
Last updated: 2026-10-29 · Written by: SaiyanMed Research Team · Reviewed against: 12 primary, registry and analytical sources, all linked below · Values marked calculated are our own arithmetic from the formulas, shown so you can check them
Quick answer: Selank is the heptapeptide Thr-Lys-Pro-Arg-Pro-Gly-Pro (TKPRPGP; C33H57N11O9; monoisotopic 751.4341 Da; average MW 751.87; CAS 129954-34-3).[1][5] Two capped versions are sold under similar names. N-acetyl Selank (Ac-TKPRPGP-OH; C35H59N11O10; 793.4446 Da) adds an acetyl group to the N-terminal amine. N-acetyl Selank amidate (Ac-TKPRPGP-NH2; C35H60N12O9; 792.4606 Da) also swaps the C-terminal acid for an amide.[5] The two capped forms differ by only 0.984 Da. A Belgian forensic lab separated all three by LC–MS/MS: the amide shifts every y-type fragment ion down by 0.984 Da.[5] The capping story is also different from Semax. Measured Selank breakdown in plasma runs about 90% from the C-terminus, by removing two residues at a time, and only about 10% from the N-terminus.[6] No peer-reviewed study has measured the stability of either capped Selank form. Acetate, TFA and HCl salts are not modifications. They change mass per milligram, not the peptide.
Key takeaways
- Selank = TKPRPGP: the tetrapeptide TKPR (tuftsin) extended at the C-terminus by PGP.calc. Sciensano describes the PGP extension as added to improve metabolic stability.[5]
- Three different molecules: Selank 751.4341, N-acetyl Selank 793.4446, N-acetyl Selank amidate 792.4606 Da (monoisotopic). Average MWs 751.87 / 793.91 / 792.93.[5] Calculated values match Sciensano to 0.0001 Da.
- A vendor label of 793.9 for the amidate is the free-acid N-acetyl mass. The amidate’s average MW is 792.93 (calculated).
- Breakdown: in the only quantitative dataset we found, about 90% of Selank was cleaved from the C-terminus (TKPRPGP → TKPRP → TKP + RP), with no free amino- or carboxypeptidase activity in vitro.[6]
- Calculated from that table (rat blood plasma, in vitro): Selank halves in about 1.4–1.6 min.
- Neither capped form is in the stability literature. N-acetyl Selank has a PubChem record and a UNII. We found no PubChem record for the amidate formula C35H60N12O9.[2][11]
- Selank has three basic sites (N-terminal amine, Lys, Arg). The N-acetyl forms have two. That sets the maximum acetate or TFA count and so the nmol per mg.
The three molecules side by side
| Property | Selank | N-acetyl Selank | N-acetyl Selank amidate |
|---|---|---|---|
| Notation | H-TKPRPGP-OH | Ac-TKPRPGP-OH | Ac-TKPRPGP-NH2 |
| Formula | C33H57N11O9 | C35H59N11O10 | C35H60N12O9 |
| Monoisotopic mass (Da) | 751.4341 | 793.4446 | 792.4606 |
| Average MW (calculated) | 751.87 | 793.91 | 792.93 |
| [M+H]+ (calculated) | 752.4413 | 794.4519 | 793.4679 |
| [M+2H]2+ (Sciensano precursor) | 376.724 | 397.729 | 397.237 |
| [M+3H]3+ (calculated) | 251.4853 | 265.4888 | 265.1608 |
| Basic sites (N-term, Lys, Arg) | 3 | 2 | 2 |
| Acidic site (C-terminal COOH) | 1 | 1 | 0 |
| Net charge at pH 7 (calculated) | ≈ +2 | ≈ +1 | ≈ +2 |
| pI (calculated) | ≈ 11.5–11.7 | ≈ 11.5–11.7 | none: no acidic group, cationic across pH 2–10 |
| nmol per mg free base (calculated) | 1330.0 | 1259.6 | 1261.2 |
| Retention time, Sciensano UHPLC (min) | 5.25 | 5.47 | 4.94 |
| Registry | PubChem CID 11765600; UNII TS9JR8EP1G | PubChem CID 133082488; UNII J3PM702O93 | no PubChem record found (checked 2026-10-29) |
Two details in this table are often missed. First, N-acetylation removes a positive charge while leaving the C-terminal negative charge, so N-acetyl Selank is the least cationic of the three at neutral pH. Second, the amidate has no acid group at all. A “pI” printed for it by an online calculator is an artefact of the algorithm. It stays positively charged until Lys and Arg deprotonate, far above any working pH (calculated).
