Endotoxin Testing in Research Peptides

What Bacterial Endotoxin Is

Endotoxin is lipopolysaccharide, or LPS, a structural component of the outer membrane of gram-negative bacteria such as E. coli. It is not a substance the organism secretes on purpose. It is released when the bacterial cell wall breaks down, whether the organism dies naturally, is killed by heat, or is destroyed during a purification or filtration step.

Because LPS is a structural molecule rather than a living cell, killing the bacteria that produced it does not remove it. A vial can be free of viable organisms and still carry a measurable endotoxin load left behind by bacteria that were present earlier in the manufacturing chain, even briefly, in raw material, in process water, or on equipment surfaces.

Key properties of LPS relevant to testing:

• It is heat-stable to a degree that exceeds standard autoclave conditions in many cases, which is why heat alone cannot be assumed to clear it.
• It is a large, amphipathic molecule that can adhere to glass and plastic surfaces, which is one reason depyrogenated glassware is a distinct spec from sterile glassware.
• It varies in potency by bacterial species and structure, which is part of why the assay is standardized against a reference material rather than measured as a raw mass.

Why Sterilization Does Not Remove Endotoxin

Sterile and pyrogen-free are not the same claim, and conflating them is one of the more consequential misreadings of a supplier document. A sterilization step, whether filtration, heat, or irradiation, is designed to eliminate viable microorganisms. None of those steps is designed to remove LPS, and several of them make the problem worse rather than better.

Heat sterilization kills bacteria but ruptures cell walls in the process, releasing more LPS into the material than was present in intact bacterial cells beforehand. Sterile filtration through a 0.2 micron filter removes intact bacteria by size exclusion, but LPS aggregates and fragments are small and structurally variable enough that filtration does not reliably clear them. A material can pass a sterility check with no viable growth and still carry a high endotoxin load from bacteria that were present and then destroyed earlier in the process.

This is why endotoxin testing exists as its own discipline, with its own assay, its own reference standards, and its own reporting units, run in addition to sterility testing, not instead of it.

The LAL Assay and Its Three Formats

The standard method for detecting and quantifying bacterial endotoxin is the Limulus amebocyte lysate assay, commonly abbreviated LAL. It uses a reagent derived from the blood cells of the Atlantic horseshoe crab, Limulus polyphemus. The lysate contains a clotting cascade that reacts specifically to LPS, and the strength of that reaction is used to quantify how much endotoxin is present in a sample.

There are three recognized formats:

Gel-clot is the oldest and simplest format and is still used as a limit test. Turbidimetric and chromogenic formats are kinetic, meaning they track the reaction over time rather than at a single endpoint, and they are what most modern quantitative COAs are built on.

Recombinant Factor C: The Synthetic Alternative

LAL depends on harvesting blood from wild horseshoe crabs, which has raised both a supply and a conservation concern within the industry for years. The reagent that answers that concern is recombinant Factor C, abbreviated rFC.

Factor C is the specific protein within the horseshoe crab clotting cascade that recognizes LPS. Rather than harvesting the whole lysate, rFC is produced by inserting the gene for Factor C into an expression system, typically insect or mammalian cell lines, and manufacturing the protein synthetically. The recombinant protein is combined with a fluorogenic substrate so that its activation by LPS produces a measurable fluorescent signal, giving a quantitative result analogous to the chromogenic LAL format.

Labs are moving toward rFC for a combination of reasons:

• It removes reliance on an animal-derived reagent and the supply variability that comes with it.
• Because it isolates a single defined protein, it does not react to the beta-glucans that can cause false positives in some LAL formats.
• Lot-to-lot consistency is generally tighter than lysate pooled from wild-caught animals.

rFC is now recognized as an approved alternative method in several major pharmacopeias, and a growing share of contract labs offer it as the default rather than as an exception.

Reading the Units: EU/mg and EU/mL

Endotoxin results are reported in Endotoxin Units, abbreviated EU, always relative to either a mass or a volume. That is the detail that makes a bare number meaningless without its unit and its basis attached.

