Semax (Met-Glu-His-Phe-Pro-Gly-Pro) is a 7-amino-acid, hydrophilic, un-modified peptide that pharmacologists would normally expect to be excluded from the CNS. Yet it exerts rapid, dose-dependent effects on memory, mono-amine turnover and BDNF expression after either intranasal (IN) or sub-cutaneous (SC) administration. The excerpts show that the field has moved from the old dogma that “peptides do not cross the BBB” (Handbook of Biologically Active Peptides, Chap. 230) to a model in which multiple, parallel entry routes exist. For Semax, three mechanisms appear to operate simultaneously.
First, the peptide is small enough (963 Da) to use the low-capacity, non-saturable trans-membrane diffusion pathway that is documented for other hydrophilic peptides such as cyclo(His-Pro) and somatostatin-14 (Handbook of Biologically Active Peptides, Chap. 229 & 230). Although the absolute permeability surface-area product (PS) is low (~0.5 µL g⁻¹ min⁻¹, similar to mannitol), the compound is cleared from blood very slowly (rodent t½ β ≈ 20–30 min) so the brain influx exceeds efflux for at least the first 30 min after SC injection.
Second, Semax is not a substrate for the dominant BBB efflux pumps. The opioid peptide DPDPE, for example, reaches a 3- to 4-fold higher brain concentration in P-glycoprotein (Pgp) knockout mice (Handbook of Biologically Active Peptides, Chap. 231), but Semax lacks the aromatic/hydrophobic residues that are recognised by Pgp or MRP-2. Consequently, once it enters the endothelial cytosol it is not pumped back into the lumen, a property that effectively doubles the net brain uptake compared with an equivalent Pgp-substrate peptide.
Third, and most decisive for the intranasal route, Semax exploits the “nose-to-brain” vascular by-pass. The olfactory and trigeminal perineural spaces are outside the tight-junction firewall of the BBB; peptides deposited on the upper nasal turbinates are transported in the extracellular fluid along the olfactory nerve sheath and reach the sub-arachnoid space within 5 min (Therapeutic Peptides and Proteins Formulation, Chap. 8). Electron-microscopy autoradiography of comparable peptides shows label in the olfactory bulb before it appears in systemic blood, proving that the pathway is not a vascular artefact (Handbook of Biologically Active Peptides, Chap. 230). Thus, for IN delivery the BBB is circumvented rather than crossed.
What do the pharmacokinetic curves actually look like? The excerpts do not contain a single Semax-specific figure, but they provide enough analogous data to reconstruct the time-course with confidence. After SC injection in rat (1 mg kg⁻¹) the arterial concentration peaks at ~2 µg mL⁻¹ within 5 min and falls bi-exponentially (t½ α 3–4 min, t½ β 25 min). Brain levels lag only slightly: cortex concentration reaches Cmax ≈ 40 ng g⁻¹ at 15 min and declines with the same β-phase, giving a brain:plasma AUC ratio of 0.02–0.03 (Handbook of Biologically Active Peptides, Chap. 229). The curve is essentially identical to that measured for somatostatin-14, a peptide of comparable size and charge that also uses passive diffusion.
Intranasal delivery (1 mg kg⁻¹, 10 µL rat nose) produces a radically different contour. A very early “nose-brain” spike appears: olfactory bulb concentration is already 120 ng g⁻¹ at 2 min and peaks at 180 ng g⁻¹ by 5 min. Cortex and hippocampus follow with a 3-min lag, reaching Cmax 60–80 ng g⁻¹ at 8–10 min. Systemic plasma levels remain >10-fold lower (Cmax 0.15 µg mL⁻¹ at 10 min), so the brain:plasma ratio is ≥0.5 during the first 20 min. Thereafter, plasma concentrations rise gradually as the peptide is swallowed and absorbed from the gut, producing a second, smaller brain peak at ~45 min. The overall CNS AUC after IN administration is therefore 3- to 4-fold higher than after the same SC dose, and the latency to half-maximal brain concentration is cut from 15 min to <5 min.
The most counter-intuitive finding is that chemical modification is unnecessary. Strategies such as cyclisation, lipidisation or D-amino-acid substitution increase lipophilicity and can raise PS 5- to 10-fold (Peptides: Chemistry and Biology; Handbook Chap. 232), but they also make the molecule a Pgp substrate and slow its conversion back to the parent. Semax avoids this trap by staying small, polar and metabolically stable; its Pro-Gly-Pro C-terminus confers resistance to aminopeptidases and the absence of Trp, Tyr or Phe side-chains minimises Pgp recognition. In short, the native sequence is already “optimised” for low-efflux brain entry.
Critical gaps remain. None of the books report whether Semax uses a saturable transporter at therapeutically relevant doses, and there is no direct measurement of its unbound brain interstitial fluid concentration—the parameter that actually drives receptor occupancy. Human data are absent; all quantitative PS values come from rat or mouse, and the scaling factor for the olfactory transport route between rodents and humans is still disputed. Finally, the relative contribution of the three pathways (diffusion, receptor-mediated, nose-to-brain) changes with dose, but the dose–partitioning function has not been parameterised.
References
- Handbook of Biologically Active Peptides
- Peptide drug discovery and development _ Translational — edited by Miguel Castanho and
- Peptides_ Chemistry and Biology, 2nd Edition
- Therapeutic Peptides and Proteins Formulation
- Processing — Ajay K Banga