The HPA axis and cognitive peptides: how selank and semax may modulate stress response

The body’s stress response is a three-gland chain. The hypothalamus puts out CRH. The pituitary, on the CRH signal, releases ACTH. The adrenal cortex, on the ACTH signal, makes cortisol. That chain is the HPA axis — hypothalamic-pituitary-adrenal — and most of what people mean by the stress response runs through it.

Two peptides from the Russian neuropeptide research tradition get marketed as the stress-and-cognition layer to that biology: selank and semax. Neither sits in the part of the chain the marketing implies. Here is what each one actually engages and where the evidence stands.

What the HPA axis is and why peptides come up in conversation with it

A short-term stressor — a deadline, a near-miss, a hard set of intervals — pushes the hypothalamus to release CRH. The pituitary reads CRH and releases ACTH into the bloodstream. The adrenal cortex reads ACTH and releases cortisol. Cortisol is what your body does to mobilise energy, suppress non-urgent housekeeping, and tighten attention onto the threat.

Cortisol is also what the chain shuts itself off with. The same cortisol acts on the hypothalamus and pituitary to dampen CRH and ACTH, closing the loop. A healthy axis turns on, does the job, and turns off.

Chronic stress is what happens when the loop fails to close cleanly. Cortisol stays high, the feedback weakens, and the downstream effects — sleep disruption, cognitive impairment, immune blunting, mood changes — start showing up as everyday symptoms. That is the biology behind a real demand for anti-stress therapeutics, and behind the marketing for peptides that claim to engage it.

The two peptides at the top of that menu are selank and semax. The first thing worth knowing about either is where, in that three-gland chain, they actually act.

Semax: structurally ACTH-derived, functionally not corticotropic

Semax is the more surprising of the two. Its structure is the ACTH(4-7) fragment — Met-Glu-His-Phe — with a Pro-Gly-Pro stabilising tail. That is, the first four amino acids of semax are lifted directly out of the pituitary hormone the HPA axis uses to drive cortisol release.

The first instinct on reading that — that semax must therefore raise cortisol — is wrong. The corticotropic activity of full ACTH lives in a different region of the molecule. The ACTH(4-7) sequence retains some of ACTH’s neural-receptor activity without the adrenal-stimulating component. Functionally, semax does not engage the cortisol arm of the axis at all.

What it does engage, in animal models, is downstream of the hypothalamus and pituitary. Dolotov and colleagues (2006, J Neurochem) reported that semax binds the ACTH receptor and increases brain-derived neurotrophic factor (BDNF) protein in rat basal forebrain and hippocampus — the two brain regions most associated with attention and memory consolidation under stress. Subsequent animal work added dopaminergic and serotonergic modulation, and a more recent 2025 paper in Br J Pharmacol extended the mechanism story to engagement of the μ-opioid receptor and functional recovery after spinal cord injury in a rat model.

The honest mechanistic frame: semax is structurally derived from the HPA-axis molecule, not a HPA-axis activator. It engages neurotransmitter and growth-factor systems in animals. Where that sits relative to the human stress experience is the part the published evidence does not yet span.

Selank: a tuftsin analog acting upstream of the receptor

Selank’s structural lineage is different. The parent molecule is tuftsin, the immunomodulatory tetrapeptide Thr-Lys-Pro-Arg that the body produces by enzymatic cleavage of immunoglobulin G. Selank takes the tuftsin sequence and appends the same Pro-Gly-Pro extension that semax carries.

Tuftsin is not a stress hormone. It is an immune signal. The selank mechanism story, developed in Russian animal work, runs through the brain’s serotonergic and GABAergic systems — the same neurotransmitter axes a benzodiazepine like phenazepam acts on, but proposed to engage upstream of the benzodiazepine receptor itself. In rats, selank produces anxiolytic behavioural effects without the sedation or cognitive impairment characteristic of benzodiazepines.

Where that interacts with the HPA axis is indirect. GABA tone modulates CRH release at the hypothalamus; serotonin modulates HPA reactivity through the limbic system. A compound that quiets anxiety-correlated activity in those systems can dampen the upstream signal that drives the cortisol response, without binding any HPA-axis receptor directly. The mechanism plausibly cushions HPA reactivity rather than mimicking or blocking any of its three glands.

The Russian clinical files behind both

Selank and semax both sit in the Russian neuropeptide research tradition, and most of their clinical-use evidence is in Russian-language journals not always indexed for Western RCT methodology details.

For semax, Gusev and colleagues (1997, Zh Nevrol Psikhiatr Im S S Korsakova) is the foundational stroke trial — a clinical and electrophysiological assessment of semax in acute hemispheric ischaemic stroke that grounded the 1999 Russian Ministry of Health authorisation. Direction reported as positive in the Russian clinical/electrophysiological framework. The trial is what a Russian regulator accepted as evidence; it is not equivalent to a Western multicenter RCT with hard primary endpoints, and the published abstract leaves the design details thin.

For selank, the comparable file is Medvedev and colleagues (2014, Zh Nevrol Psikhiatr Im S S Korsakova). Seventy adults with generalised anxiety disorder and adjustment disorder were randomised between selank and phenazepam, a Russian benzodiazepine. Reported result: comparable anxiolytic effect to phenazepam, with better tolerability — no sedation, no cognitive impairment, no dependence signal. The design caveats matter: this was an open-label Russian-language head-to-head study with no placebo arm, classified mixed by Western RCT standards. The trial did not reach the placebo-controlled bar that would let it stand as evidence of absolute efficacy. What it shows is selank’s performance against an active benzodiazepine comparator. What it does not show is selank’s performance against an inert control — and the gap between the two is the missing data point.

Mendias and Awan (2026, Sports Medicine) places both compounds in the broader sports-medicine framing — non-approved peptides operating outside FDA-recognised drug application frameworks, with scarce Western controlled human safety and efficacy data. That is the regulatory shape of where the buyer’s evidence base actually sits.

Same family, two files

What the HPA-axis label collapses is two different molecules with two different mechanism stories, two different Russian clinical contexts, and two different regulatory positions.

Semax’s file is a structural ACTH(4-7) backbone, an English mechanism paper showing BDNF and ACTH-receptor binding in rats, a 1997 Russian stroke trial that grounded the Russian authorisation, and a July 24, 2026 FDA PCAC docket slot for the reviewed indications cerebral ischemia, migraine, and trigeminal neuralgia — not cognitive enhancement, not stress management, not the nootropic case the consumer market sells the molecule for. The prescribing question for semax in the US, post-July, will be about its three FDA-evaluated neurological indications.

Selank’s file is a tuftsin-analog backbone, an animal mechanism story in GABA and serotonin systems, a 2014 Russian open-label comparison to a benzodiazepine that is the best-published clinical comparison available, and no PCAC docket slot at all. Selank is neither on the July 2026 Wave 1 docket nor on the February 2027 Wave 2 docket. Its US regulatory position runs entirely through Section 503A compounding under state pharmacy authority, with no FDA evaluation of any specific indication scheduled. The prescribing question for selank in the US is structurally open — and likely to stay that way until either a Western trial reads out or the FDA picks the molecule up.

Two peptides, two evidence files, two different conversations with a supervising physician. Wolverine Health is being built so those conversations actually happen — physician-supervised peptide protocols, US-licensed compounding pharmacies, every batch third-party tested. We covered how PCAC reviews actually work in What is the PCAC. Join the waitlist for a note the day either file moves.