Proteomic analysis of the supernatant of red blood cell units: the effects of storage and leucoreduction

Abstract

Background: Red blood cell (RBC) transfusion is a life-saving intervention for critically ill patients; however, it has been linked to increased morbidity and mortality. We hypothesize that a number of important proteins accumulate during routine storage of RBCs, which may explain some of the adverse effects seen in transfused patients.

Study design: Five RBC units were drawn and divided (half prestorage leucoreduced (LR-RBC) and half left as an unmodified control (RBC). The supernatant was separated on days 1 and 42 of storage and proteomic analyses completed with in-gel tryptic digestion and nano-liquid chromatography tandem mass spectrometry.

Results: In RBC supernatants, 401 proteins were identified: 203 increased with storage, 114 decreased, and 84 were unchanged. In LR-RBC supernatant, 231 proteins were identified: 84 increased with storage, 30 decreased, and 117 were unchanged. Prestorage leucoreduction removed many platelet- and leucocyte-derived structural proteins; however, a number of intracellular proteins accumulated including peroxiredoxins (Prdx) 6 and latexin. The increases were confirmed by immunoblotting, including the T-phosphorylation of Prdx-6, indicating that it may be functioning as an active phospholipase. Active matrix metalloproteinase-9 also increased with a coinciding decrease in the metalloproteinase inhibitor 1 and cystatin C.

Conclusion: We conclude that a number of proteins increase with RBC storage, which is partially ameliorated with leucoreduction, and transfusion of stored RBCs may introduce mediators that result in adverse events in the transfused host.

Keywords: leucoreduction; proteins; red blood cell; storage; supernatant; transfusion.

Figure 1. Representative 1D-PAGE profile of RBC and LR-RBC supernatants after days 1 and 42 of storage

Quantitative densitometric analysis of the ~16 kDa MW band (hemoglobin band) by equal protein concentration (textured fill), and corrected for equal volume of supernatants (solid fill). The numbers of the left side of the gel indicate molecular size in kDa.

Figure 2. Venn diagram of total proteins identified in the supernatant of RBCs (A) and LR-RBCs (B) stored days 1 and 42

These diagrams represent a breakdown of the total proteins identified. The numbers on the left (D1) and right (D42) are the number of protein unique to the RBC or LR-RBC supernatant. The numbers in the center are the total number of proteins that are in both D1 and D42 supernatant samples.

Figure 3. Heat map of select identified proteins across RBC and LR-RBC supernatants

A. Proteins are grouped by patterns of relative abundance change for both RBC and LR-RBC supernatants at day (D1) and D42 of 5 stored units. Gene names are on the right of the heat maps with the protein identification below (B). The map reads from the very little (blue) to a 1:1 ratio (white) to the greatest amount (red).

Figure 4. Latexin accumulates in the plasma fraction during routine storage of LR-RBC supernatant

A representative immunoblot for latexin demonstrated significant immunoreactivity in the plasma fraction of stored, day (D)42, LR-RBC supernatant which was not present on D1 in the identical samples. Albumin is shown as loading control. These data illustrated that latexin accumulates during routine storage of LR-RBC supernatant and is likely a red blood cell protein (n=3).

Figure 5. Peroxiredoxin 6 accumulates in the plasma fraction during routine storage of LR-RBC supernatant, and is T-phosphorylated

Panel A: a representative immunoblot which demonstrates minimal amounts of Prdx6 in the plasma fraction of day (D)1 LR-RBC supernatant. The immunoreactivity significantly increases (C) during storage and is released into the plasma fraction of stored, D42 LR-RBC supernatant indicating it is likely a protein inherent to the red blood cell.so that there are obvious amounts in the plasma fraction. Albumin is shown as a loading control. Panel B: a representative western blot depicts the identical blot in panel A that has been stripped and re-probed with phospho-threonine. There is phospho-threonine immunoreactivity in the bands that demonstrated Prdx6 immunoreactivity suggesting the Prdx6 was T-phosphorylated and may be an active phospholipase. Panel C: The mean densitometry for panels A and B calculated with ImageJ software.

Figure 6. Gelatin zymograph showing gelatinase activity during RBC storage

Representative matrix metalloproteinase activity, MMP-8 and MMP-9, in RBC supernatant and LR-RBC supernatant units as measured on days 1 and 42 of storage.