Tumor necrosis factor α primes cerebral endothelial cells for erythropoietin-induced angiogenesis

Abstract

Erythropoietin (EPO) enhances angiogenesis in the ischemic brain. Stroke induces secretion of tumor necrosis factor α (TNF-α). We investigated the effect of TNF-α on EPO-induced in vitro angiogenesis in cerebral endothelial cells. Using a capillary-like tubular formation assay, we found that transient incubation of primary rat cerebral microvascular endothelial cells (RECs) with TNF-α substantially upregulated EPO receptor (EPOR) expression and addition of EPO into TNF-α-treated RECs significantly augmented the capillary-like tube formation. Blockage of TNF receptor 1 (TNFR1) suppressed TNF-α-upregulated EPOR expression and abolished EPO-induced tube formation. Attenuation of endogenous EPOR with small interfering RNA (siRNA) also inhibited EPO-enhanced tube formation. Treatment of RECs with EPO activated nuclear factor-kappa B (NF-κB) and Akt. Incubation of the TNF-α-treated endothelial cells with EPO activated vascular endothelial growth factor (VEGF), VEGF receptor 2 (VEGFR2), angiopoietin 1 (Ang1), and Tie2. Blockage of VEGFR2 and Tie2 resulted in reduction of EPO-augmented tube formation. These data indicate that interaction of TNF-α with TNFR1 sensitizes cerebral endothelial cells for EPO-induced angiogenesis by upregulation of EPOR, which amplifies the effect of EPO on activation of the VEGF/VEGFR2 and Ang1/Tie2 pathways. Our results provide the evidence for crosslink between TNF and EPOR to coordinate the onset of angiogenesis in cerebral endothelial cells.

Figure 1

The effect of recombinant human tumor necrosis factor (rhTNF) and recombinant human erythropoietin (rhEPO) on the capillary-like tube formation. (A–E) Representative images of the capillary-like tube formation of rat brain microvascular endothelial cells (RECs) in control (A), rhTNF alone at a dose of 2.5 ng/mL (B), or 10 ng/mL (C), and rhEPO alone at a dose of 1 ng/mL (D), or 100 ng/mL (E). (F) Quantitative data of total lengths of the tube in mm/mm² (n=6/group). ^*P<0.05 versus the control group. Bar=100 μm.

Figure 2

Recombinant human tumor necrosis factor (rhTNF) sensitizes rat brain microvascular endothelial cells (RECs) for recombinant human erythropoietin (rhEPO)-induced capillary tube formation. (A–I) Representative images of the capillary-like tube formation of RECs in control (A), rhTNF alone (B), rhEPO alone (C), rhTNF and rhEPO (D), and rhTNF and rhEPO with antibody against TNF receptor 1 (TNFR1) (E), small interfering RNA (siRNA)-EPO receptor (EPOR) (F), scramble control (G), SU4832 (H), or antibody against anti-Tie2 (I). (J) Quantitative data of total lengths of the tube in mm/mm². (K, L) Real-time RT-PCR (K) and Western blot (L) analysis of EPOR mRNA and protein levels in RECs transfected with siRNA-EPOR (siEPOR) and scramble control (ssEPO). Glyceraldehyde-3-phosphate dehyrogenase (GAPDH) and β-actin were used as internal controls for mRNA and proteins, respectively. Bar=100 μm. ^*P<0.05 and ^#P<0.05 versus the control and TNF+EPO groups, respectively (n=6/group).

Figure 3

Effects of recombinant human tumor necrosis factor (rhTNF) on apoptotic cells and erythropoietin receptor (EPOR), NF-κB, and Akt levels. (A) Quantitative data of the percentage of apoptotic cells. Real-time RT-PCR (B) and Western blot (C) show EPOR expression in rat brain microvascular endothelial cell (REC) treated with rhTNF at different concentrations, rhTNF with the antibody against TNFRI (+anti-TNFRI) or against TNFRII (+anti-TNFRII). (D) Western Blot analysis of phosphorylated NF-κB and phosphorylated Akt in RECs treated with rhTNF at different times. (E) Western Blot analysis of EPOR levels in REC treated with rhTNF, rhTNF with the NF-κB inhibitor, SN50 (+SN) or with the phosphatidylinositol 3-kinase (PI3K) inhibitor, LY294002 (+LY). Glyceraldehyde-3-phosphate dehyrogenase (GAPDH) and β-actin were used as internal controls for mRNA and proteins, respectively. ^*P<0.05 and ^#P<0.05 versus the control and rhTNF groups, respectively (n=6/group).

Figure 4

Rat brain microvascular endothelial cells (RECs) treated with carbamylated erythropoietin (CEPO) promote capillary tube formation. (A–D) Representative images of the capillary-like tube formation of RECs in control (A), CEPO alone at a dose of 1 ng/mL (B), or 10 ng/mL (C), and CEPO (1 ng/mL) + recombinant human tumor necrosis factor (rhTNF) (5 ng/mL) (D). (E) Quantitative data of capillary tube formation. Bar=100 μm. ^*P<0.05 versus the control group (n=6/group).

Figure 5

Effect of recombinant human tumor necrosis factor (rhTNF) and recombinant human EPO (rhEPO) on expression of vascular endothelial growth factor (VEGF)/VEGFR2 and angiopoietin 1 (Ang1)/Tie2. Real-time RT-PCR (A) and Western blot (B and C) show VEGF/VEGFR2 and Ang1/Tie2 mRNA and protein levels, respectively, in rat brain microvascular endothelial cells (RECs) treated with rhTNF alone (5 ng/mL), rhEPO alone (10 ng/mL), rhTNF and rhEPO, rhTNF and rhEPO with the antibody against TNF receptor 1 (TNFR1) (+anti-TNFR1), rhTNF and rhEPO with small interfering RNA (siRNA)-EPO receptor (EPOR) (+siEPOR), or rhTNF and rhEPO with scramble control (+ssEPO). Glyceraldehyde-3-phosphate dehyrogenase (GAPDH) and β-actin were used as internal controls for real-time RT-PCR and Western blot analysis, respectively. ^*P<0.05 and ^#P<0.05 versus the control and TNF+EPO groups, respectively (n=6/group).