A Comparison of the Ability of AMD3100 Versus G-CSF to Induce Angiogenesis and Mobilize Endothelial Progenitor Cells (EPCs).

There is emerging evidence that circulating EPCs are able to home to sites of vascular injury and facilitate revascularization. However, under normal conditions, the number of EPCs in the blood is small, potentially limiting revascularization following acute injury. Previous reports suggest that certain cytokines including G-CSF can mobilize EPCs into the blood, potentially circumventing this limitation. However, the kinetics and magnitude of EPC mobilization by G-CSF have not been carefully defined and the angiogenic capacity of mobilized EPCs unclear. Herein, we characterize the ability of G-CSF treatment to improve revascularization in a mouse model of acute hindlimb ischemia. Since it is possible that EPC mobilization by G-CSF is delayed (maximal mobilization of hematopoietic stem cells (HSC) occurs after 4-5 days), we also studied AMD3100, a CXCR4 antagonist, that has been shown to lead to rapid and robust mobilization of HSC. Four groups of mice (n=7–10, each) were treated with saline alone, G-CSF (250μg/kg per day for 7 days by miniosmotic pump), AMD3100 (5mg/kg SQ on days 1–3), or a combination of both G-CSF and AMD3100; treatment was initiated immediately after induction of unilateral hindlimb ischemia.

Revascularization was monitored by laser Doppler imaging and the ratio of perfusion between ischemic and non-ischemic calculated. At 2 weeks post surgery, a significant improvement in ischemic limb perfusion was observed in G-CSF and G-CSF+AMD3100 treated mice compared with saline treated mice with a trend towards improved perfusion noted in the AMD3100 treated mice [ratio of ischemic vs. non-ischemic limb perfusion + SEM: 71.9 ± 5.2% (saline treated); 89.2 ± 4.7%, p<0.05 (G-CSF); 80.5 + 6.6%, p=NS (AMD3100); 94.5 ± 3.6%, p<0.01 (G-CSF+AMD3100)]. In all treatment groups, the kinetics of revascularization was more rapid compared with control mice. In particular, the combination of G-CSF and AMD3100 resulted in the most rapid increase in perfusion with significant differences observed as early as 2 days post surgery (28.2 ± 2.5% compared with 15.3 ± 1.4% for control mice, p<0.01).

As described more completely in a separate abstract, we have initiated a phase II study to compare the safety and efficacy of AMD3100 versus G-CSF for mobilization and transplantation of HSC for HLA-matched sibling allografting. Donors are sequentially mobilized with AMD3100 and G-CSF. To compare the ability of AMD3100 and G-CSF to mobilize EPCs, we have begun to quantify the number of EPCs in the blood at baseline and after treatment with AMD3100 or G-CSF. A culture-based method was used to identify both early outgrowth EPCs (scored on day 7; predominantly CD14⁺, CD31⁺, monocytic cells) and late outgrowth EPCs (scored on day 14–21; predominantly CD45⁻, CD14⁻, CD31⁺ endothelial cells). To date only a single patient has been analyzed completely. Interestingly, treatment with G-CSF leads to a more robust mobilization of early EPCs (fold increase compared with baseline: 4.7-fold and 25-fold for AMD3100 and G-CSF, respectively). In contrast, treatment with AMD3100 leads to a more robust mobilization of late EPC. Surprisingly, no late outgrowth EPC were detected in 40 ml of baseline or G-CSF treated blood, whereas 26 were detected in 40 ml of AMD3100 treated blood. Studies are underway to confirm these findings with additional donors.

Collectively, these data show that treatment with AMD3100 alone or in combination with G-CSF leads to enhanced revascularization following acute vascular injury and the rapid mobilization of EPCs.