Endotoxin Testing: What the LAL Assay Measures

The limulus amebocyte lysate (LAL) assay measures bacterial endotoxin in a sample. It does this by using a reagent, made from the blood cells of the horseshoe crab, that clots or changes colour in proportion to the endotoxin present. The result is expressed in endotoxin units (EU), normally per millilitre of solution or per milligram of material. It does not measure purity, identity or sterility. It answers one narrow question, and that question is separate from all of those.

This article explains what endotoxin is, how the assay works, what the main method formats report, and why a high chromatographic purity figure says nothing about endotoxin. It also covers the limits of the measurement, because those limits matter as much as the number.

What endotoxin is

Endotoxin is the common name for lipopolysaccharide (LPS). LPS is a large molecule found in the outer membrane of Gram-negative bacteria. It has three parts: a lipid component called lipid A, a core oligosaccharide, and a variable polysaccharide chain on the outside. Lipid A is the part responsible for most of the biological activity.

The word "endotoxin" reflects the older view that the toxin is held inside the cell and released when the cell breaks down. In practice, LPS is also shed by living bacteria. Whether the organisms are alive or dead does not change whether the molecule is there.

Endotoxin is pyrogenic. When it enters the bloodstream it can trigger an inflammatory response in mammals, including fever. This is why it is a concern for injectable and parenteral products specifically. The route of exposure matters. The same quantity that is of little consequence in one context is relevant in another.

Why endotoxin persists after sterilisation

Endotoxin is chemically robust. It is heat stable, and ordinary steam sterilisation conditions are not designed to destroy it. Removing endotoxin from glassware and equipment, called depyrogenation, generally requires much more severe conditions than sterilisation does, such as prolonged dry heat at high temperature.

Filtration is a similar story. A sterilising-grade filter is built to retain viable organisms. Endotoxin exists as individual molecules, fragments and aggregates of varying size, and these can pass through such a filter. A solution can therefore be filtered to a sterile state and still contain endotoxin that came from bacteria present earlier in manufacture, in water, in raw materials or on equipment.

The practical consequence is simple. Endotoxin is a property of the history of the material, not only its current microbial state. It has to be tested for directly.

How the LAL assay works

The assay has a biological origin. In the 1950s, Frederik Bang observed that the blood of the Atlantic horseshoe crab, Limulus polyphemus, clotted in the presence of Gram-negative bacteria. Work with Jack Levin later showed that endotoxin was the trigger. The blood cells, called amebocytes, were found to contain the clotting machinery, and a lysate of them became a laboratory reagent.

The reaction is a serine protease cascade:

• Endotoxin activates Factor C, the first enzyme in the sequence.
• Activated Factor C activates Factor B.
• Activated Factor B converts the proclotting enzyme into the clotting enzyme.
• The clotting enzyme cleaves a protein called coagulogen into coagulin, which forms a gel.

Each LAL method is a different way of reading the end of that cascade. Some detect the gel itself. Others detect the cloudiness that precedes it. Others replace the natural target with a synthetic substrate that releases a coloured molecule.

There is a second pathway in the lysate, triggered by (1,3)-beta-D-glucans rather than endotoxin. Glucans from cellulose-containing materials can therefore produce a false positive in an unmodified LAL reaction. Reagents are often formulated to suppress this pathway, and it is one of the reasons the assay needs controls.

The three main LAL methods

Gel-clot

The gel-clot method is the oldest format. The sample is mixed with lysate and incubated. The tube is then inverted. A firm gel that stays intact means the endotoxin concentration reached or exceeded the labelled sensitivity of the reagent. No firm gel means it did not.

The basic test is a limit test with a positive or negative answer. When the sample is diluted in a series, the method can give a semi-quantitative estimate based on the last dilution that still clots. It is simple and needs little equipment, but it is less precise than the instrumental methods and the endpoint depends on a human reading the tube.

Turbidimetric

Turbidimetric methods measure how cloudy the reaction mixture becomes as coagulin forms. An instrument records optical density over time.

The kinetic version measures the time taken to reach a set turbidity. A sample with more endotoxin reaches that threshold sooner. The result is read against a standard curve. The endpoint version measures turbidity at a fixed time.

Chromogenic

Chromogenic methods add a synthetic peptide substrate that the activated enzyme cleaves. The cleavage releases a coloured compound, commonly p-nitroaniline, which is yellow. Absorbance is read at around 405 nm.

Like the turbidimetric format, it can run kinetically or as an endpoint measurement. Results are quantitative and are read from a standard curve built from a reference endotoxin.

What the numbers mean

All three formats report in endotoxin units. An EU is defined relative to a reference standard endotoxin, so it is a unit of biological activity and not a unit of mass. Different endotoxins can have different potency per unit of mass. This is why results are given in EU and not nanograms.