Calculated: why 0.984 Da is the number to watch
Acetylation adds C2H2O (+42.0106 Da). Amidation swaps OH for NH2 (−0.9840 Da). So the amidate is 0.9840 Da lighter than N-acetyl Selank. At the [M+2H]2+ charge state the gap is 0.492 m/z units (397.2376 vs 397.7296). In ppm, 0.984 Da at 793 Da is about 1240 ppm. Any modern high-resolution instrument resolves that easily.calc.
The trap is isotopes. The M+1 peak of the amidate sits 1.0034 Da above its monoisotopic peak (the 13C spacing), only 0.0193 Da from the N-acetyl monoisotopic mass. Seven carbon-rich residues give an M+1 of about 41–43% of the main peak.calc. On a unit-resolution instrument, an amidate sample therefore shows a strong peak at the N-acetyl nominal mass, and a mix of the two looks like one compound with an odd isotope pattern. Check the monoisotopic peak at ≤10 ppm and the isotope ratio, not just the nominal mass.
MS/MS fingerprints: what Sciensano actually saw
Sciensano (Belgium’s public health institute) identified Selank in a seized sample and built custom reference standards (>95% purity) for Selank, N-acetyl Selank and N-acetyl Selank amidate.[5] Their main product ions, with our calculated assignments, are below.
| Ion (calculated assignment) | Selank | N-acetyl Selank | NA-Selank amidate | Calculated m/z (free acid / amide) |
|---|---|---|---|---|
| y1 (P) | 116.07 | 116.07 | 115.09 | 116.0706 / 115.0866 |
| Lys-related ion | 129.10 (base) | 129.10 (base) | 129.10 (base) | 129.1022 (C6H13N2O+) |
| y2 (GP) | 173.09 | 173.09 | 172.11 | 173.0921 / 172.1081 |
| b2−H2O (Ac-TK) or internal PR | — | 254.16 | 254.16 | 254.1499 / 254.1612 |
| y3 (PGP) | 270.14 | 270.14 | 269.16 | 270.1448 / 269.1608 |
| y4 − NH3 | 409.22 | 409.22 | 408.23 | 409.2194 / 408.2354 |
| y4 (RPGP) | 426.25 | — | — | 426.2459 / 425.2619 |
| y5 (PRPGP) | 523.30 | 523.30 | 522.31 | 523.2987 / 522.3147 |

What the table shows:
- Every y ion of the amidate is 0.98 lower (116→115, 173→172, 270→269, 409→408, 523→522). That is the C-terminal amide, observed directly. It is the most useful single check.
- Selank and N-acetyl Selank share their y ions. The acetyl is on the N-terminal residue, which y ions never include. Their MS/MS spectra look alike, so you tell them apart by precursor mass (+42.01 Da) and retention time.
- b ions are almost absent. With Pro at positions 3, 5 and 7 and the basic Arg near the C-terminus, the charge and cleavage sit on the C-terminal side. This makes the N-terminal acetyl hard to see in MS/MS. The base peak at 129.10 matches a Lys-containing ion (C6H13N2O+, 129.1022) and is shared by all three.calc.
- A note on 254.16: Sciensano labels this ion b2−H2O. For Ac-TK, b2−H2O calculates to 254.1499. The internal Pro-Arg ion calculates to 254.1612, which is closer to the reported 254.16. Because the table gives only two decimals, we treat the assignment as open. Either way it signals an N-terminal or internal fragment, and it does not change their identifications.calc.
Retention order: the amidate elutes before Selank, unlike Semax
On Sciensano’s reversed-phase method the amidate eluted first (4.94 min), Selank second (5.25) and N-acetyl Selank last (5.47).[5] Two effects explain the order. The amidate keeps two positive charges and loses the negative C-terminus, so it is the most cationic and most polar at acidic mobile-phase pH. N-acetyl Selank loses one positive charge, so it is the least charged and binds the column longest. Retention times are method-specific, so use them only against a co-injected reference on the same system. In the Semax case on the same method, the parent eluted first (8.70 min), then the amidate (10.26) and N-acetyl Semax (10.89). So “capped forms elute later” is not a general rule: in Selank’s case, losing the C-terminal negative charge outweighs the acetyl. See our Semax vs N-acetyl Semax amidate guide.
