Human whole blood
Human whole blood was obtained from healthy volunteer donors. All blood donations were approved by the Ethics Committee of the University for Continuing Education Krems (EK GZ 13/2015–2018). Written informed consent was obtained from all blood donors. All experiments were conducted in accordance with the guidelines of the Declaration of Helsinki of the World Medical Association.
Standard LAL assay protocol for plasma and serum samples
LPS was quantified using the kinetic chromogenic Limulus amebocyte lysate (LAL) assay (Charles River, Wilmington, MA) according to the instructions of the manufacturer. Samples were diluted 1:10 with pyrogen-free water in pyrogen-free tubes (Charles River) and incubated at 70°C for 15 min to denature plasma proteins and proteases interfering with the LAL assay 34,35. Aliquots of 100 µL of each standard and sample were transferred into a 96-well flat-bottomed microplate (Charles River) in duplicate and incubated at 37°C for 10 min in a microplate reader (Tecan Sunrise, r-biopharm, Darmstadt, Germany). Following the addition of 100 µL LAL reagent per well, the kinetic color development at 405 nm was recorded, and the LPS concentration in the individual samples was calculated from the standard curve.
Improved protocol for the LAL assay
Plasma or serum samples were diluted 1:10 in pyrogen-free water containing 5 mM Mg++ (Merck, Darmstadt, Germany) and 10 IU/ml heparin (Gilvasan Pharma GmbH, Vienna, Austria) and incubated for 3h at room temperature. During incubation, endotoxin-inactivating cationic plasma components bind to the polyanion heparin and are inactivated. All subsequent steps were performed as described in the standard protocol above. The linearity of the improved protocol was tested in serum from five healthy donors spiked with 50, 25, 12.5, 3.13, and 0.78 ng/ml LPS.
Endotoxin neutralizing capacity
The recovery of endotoxin was determined in a dilution series of human serum spiked with LPS. Blood from six healthy donors was drawn in Vacuette® CAT Serum Clot Activator tubes (Greiner bio-one, Kremsmünster, Austria) without addition of anticoagulant and centrifuged after clotting (3500 g, 10 min). Serum was collected and diluted with physiological saline containing 1.2 mM Ca++ and 0.6 mM Mg++ (Pharmacy Bad Ischl, Austria) to obtain a dilution series containing 0, 10, 20, 40, 60, 80, and 100% serum. All samples were spiked with 50 ng/ml LPS from E. coli O55:B5 (Sigma Aldrich, St. Louis, MO), incubated overnight at room temperature with gentle rolling, and analyzed using the standard LAL assay protocol.
Influence of heparin concentration on endotoxin recovery
To determine the optimal heparin concentration for suppressing the endotoxin neutralizing effect of plasma, serum was obtained from six healthy donors as described above, spiked with increasing amounts of heparin (0, 5, 10, 20, 30, 40, 50, 75 and 100 IU/ml) and with 50 ng/ml LPS from E. coli, incubated overnight at room temperature with gentle rolling, and analyzed using the standard LAL assay protocol.
Influence of exposure time to heparin on endotoxin recovery
Serum from six different donors was incubated overnight at room temperature with 50 ng/ml LPS from E. coli with gentle rolling. Aliquots of 980 µl LPS-spiked serum were incubated with 20 µl heparin for 10 min, 30 min, 1h, 2h, 4h, 6h, and 24h. After incubation, all samples were diluted 1:10 in pyrogen-free water and analyzed using the standard LAL assay protocol. This direct addition of heparin to the LPS-spiked serum was compared to the use of a heparin-containing diluent, i.e., the LPS-spiked samples were diluted 1:10 in pyrogen-free water spiked with 10 IU/ml heparin, 1.2 mM Ca++ and 0.6 mM Mg++.
