MSACL 2022 Abstract

Introduction
The lipidome comprises abundant lipids and very-low concentrated lipid mediators. Both can act as signaling molecules in the human body and are involved in a variety of pathophysiological processes. They are subject to ongoing investigations for example on inflammatory or psychiatric diseases and can indicate distinctive differences between healthy controls and patients. However, the suitability of lipids and lipid mediators as biomarkers in an easily accessible matrix like blood is still discussed intensively. One of the main points of discussion is the alteration of the lipidome by exogenous factors like processing conditions and sample type (plasma or serum). Data quality relies heavily on pre-analytical and analytical workflows and while the latter is commonly centralized and validated, sample generation is often multi-centered and due to its complexity prone to deviations. Therefore, it is important to identify critical steps especially during the pre-analytical phase to define a protocol suitable to preserve physiological concentrations in blood-based samples, but still feasible for implementation in the clinical routine.

Methods
We accessed the stability of the lipidome regarding the processing time and conditions after blood draw, including whole blood (WB) storage, and plasma or serum storage after centrifugation at room temperature (RT) or on ice (FT). Furthermore, we investigated the influence of the anticoagulant NaF/citrate as an alternative to the commonly used anticoagulant K3EDTA. To verify the quality of blood-based samples we screened potential quality markers such as 12-HETE, 2-AG, and S1P-d18:2 for their value to indicate altered processing conditions.
The quantification of low-concentrated lipid mediators (e.g. eicosanoids, endocannabinoids (ECs), and lysophosphatidic acids (LPAs)) was performed via targeted LC-MS/MS analytics and the non-targeted screening for more abundant lipids was achieved using LC-HRMS (e.g. acylcarnitines, diglycerides (DGs), lysophosphatidylcholines (LPCs) and -ethanolamines (LPEs)).

Results
Pre-analytical sample handling conditions can have varying impacts on the lipidome. Lipids such as acylcarnitines, DGs, LPCs, and LPEs are prone to an ex vivo increase of concentrations when K3EDTA WB is stored at RT (≤ 1 h). This is also the case for storage of K3EDTA plasma at RT with exception of acylcarnitines. Stability can be drastically increased for most lipids (≤ 4 h) if K3EDTA WB as well as plasma are handled at FT. However, several lipid mediators from the groups of ECs, LPAs, eicosanoids, and sphingolipids are prone to ex vivo alterations even under FT conditions. Therefore, blood processing for lipid mediators like 15-HETE and AEA should not exceed a total processing time of 1 h. Furthermore, the stability of ECs can be increased by using NaF/citrate as an anticoagulant. On the contrary, the use of serum should be avoided due to mandatory clotting at RT. To evaluate adherence to the given sampling protocol different quality markers might be helpful as for example the ratio of the isomers 2-AG and 1-AG is negatively correlated with prolonged storage of K3EDTA WB and plasma at RT, and increased S1P-d18:2 levels could be used to indicate prolonged storage of K3EDTA WB at RT.

Conclusion
The results of this study highlights the impact of pre-analytical sample processing conditions on the lipidome and the need for highly standardized workflows to secure sample quality. Temperature and storage time of K3EDTA WB until centrifugation are the most critical parameters and should be minimized for optimal stability of the lipidome. Overall sample storage at FT until final storage at < -70 °C is recommended. Total processing time after blood draw until final freezing should be ≤ 1 h. In the future, quality markers could be implemented to better evaluate data quality and facilitate clinical biomarker research targeting the lipidome.