Telling Selank apart from a scrambled sequence
Mass cannot distinguish TKPRPGP from the isomer KTPRPGP: same formula, same mass. Sciensano solved this with enzyme digests.[5] Lys-C cut after Lys and gave TK ([M+H]+ 248.160) plus PRPGP ([M+2H]2+ 262.153). Arg-C cut after Arg and gave TKPR ([M+2H]2+ 251.160) plus PGP (270.144). Only TKPRPGP gives both patterns. For the amidate, the same logic predicts Ac-TK ([M+H]+ 290.1710) and PRPGP-NH2 (522.3147) from Lys-C. One digest therefore confirms both caps and the order.calc. This prediction is ours and has not been published.
How Selank breaks down: the measured data
The best quantitative source is a 2005 study from the Institute of Molecular Genetics (Moscow) using evenly tritium-labelled Selank.[6] In rat blood plasma, about 90% of biodegradation came from removing C-terminal dipeptides. This gave TKPRP first and then TKP and RP, which persisted. Dipeptidyl aminopeptidases accounted for about 10%. Amino- and carboxypeptidases were not involved in vitro. Breakdown in blood in vivo was similar to that in plasma in vitro, but minor products also came from aminopeptidase (4%) and carboxypeptidase (0.4%) action were detected. Rat plasma degraded Selank about three times faster than human plasma, and the sets of products were similar. A 2006 paper from the same group lists TKPRP, TKP, RP and GP as the major products.[7]

| Time (min) | TKPRPGP | TKPRP | TKP | RP | PRPGP |
|---|---|---|---|---|---|
| 0.3 | 895 | 109 | 0 | 0 | 12 |
| 1 | 731 | 151 | 39 | 38 | 37 |
| 1.5 | 555 | 171 | 82 | 69 | 46 |
| 2 | 410 | 164 | 125 | 123 | 53 |
| 2.5 | 321 | 145 | 137 | 185 | 35 |
| 4 | 158 | 118 | 166 | 207 | 37 |
| 23 | 8 | 6 | 19 | 31 | 5 |
Our calculations from the table:calc.
- Concentration falls from 895 to half (447.5 nM) between 1.5 and 2 min. Interpolated, that is about 1.56 min after the first sample. A first-order fit of the 0.3–4 min points gives t½ ≈ 1.43 min (k ≈ 0.49 min−1).
- By 4 min, 17.7% remains; at 23 min, 0.9%.
- TKPRP peaks at about 1.5 min and then declines. That is the signature of an intermediate cut again, consistent with a second dipeptide removal (RP).
- PRPGP (the N-terminal dipeptidyl product) never exceeds 53 nM, against 171 nM for TKPRP. That fits the 90/10 split.
- The table is rat plasma. Applying the abstract’s statement that rat plasma hydrolyses Selank about three times faster than human plasma gives roughly 4.3–4.7 min for human plasma. That is our extrapolation, not a measurement.
What the caps would and would not block
| Pathway (measured share) | Product | [M+H]+ (calc.) | N-acetyl blocks? | C-amide blocks? |
|---|---|---|---|---|
| C-terminal dipeptide removal (~90%) | TKPRP + GP | 598.3671 / 173.0921 | no: acts at the other end | plausible but untested for Selank |
| second C-terminal cut | TKP + RP | 345.2132 / 272.1717 | no | not applicable after the first cut |
| N-terminal dipeptidyl cleavage (~10%) | TK + PRPGP | 248.1605 / 523.2987 | expected (enzyme needs free amine), untested | no |
| aminopeptidase (in vivo, 4%) | single residues | — | expected, untested | no |
| carboxypeptidase (in vivo, 0.4%) | single residues | — | no | expected, untested |
This is the core difference from Semax. Semax degrades from both ends, so two caps address two measured routes. For Selank, N-acetylation addresses at most the 10% route. The protective case rests entirely on whether a C-terminal amide slows the dipeptidyl carboxypeptidase that removes GP, and nobody has published that measurement. Evidence from other peptides is mixed. In a full serum study of short antimicrobial hexapeptides, C-terminal amidation “makes little difference to its proteolytic degradation in human serum”, while N-terminal acetylation “significantly increases” protease resistance.[10] We would not extrapolate those numbers to Selank, but they caution against assuming that an amide equals protection.
For context only: Selank and its fragments inhibit enkephalin-degrading enzymes in human serum in vitro (IC50 about 20 µM in one study).[9] The same group reports Selank as more specific for carboxypeptidases and dipeptidyl carboxypeptidases.[8] Neither study tested capped forms.
