Facing up to the challenge of Low Endotoxin Recovery: better understanding LER and how to detect endotoxins in medicinal products.
There is a range of tests available for detecting endotoxins including the Rabbit Pyrogen Test (RPT) and tests based on Limulus Amebocyte Lysate (LAL), which is derived from the blood cells of the horseshoe crab. The sensitivity, reliability, and ease-of-use of the LAL test has made it the preferred method in most laboratories. However, an inhibitory phenomenon known as Low Endotoxin Recovery (LER) has raised questions about its use and caused considerable concern in the industry.
Better Understanding LER
LER is the masking of endotoxins in undiluted materials, thought to be attributable to combinations of specific excipients. This differs from the inhibition or interference of endotoxin tests caused by pH, high divalent ion concentrations, chelators, serine proteases, and glucan, which can usually be overcome using a pretreatment such as dilution. Since LER was first highlighted by Chen and Vinther in 2013, it has been the subject of much discussion. There has been significant debate about the mechanism behind LER.
It is known that the LPS molecules that make up endotoxins tend to aggregate under certain conditions. For example, in nature they often form a part of bacterial membranes due to their molecular structure, which is composed of both hydrophilic and hydrophobic sections. This is also true of purified LPS molecules. These aggregate to form micelles with the hydrophobic cores of LPS molecules arranged to avoid the aqueous external solution.
Aggregates are thought to be the form detected by the LAL test as they are far more potent than monomers for activating Factor C, the first enzyme within the LAL enzymatic cascade. (2) However, there is evidence to suggest that both monomers and aggregates can activate Factor C, as the mechanism of action is often complex.
The theory assumes that the activity of LPS depends on its supramolecular structure and that, in a solution, this is in equilibrium between the monomer and aggregated states. This equilibrium can be influenced by the presence of surfactants (like polysorbate) and chelating agents (like citrate) and could explain the link between these factors and LER.
Increasing the concentration of chelating agents disturbs the ionic interactions between LPS molecules, reducing the rigidity of aggregates. However, in the absence of a surfactant they can still be detected using LAL-based tests, as the aggregate structure is maintained via the hydrophobic interactions between the lipid parts.
The introduction of surfactants, such as polysorbate, worsens the situation, as they can intercalate into the destabilized aggregates and disaggregate the LPS via the formation of mixed micelles. In other words, an increase in the surfactant concentration provides a mechanism by which the lipid part of the LPS can 'escape' the aqueous phase or solution. Unlike original LPS aggregates, the monomeric LPS (mixed micelles) do not trigger Factor C, as the LPS aggregates are dispersed and hydrophobic lipids are not 'accessible' and therefore cannot be detected. So, if the LPS aggregates start to breakdown (e.g. in the presence of chelating agents) and surfactants are present to form mixed micelles, the endotoxin units become masked from detection by the LAL test.
This LER explanation fits with previous observations: as the surfactant concentration increases, there is a point at which the LPS aggregates start to break down, but still exist in partial aggregates. (3) This better exposes the lipid molecules to Factor C and leads to an increase in the efficiency of endotoxin detection. However, as the concentration increases further, this drives the full breakdown of the aggregates into monomers and enables the formation of mixed surfactant-LPS micelles, which then leads to a sharp reduction in the amount of endotoxin recovered from a sample.
Demasking the Effect of LER
It follows that if LER occurs as a result of a physical state equilibrium, then endotoxin masking should be reversible by pushing the equilibrium towards the aggregated state. Sample treatments to achieve this could include adjusting the pH to influence hydrogen bonding or adding Mg2+/Ca2+ ions to the solution to saturate the chelating agent and prevent destabilization of the LPS aggregates.
The addition of metallo-modified polyanionic dispersants, such as Pyrosperse[TM], can also prevent the surfactant from interfering with the LPS aggregation state. Proprietary proteases may also help in those cases where protein excipients are thought to be causing LER. However, it is important to recognize that commercial proteases may be themselves contaminated with endotoxin.
The dynamic equilibrium between aggregates and monomers is inherently dependent on the concentration of a number of factors, which are likely to be specific to each new drug formulation. As such, the demasking protocol will need to be optimized for each drug and may require a combination of strategies.
The recommended protocol requires an initial step to determine if LER is truly occurring. Next, it will be necessary to identify the combination and concentrations of demasking factors (cations or dispersants, for example) required to trigger the expected recovery rate of a known endotoxin spike-in control. Finally, the optimized mix should be included as part of the sample preparation stage when performing routine endotoxin testing for that formulation.
LER and the failure to detect endotoxins in medicinal products has clear implications for patient health. While no serious incidents have yet been reported, the infusion of many biologies can trigger side-effects such as fever--the cause of which is difficult to determine. It is essential the pharmaceutical industry remains vigilant about the risks of endotoxin contamination and that steps are taken to overcome the challenge of LER. Fortunately, new demasking protocols are providing opportunities to overcome the phenomenon and safeguard the quality of pharmaceutical products.
(1.) Rietschel ET, et al. Bacterial endotoxin: molecular relationships of structure to activity and function. FASEB J. 1994 Feb;8(2):217-25.
(2.) Muller M et al. Aggregates Are the Biologically Active Units of Endotoxin, doi: 10.1074/jbc.M401231200, 2004 Apr 19.
(3.) Nakamura T, et al. Interaction Between Lipopolysaccharide and Intracellular Serine Protease Zymogen, Factor C, from Horseshoe Crab (Tachypleus tridentatus) Hemocytes. J. Biochem. 103, 1988: 370-374.
By Lakiya Wimbish, Lonza, and Johannes Reich, University of Regensberg