Mode of Action of the SDF-1/CXCL12 Inhibiting Spiegelmer^® Nox-A12 and Its Impact On Chronic Lymphocytic Leukemia (CLL) Cell Motility and Chemosensitization

Abstract 318

NOX-A12 is a novel Spiegelmer^®-based antagonist of SDF-1/CXCL12, a chemokine involved in the regulation of chronic lymphocytic leukemia (CLL) cell trafficking. Spiegelmers^® are mirror-image oligonucleotides that are identified to specifically bind to proteins in a manner conceptually similar to antibodies. Unlike aptamers, however, Spiegelmers^® are built from the non-natural L-isomer form of nucleotides which confers resistance to the action of nucleases and avoids potential immunogenicity.

CXCL12 is constitutively secreted and presented by bone marrow stromal cells (BMSC) via glycosaminoglycans (GAG) and acts as a homing factor for normal and malignant hematopoietic cells to the bone marrow (BM) and secondary lymphoid tissues via CXCR4 receptors that are expressed at high levels on circulating CLL cells. The microenvironment in the BM and secondary lymphoid tissues, in particular the CXCL12-CXCR4 axis, favors survival and chemotherapy-resistance of leukemic cells. We therefore investigated the effects of NOX-A12 in an in vitro co-culture system to model the interaction of CLL cells with their microenvironment.

Surprisingly we observed that NOX-A12 increased pseudoemperipolesis in vitro, i.e. spontaneous leukemia cell migration beneath BMSC. Interestingly, this NOX-A12 induced trans-migration of CLL cells was completely inhibited by the CXCR4 antagonist AMD3100, suggesting a CXCL12/CXCR4 dependent mechanism. We postulated that this observation might result from a direct effect of NOX-A12 on CXCL12 release by the stromal cells. Therefore, we investigated this hypothesis in different BMSC lines (MS-5, R15C, and TSt-4) and we found that NOX-A12 induced a significant CXCL12 release in all three tested cell lines.

We asked whether this NOX-A12 dependent increase of CXCL12 of BMSCs is due to release from either intracellular or extracellular storages. Intracellular staining of CXCL12 using flow cytometry did not reveal significant changes when BMSCs were incubated with NOX-A12. Furthermore, the transcription of CXCL12 was not found to be altered after NOX-A12 incubation over a period of three days as shown by quantitative RT-PCR. Rather, CXCL12 is released from extracellular storages of BMSCs. First hints were obtained through a rapid CXCL12 release within five minutes of incubation with NOX-A12. To confirm that CXCL12 is bound to the extracellular surface (by GAGs like heparin) and is being detached by NOX-A12 we first incubated BMSCs with NOX-A12, followed by a wash step and the addition of recombinant CXCL12. Recombinant CXCL12 was bound by BMSCs that were pre-incubated with NOX-A12 but not with a non-functional control (revNOX-A12), indicating that NOX-A12 strips off CXCL12. To corroborate the findings we incubated the BMSCs with heparin which also led to the release of CXCL12 in a dose dependent manner. Of note, the EC₅₀ of heparin regarding CXCL12 release was much higher compared to the EC₅₀ of NOX-A12 (≈ 12 μM vs. 5 nM) revealing the high affinity of NOX-A12 to CXCL12. The competition of NOX-A12 with heparin regarding CXCL12 binding was confirmed by Biacore experiments.

Based on these findings, we developed a novel adapted co-culture approach to examine the ability of NOX-A12 to chemosensitize CLL cells. In this setting, we first strip off CXCL12 from BMSCs by NOX-A12 and subsequently add CLL cells which will be either non-treated or treated with chemotherapy (fludarabine combined with bendamustine). We found that NOX-A12 slightly decreased CLL cell viability. As expected, a strong viability decrease was observed with chemotherapy, which could be even further decreased by the combination with NOX-A12, suggesting synergistic effects.

In conclusion, we propose that NOX-A12's mode of action is the release of extracellular bound CXCL12 and its subsequent inhibition. Since CXCL12 induces leukemia cell trafficking and homing to tissue microenvironment and also favors leukemia cell survival, we believe that targeting CXCL12 is an attractive approach to remove the protective effects of CXCL12-secreting BMSCs in order to sensitize CLL cells for subsequent chemotherapy. Thus, NOX-A12 represents a very promising agent to significantly improve the treatment of CLL. The compound is currently being tested in a Phase IIa study in relapsed CLL patients.