Salt forms: same peptide, different milligrams
PubChem lists Selank acetate (C35H61N11O11, 811.9) and Selank diacetate (C37H65N11O13, 872.0) as separate records.[3][4] A salt changes how much of the powder is peptide. It does not change the sequence or the caps. The counter-ion count is limited by the basic sites: up to three for Selank and two for either N-acetyl form.calc.
| Form | Formula weight | Peptide content | nmol peptide per mg |
|---|---|---|---|
| Selank free base | 751.87 | 100% | 1330.0 |
| Selank ·1 acetate | 811.93 | 92.6% | 1231.6 |
| Selank ·2 acetate | 871.98 | 86.2% | 1146.8 |
| Selank ·3 acetate | 932.03 | 80.7% | 1072.9 |
| Selank ·3 TFA | 1093.94 | 68.7% | 914.1 |
| Selank ·3 HCl | 861.26 | 87.3% | 1161.1 |
| NA-Selank amidate free base | 792.93 | 100% | 1261.2 |
| NA-Selank amidate ·2 acetate | 913.03 | 86.8% | 1095.3 |
| NA-Selank amidate ·2 TFA | 1020.97 | 77.7% | 979.5 |
Across these forms, the same 1 mg can contain anywhere from about 914 to 1330 nmol of peptide, a 45% spread. That spread is larger than many of the effects people try to measure. Report the salt form and net peptide content with any data.
Identification decision tree
- Accurate mass of the precursor. 376.724 (2+) → Selank family, go to step 3. 397.730 (2+) → N-acetyl Selank. 397.238 (2+) → N-acetyl Selank amidate. Use ≤10 ppm and check that the monoisotopic peak is the lowest in its cluster.
- For 397.x: look at the y ions. 116/173/270/523 → free acid (N-acetyl Selank). 115/172/269/522 → amide.
- For Selank: confirm the sequence order with a Lys-C or Arg-C digest (TK + PRPGP; TKPR + PGP) to rule out scrambled isomers.
- Retention time against a co-injected reference on your own method. Order on Sciensano’s method: amidate < Selank < N-acetyl.
- Counter-ion and content: ion chromatography or 19F NMR for TFA, and amino-acid analysis for net peptide. Mass spectrometry cannot see the salt.
- Purity: HPLC-UV at 210–220 nm. Watch for TKPRP, TKP and their capped analogues as degradation markers.
Limits of each technique are covered in What mass spectrometry can and cannot confirm.
Myth vs evidence
| Claim seen online | What the evidence shows |
|---|---|
| “Selank has a half-life of 20–30 minutes.” | We found no primary source for this. The one quantitative plasma dataset (rat blood plasma, in vitro) gives a calculated halving time of about 1.5 min. Human plasma was reported to be about three times slower, which would still be only minutes.[6] |
| “N-acetyl Selank amidate MW 793.9.” | 793.9 is N-acetyl Selank (free acid). The amidate is 792.93 (average), 792.4606 monoisotopic. |
| “Selank, like Semax, is degraded mainly at the N-terminus, so N-acetylation protects it.” | About 90% of measured Selank breakdown is C-terminal dipeptide removal. |
| “The amide cap makes it fully protected.” | Untested for Selank. In other short peptides amidation made little serum difference.[10] |
| “Selank acetate is a different molecule.” | It is the same peptide plus acetic acid. Only mass per mg changes. |
| “PGP makes the C-terminus stable.” | PGP was added for stability,[5] yet GP and then RP are still removed as the main pathway.[6] |
Registry notes
- Selank: PubChem CID 11765600, CAS 129954-34-3, UNII TS9JR8EP1G.[1][11]
- N-acetyl Selank: PubChem CID 133082488 (C35H59N11O10, 793.9), UNII J3PM702O93.[2][12] One PubChem synonym, “Ac-Thr-Lys-Pro-Arg-Gly-Pro”, omits a proline. That is a registry text error. The structure and formula are seven residues.
- N-acetyl Selank amidate: no PubChem compound record for C35H60N12O9 as of 2026-10-29. The only peer-reviewed identity data we found are Sciensano’s.[5]
Glossary
- N-acetylation: an acetyl group (CH3CO–) on the N-terminal amine; +42.0106 Da; removes a positive charge.
- C-terminal amidation: –COOH replaced by –CONH2; −0.9840 Da; removes a negative charge.
- y ion: an MS/MS fragment that keeps the C-terminus; shifts with a C-terminal amide.
- b ion: a fragment that keeps the N-terminus; shifts with an N-acetyl.
- Dipeptidyl carboxypeptidase: an enzyme that removes two residues at a time from the C-terminus.
- Dipeptidyl aminopeptidase: an enzyme that removes two residues at a time from the N-terminus.
- Counter-ion: the acid (acetate, TFA, chloride) paired with basic sites in a peptide salt.
