The Semax research, organized.

What the research actually covers

The Semax literature divides into four areas. The deepest is neuroprotection in stroke models: multiple independent research teams have used rodent models of cerebral artery occlusion and shown Semax limiting inflammatory gene programs and preserving tissue around the injury site — affecting hundreds of genes in some studies. The second area is neurotrophic signaling: Semax rapidly and selectively boosts BDNF and NGF gene expression across multiple brain regions, with protein-level increases measurable within hours. Third comes a growing Alzheimer's-relevant line: two recent studies show Semax stripping copper from amyloid-beta complexes in vitro, cutting amyloid toxicity, and a 2025 mouse study found a 2.8-fold reduction in cortical plaque load. Fourth, antidepressant-like and antistress effects in chronic-stress and early-life-isolation rodent models.

This page presents each finding as a numbered step-card showing the exact dose, species, route, and citation — the format that makes the evidence checkable at a glance. The signal is real; the gap to human clinical proof is also real, and the walkthrough is explicit about both.

Mechanism: the BDNF and neurotrophin axis

The defining mechanistic feature of Semax in the research literature is its reliable induction of BDNF (brain-derived neurotrophic factor) and related neurotrophins across multiple brain regions.

Finding 01 — Dolotov 2006: Intranasal Semax at 50 microg/kg produced rapid, dose-dependent BDNF protein increases in rat basal forebrain within three hours of administration, with effects sustained over 24 hours. Binding to specific ACTH-like receptor sites was confirmed ^[1]. This study established the mechanism's existence at the protein level.

Finding 02 — Agapova 2007: In intact male Wistar rats, a single intranasal dose of 50 microg/kg elevated hippocampal BDNF and NGF mRNA within one hour. The effect was region-specific: frontal cortex NGF mRNA decreased, suggesting the compound modulates neurotrophin gene transcription in a spatially selective manner rather than producing global upregulation ^[2]. The distinction between hippocampal induction and frontal cortical reduction is an important nuance for interpreting cognitive-enhancement claims.

Finding 03 — Dmitrieva 2009: In a permanent MCAO (middle cerebral artery occlusion) rat model, Semax at 100 microg/kg IP produced a time-dependent cascade of neurotrophin gene activation. BDNF and TrkC mRNA were elevated at three hours post-stroke; NT-3 and NGF followed at 24 hours; NGF was again elevated at 72 hours. The effects were cortex-selective and distinct from the activity of the PGP metabolite alone ^[3]. This temporal cascade is significant: the compound appears to initiate a sequenced neurotrophin response rather than a single undifferentiated signal.

Neuroprotection in stroke models: JNK, CREB, and the penumbra

The bulk of the Semax rodent literature uses the tMCAO (transient middle cerebral artery occlusion) model — a standard experimental stroke in which a major cerebral artery is temporarily blocked, creating a core infarct surrounded by salvageable penumbral tissue. Semax has been studied in this model across multiple independent research groups, consistently showing suppression of pro-apoptotic and pro-inflammatory signaling.

Finding 04 — Sudarkina 2021: In a tMCAO rat model (90-minute occlusion), Semax at 100 microg/kg IP (administered at occlusion, 1.5 hours, and 5 hours post-reperfusion) reduced phosphorylated JNK by more than 1.5-fold in cortex and subcortex at 24 hours post-ischemia. Simultaneously, pCREB increased by more than 1.5-fold in subcortical structures, and MMP-9 and c-Fos were each reduced by more than 1.5-fold in the frontoparietal cortex ^[4]. The simultaneous JNK suppression and CREB enhancement — pro-apoptotic down, pro-plasticity up — is the structural basis for the compound's neuroprotective characterization.

Finding 05 — Medvedeva 2014: A genome-wide transcriptional analysis in a focal cerebral ischemia rat model found that Semax at 100 microg/kg IP (administered at 15 minutes, 1, 4, and 8 hours post-stroke) altered expression of 96 genes at three hours and 68 genes at 24 hours. More than 50% of altered genes were involved in immune processes, including upregulation of chemokines CXCL13, CXCL9, CXCL10, CCL5, CCL7, and CCL19. Twenty-four vascular genes were altered at three hours, and 12 calcium-regulation genes were modulated ^[5]. The scale of transcriptional reprogramming documented here — nearly 100 genes in three hours — illustrates why Semax is characterized as having broad immunomodulatory and vascular effects, not merely localized neuroprotection.

Finding 06 — Filippenkov 2020: In another tMCAO rat model, Semax at approximately 100 microg/kg IP (at 90 minutes, 2.5 hours, and 6.5 hours post-ischemia) identified 394 differentially expressed genes at 24 hours. The pattern showed suppression of inflammatory gene programs (Ccl6, Ccl9, Fos, Hspa1a/b) and activation of neurotransmission-related genes (Gpr6, Gpr88, Drd2, Cplx2, Gabra5) ^[6]. The shift from inflammatory to neurotransmission gene programs in peri-infarct cortex is a consistent theme across the transcriptional studies.

Finding 07 — Filippenkov 2023: A 2023 study using the same tMCAO model compared Semax to the newer ACTH(6-9)PGP analog, finding that Semax at 100 microg/kg IP modulated 131 differentially expressed genes in the dorsolateral frontal cortex at 4.5 hours post-stroke. MHC class II genes (RT1-Ba, RT1-Da, RT1-Db1) were specifically downregulated — an immunomodulatory mechanism shared with the ACTH(6-9)PGP analog, suggesting the broader ACTH-derived peptide family shares this immune-axis action ^[7].

