What an endotoxin test measures, and what it misses
An endotoxin test measures the ability of a sample to trigger an enzymatic cascade that horseshoe crab blood — or a recombinant version of the first enzyme in it — runs when it encounters bacterial lipopolysaccharide. It is a biological activity assay, not a chemical assay, and this is the single most useful thing to understand about it. It does not identify a molecule, does not weigh anything and does not report a percentage of the sample. It reports how strongly the sample provoked a reaction, calibrated against a reference standard and expressed in endotoxin units. Because the readout is reactivity rather than composition, an endotoxin result is orthogonal to a chromatographic purity figure: neither one predicts the other, and a certificate reporting only purity has said nothing at all about endotoxin.
Why does a purity figure carry no endotoxin information?
Because the two measurements are looking for different things by different physics.
Reversed-phase HPLC with ultraviolet detection reports the areas of peaks that absorb at the detection wavelength. Lipopolysaccharide is not a peptide, does not behave like one chromatographically, and is not counted as an impurity peak in a peptide purity calculation. A sample can therefore return a very high area percentage and still carry endotoxin, because the purity method was never asked the question.
The reverse holds too. An endotoxin figure says nothing about whether the peptide is the right peptide, whether it has degraded, or how much of it is present. These are separate determinations that happen to appear on the same page when a laboratory runs both.
Why does endotoxin matter for in-vitro work?
Because it is biologically active in exactly the cell types most often used in laboratory research, and its effects are easy to mistake for the effects of the material under study. The published record contains worked examples of that mistake being made and then caught.
The clearest demonstration of this failure mode is old and still instructive. Brooks and colleagues studied particulate prosthetic materials by adding them to monocytic cells and measuring cytokine release — a standard design. Four cement formulations tested free of endotoxin by standard gel-clot assay nonetheless produced different levels of cytokine release from macrophages. A more sensitive kinetic assay showed that those differences correlated with minor endotoxin contamination, and when the experiment was repeated with particles carrying low or undetectable endotoxin by the kinetic method, the apparent differences between formulations disappeared entirely (J Bone Joint Surg Br, 2002;84(2):295–9; PMID 11922375, DOI 10.1302/0301-620x.84b2.12061). The reported biological difference between the materials was an artefact of contamination.
The same problem recurs in modern model systems. Heinrich and colleagues examined high endotoxin levels in commercially available gelatin used in a 3D bioprinted macrophage–cancer cell co-culture, and reported that endotoxin significantly influenced the metabolic activity of both cell types, provoked a strong inflammatory reaction in macrophages, inhibited the paracrine effects of the co-culture, and — the consequential finding — drastically altered the apparent efficacy of two macrophage-modulating compounds. They describe the presence of endotoxin in such biomaterials as "often an overlooked problem" that can lead to misinterpretation of potency (Biomater Adv, 2023;144:213220; PMID 36476713, DOI 10.1016/j.bioadv.2022.213220).
Both studies point the same way. Endotoxin does not merely add noise to a cell-based experiment; it can generate a clean, reproducible, entirely spurious result.
Can an endotoxin test miss endotoxin that is present?
Yes, and this is the part least often stated on a certificate.
The best-documented mechanism is masking, usually discussed as low endotoxin recovery. Reich and colleagues investigated endotoxin detection in samples containing non-ionic surfactants and found that the process is kinetically controlled and temperature-dependent, that low recovery arose specifically when non-ionic surfactants and metal-complex-forming components were present together, and that "even hazardous amounts of endotoxin can remain undetectable" in such compositions. Their conclusion is that this is genuine masking — the supramolecular structure of the endotoxin is altered so it no longer binds Factor C efficiently — rather than simple test interference (Biologicals, 2016;44(5):417–22; PMID 27464990, DOI 10.1016/j.biologicals.2016.04.012).
Adsorption is a second route to a falsely low result. Yokota and colleagues observed poor recovery of spiked endotoxin in the presence of a cationic, hydrophobic recombinant protein, and traced it to endotoxin adsorbing onto the glass of the incubation tubes, with the protein mediating the adsorption. They note the general implication: this may occur with any substance that binds both endotoxin and glass (J Pharm Biomed Anal, 2000;22(5):757–61; PMID 10815718, DOI 10.1016/s0731-7085(00)00292-2).
The Brooks study adds a third: endotoxin adsorbed onto the surface of particles was not detected when the assay was run on the water the particles had been soaked in, and the most sensitive approach was to assay the particle suspension directly. Sample handling determines what the assay can see.
The practical consequence is that a negative endotoxin result is a statement about a specific sample preparation under a specific method, and its strength depends on whether recovery was demonstrated in that matrix.
Does the choice between LAL and recombinant Factor C matter?
It matters for what can interfere. Limulus amebocyte lysate detects endotoxin through a Factor C–mediated clotting cascade, but that lysate contains other components, including a Factor G pathway that responds to glucans and can produce a response not caused by endotoxin. Recombinant Factor C uses only the first enzyme, so those alternative pathways are absent.
Piehler and colleagues addressed the robustness question by comparing LAL and rFC across a proficiency testing programme run from 2014 to 2019, covering unknown samples, multiple laboratories, operators and reagent lots — noting that pharmaceutical users had validated rFC for their own products but that long-term comparative data spanning those variables had not been published (Microorganisms, 2020;8(3):418; PMID 32188126, DOI 10.3390/microorganisms8030418). The relevant point for reading a certificate is simply that "endotoxin tested" is not one method, and the method should be named.
Is a stated endotoxin limit a compendial standard?
Not on its own. Pharmacopoeial endotoxin limits are not universal constants; they are values calculated for a particular finished drug product from parameters specific to it, which is why no single figure expressed per milligram of substance is a compendial limit in the abstract. A bare threshold quoted against a research chemical is therefore a supplier specification, not a pharmacopoeial one — the framework that generates pharmacopoeial limits operates on drug products and does not extend to material supplied for laboratory research. Treating the two as interchangeable reads a regulatory standard into a commercial number.
What do our own certificates report?
Purity, method, batch number, test date and the issuing laboratory. Nothing else, and specifically not endotoxin.
The five published certificates — BPC-157 batch 2026-03, GHK-Cu batch 2026-03, Retatrutide batch 2026-03, MOTS-c batch 2026-03 and NAD+ batch 2026-03 — are chromatographic purity determinations by a named third-party laboratory on a stated date. No endotoxin determination has been made on those batches, and no figure should be inferred for them. Where a laboratory has not run the assay, the honest statement is that the result does not exist, not that the material is presumed clean.
Veridian Research supplies these materials strictly for in-vitro laboratory research. They are not drugs and are not approved for human or veterinary use.
The short version
An endotoxin test measures biological reactivity to bacterial lipopolysaccharide, expressed in endotoxin units against a reference standard. It is independent of chromatographic purity in both directions, so a purity certificate carries no endotoxin information. The assay can be defeated by its own sample matrix through masking by surfactants and complexing agents, through adsorption onto glass or particle surfaces, and through preparation choices that keep the contamination out of what is measured. Where endotoxin is undetected, what has been established is that this method, in this matrix, did not detect it — and for cell-based work, the published record shows that overlooked contamination has produced entire results that were not real.