23 |
python3 openfold3/run_openfold.py predict --query_json {query.json} --runner_yaml examples/example_runner_yamls/mlx_runner.yml --output_dir {output_dir} --num_diffusion_samples 1 ▸citationbibtex
@peptide{pep10588,
sequence = {HSDGIFTDSYSRYRKQMAVKKYLAAVLGKRYKQRVRNK},
target = {pac1r},
author = {peptidemodel},
year = {2026},
status = {synthesized}
}