Sterile, endotoxin-free and pyrogen-free are three different claims, and a material can satisfy one and fail another. Sterility is about living organisms. Endotoxin is about a molecule that survives after the organisms are dead. Pyrogen is the broadest of the three, covering anything that provokes a fever response. Autoclaving delivers the first and does nothing for the second, which is the single most useful thing to know about the group.
What each term actually covers
| Term | What it refers to | Removed by autoclaving? | Typical test |
|---|---|---|---|
| Sterile | No viable organisms | Yes | Growth-based sterility testing |
| Endotoxin-free | Below a stated limit of bacterial lipopolysaccharide | No | LAL, USP <85> |
| Pyrogen-free | Below a limit for anything fever-inducing | Partly | Monocyte activation test |
Endotoxin is lipopolysaccharide from the outer membrane of Gram-negative bacteria. It is a structural molecule rather than a living thing, it is thermally stable well beyond autoclave conditions, and it is biologically active at picogram quantities in sensitive systems. Killing the bacteria leaves it behind — intact, and now released from the cell that carried it.
Why does autoclaving not remove endotoxin?
Because it is not alive and it is not fragile. An autoclave cycle at 121 °C denatures proteins and kills organisms, but lipopolysaccharide requires substantially harsher treatment — dry heat at 250 °C for extended periods is the usual depyrogenation route, which glassware tolerates and a peptide solution does not. Worse, autoclaving a contaminated solution lyses the bacteria and liberates endotoxin that was bound in intact cell walls, so the result can be a sterile solution with a higher free endotoxin level than before. Sterile and endotoxin-free are independent properties.
Does filtering through 0.22 micron remove endotoxin?
No, and this is the assumption that most often goes wrong. A 0.22 µm membrane retains bacteria, which is what it is for. Endotoxin molecules are orders of magnitude smaller and pass straight through; they also aggregate into micelles of variable size, so retention is unpredictable rather than simply absent. Sterile-filtering a solution produces a sterile solution whose endotoxin content is essentially unchanged. Removing endotoxin requires a method aimed at it specifically — affinity resins, ultrafiltration with an appropriate cut-off, or starting from materials that never carried it.
How endotoxin is measured
The standard assay is LAL, based on a clotting cascade in horseshoe crab amoebocyte lysate that is triggered by lipopolysaccharide with remarkable sensitivity. Results are reported in endotoxin units per millilitre or per milligram, against a reference standard.
Two details make a reported figure readable. First, the units: EU/mL describes a solution and EU/mg describes the material, and converting between them requires knowing the concentration — so a figure quoted without its basis is ambiguous. Second, the method: USP <85> is the compendial chapter, and a certificate citing it is describing a defined procedure with defined controls rather than an in-house variant.
What endotoxin limit should a peptide meet?
It depends entirely on the application, which is why a bare number is less useful than it looks. A compound used in a cell culture experiment measuring immune activation needs a far lower limit than one used in a chemical assay, because endotoxin activates the pathways being measured — at picogram levels it will produce a response that reads as the compound's effect. The right question is not "is this low" but "is this low relative to what the readout is sensitive to". A certificate reporting a measured value against a stated specification lets that judgement be made; one reporting "endotoxin-free" with no figure does not.
Why does endotoxin matter so much in cell-based work?
Because it is an agonist in its own right. Lipopolysaccharide is recognised by TLR4 and drives a strong inflammatory signalling programme, so a contaminated preparation produces cytokine release, NF-κB activation and transcriptional changes that look exactly like a biologically active compound. This is the classic false positive in peptide cell work: an apparently immunomodulatory result that is the contaminant rather than the molecule. For compounds studied for immune activity specifically, it is the first control to run rather than the last. It is also why water grade matters — laboratory deionised water can be chemically pure and still carry endotoxin, because purity and endotoxin are different specifications.
Where contamination enters
Rarely from the synthesis itself. Solid-phase peptide synthesis runs in organic solvents that are inhospitable to bacteria, so a crude peptide is usually a low-endotoxin starting point. The entry points are downstream: water used in purification and lyophilisation, glassware and plasticware, buffers, and handling after the vial is opened.
That last one is worth emphasising because it is the one under your control. A certificate describes the material as it was released. Reconstituting with water that carries endotoxin introduces it, and no certificate can speak to that.
Is a sterile peptide the same as an aseptically handled one?
No, and the difference is about the future rather than the past. Sterility describes the material at a moment; aseptic technique is what keeps it that way through every subsequent entry. A sterile vial entered with a used needle is no longer sterile, and nothing about its certificate changes to say so. This is the practical reason a preservative-containing diluent exists for multi-entry vials — not to make technique unnecessary, but to hold back growth introduced despite it.
Should every peptide certificate report endotoxin?
It is a reasonable expectation for anything intended for cell-based work, and its absence is informative rather than damning. Endotoxin testing is an additional assay with an additional cost, and not every research application needs it. What matters is knowing which situation you are in: a certificate silent on endotoxin means the figure is unknown, not that it is low. Reading each panel for the question it answers is what keeps an absent test from being read as a passed one.
All products referenced here are supplied for laboratory and research use only. They are not drugs, foods, supplements or cosmetics, and are not for human or veterinary use.
