Introduction: We successfully identified a subset of CD11b+ monocytes expressing CX3CR1, the only receptor of fractalkine (Fkn; CX3CL1), in G-CSF mobilized peripheral blood stem cell collection (PBSC). We found that Fkn-treated CD11b+CX3CR1+ monocytes express pigment epithelium-derived factor (PEDF) once Fkn activates Fkn/CX3CR1 signaling. PEDF is a glycoprotein, which belongs to the serpin family and involves various physiologic processes such as angiogenesis, cell proliferation, and survival. In this study, we investigate the role of PEDF and its mechanism of action in promoting angiogenesis. Materials andmethods: To mobilize mononuclear cells (MNCs) including CD11b+CX3CR1+ monocytes into peripheral blood (PB), heathy donors were subcutaneously injected with G-CSF (10μg/kg) for 5 days. Apheresis MNCs were collected from donor PB using a COBE spectra cell separator (COBE, Lakewood, CO) 5 days after daily G-CSF injection. Then, CD11b+CX3CR1+ cells were isolated with a MoFlo™ XDP Cell Sorter (Beckman Coulter, Brea, CA). Human umbilical vein endothelial cells (HUVECs) were isolated from human cords and cultured at 37 °C in 5% CO2 on top of collagen in four different settings: (1) untreated (HUVECs alone) (2) VEGF (1ng/mL) treated HUVECs; (3) PEDF (3,30 and 300 ng/mL) treated HUVECs; (4) VEGF (1ng/mL)+PEDF (3,30 and 300 ng/mL) treated HUVECs. To assess dose-dependency, PEDF was administered at different dosages (0, 3, 30 and 300 ng/mL) for 7 days in either the presence or absence of VEGF (1ng/mL). HUVEC growth areas were measured by image J. Results: CD11b+CX3CR1+ monocytes were found at 19.6±3.58% of total MNCs in human G-CSF mobilized PBSC. Fractalkine treatment (50ng/mL for 30 minutes) of these monocytes promoted endothelial cell proliferation of HUVECs and increased PEDF expression according to the angiogenic protein array results. Administration of PEDF inhibitor significantly decreased HUVEC proliferation and vascular structure formation compared to the control group (p <0.001). We further investigated pro-angiogenic effects of PEDF on HUVEC proliferation. In PEDF (3,30 and 300 ng/mL) treated HUVECs, PEDF itself has minimal effect on HUVEC proliferation. However, EC sprouting at the edge of vessel-like structures was greater in the 3 and 30 ng/mL. Interestingly, there was no EC sprouting with 300 ng/mL PEDF. PEDF (3 and 30 ng/mL) itself showed little effect on HUVEC proliferation. However, co-culturing with VEGF (1ng/mL) and low PEDF concentrations (3 and 30 ng/mL) significantly increased EC proliferation (EC proliferation area (mm2), 1.7±0.4 (VEGF only) vs 3.0±0.6 (VEGF+3 ng/mL PEDF) and 2.5±1.1 (VEGF+30 ng/mL PEDF); p=0.002 and 0.048, respectively). Higher concentration of PEDF did not induced EC proliferation even in the presence VEGF (1.7±0.4 (VEGF only) vs 1.5±0.9 (VEGF+300 ng/mL PEDF); p=0.192). The results suggest that low concentrations of PEDF (3 and 30 ng/mL) has no effect on HUVEC proliferation, but has a potent effect on HUVEC migration regardless of VEGF activity. PEDF is not pro-angiogenic by itself, but can markedly enhance HUVEC proliferation in PEDF (3 and 30 ng/mL) treated HUVECs with VEGF. Conclusion: Fractalkine-treated CD11b+CX3CR1+ monocytes promote angiogenesis by increasing PEDF expression, which acts as a pro-angiogenic factor only under conditions where VEGF is present.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.