Advanced glycation end products on stored red blood cells increase endothelial reactive oxygen species generation through interaction with receptor for advanced glycation end products

Abstract

Background: Recent evidence suggests that storage-induced alterations of the red blood cell (RBC) are associated with adverse consequences in susceptible hosts. As RBCs have been shown to form advanced glycation end products (AGEs) after increased oxidative stress and under pathologic conditions, we examined whether stored RBCs undergo modification with the specific AGE N-(carboxymethyl)lysine (N(ε) -CML) during standard blood banking conditions.

Study design and methods: Purified, fresh RBCs from volunteers were compared to stored RBCs (35-42 days old) obtained from the blood bank. N(ε) -CML formation was quantified using a competitive enzyme-linked immunosorbent assay. The receptor for advanced glycation end products (RAGE) was detected in human pulmonary microvascular endothelial cells (HMVEC-L) by real-time polymerase chain reaction, Western blotting, and flow cytometry. Intracellular reactive oxygen species (ROS) generation was measured by the use of 5-(and 6-)chloromethyl-2',7'-dichlorodihydrofluorescein diacetate, acetyl ester-based assays.

Results: Stored RBCs showed increased surface N(ε) -CML formation when compared with fresh RBCs. HMVEC-L showed detectable surface RAGE expression constitutively. When compared to fresh RBCs, stored RBCs triggered increased intracellular ROS generation in both human umbilical vein endothelial cells and HMVEC-L. RBC-induced endothelial ROS generation was attenuated in the presence of soluble RAGE or RAGE blocking antibody.

Conclusions: The formation of the AGE N(ε) -CML on the surface of stored RBCs is one functional consequence of the storage lesion. AGE-RAGE interactions may be one mechanism by which transfused RBCs cause endothelial cell damage.

Figure 1. N^ε-CML expression on stored RBC

(A) N^ε-CML is detectable on erythrocytes from a 21 d stored PRBC unit. 3 independent studies were performed, data is representative of 1 experiment. (B) N^ε-CML on fresh erythrocytes obtained from healthy donors (triangles) or on erythrocytes obtained from 42d old PRBCs units (circles). Erythrocytes from stored PRBC units showed increased N^ε-CML content when compared with fresh erythrocytes (p=0.003). Two-tailed Student’s t test was used to determine significance using SigmaPlot 10 software (Systat Software Inc., Chicago, IL, USA).

Figure 2. RAGE Expression on Human Pulmonary Microvascular Endothelial Cells

(A) RNA gel of RAGE mRNA in Human Pulmonary Microvascular Endothelial Cells (HMVEC-L) or Human Embryonic Kidney Cells (293) demonstrates increased RAGE transcripts in HMVEC-L (B) Western blot of extracts from HMVEC-L or 293 cells demonstrates increased RAGE expression in HMVEC-L. Densitometry is normalized to ß Actin. (C) FACS analysis demonstrates surface expression of RAGE on HMVEC-L, PECAM expression serves as a positive control. FACS analysis was repeated 3 times.

Figure 3. Endothelial Cell ROS Generation Following Incubation with Stored Erythrocytes is attenuated by RAGE blockade

(A) HUVEC were incubated with fresh or stored erythrocytes and ROS generation was measured as described in the methods. Incubation with fresh erythrocytes did not increase EC ROS generation (p=0.143), whereas incubation with stored erythrocytes increased EC ROS generation (*p=0.005 for LR, ^ p=0.002 for non-LR erythrocytes). When compared with fresh erythrocytes, both the LR and non-LR erythrocytes increased EC ROS generation (**p=0.026 for LR, ^^p =0.009 for non-LR erythrocytes). (B) Immortalized Human Umbilical Vein Endothelial Cells (IVECs) were incubated with erythrocytes, with or without RAGE blocking antibody or IgG control (100 ug/mL) as described in the methods. Endothelial cell ROS generation was increased over baseline following incubation with erythrocytes (*p=0.001). This effect was attenuated with RAGE blocking antibody (+p=0.004) but not in the presence of IgG control antibody (p=0.428). (C and D) Measurement of DCF fluorescence generated by HMVEC-L following incubation with fresh erythrocytes, stored erythrocytes or stored erythrocytes in the presence of soluble RAGE (sRAGE, 75 μg/mL). (C) Fresh erythrocytes did not increase ROS generation over baseline (p=0.100). Stimulation of endothelial cells with stored erythrocytes increased ROS generation when compared with ECs alone (**p=0.001) and ECs stimulated with fresh erythrocytes (+p=0.024). In the presence of sRAGE, EC ROS generation following stimulation with stored erythrocytes was significantly attenuated (*p=0.002). (D) Phase (left panel) and fluorescence (right panel) images of (subpanel a) EC with or without RBC + sRAGE as indicated in figure.