Two reporting conventions show up on peptide COAs:

• EU/mg reports endotoxin relative to the mass of peptide, typically for bulk lyophilized powder before it has been put into solution. This is the figure most relevant to a raw material spec.
• EU/mL reports endotoxin relative to a volume of solution, which only means something once you know the concentration of peptide in that solution, since the same vial reconstituted to a different volume changes the EU/mL figure without changing the total endotoxin present.

A number like "less than 1.0" is not a result on its own. Less than 1.0 EU/mg and less than 1.0 EU/mL describe very different amounts of endotoxin once mass and volume are factored in, and a COA that states a bare number with no unit attached should be read as incomplete rather than favorable. When comparing figures across two suppliers, confirm both are expressed on the same basis before treating one as better than the other.

Endotoxin Testing Is Not Sterility Testing

These are two separate tests, run by two separate methods, answering two separate questions, and a result on one says nothing about the other.

A sterility check asks whether viable microorganisms are present, typically by incubating a sample in growth media for an extended period and checking for turbidity or colony growth. It is a pass or fail biological growth test.

Endotoxin testing asks how much LPS is present, using LAL or rFC, regardless of whether any organism is currently alive. A material can be sterile and still carry a substantial endotoxin load. In principle a material could also show detectable bioburden on a sterility check while still reading low on endotoxin, if contamination happened recently and had not yet built an LPS load. Neither test substitutes for the other, and a COA that reports one but not the other has only answered half the contamination question.

Where Contamination Actually Enters

Endotoxin does not appear in a product from nowhere. It comes from a specific set of upstream points, and identifying which one is weak is the practical value of tracking a result across multiple lots rather than reading one COA in isolation.

• Water. High-purity water systems are the most common vector. Systems that are not properly maintained, or that use a lower-grade water at any process step, are a frequent source.
• Glassware and labware. Standard washing and sterilization does not depyrogenate glass. Vials and equipment need a specific dry-heat depyrogenation cycle, distinct from autoclaving, to destroy LPS bound to a surface.
• Raw material. The starting peptide material itself, particularly material produced by fermentation-adjacent or recombinant expression routes, can carry endotoxin from the host organism used in production.
• Handling. Human contact, ambient air, and unclean work surfaces during formulation, filling, and lyophilization all introduce risk, which is why the level of environmental control in a facility matters as much as the raw material spec.

Why Lyophilized Material Still Needs the Test

Lyophilization, or freeze-drying, removes water but does not remove or degrade LPS. It is a common misconception that because a lyophilized peptide is a dry, stable-looking cake, the endotoxin question is closed. Endotoxin present before lyophilization is still present after it, essentially unchanged, since the molecule is stable across a wide range of conditions including freezing and drying.

This is why a legitimate COA tests the lyophilized powder directly rather than assuming a clean result from an earlier process step covers the finished vial. If a supplier's documentation only shows testing on a raw material or an intermediate rather than the final lyophilized lot, that is a gap worth asking about directly.

What a Missing Endotoxin Line Means, and Interference Controls

When a COA has no endotoxin figure at all, the honest interpretation is that endotoxin was not tested, not that it was found to be absent. Absence of a bad result is not the same as a good result, and a document that reports purity and identity while skipping endotoxin entirely should be treated as incomplete rather than reassuring.

A separate detail worth understanding about a well-run assay: LAL and rFC methods are both susceptible to interference from the sample matrix itself. Certain compounds can either mask endotoxin that is present, called inhibition, or artificially amplify the signal, called enhancement, producing a false negative or false positive result. A properly validated test includes spike-recovery controls, where a known amount of endotoxin is added to a diluted sample to confirm the assay recovers close to the expected value in that specific matrix. A COA or method summary that never mentions an inhibition or enhancement control, particularly for a novel peptide matrix, is one that has not demonstrated the assay actually works on that specific material, only that the assay ran.