- Net peptide content: the mass fraction of a powder that is the peptide itself.
Frequently asked questions
Is N-acetyl Selank the same as N-acetyl Selank amidate?
No. N-acetyl Selank (Ac-TKPRPGP-OH, 793.4446 Da) has a free C-terminal acid. The amidate (Ac-TKPRPGP-NH2, 792.4606 Da) has a C-terminal amide. They differ by 0.984 Da and can be separated by accurate mass and by their y-ion series.
What is the molecular weight of N-acetyl Selank amidate?
The formula is C35H60N12O9. Monoisotopic mass 792.4606 Da; average MW 792.93 (calculated). A label of 793.9 corresponds to N-acetyl Selank, the free acid.
How is Selank degraded in plasma?
In the one quantitative study we found, about 90% of degradation removed C-terminal dipeptides (TKPRPGP to TKPRP, then TKP and RP), and about 10% came from dipeptidyl aminopeptidases. Rat plasma was about three times faster than human.
Does N-acetylation make Selank more stable?
It has not been measured. N-acetylation can only address the N-terminal routes, which made up about 10% of measured breakdown. The main C-terminal route is unaffected by an N-terminal cap.
How can a lab tell Selank, N-acetyl Selank and the amidate apart?
Use the accurate precursor mass ([M+2H]2+ 376.724, 397.730 and 397.238), then the y ions (the amidate’s are 0.98 lower), then retention time against a reference. Sciensano’s method eluted the amidate first and N-acetyl Selank last.
Is Selank acetate a different compound from Selank?
No. It is Selank paired with acetic acid. The peptide is identical; only the formula weight and nmol per mg change (for example 1330 nmol/mg for the free base vs about 1232 for the mono-acetate, calculated).
Why does the amidate have no isoelectric point?
Without a C-terminal acid it has no acidic group, so it stays positively charged across normal working pH. Calculated pI values for the free-acid forms are about 11.5–11.7.
Can mass spectrometry confirm the Selank sequence order?
Not from the mass alone, since scrambled isomers have the same mass. Sciensano used Lys-C and Arg-C digests to confirm the order TKPRPGP.
Where to go next
For the general profile of the parent peptide, read What is Selank. For how sequences, fragments, terminal modifications and salt forms fit together, see the peptide identity hub.
References
- PubChem CID 11765600: Selank (Thr-Lys-Pro-Arg-Pro-Gly-Pro), CAS 129954-34-3
- PubChem CID 133082488: N-acetyl Selank (C35H59N11O10)
- PubChem CID 155489759: Selank acetate (C35H61N11O11)
- PubChem CID 155804769: Selank diacetate (C37H65N11O13)
- Vanhee C, Francotte A, Janvier S, Deconinck E. The occurrence of putative cognitive enhancing research peptides in seized pharmaceutical preparations. Drug Test Anal 2020;12:371–381 (Sciensano author copy; PubMed 31667971)
- Zolotarev YuA, Sokolov OYu, Kozik VS, et al. Pharmacokinetics of peptide anxiolytic Selank in rat blood and brain after intranasal administration (poster P.3.026). Eur Neuropsychopharmacol 2005;15(Suppl 3):S154
- Zolotarev YuA, Dadaian AK, Dolotov OV, et al. Evenly tritium-labeled peptides and their in vivo and in vitro biodegradation. Bioorg Khim 2006;32:183–191
- Zolotarev YuA, Sokolov OYu, Kost NV, et al. Leu-enkephalin homogeneously labeled with tritium in studying the Selank inhibiting effect on the enkephalin-degrading enzymes of human plasma. Bioorg Khim 2004;30:234–240
- Kost NV, Sokolov OYu, Gabaeva MV, et al. Semax and selank inhibit the enkephalin-degrading enzymes from human serum. Bioorg Khim 2001;27:180–183
- Nguyen LT, Chau JK, Perry NA, et al. Serum stabilities of short tryptophan- and arginine-rich antimicrobial peptide analogs. PLoS One 2010;5:e12684
- FDA Global Substance Registration System, UNII TS9JR8EP1G: Selank
- FDA Global Substance Registration System, UNII J3PM702O93: N-acetyl Selank
Research Use Only Disclaimer
This article is for laboratory and analytical education only. The compounds discussed are research chemicals for in-vitro research use only (RUO). They are not drugs, foods, supplements or cosmetics and are not for human or animal use. Nothing here is medical advice, a protocol for administration, or a claim about any effect. Animal, cell and analytical literature is cited for scientific context only.