In Vivo Mobilization of Leukemic Human Precursor-B-ALL Cells by the CXCR4-Antagonist AMD3100 Is Via Secretion of SDF-1 and Synergistically by Catecholamine Action.

Introduction Mobilization of leukemic cells from the bone marrow (BM) to the circulation in order to better kill them with DNA damaging chemotherapy agents is emerging as a new experimental therapeutic intervention, however the mechanism is not entirely clear. Currently CXCR4-antagonists such as the mobilizing agent AMD3100 (AMD) are becoming available for clinical usage. The aim of this study is to explore mechanisms of human precursor-B-ALL cell mobilization from the BM in a functional, pre-clinical immune deficient mouse model.

Methodology Immunodeficient mice were stably engrafted with the childhood pre-B-ALL leukemic cell line G2 (4 weeks after transplantation in NOD/SCID mice) and with primary childhood precursor-B-ALL cells from 4 patients (4-8 weeks after transplantation in NOD/SCID IL2R {gamma} null and NOD/SCID/B2m(null) mice). Two of the patients had a translocation (t4;11) (pro-B-ALL). All human leukemias were engrafted without prior irradiation of the mice. This approach prevents possible irradiation damage to the host microenvironment and thereby leads to a model which better mimics growth of human leukemias. To accommodate for differences in the level of leukemic BM engraftment (FACS analysis for huCD45+ cells), we assessed the leukemia mobilization level by calculating a leukemia mobilization index: WBC x % leukemic cells in the PB / % leukemic cells in the BM.

Results and Discussion Treatment with AMD leads to a significant mobilization of all transplanted leukemias with a mobilization level of between 3 – 8 times above baseline. As we recently showed for mobilization of normal murine progenitors, AMD induces a strong release of SDF-1 from the BM (Dar et al. ASH 2006). To examine if this is also instrumental for the leukemia mobilization process, we inhibited SDF-1 action by injection of neutralizing CXCR4 antibodies (clone 12G5) in leukemic chimeras. This led to an abrogation of AMD-induced leukemia mobilization. Pointing towards the same mechanism, 3 daily injections of fucoidan, a known SDF-1 releasing agent, also led to significant leukemia mobilization in G2 and precursor-B-ALL chimeras. Recently we demonstrated that human hematopoietic stem and progenitor cells express receptors for catecholamines, such as dopamine and epinephrine (Epi) and that treatment with catecholamines leads to mobilization of murine progenitor cells (Spiegel et al. Nat. Immunol. 2007). Accordingly, we examined the effect of neurotransmitters. First, we found that the G2 cell line and all 4 examined precursor-BALL samples express the catecholamine receptors D3, D5 and beta-2. The expression is dynamic, as it was, in part, increased after engraftment of immunodeficient mice. Treatment of chimeras with high doses of Epi alone led to leukemia mobilization in vivo similar to AMD-induced mobilization. In combination with AMD, lower doses of norepinephrine increased leukemia obilization synergistically and significantly, resulting in dramatic leukemia mobilization up to 20 times above baseline. Unexpectedly and in contrast to normal cells, treatment of chimeras with the beta-2 agonist clenbuterol was accompanied by inhibition of AMD-induced mobilization of leukemic cells. These observations suggest similarities and differences in the activation of catecholamine receptors in the mobilization process of normal and leukemic cells.

Conclusions Our results show that SDF-1 has a crucial role in AMD-induced leukemic cell mobilization. Human leukemias can be mobilized by catecholamine action synergistically with AMD in immunodeficient mice. This approach could be potentially used for future mobilization protocols of leukemia in combination with established chemotherapy to improve eradication of minimal residual disease of leukemia.