Adsorption and Labware Surface Loss: Where the Certificate Stops
A batch certificate makes one quantitative promise about content, and it is a promise about the vial. Our KPV 10MG reference material certifies 9.57 mg of content mass against a 10.00 mg expected value on report DBAV-KPV-10-062226, lot 5814, at 98.70% purity by HPLC-UV. That number was produced by a third-party laboratory weighing and running the contents of a vial from that lot.
Every number downstream of it belongs to whoever opens the vial. And there is a well-documented, quantitatively large mechanism that operates entirely in that downstream territory, removes peptide from solution before any measurement is taken, and appears on no certificate anywhere in this industry: the peptide sticks to the container.
What was actually measured, and in what
The most systematic study in this area is by Kristensen, Henriksen and Andresen, published in PLoS ONE in 2015. They quantified adsorption of three peptides — mastoparan X, melittin and magainin 2 — to borosilicate glass vials, standard polypropylene tubes, Protein LoBind tubes and quartz cuvettes, using analytical HPLC with lipid vesicles added at the final step to release peptide from the autosampler insert walls so that the loss being measured was the loss that had already happened.
The headline result is stark. In 220 µL solutions incubated for one hour in 10 mM HEPES, 100 mM NaCl at pH 7.4, “for 1 µM peptide solutions incubated in borosilicate glass vials or polypropylene tubes, only 10–20% peptide was recovered.” The paper’s abstract puts the general case at “90% or more of the peptides might be lost from solution due to rapid adsorption to the walls of the sample containers.”
Before that number gets quoted anywhere, note what the three peptides are. Mastoparan X, melittin and magainin 2 are α-helical cationic membrane-active peptides — antimicrobial and cell-penetrating compounds selected for study precisely because they bind to amphipathic surfaces. They are, by construction, close to a worst case. Chico, Given and Miller had shown as early as 2003 in Peptides that for cationic cell-permeable constructs the binding to plastic and glass “was due entirely to the cationic vector moieties,” and warned that under some conditions this non-specific binding “could be mistaken for cellular penetration.”
That is the honest frame for the whole literature: the direction is general, the magnitude is class-specific, and a supplier who quotes “80 to 90 percent loss” as a flat fact about research peptides is over-reading a paper about three membrane-lytic peptides.
The concentration effect runs opposite to intuition
The most useful finding in the 2015 study is also the least intuitive. Recovery improved as concentration rose. Kristensen and colleagues observed “increasing percentages of peptide were recovered for increasing peptide concentrations, indicating that container walls became increasingly saturated with peptide.”
The wall has a finite number of binding sites. At high concentration those sites take a negligible fraction of the total; at low concentration they take most of it. Fractional loss is therefore worst exactly where it is hardest to notice — in the dilute working solution, at the end of a dilution series, in the well where the measurement happens.
Their own protocol reflects this: stock solutions were deliberately handled at “a high concentration of at least 100 µM” in low-binding tubes, specifically to saturate pipette tip walls and minimize relative loss during pipetting. The concentrated stock is not just convenient. It is the control.
Geometry, transfers, and speed
Three further results define the shape of the problem.
Surface-area-to-volume ratio matters. The same 2 µM solution recovered better at 2 mL than at 220 µL, because the ratio of wall to liquid fell. Quartz cuvettes were the worst container tested: recovery “was ≤ 50% for all three peptides” in 2 mL 2 µM solutions, with the magnetic stirring bar sharing the blame.
Losses compound across transfers. Solutions moved successively through up to four containers of the same type, with a one-hour incubation between each step. After four borosilicate glass vials or four polypropylene tubes, “only a small percentage of peptide, close to 0%, was recovered.” Each transfer presents a fresh unsaturated surface.
It is essentially instantaneous. Recovery at 10 seconds, 1 hour and 24 hours was only weakly different, so adsorption “is a fast process that occurs during the first few seconds after addition of peptide to the solutions, probably while the solutions are still being vortexed.” There is no window in which to act, and incubation time is not a variable worth optimizing.
Worse, some of it does not come back. Lipid vesicles added afterwards induced desorption quickly but incompletely from borosilicate glass, which the authors read as indicating mastoparan X “was partly irreversibly adsorbed to the walls,” noting that extended vortexing induced no further release.
Three fixes that do not work
Adjusting ionic strength. Solutions were prepared at 2, 100 and 149 mM NaCl. In glass there was at most “a slight trend” toward better recovery at higher salt; in polypropylene and low-binding tubes there was no clear dependence at all. The authors’ conclusion was that ionic strength cannot be tuned to prevent the effect.
Siliconizing. Goebel-Stengel, Stengel, Taché and Reeve tested eight radiolabeled endocrine peptides — ghrelin, sulfated CCK-8, corticotropin-releasing factor, GLP-1, insulin, leptin, nesfatin-1 and peptide YY — in Analytical Biochemistry in 2011 across treated and untreated glass and plastic over a 48-hour incubation. Their finding was direct: “siliconization decreased, whereas the addition of BSA improved recovery from surfaces tested.” A treatment widely assumed to help made things worse.