Influence of Ca++ and Mg++ on endotoxin recovery
To assess the influence of Ca++ or Mg++ on endotoxin recovery in serum, diluents containing 0.1 mM, 5 mM, 10 mM, or 15 mM Ca++ or Mg++ in pyrogen-free water were prepared. In addition, a diluent containing Ca++ and Mg++ at a 2:1 ratio was prepared, reflecting the ratio of divalent ions in human blood. Serum from six different donors was incubated with 50 ng/ml LPS from E. coli overnight with gentle rolling. The LPS-spiked sera were diluted 1:10 with the diluent described above, and endotoxin was quantified using the standard protocol of the LAL assay.
LPS recovery in serum samples for standard and improved protocols
Sera from ten donors were spiked with 25 EU/ml of an endotoxin standard (BRP, Sigma-Aldrich, St. Louis, USA) and endotoxin was quantified using the standard and the improved protocols. Pyrogen-free spiked water with 25 EU/ml endotoxin standard BRP served as reference. The endotoxin concentration obtained using the two protocols was compared to the endotoxin concentration obtained for the aqueous solution, and the LPS recovery (%) was calculated as endotoxin concentration obtained for serum/endotoxin concentration obtained for water.
Serum samples were obtained from 20 male and 20 female volunteer donors and stored at -80°C. Prior to analysis, they were spiked with 50 ng/ml LPS from E. coli and incubated overnight at room temperature. On the next day, endotoxin measurement was performed using both, the standard protocol and the improved protocol.
Influence of clot formation on LPS recovery in serum samples
To determine whether the formation of blood clots during the collection of serum affects LPS recovery, e.g., by binding of LPS to the clot, whole blood samples were drawn from five healthy donors into Vacuette CAT Serum Clot Activator tubes (without anticoagulant) and into Vacuette LH Lithium Heparin tubes (both from Greiner bio-one). Immediately after blood collection, an aliquot of the non-anticoagulated blood was spiked with 30 ng/ml LPS (E. coli). When clotting was completed, the serum was collected by centrifugation (3500 g, 10 min), and endotoxin was quantified using the improved LAL assay protocol (see above). The spike concentration of 30 ng/ml LPS corresponds to a serum level of 50 ng/ml LPS at a hematocrit of 40% and assuming that LPS only distributes in plasma. For comparison, 50 ng/ml of LPS was added directly to serum and analyzed as described above.
Influence of different anticoagulants on LPS recovery
Next to testing the influence of clot formation during collection of serum samples on LPS recovery, we assessed the influence of different anticoagulants.
Blood from five healthy donors was drawn into Vacuette® CAT Serum Clot Activator tubes and into Vacuette 9 NC Coagulation 3.2% trisodium citrate tubes (both from Greiner bio-one). The serum was spiked with 50 ng/ml LPS from E. coli. Blood cells in the whole blood samples anticoagulated with citrate were pelleted by centrifugation (3,500 g, 10 min) and washed three times with physiological saline. The LPS-spiked serum and blood cells from the same donor were mixed at a 1:1 ratio, incubated overnight at 37°C with gentle shaking, and the “reconstituted blood” was further incubated in tubes containing heparin (Vacuette LH Lithium Heparin), citrate (Vacuette 9 NC Coagulation trisodium Citrate 3.2%), EDTA (Vacuette K2E K2EDTA), or no additives for one hour (Supplementary Information S1). After incubation, the samples were centrifuged (3,500 g, 10 min), and LPS was quantified using the improved protocol. LPS-spiked serum was used as control.
Statistical analysis
Calculations of means, standard deviation (SD), and standard error of the mean (SEM) were carried out using Microsoft Excel 2010 (Microsoft, Redmond, WA). All other statistical tests were carried out using GraphPad Prism 9.3.1 (GraphPad Software, Boston, MA). The Kolmogorov-Smirnov Test was applied to check for normal distribution. Normal distributed data were compared using the t test. For non-normally distributed data, the Mann-Whitney Rank Sum Test was used. P-values of ≤ 0.05 were considered as statistically significant.