Limitations of Limulus amebocyte lysate test for endotoxin control in raw materials for liposomal nanoformulations

Aim: To evaluate the applicability of Limulus amebocyte lysate (LAL) assay for endotoxin determination in lipid compounding liposomal nanoformulations. Materials & methods: Spiked cholesterol, hydrogenated soy phosphatidylcholine and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[amino(polyethylene glycol)-2000] (DSPE-PEG 2000) samples with endotoxins, simulating contaminated samples or in-process contamination were analyzed by chromogenic LAL assay. Results: Recovery of spiked endotoxins was achieved from DSPE-PEG 2000 suspended in water, whereas recovery was not achieved from spiked cholesterol and hydrogenated soy phosphatidylcholine suspended in methanol, and from multilamellar vesicles. Conclusion: Endotoxins, when in contact with organic solvents, no longer react in the LAL assay as they do in aqueous media. This indicates limitations of the LAL assay for endotoxin control in raw materials for liposomal nanoformulations. Raw materials for liposome nanoformulations have low solubility in water. However, DSPE-PEG 2000 forms a homogeneous suspension in water, which does not interfere with the chromogenic Limulus amebocyte lysate (LAL) assay, according to the criteria established by several pharmacopeias (US, European and Japanese). Cholesterol (Chol) and hydrogenated soy phosphatidylcholine (HSPC) should be dissolved in organic solvents in order to determine the endotoxins eventually entrapped in them. Then, to mimic contaminated samples for interference testing, which are dissolved in MeOH for analysis by LAL assay, control standard endotoxin (CSE) resuspended in MeOH should be added. However, CSE resuspended in MeOH does not react in LAL assay as it does when resuspended in aqueous media. When CSE comes into contact with organic solvents, it forms supramolecular structures that can influence its reactivity with the LAL reagent. Independent solutions of different concentrations do not react proportionally to the amounts of CSE, whereas solutions prepared by dilution from the highest concentrations do react proportionally. Therefore, if any sample is dissolved in organic solvents, as is the case with Chol and HSPC for the chromogenic LAL assay, the results cannot be relied upon for accurate endotoxin measurement. CSE resuspended in water is not affected by 2.5 and 5% MeOH. However, if C12E10 is added, the CSE no longer reacts proportionally in the LAL assay. Inhibition/enhancement controls of Chol and HSPC. Inhibition/enhancement controls should be carefully designed to mimic actual contaminated samples to ensure accurate measurement of endotoxins. These controls should also be conducted by spiking with different concentrations of endotoxins if they are assaying under non-aqueous matrices. LPS was not detected in MLV prepared by spiking the lipid solution in EtOH but was detected in MLV prepared by spiking the aqueous solution (5% dextrose). The lipid medium affects endotoxin detection. Heating the spiked MLV samples at 75°C and 92°C for 30 min did not allow endotoxin detection.