Pre-saturating the container. Kristensen’s team flagged the risk explicitly. Because adsorption is at least partly reversible, presaturated walls “could serve as a reservoir of peptides that might be released into solution in response to some specific experimental step,” producing a concentration higher than intended rather than lower.
What the data does support
Low-adsorption plasticware is the single best-supported intervention. Kraut and colleagues compared standard plastic, glass and low-adsorption plastic tubes over 28 days in the Journal of Proteome Research in 2009 and found standard plastic unsuitable over that period, glass better, and low-adsorption plastic optimal “whatever the peptide concentration stored” — with hydrophobic peptides showing the poorest recovery. Bark and Hook had reported the same direction in the same journal in 2007, finding “highly improved consistency and lower errors in quantitation” from low-retention tubes, along with qualitative differences in which peptides were detected at all.
But low-binding is a category, not a specification. Zhang and Gao evaluated microtube binding preference for nanoproteomics in 2022 and found “a protein recovery rate ranging from less than 10% to over 90% for different microtubes,” with storage temperature emerging as another key factor and binding preference tracking isoelectric point, hydrophobicity, length and charge. Two tubes both marketed as low-binding are not interchangeable.
Carrier protein remains the most effective single addition where the assay tolerates it. Goebel-Stengel’s group reported that lyophilizing solutions containing labeled peptide and 1% BSA in the container best suited to each peptide “rendered more than 89% recovery for all peptides” — recovering a set of compounds that had otherwise varied widely.
Where this sits relative to a certificate of analysis
It sits entirely outside it, and that is the point worth being plain about.
A certificate answers what is in the vial. It does not and cannot answer what reaches the assay, because that depends on the dilution series, the labware, the buffer and the number of transfers — all of which happen after the material leaves us. Our content uniformity piece covers the sampling question inside the vial; our reconstitution article covers the first dilution. This article covers every dilution after it.
There is a structural point in there too. A 50 mg fill diluted to 1 µM passes through more serial transfers than a 5 mg fill does, so larger reference vials carry a longer adsorption path, not a shorter one. A tripeptide such as KPV is cationic and short, which puts it on the same side of the ledger as the peptides in these studies — but nobody has published a recovery curve for it, and we are not going to invent one. What can be said is the direction and the mechanism, and that the fix is the researcher’s labware rather than anyone’s certificate.
The domestic-supply angle, stated narrowly
None of this is solved by where material ships from. What domestic supply does affect is the number of unmeasured intervals in a compound’s history. Material shipped same-day from our Santa Barbara facility, held by a single Wyoming entity with a batch certificate tied to the lot in the box, gives a US laboratory a documented starting mass and a short custody chain. Adsorption then begins in that laboratory’s own tubes, where it can be controlled by the choices above.
That is a narrower claim than the category usually makes, and it is the one the evidence supports.
References
- Kristensen K, Henriksen JR, Andresen TL — adsorption of cationic peptides to solid surfaces of glass and plastic. PLoS ONE, 2015; 10(5):e0122419. DOI 10.1371/journal.pone.0122419. PMID 25932639.
- Goebel-Stengel M, Stengel A, Taché Y, Reeve JR — the importance of using the optimal plasticware and glassware in studies involving peptides. Analytical Biochemistry, 2011; 414(1):38–46. DOI 10.1016/j.ab.2011.02.009.
- Kraut A, Marcellin M, Adrait A, Kuhn L, Louwagie M, Kieffer-Jaquinod S, Lebert D, Masselon CD, Dupuis A, Bruley C, Jaquinod M, Garin J, Gallagher-Gambarelli M — defining optimal storage conditions for proteomics samples. Journal of Proteome Research, 2009; 8(7):3778–3785. DOI 10.1021/pr900095u.
- Bark SJ, Hook V — differential recovery of peptides from sample tubes and the reproducibility of quantitative proteomic data. Journal of Proteome Research, 2007; 6(11):4511–4516. DOI 10.1021/pr070294o.
- Zhang Z, Gao Y — evaluation of the binding preference of microtubes for nanoproteomics sample preparation. Journal of Proteome Research, 2022; 22(1):279–284. DOI 10.1021/acs.jproteome.2c00477.
- Chico DE, Given RL, Miller BT — binding of cationic cell-permeable peptides to plastic and glass. Peptides, 2003; 24(1):3–9. DOI 10.1016/s0196-9781(02)00270-x.
- Maple Research Labs batch certificate DBAV-KPV-10-062226, lot 5814, issued by Testides Analytical; read directly from the certificate linked on the product page.
Citation metadata in references 1–6 was verified against PubMed records, and reference 1 was read in full text.
All materials supplied by Maple Research Labs are provided for research use only. They are not for human use, not for veterinary use, and not for diagnostic or therapeutic application. Nothing in this article describes administration to people.