Avoiding ambient air in test tubes during incubations of human whole-blood minimizes complement background activation
Peer reviewed, Journal article
MetadataShow full item record
Original versionJIM - Journal of Immunological Methods. 2020, 1-5. 10.1016/j.jim.2020.112876
Background In vitro, the complement system can be studied in test tubes incubated with anticoagulated human whole-blood. Background activation of complement may mask inflammatory signals. Air bubbles are known to activate complement. We examined if removing ambient air from test tubes before incubation reduced background complement activation. Methods Blood from twelve donors was anticoagulated with the thrombin inhibitor lepirudin and incubated with either no air, ambient air or air bubbles in polypropylene tubes at 37 °C for 180 min on a roller mixer. After incubation, EDTA was added, plasma isolated and analyzed for seven complement activation products using ELISA. Results are presented as means with 95% confidence intervals. Results Blood incubated without air had significantly lower complement activation compared to blood incubated with ambient air; C4d 273 (192–364) vs. 379 (263–494) ng/mL (p = 0.002), C4bc 8.2 (4.1–13) vs. 12 (3.2–21) CAU/mL (p = 0.01), C3a 1351 (873–1838) vs. 2944 (2315–3572) ng/mL (p = 0.0005), C3bc 31 (17–46) vs. 68 (52–84) CAU/mL (p = 0.002), C3bBbP 134 (97–171) vs. 427 (358–506) CAU/mL (p < 0.0001), C5a 3.5 (1.9–5 0.2) vs. 15 (1.8–27)) ng/mL (p = 0.003), TCC 4.6 (2.8–6.3) vs. 9.9 (7.3–12) CAU/mL (p = 0.006). At the end of the experiment blood incubated with air bubbles had a higher complement activation than blood incubated with ambient air with an average 26 fold increase (range 1.6–59) from baseline of all activation products; C4d 551 (337–766) ng/mL, C4bc 21 (5.0–36) CAU/mL, C3a 3983 (3518–4448) ng/mL, C4bc 103 (86–121) CAU/mL, C3bBbP 626 (543–708) CAU/mL, C5a 10 (2.8–18) ng/mL and TCC 10 (6.0–14) CAU/mL. Conclusion Avoiding air in test tubes during whole-blood experiments reduced background complement activation substantially and represents an important improvement to the lepirudin whole-blood model. This could also apply to other in vitro models.