Chemically Induced Degradation of MALT1 to Treat B-Cell Lymphomas

MALT1 is a protease and scaffold protein involved in signal transduction to NF-κB downstream of several receptors including B-cell (BCR) and T-cell receptors (TCR). MALT1 is aberrantly activated in ABC DLBCL by mutations in upstream genes in the BCR and TLR pathways (CD79A/B, CARD11, MYD88) and is critical for proliferation and survival. Recent studies by our lab, and others, identified inhibitors of MALT1 protease activity that revealed MALT1 is therapeutically targetable in ABC DLBCL. MALT1 is also essential for CLL, MCL and certain solid tumors (most notably lung cancer and glioblastoma). A first-in-man clinical trial recently started to evaluate MALT1 protease inhibition in B-cell non-Hodgkin's lymphomas. However, chronic inactivation of MALT1 protease activity suppressed T regulatory cells in vivo in protease dead murine models causing fast progressing autoimmune disease and death. Loss of MALT1, on the other hand, has also potent anti-tumoral effects but does not lead to autoimmunity in murine models. These findings prompted us to study alternative MALT1 targeting therapeutic approaches aimed to target its scaffolding activity.

We have developed a series of proteolytic targeting chimera (PROTAC) compounds against MALT1. PROTACs are bifunctional compounds that induce selective proteolysis by targeting proteins of interest to E3 ligases for directed proteosomal degradation. Our MALT1 PROTACs are based on an allosteric MALT1 inhibitor that binds reversibly to MALT1 and fused to a Cereblon (CRBN) binding moiety, with the intent of bringing the CRBN E3 ligase complex in close proximity of MALT1 and promoting its ubiquitination and proteasomal degradation. We performed structure-activity relationship analysis and studied: 1) three linker attachment points in the MALT1 binding moiety and evaluated their MALT1 enzymatic inhibitory activity and binding to MALT1; 2) the effect of distinct linker length and polarity in MALT1 degradation and, 3) compared the effect of Lenalidomide, Pomalidomide and CC-220, which have increasing CRBN affinities, as alternative CRBN-binding moieties. Six out of eighteen compounds presented higher than 50% MALT1 degradation at 1 μM compared to vehicle treated cells in a MALT1-dependent cell line, OCI-Ly3. The parental allosteric compound, on the other hand, did not affect MALT1 levels compared to vehicle treated cells and was used as a negative control for MALT1 degradation. Compounds that actively degraded MALT1 over 50% preserved selective killing of ABC DLBCL over GCB DLBCL, same as the parental MALT1 inhibitor. GI50 of active compounds in OCI-Ly3 was 2-6 μM while it was greater than 20 μM for the MALT1-independent cell line OCI-Ly1.

We chose two of our most effective and selective compounds to validate MALT1 PROTACs mechanism of action. Unlike their parental MALT1 targeting allosteric compound, MALT1 PROTACs effectively degraded MALT1 in a CRBN-dependent manner as shown in a 293T-CRBN knockout cell line or in OCI-Ly3 cells by treatment with 1 μM MLN4924. MLN4924 inhibits NEDD8-activating enzyme which is essential for the CRBN complex to function. Notably, MALT1 PROTACs degraded MALT1 in OCI-Ly1 cells (FC=-2.5) and Raji cells (FC=-1.7), where MALT1 is inactive. MALT1 degradation by PROTACs was not affected by activation in Raji cells, since PMA/ionomycin treatment did not alter the effect of MALT1 PROTACs on MALT1 levels. Therefore, MALT1 PROTACs can degrade MALT1 independent of its activation state. Moreover, unlike MALT1 protease inhibitors, MALT1 PROTACs potently suppress NF-κB activation, which is dependent on MALT1 scaffolding activity, as assessed by WB of phopho and total IκB in ABC DLBCL cell lines.

Our data shows that MALT1 PROTACs could be excellent agents for the treatment of ABC DLBCL and other lymphomas, providing an alternative to enzymatic targeting that might prove useful to avoid autoimmunity or overcome resistance mechanisms.