Cytosolic phospholipase A₂ type IVA is present in human red cells

In their 1997 review Minetti and Low¹  suggest that the cell signaling components of erythrocytes allow them to respond to their cellular environment and interact with other blood or endothelial cells. Phospholipase A₂ group IVA (IVAPLA₂) is a widely distributed enzyme²  with cell signaling importance. It is up-regulated by a rise in intracellular calcium, triggering phosphorylation and transfer of active enzyme from cytosol to membrane. On membrane binding it preferentially releases the fatty acid arachidonic acid (AA) from the sn2 position of the membrane phospholipid. Both enzymic products, AA and lysophospholipid, have second messenger properties,³  whereas AA is the source for the eicosanoids. A study⁴  of arachidonate metabolism in mammalian red cells concluded that the AA-derived 12-hydroxyeicosatetraenoic acid metabolite, promoted by Ca²⁺ ionophore stimulation, came from red cells and not from contaminating platelets and leukocytes.

IVAPLA₂ is activated by mitogen-activated protein (MAP) kinases,^5,6  and the 2 MAP kinase isoforms extracellular signal regulated kinase 1 (ERK1) and ERK2 required for IVAPLA₂ phosphorylation are present in human red cells.⁷  IVAPLA₂ is present in neutrophils⁸  and platelets,⁹  and it has been measured in eosinophils.¹⁰  As the red cell and the leukocytes differentiate from the same common pluripotent stem cell, it is possible that the mature red cell may also contain this enzyme. IVAPLA₂ has not previously been observed in human red cells. Detection of IVAPLA₂ in the red cell is difficult, as substrate assays lack specificity for the group IVA enzyme and hemoglobin masks detection by Western blot. We have used an affinity purification technique prior to Western blotting to overcome this difficulty.

The human monocytic cell line U937 (American Type Culture Collection [ATCC] CRL2093) was cultured in roller bottles and harvested at 3.6 × 10¹⁰ cells, and a cell cytosol was prepared.¹¹  A human red cell lysate was prepared from 100-mL date-expired, leukocyte- and platelet-depleted red cell transfusion unit, frozen, and stored overnight at -80°C to lyse the cells. The IVAPLA₂ from both the U937 cytosol and the red cell lysate was separately affinity purified by chromatography on Prosep-G columns (Millipore, Bedford, MA) that had anti-IVAPLA₂ (Binding Site, Birmingham, United Kingdom) coupled to it. The columns were washed with 25 mM Tris (tris(hydroxymethyl)aminomethane) buffer, pH 7.4, containing 500 mM sodium chloride, and the bound IVAPLA₂ was eluted under denaturing conditions with 2% sodium dodecyl sulfate (SDS).

To assess the possible contribution of leukocytes and platelets to the IVAPLA₂ in the red cell transfusion unit, leukocyte- and platelet-rich preparations were examined. An expired transfusion pack of human platelets was washed with phosphate-buffered saline (PBS), resuspended in 10 mL PBS, and frozen at -80°C. Leukocytes were isolated from 60 mL EDTA (ethylenediaminetetraacetic acid) blood by centrifuging at 200g to remove platelets. The buffy coat was washed with water to remove the red cells, resuspended in 0.5 mL PBS, and frozen at -80°C. Table 1 shows the cell composition of the blood preparations.

The cell lysates were Western blotted after polyacrylamide gel electrophoresis in 4% to 12% gradient gels, blotted on nitrocellulose membrane, and probed with a monoclonal anti-IVAPLA₂ (Santa Cruz Biotechnology, Santa Cruz, CA), then with peroxidase-conjugated rabbit antimouse immunoglobulin G (IgG; Dako, Glostrup, Denmark) (Figure 1).

The IVAPLA₂ present in the blood cell preparations and the affinity column eluates were measured by double dilution titers on Western blots by using IVAPLA₂ expressed as a glutathione-S-transferase (GST)-tagged protein in Spodoptera frugiperda (Sf9) insect cells¹¹  as a standard covering the range 0.62 to 5.0 μg IVAPLA2/mL; results are shown in Table 2.