Finding 08 — Romanova 2006: Intranasal Semax administered for six days following photoinduced focal cortical ischemia in rats significantly reduced infarct volume and improved memory retention in conditioned passive avoidance testing ^[8]. This dual outcome — smaller infarct plus better behavioral performance — in a model distinct from MCAO strengthens the neuroprotective and antiamnesic characterization across stroke model types.

Alzheimer's disease-relevant findings: amyloid-beta and copper chelation

Two independent research lines have converged on Semax as a potential modulator of amyloid-beta aggregation — an Alzheimer's-relevant mechanism independent of the stroke neuroprotection literature.

Finding 09 — Sciacca 2022: In vitro studies using differentiated SH-SY5Y neuronal cells and artificial membrane models found that Semax at 20-100 micromolar concentrations inhibited copper-induced amyloid-beta fiber formation in a concentration-dependent manner. At a 5:1 molar ratio (Semax:Abeta), amyloid fibril formation was nearly completely abolished. The mechanism: Semax forms Cu2+ complexes with a conditional dissociation constant of 1.3 x 10^-15 M, stripping copper from Abeta complexes. MTT assay showed up to 90% reduction in Abeta1-42-induced neuronal toxicity; membrane disruption was reduced from 68% (Abeta alone) to 15-31% (Semax-treated) ^[9].

Finding 11 — Radchenko 2025: In transgenic APPswe/PS1dE9 Alzheimer's-model mice, intranasal Semax at 50 microg/kg (every other day for one month, starting at six months of age) reduced cortical amyloid plaque load by 2.8-fold at 7.5 months and 2.2-fold at 8.5 months. The effect showed preferential reduction of small, newly-formed plaques under 100 square micrometers. Behavioral assessments showed improved performance in open field, novel object recognition, and Barnes maze testing ^[11].

A 2025 follow-up study extended the copper-chelation mechanistic work, confirming that Semax strips Cu2+ from amyloid-beta complexes, silences redox cycling, and reduces reactive oxygen species-driven neurotoxicity. Together, the two copper-chelation studies and the transgenic mouse study constitute a coherent Alzheimer's disease-relevant research line — though all data remain at the in vitro and animal-model stage.

Antidepressant-like findings and stress modulation

Finding 10 — Inozemtseva 2024: Male Sprague-Dawley rats subjected to chronic unpredictable stress were treated with Semax at 60 nmol/kg daily IP. Chronic stress produced anhedonia (reduced sucrose preference), body weight suppression, adrenal hypertrophy, and decreased hippocampal BDNF levels. Semax reversed all four measures: sucrose preference was restored, body weight normalized, adrenal mass normalized, and hippocampal BDNF returned to control levels ^[10]. The 2024 publication date means this is among the most recent independent replications of Semax's antidepressant-like and antistress profile.

Finding 14 — Volodina 2012: Neonatal isolation in rat pups — a developmental stress protocol — produces lasting disruptions in adrenocortical stress response, glycemic regulation, and food motivation. Semax at 50 microg/kg intranasal (postnatal days 15-28) corrected all three effects in adult animals: the corticosterone response to acute stress was normalized, fasting blood glucose was restored, and food motivation improved ^[14]. The fact that the peptide was administered during a critical developmental window and produced lasting adult-stage normalization suggests effects on neuroendocrine programming, not just acute symptom suppression.

Finding 16 — Kamensky 2006: Semax activated serotonergic systems in rodents, with significantly elevated striatal 5-HIAA (the primary serotonin metabolite) following administration. When co-administered with D-amphetamine, Semax dramatically enhanced extracellular dopamine levels and locomotor activity beyond amphetamine alone ^[16]. The monoamine modulation literature is consistent with — and may help explain — the antidepressant-like behavioral effects documented in the chronic stress models.

Spinal cord injury and the Oprm1 / USP18 / FTO pathway

Finding 12 — Liu 2025: The most structurally novel finding in the 2025 literature describes a newly identified mechanism for Semax in spinal cord injury recovery. In female C57BL/6 mice following T9-T10 spinal cord impact injury, Semax promoted functional recovery by targeting the mu-opioid receptor gene Oprm1, enhancing USP18-mediated deubiquitination of FTO protein, and consequently reducing lysosomal membrane permeabilization, neuroinflammation, oxidative stress, and pyroptosis (an inflammatory form of programmed cell death). Behavioral outcomes included improved Basso scores, footprint analysis, and inclined plane test performance ^[12].

This finding is significant for two reasons. First, it identifies an entirely different mechanistic entry point — the Oprm1/USP18/FTO deubiquitination axis — distinct from the BDNF/TrkB pathway and the immunomodulatory transcriptional effects documented in the stroke literature. Second, it demonstrates activity in a non-cerebral injury model (spinal cord, not brain), broadening the documented scope of the compound's neuroprotective effects beyond ischemic stroke.

The USP18/FTO pathway connection is particularly novel: USP18 is a deubiquitinating enzyme with known roles in interferon signaling; FTO is an RNA demethylase involved in m6A RNA modification. Semax's apparent ability to modulate this axis — via the mu-opioid receptor — represents a mechanistic territory not previously associated with ACTH-derived peptides and warrants independent replication.