CD38 and ZAP-70 Mark Chronic Lymphocytic Leukemia (CLL) Cells with High Migratory Potential and Regulate Chemotaxis.

The absence of mutations in the IgVH genes along with the expression of the cell surface receptor CD38 and the cytoplasmic kinase ZAP-70 are independent molecular markers of unfavourable outcome in CLL patients. Our group has been interested in unravelling the role of CD38 in CLL not only as disease marker but also as pathogenetic agent. We have shown that CD38 can positively affect proliferation and survival of CLL cells through binding CD31, the ligand expressed by stromal, endothelial and nurse-like cells (NLC). Starting points of our current investigation are that

• NLC produce high levels of the attractant stromal cell-derived factor-1 (SDF-1α/CXCL12),

• CD38 is an essential mediator of migration and homing in murine neutrophils, and

• ZAP-70 is a key element in CXCR4 (SDF-1α receptor) pathway in T cells.

The consequent working hypothesis is that CD38 and ZAP-70 intertwine to influence CLL cell migration. The analysis of a cohort of 56 clinically and molecularly characterized CLL patients demonstrated that CD38⁺/ZAP-70⁺ samples migrate significantly better towards CXCL12 than all the other CLL subgroups, including the CD38^+/ZAP-70⁻ one, suggesting a functional association between CD38 and ZAP-70. CD38 engagement by agonistic monoclonal antibodies (mAbs) leads to phosphorylation of the activatory tyrosines present in ZAP-70 in 26 CLL cell cases and in an ad hoc modified B line expressing a ZAP-70-myc construct. More importantly, ZAP-70 is required for CD38 signalling, as the CD38 pathway is impaired in CD38⁺/ZAP-70⁻ CLL cells. Microarray analyses performed on a subgroup of 20 CLL patients indicate the existence of a genetic signature marking CLL samples endowed with a high migratory potential. These studies resulted in the identification of a panel of 24 differently expressed genes, many of which are known to be involved in adhesion and migration. Taken together, the above data point to a direct role for the CD38/ZAP-70 pathway in CLL migration towards CXCL12. Indeed, CXCL12-mediated chemo taxis of CLL cells and of the NALM-6 pre-B cell line can be efficiently blocked by anti-CD38 mAbs in a dose-dependent fashion. Out of the 9 different anti-CD38 mAbs tested, only the ones binding epitopes close to the C terminus of the molecule interfere with migration, while the ones binding closer to the plasma membrane do not. Pre-treatment of CLL and NALM-6 cells results in the almost complete block of the CXCR4 signalling pathway. This was shown by monitoring Ca²⁺ mobilization, actin polymerization and ligand-induced internalization. Further, upon exposure to CXCL12, CD38 and CXCR4 become physically associated, as shown using confocal microscopy and co-immunoprecipitation approaches. CD38 however, does not internalize upon CXCL12 biding to CXCR4. Collectively, these data suggest that the blocking effects of the anti-CD38 mAbs are due to steric hindrance. In conclusion, we believe that these results offer biological evidence explaining why the combination of CD38 and ZAP-70 results in a more efficient identification of CLL patients with aggressive disease. Further, it may provide a rationale for using anti-CD38 ligands or inhibitors as an adjunct tool for the therapy of aggressive CLL.