Inhibiting SOX11-DNA Interaction in Mantle Cell Lymphoma

Mantle Cell Lymphoma (MCL) is a fatal subtype of Non-Hodgkins Lymphoma. MCL is characterized by cell cycle dysregulation due to Cyclin D1 (CCND1) overexpression. Murine models overexpressing CCND1 do not develop B-cell proliferation. The SOX11 transcription factor is overexpressed in the majority of nodal human MCL and has been implicated in MCL pathogenesis. We have developed a transgenic Eµ-SOX11-EGFP mouse model that overexpresses SOX11 with aberrant B cell proliferation, lymphadenopathy, spenomegaly and hepatomegaly phenocopying the characteristics of patients with MCL.

Using Mass Cytometry (CyTOF), we further characterized this B cell population to be CD23-, CD21/35 dim, CD138-, high surface IgM, and variable IgD expression, an immunophenotype identical to human MCL. SOX11 overexpression drives B cell receptor (BCR) signaling in murine MCL cells, assayed by phosphorylation of p-BTK and p-PLC gamma.

To further the study of the role of SOX11 in MCL, compounds inhibiting the binding interaction of SOX11 and DNA are highly desirable and we aim at their discovery by using structure-based virtual screening predictions and experimental validations. No small-molecule inhibitor of SOX11-DNA binding has been developed to date and there is an unmet need in MCL therapeutics for novel and successful treatments.

Transcription factors have been considered "undruggable" but since SOX11 binds the minor groove in DNA, we hypothesized that there may be pockets created by SOX11-DNA binding that could be inhibited by small molecules. We therefore built a human SOX11 homology model using the crystal structure of murine SOX4 (98% homology). We first docked molecules from the NIH Chemical Genomics Center (NCGC) Pharmaceutical Collection (NPC). From this virtual screening, 70 compounds were purchased for experimental testing. Using a fluorescence anisotropy assay, Flavitan was identified as a potential inhibitor of the SOX11-DNA interaction. Based on inferences from the predicted binding mode of Flavitan on SOX11, we performed another virtual screening of ~12 million molecules from the ZINC database followed by assaying 26 top-scoring compounds, as well as, 16 derivatives of one of them, for SOX11 binding. We thus discovered two compounds, R and T, with binding constants in the low micromolar range. Importantly, compounds R and T were found not to intercalate DNA in a Topoisomerase I based assay.

To assess the biological activity of these molecules, we tested whether BCR signaling in MCL cell lines expressing SOX11 would be inhibited by treatment with these compounds. Compounds R and T significantly inhibited p-BTK as compared to vehicle control. Ibrutinib, an FDA approved BTK inhibitor, was used as a positive control in these experiments.

To determine whether BCR pathway inhibition through compounds R and T would be cytotoxic in SOX11 positive MCL cell lines, we treated Z138 and JEKO cells with these small-molecules and assayed Annexin-V and PI staining 48 hours later by flow-cytometry. Our results indicate that compound R has greater single agent cytotoxic activity than compound T and Ibrutinib. Compounds R and T also showed synergistic cytotoxicity in combination with Ibrutinib in SOX11 expressing Z138 and JEKO cells, but not in JVM2, a SOX11 negative MCL cell line.

In summary, through a combination of in silico predictions and experimental validation, we have identified small-molecule inhibitors of SOX11-DNA binding as candidate leads for optimization for anti-MCL therapy.