Endotoxin in Peptide Research: What It Is and How It Is Measured

What endotoxin is

Endotoxin is the common name for the lipopolysaccharide (LPS) found in the outer membrane of Gram-negative bacteria such as Escherichia coli. A typical LPS molecule has three parts: a hydrophobic anchor called lipid A, which carries most of the biological activity, a core oligosaccharide, and a long, variable polysaccharide chain known as the O-antigen.^[1] Bacteria shed LPS as they grow and release it when they die, so a solution can contain endotoxin long after every bacterium in it has gone.

Two properties make endotoxin a practical problem in the laboratory. It is heat-stable, so ordinary autoclaving does not destroy it, and it is amphiphilic, so it forms aggregates in water and sticks to surfaces. It also travels with the materials that are easiest to overlook: water, buffers, glassware and plastic.

Why it matters for cell-based work

Animal cells detect LPS through a receptor called Toll-like receptor 4 (TLR4). The link was established genetically when two mouse strains that do not respond to LPS were shown to carry mutations in the Tlr4 gene, one a missense change and the other a null mutation.^[2] Lipid A binding to TLR4 starts an inflammatory signalling cascade in cells such as monocytes, macrophages and dendritic cells.^[1]

That sensitivity is what makes endotoxin a confounder. A review in immunology described how endotoxin carried over in bacterially produced recombinant proteins has been responsible for effects first attributed to the proteins themselves, and argued that conventional methods of detecting and removing LPS can still leave enough to mislead.^[3] A later study measured commercial recombinant proteins, many sold with a stated limit below 1 endotoxin unit, and found that LPS in the range of 0.002 to 2 ng/mL was enough to activate some cell types; primary CD1c+ dendritic cells were especially sensitive, and the authors recommended screening reagents before use with such cells.^[4]

Synthetic peptides made by solid-phase synthesis are not produced in bacteria, but that does not make a prepared solution free of endotoxin. Water, buffers, purification equipment and containers can all introduce it. For any experiment with immune or inflammatory readouts, endotoxin is a variable to record rather than assume away.

Sterile is not the same as endotoxin-free

A sterile solution contains no living microorganisms. An endotoxin-free solution contains no measurable LPS. The two are easy to confuse because they are often achieved together, but the usual sterilizing steps do not remove endotoxin. Autoclaving leaves it intact, and filtration through a 0.22 µm membrane removes bacteria but lets much of the LPS through, because individual LPS molecules and most of their aggregates are far smaller than the pores.

The same distinction applies to the diluent used to prepare a stock solution. Water labelled sterile describes the absence of organisms; whether it has been tested for endotoxin is a separate property that is stated separately. The bacteriostatic versus sterile water guide covers the other properties of common laboratory waters.

How the LAL test works

The standard detection method uses an extract of blood cells (amebocytes) from the horseshoe crab, called Limulus amebocyte lysate. When the lysate meets endotoxin it forms a gel. Biochemical work has shown why: the cells contain a cascade of serine protease zymogens in which endotoxin activates factor C, factor C activates factor B, factor B activates a proclotting enzyme, and that enzyme converts a protein called coagulogen into an insoluble coagulin gel.^[5] A second branch, started by factor G, responds to (1,3)-β-D-glucan, a polysaccharide from fungi and some cellulose-based materials, and also ends in gel formation.^[5] That glucan branch is one reason a LAL test can read positive when no endotoxin is present.

• Gel-clot. The sample and lysate are mixed in a tube and inverted after incubation. A firm gel is a positive result at the stated sensitivity of the reagent. The method is simple and gives a pass or fail result, or a range when run across a dilution series.
• Turbidimetric. The rise in cloudiness as the gel forms is followed with a reader. Faster clouding means more endotoxin, and a standard curve converts the timing to a value.
• Chromogenic. The clotting enzyme cuts a synthetic substrate that releases a coloured product. The colour, read as an end point or over time, is compared with a standard curve.

Results are reported in endotoxin units (EU), usually per millilitre or per milligram of material. The unit is defined against a reference endotoxin standard rather than by mass; a common rule of thumb is that 1 EU corresponds to roughly 0.1 ng of reference E. coli endotoxin, though that conversion is approximate and depends on the preparation.

Recombinant factor C

Because the lysate comes from wild horseshoe crabs, a recombinant alternative has been developed. Recombinant factor C (rFC) reproduces only the first enzyme of the cascade, and a fluorescent substrate reports its activity. A validation study across several products found an end-point fluorescence method using rFC to be equivalent or superior to the compendial bacterial endotoxins test method, and argued for its wider inclusion in testing standards.^[6] Since the reagent contains factor C alone, it lacks the factor G branch and so does not respond to glucans in the way a full lysate can.

Interference and sample preparation

Both LAL and rFC are enzyme assays, and anything that changes enzyme activity changes the result. Extreme pH, high salt, chelating agents, detergents and high concentrations of the material being tested can all suppress or enhance the reaction. The usual remedy is dilution: the sample is diluted in endotoxin-free water until interference disappears, while staying above the level at which the assay can still detect the limit of interest. A serial dilution worked out with the dilution calculator keeps those volumes clear.

To show that a given sample does not interfere, laboratories run a spiked control: a known amount of endotoxin is added to the diluted sample and the amount recovered is compared with the amount added. Recovery well outside the expected range, commonly 50 to 200 percent, means the result for that sample cannot be relied on at that dilution. Peptides that bind lipids or carry many positive charges deserve particular attention here, because they can bind LPS and mask it.

Keeping endotoxin out

Removing endotoxin from a finished solution is difficult. A review of removal methods covered adsorption, two-phase partitioning, ultrafiltration and chromatography, noted that efficient and cost-effective removal remains challenging, and observed that the protein of interest is often lost along with the LPS.^[7] In practice, prevention is easier than removal.

1. Use water and buffers that are specified as tested for endotoxin, and record that specification.
2. Use plasticware supplied as pyrogen-free, or glassware depyrogenated by dry heat; dry-heat cycles at about 250 °C are commonly used, because autoclaving is not enough.
3. Keep a separate set of containers for endotoxin-sensitive work, since LPS adsorbs to surfaces and is hard to wash off.
4. Prepare small single-use aliquots so a stock is opened as few times as possible.
5. Include an endotoxin-only control, such as a known LPS concentration, in cell experiments with inflammatory readouts so an LPS-like response can be recognised.
6. Write the endotoxin status of every reagent, if known, in the sample log, alongside concentrations worked out with the concentration calculator.

Endotoxin testing is a separate measurement from identity and purity testing. An HPLC chromatogram or a mass spectrum describes the peptide; neither detects LPS. This guide describes methods only and is not a statement about any particular material. Terms used here are defined in the glossary, and questions can be sent to support.