Novel Mouse Models of Chronic Lymphocytic Leukemia (CLL) Unravel the Molecular Mechanisms Controlling Bone Marrow Involvement by Leukemic B Cells.

Abstract 360

In CLL, the bone marrow (BM) represents a typical site of involvement and relapse, suggesting a preferential homing of leukemic cells to this anatomical site compared to other lymphoid organs, though the mechanisms controlling CLL cell migration and accumulation within the BM are unclear.

In order to define the rules driving in vivo CLL cell re-circulation between the blood and tissutal compartments, we specifically generated two different mouse models and investigated the role played by HS1; this molecule, other than being a putative prognostic factor in CLL, is also involved in cytoskeleton reorganization of lymphocytes, and, potentially, in the control of cellular shape, migration and homing.

First, we established a novel transplantable xenograft murine model of CLL by engrafting the cell line MEC1 into RAG2^-/-γ_c^-/- mice, at a variance with previous studies in nude mice where MEC cells failed to engraft. Likely due to the lack of B, T and NK cells (while nude mice retain NK cells), RAG2^-/-γ_c^-/- animals were successfully transplanted with the CLL cell line through either subcutaneous or intravenous routes, resulting in a systemic blood and tissutal involvement. When subcutaneous MEC1 cells silenced for HS1 expression were injected in these animals, we observed a preferential localization in the tumor draining axillary and inguinal lymph nodes and especially in the BM, when compared to controls.

As we have previously demonstrated that CLL cases with hyper-phosphorylated HS1 show a worse clinical outcome, we took advantage of this mouse model to investigate the in vivo homing ability of primary CLL cells from patients showing different HS1 phosphorylation patterns. Purified leukemic cells from 4 patients with hyper-phosphorylated HS1 were labeled with high concentration of the dye CSFE, and each sample was paired and admixed with purified CLL cells obtained from patients with low levels of HS1 phosphorylation and separately labeled with low CSFE concentration. Each pair of samples was injected i.v. into RAG2^-/-γ_c^-/- mice recipients. When we analyzed the different organs of the animals by flow-cytometry, the differential expression of CFSE fluorescence (CFSE-high vs CFSE-low) allowed us to distinguish between the two leukemic cell populations with opposite HS1 phosphorylation status. In 3/4 experiments, CLL cells with hyper-phosphorylated HS1 revealed a preferential homing to the BM.

Based on these results and on the in vitro evidence that B lymphocytes from HS1^-/- mice have an impaired spontaneous migration, we have crossed HS1^-/- (H^-/-) mice with the Eμ-TCL1 transgenic (T^tg) mouse, an animal model that between 13 and 18 months of age develops a disease resembling human CLL. In the H^-/-/T^tg mice, monoclonal CD19⁺CD5⁺ cells became evident earlier (at 7-13 months of age) and in significantly higher proportion as compared to Eμ-TCL1 transgenic mice. Cells preferentially localized in the BM where leukemic cells are usually observed at low frequencies in the Eμ-TCL1 mouse (mean value: 28%±16 vs 5%±2, respectively, p=0,008).

These findings suggest that HS1 may have a relevant role in both normal and leukemic B-cells and in particular is crucial for cell migration, through its involvement in cytoskeleton organization. Accordingly, we also provide evidence that, in the absence of HS1, cells fail to form actin-myosin complexes, leading to an instability of the cell signalling complex. Our findings suggest a relationship between the expression of HS1 and the development and progression of CLL, most notably in terms of BM involvement, indicating that specific abnormalities in the cytoskeleton organization may be pivotal in regulating leukemic migration and infiltration in selected anatomical sites.

This points at HS1 as a target for development of novel cancer treatments, aiming at interfering with the lymphoid tissue infiltration and invasion which is characteristic of the disease. In addition, these animal models could become very useful for evaluating the biological basis of CLL growth and dissemination as well as the efficacy of new therapeutic agents.