Recombinant Factor C

The natural lysate depends on harvesting horseshoe crab blood. That has raised conservation and supply concerns, and it has also made reagent consistency a recurring topic. Recombinant Factor C (rFC) is an alternative. It is a manufactured form of the first enzyme in the cascade. When endotoxin activates it, it cleaves a fluorogenic substrate, and the fluorescence is measured. No animal-derived lysate is involved.

Pharmacopoeias have been moving toward recognising this approach. The European Pharmacopoeia has a general chapter, 2.6.32, covering a test for bacterial endotoxins using recombinant Factor C. Pharmacopoeial texts continue to be revised, so the current published text of the relevant chapter is the authority on what is accepted and under what conditions.

The pharmacopoeial framing

The bacterial endotoxins test is a standard, described test. The United States Pharmacopeia describes it in general chapter <85>. The European Pharmacopoeia describes it in general chapter 2.6.14. These chapters set out the reagents, the controls, the validation of the method for a given product, and how results are interpreted.

A few features of that framing are worth stating plainly.

• The test has to be shown to work for the specific material. Some samples inhibit or enhance the reaction, which would distort the result. Positive product controls, in which a known amount of endotoxin is spiked into the sample, are how this is checked.
• Limits are not universal. They are set for a product and depend on the intended use, the route of administration and the dose. They are expressed as a concentration, for example EU per mL or EU per mg, so the same absolute amount can be acceptable at one dose and not at another.
• Using a pharmacopoeial method does not by itself say anything about who performed it. Accreditation of a laboratory is a separate matter. In general terms, accreditation means an independent body has assessed a laboratory against a defined standard for a defined scope of tests. It applies to that scope and nothing wider.

This article does not give a numerical limit for any product. A limit only has meaning when the material, the dose and the route are all specified.

Why endotoxin is not a purity question

Chromatographic purity is usually reported as an HPLC area percent. The method separates the components of a dissolved sample and records them with a UV detector. The percentage is the area of the main peak as a share of the total area of detected peaks. It describes the composition of what absorbs UV light under the conditions used. The detail of that figure is covered in what 99 percent purity means.

Endotoxin sits outside that measurement for two reasons.

• Scale. Biologically meaningful quantities of endotoxin are tiny in mass terms. They fall far below anything that would register as a distinct peak against a main component present at milligram quantities.
• Detection. LPS does not behave as a peptide in a reversed-phase UV method. The assay was never designed to quantify it.

A sample can show a very high area percent and still carry endotoxin. The reverse is also true. A sample with a lower area percent from synthesis by-products may be free of detectable endotoxin. The two results describe different properties, and neither predicts the other. For a wider view of how the individual tests fit together, see peptide testing.

Endotoxin versus sterility

Sterility testing asks whether viable microorganisms are present. It typically involves incubating a sample or a filtered portion of it in growth media and observing whether anything grows.

Endotoxin testing asks whether the toxin is present. It does not need living organisms. A sample can be sterile and carry endotoxin, because the organisms that produced the toxin are gone. A sample can also contain a low count of viable organisms and, at the moment of sampling, little free endotoxin.

These are different tests answering different questions. A pass on one does not imply a pass on the other.

Limits of the measurement

An endotoxin result describes the sample that was tested. It has limits that should be understood before the number is used for anything.

• Sampling. Contamination is not always evenly distributed through a batch. A result applies to the vial or portion that was submitted.
• Interference. Peptides, buffers and excipients can inhibit or enhance the reaction. Without a valid recovery check, a low result may reflect suppression and not absence.
• Sensitivity. Every method has a lower limit of quantification. A result reported as below that limit means below what the method can measure, not zero.
• Method differences. Gel-clot, turbidimetric, chromogenic and rFC methods are calibrated against reference standards, but they are not identical. Results from different methods are comparable only within the limits of that calibration.
• Scope. An endotoxin result says nothing about identity, content, sterility, residual solvents or safety.

A certificate describes the sample submitted and nothing beyond it. Purity is composition, not safety. Endotoxin is one measured property among several, and it does not add up to a conclusion about a product on its own. The way those individual results are laid out on a report is explained in how to read a peptide CoA.

Where Novagen fits

Novagen is a third-party analytical laboratory. It reports what was measured in the sample submitted. It publishes a certificate register at novagenlaboratories.com, and any certificate can be checked against that register on the verification page. The current quoted turnaround is 21 days from receipt of samples. Questions about a specific analysis can be sent to admin@novagenlaboratories.com.

The useful habit is the same whichever laboratory produces a report. Check which question each test answers. Check what the result is reported against. Check what it does not cover.