Genetic Regulation of Carnitine Metabolism Controls Lipid Damage Repair Mechanisms and Hemolytic Propensity of Human Red Blood Cells during Aging In Vivo and in Vitro

Large scale genomics studies and vein-to-vein databases have started to reveal that donor biology and genetics influence red blood cell (RBC) storability and transfusion outcomes. To further delve into this concept, here we performed metabolomics analyses of 13,091 packed RBC units from donors enrolled in the Recipient Epidemiology and Donor Evaluation (REDS) RBC Omics study. End of storage (day 42) units were tested for metabolomics and hemolytic propensity. Donors ranking in the 5^th and 95^th percentile were contacted again and invited to donate a second unit of blood, which was stored again for 42 days (n=643). Correlation of end of storage metabolomics measurements for the first (index) and second (recalled) donation identified a core of metabolites involved in carnitine synthesis and acyl-carnitine metabolism as the most reproducible within the same donor across multiple donations (Figure 1.A). Carnitine and its precursors, methyl- and trimethyl-lysine were the most significantly reproducible of all the metabolites tested in this study (Figure 1.A). Association of L-carnitine measurements to genomics data (i.e., 879,000 single nucleotide polymorphisms that were assayed via a precision transfusion medicine array developed for this study - Figure 1.B) identified non-synonymous coding polymorphisms in the carnitine transporter SLC22A16 as a critical genetic factor contributing to inter-donor heterogeneity in end of storage carnitine levels (Figure 1.B). Donors carrying two alleles of this SNP were characterized by the lowest L-carnitine levels, and associated depletion of the whole carnitine pool. Functionally, stored RBCs with the lowest levels of carnitine pools were characterized by significant elevation in in vitro hemolysis and the highest degree of vesiculation, in parallel to increases in lipid peroxidation markers (hydroxy-eicosatetraenoic - HETEs and hydroxy-octadecadienoic acids – HODEs).

We leveraged a novel CFDA-SE staining protocol, which allows flow cytometry sorting of end of storage RBCs in two highly enriched preparations: a morphologically-altered subpopulation (CFDA-SE^high) and a morphologically-normal RBC subpopulation (CFDA-SE^low). The morphologically-altered subpopulations contains mostly storage-induced micro-erythrocytes (SMEs), i.e., RBCs with decreased surface/volume ratio upon storage-induced vesiculation. This RBC sub-population has increased susceptibility to extravascular hemolysis via splenic sequestration upon transfusion in vivo. Metabolic characterization of SMEs (i.e., CFDA-SE^low) vs morphologically-normal end of storage RBCs (i.e., CFDA-SE^low) showed that the former are characterized by a complete depletion of acyl-carnitine pools compared to the latter or to fresh RBCs. We thus propose a model where storage promotes lipid peroxidation, which is in turn counteracted by activation of the Lands cycle, a pathway that relies on acyl-carnitine pools to repair lysophospholipids secondary to oxidation of fatty acids in membrane lipids. To determine the translational relevance of this mechanism to RBCs as they age in the bloodstream in vivo, we separated packed RBCs via Percoll density gradients and determined total carnitine pools in each fraction. Our results indicate that the oldest circulating RBCs are not only the smallest, but are also characterized by the lowest levels of carnitine pools. Incubation of the youngest, oldest and intermediate RBC populations with radio-labeled palmitate showed that incorporation of labeled hexadecenoic acid into phosphatidylethanolamines is constrained by carnitine pools in old and average age RBCs, compared to the youngest population. Altogether, our results suggest an important role for L-carnitine levels during RBC aging in vivo and in vitro. As such, carnitine supplementation may represent a viable strategy to boost RBC storage quality and modulate hemolytic propensity, especially in RBCs from donors harboring genetic polymorphism associated with partial ablation of the carnitine transport system.

Figure 1 – Levels of carnitine metabolites are reproducible across multiple donations from the 643 REDS RBC Omics donors across two consecutive donations (index and recalled in A), a phenomenon in part explained by genetic polymorphisms to the carnitine transporter SLC22A16 as gleaned by L-carnitine measurements in end of storage units from 13,091 blood donors (B).