The immunoblot (Figure 1) shows that the IVAPLA₂ present in U937 cells (lane 2) is effectively bound to the column, as no IVAPLA₂ was detected in the eluate (lane 3). SDS elution from the column showed the presence of a 90-kDa protein (lane 4) which was also present in the affinity-purified red cells (lane 5) and in the unpurified platelets (lane 7). Column recovery of IVAPLA₂ from the U937 lysate was 29%. Identification of the IVAPLA₂ depends on 2 antibodies. The affinity purification antibody is against the last 24 amino acids of the molecule's C terminus and has no homology with any other protein, whereas the probing antibody is against the first 216 amino acids at the N terminus. Figure 1 provides clear evidence that IVAPLA₂ present in U937 cells before and after affinity purification is also detected in red cells and platelets. Failure to detect IVAPLA₂ in the leukocyte preparation (lane 8) is probably a result of insufficient numbers of IVAPLA₂-containing leukocytes in the preparation (Table 1).

In the immunoblot the IVAPLA₂ from the 3 different sources migrated on polyacrylamide gel electrophoresis (PAGE) as a 90-kDa protein. The molecular weight of IVAPLA₂ on the basis of amino acid sequence is 85 kDa.¹²  Molecular weights reported for this protein have ranged from 110 kDa when purified from U937 cells,¹³  100 kDa when expressed and purified from Sf9 insect cells,¹¹  94 kDa in Chinese hamster ovary cells,¹⁴  and 90 kDa when purified from human platelets.¹⁵  Although our findings agree with the last report, there has never been an adequate explanation proposed for the differences in molecular weights reported for this protein.

In the blood used for affinity purification, leukocytes were reduced to less than 3% and platelets to less than 0.4% of that found in whole blood; nevertheless, it is possible that the remaining leukocytes and platelets could have contributed to the IVAPLA₂ isolated by the affinity purification from red cells. By using the results from the Western blot titering experiment we were able to estimate the IVAPLA₂ content of the nonerythroid cells (Table 2).

The U937 cell contained 292 fg IVAPLA₂/cell, consistent with overexpression of the protein; platelets contained 2.8 fg/cell. Because we could not detect IVAPLA₂ in the leukocyte, we assumed the amount present to be less than the detection limit of the titering experiment (ie, 0.62 μg/mL), and by using this figure we estimated the leukocyte has less than 7.8 fg/cell. However, the true figure is likely to be much less than this level. This figure now allows us to estimate the possible contribution of leukocytes and platelets to the IVAPLA₂ measured in the red cell affinity-purified material. In 100 mL red cells applied to the affinity column; platelets contributed 0.1 × 10⁹ × 2.8 × 10^-15 g = 0.28 × 10^-6 g (ie, 0.28 μg). Leukocytes contributed 0.02 × 10⁹ × 7.8 × 10^-15 g = 0.16 × 10^-6 g (ie, 0.16 μg). Total platelet and leukocyte contribution = 0.44 μg.

This amount is 1.5% of the 29.6 μg eluted from the red cell affinity column. We, therefore, conclude that the remaining platelets or leukocytes in the red cell transfusion pack only contributed a very small amount to the IVAPLA₂ found and that the detected IVAPLA₂ was derived from the red cells. The estimate of IVPLA₂ in the red cells can now be calculated as 0.16 fg/cell and assumes an equivalent recovery from the affinity purification as for U937 cells. This estimate is of the same order as found in eosinophils by another technique (ie, 0.38 fg/cell).¹⁰ 

In conclusion this report provides clear evidence that IVAPLA₂ is present in human red cells. Its function within the red cell is unclear, but it may provide a large mobile store for AA, allowing its release at sites in which there is a metabolic interaction between red cells and other cells, for example, platelets.¹⁶ 

We acknowledge the generous support of the South Glasgow University Hospitals NHS Trust and Dr A. S. Hutchison, Consultant Clinical Biochemist and Head of Department in offering research facilities in the laboratory. We thank Professor C. C. Leslie, Department of Pediatrics, National Jewish Medical and Research Centre, Denver, CO, for her generous gift of GST-cPLA₂. We are indebted to Mr H. Oliphant, Mr A. Kyle, and the staff of the Haematology Department, Victoria Infirmary, for their help in the cell preparation work.