Bacteria-induced release of white cell--and platelet-derived vascular endothelial growth factor in vitro

Poor prognosis after resection of primary colorectal cancer may be related to the combination of perioperative blood transfusion and subsequent development of infectious complications. White blood cell--and platelet-derived cancer growth substances, including vascular endothelial growth factor (VEGF), may be involved in this process. Therefore, we studied the in vitro release of VEGF from white blood cells and platelets stimulated by bacterial antigens and supernatants from stored red cell components.

Eight units of whole blood (WB) and eight units of buffy-coat-depleted red cell (SAGM) blood were donated by healthy blood donors. Subsequently, half of every unit was leucocyte depleted by filtration, and all 32 half-units were stored under standard conditions for 35 days. Just after storage, and on days 7, 21 and 35 during storage, aliquots of the supernatants were removed from the units and frozen at -80 degrees C. WB from other healthy donors was stimulated for 2 h with sodium chloride (controls), with Escherichia coli lipopolysaccharide (LPS) alone, or with LPS plus supernatants from the non-filtered or prestorage leucofiltered WB units (diluted 1:10), or from non-filtered or prestorage leucofiltered SAGM blood units (diluted 1:20) stored for 0, 7, 21, or 35 days, respectively. Similar assays were performed using Staphylococcus aureus-derived protein A as a stimulatory antigen. The concentration of VEGF was determined in supernatants from stored blood and in assay supernatants by using enzyme-linked immunosorbent assay (ELISA).

The concentration of VEGF increased significantly (P < 0.0001) in a storage time-dependent manner in non-filtered WB and SAGM blood, while the increase was abrogated by prestorage leucofiltration. The supernatant concentration of VEGF was significantly increased in LPS-stimulated (P = 0.002) and in protein A-stimulated (P < 0.0001) assays compared with controls. Addition of supernatants from stored, non-filtered WB or SAGM significantly increased the assay supernatant VEGF concentration storage-time dependently (P = 0.006) in LPS assays. In protein A assays, only supernatants from non-filtered WB significantly increased the assay supernatant VEGF concentration storage-time dependently (P = 0.022). This additional effect by supernatants from stored blood components was not observed with prestorage leucofiltered blood.

Extracellular VEGF may accumulate in non-filtered red cell components, but this can be prevented by prestorage leucocyte depletion using filtration. In addition, bacterial antigens appear to induce release of VEGF from white blood cells and platelets. Addition of supernatants from stored, non-filtered WB or SAGM blood may increase the VEGF levels in a storage time-dependent manner, while prestorage leucofiltration may prevent further increase by supernatants.