Background: Hematopoietic stem cells (HSC) interact with stromal cells, osteoblasts and matrix proteins in the hematopoietic niche. This interaction plays an important role in HSC trafficking, proliferation and differentiation. Significant data support the roles of both the SDF-1/CXCR4 and the VCAM1-VLA-4 axes in stem cell homing and mobilization. Two recent reports have shown that Natalizumab, an anti-VLA-4 monoclonal antibody used in the treatment of Multiple Sclerosis, induces mobilization of CD34+ HSC over several days. In the present study we tested the specificity and efficacy of a novel VLA-4 small molecule inhibitor, AMD15057, in murine preclinical studies of normal HSC mobilization.Methods: Fibronectin adhesion assays were performed with Jurkat cells to test the specificity of AMD15057. Ninety-six well plates were coated with fibronectin or BSA and the inhibition of Jurkat cell (VLA-4 +) adhesion by AMD15057 was determined in the presence or absence of PMA activation. To evaluate the mobilization of HSC progenitors, 8 week old 129/B6 F1 mice (n=6 per group, 3 experiments) were left untreated (unmobilized) or treated with AMD15057 (0.1,1,3 mg/kg iv or 3,5,7 mg/ kg sc), AMD3100 (1,3 mg/kg iv or 1,3,5 mg/kg sc), G-CSF (250 mg/kg/d × 5 d) or combinations. Total WBC and CFU-GM were determined at different time points for each mobilization regimen. For long-term competitive repopulation cell assays (LTRC), PBMCs from 3 Ly5.2+ unmobilized or mobilized mice (700uL PB each) were pooled, mixed with 5×105 Ly5.1+ BM mononuclear cells (MNC) (3:1 ratio) and injected into lethally irradiated Ly5.1+/Ly5.2+ compound heterozygote recipient mice. Chimerism was evaluated monthly for 6 months by flow cytometry. Secondary transplants to evaluate long term repopulation activity was performed by injecting lethally irradiated heterozygotes with 106 BM pooled MNC from the primary recipients.Results: The adhesion of untreated and PMA treated Jurkat cells to fibronectin coated wells was decreased by 62% (3.2 SD) and 69% (3.4 SD) (p<0.001) respectively, upon the addition of 1ug/ml AMD15057 compared to vehicle control. In vivo, a single iv or sc injection of AMD15057 resulted in maximum mobilization of CFU-GM within 0.5–0.75 hr. This effect was dose dependent for both sc and iv administrations. Maximum and comparable peak mobilization (13-fold iv; 9-fold sc, pNS) and kinetics of murine HSC mobilization was seen in mice receiving 1mg/kg iv and 5mg/kg sc AMD15057. In contrast, iv dosing of a small molecule inhibitor of CXCR4, AMD3100 (3mg/kg optimal dose) resulted in more rapid peak mobilization (10-fold) in <1hr compared to 3hr peak mobilization (20-fold) after sc dosing (5mg/kg optimal dose). Combination of AMD15057 (1mg/kg iv) with AMD3100 (5mg/kg sc) resulted in synergistic mobilization of CFU-GM (60-fold) when compared to each agent alone (p<0.01). In addition, when AMD3100 and AMD15057 were administered to mice after 5 days of mobilization with G-CSF (17-fold), a dramatic, rapid and reversible mobilization of CFU-GM was observed (200-fold) which was significantly higher than G-CSF+AMD3100 (90-fold) and G-CSF+AMD15057 (90-fold). LTRC assays confirmed that both AMD3100 and AMD15057 induced the rapid mobilization of short and long term repopulating cells and that this effect was synergistic when both agents were co-administered and exceed the LTRC seen after G-CSF mobilization. Secondary transplants confirmed the long term repopulating capacity of HSC mobilized with AMD15057.Conclusions: The VCAM1/VLA-4 axis is involved in HSC trafficking. AMD15057 is effective in blocking the interaction between VLA4 and its ligand fibronectin. AMD15057 induces rapid and reversible mobilization of normal progenitors and HSC which have the long term repopulating capacity. Finally, a dramatic synergistic effect was observed when AMD15057 was combined with AMD3100, G-CSF and the combination. The results provide a plausible foundation for replacing G-CSF with small molecule inhibitors of CXCR4 and VLA-4 for rapid and reversible HSC mobilization in humans.