(DCP003) APPLICATION OF A HIGH-THROUGHPUT SCREENING APPROACH TO REPURPOSE PREVIOUSLY-APPROVED DRUGS TO TREAT TYPE 1 DIABETES

Background: Type 1 diabetes (T1D) accounts for approximately 10% of all diagnosed cases of diabetes, and people living with T1D require lifelong insulin injections due to the profound loss of insulin-producing beta-cells. Recently, the US Food and Drug Administration (FDA) approved Teplizumab for treating T1D, but the high cost of this monoclonal antibody may compromise its broad use. The success of Teplizumab indicates that depleting autoreactive T cells can effectively protect beta-cells from undergoing cell death. An alternative strategy is to repurpose previously-approved drugs for T1D management. Drug repurposing reduces research and development costs and accelerates drug discovery; however, proper drug targets or assays are needed. We previously demonstrated that scaffold protein 14-3-3(z)eta promotes cell survival through its inhibitory actions on pro-apoptotic BCL-2 proteins, including BAD. Thus, we hypothesize that small molecules that disrupt 14-3-3z:BAD interactions could induce apoptosis in T cells, representing a promising drug target for finding novel anti-T1D agents.

METHODS AND RESULTS: Confocal microscopy was used to confirm that inhibiting 14-3-3z leads to the translocation of BAD from the cytoplasm to mitochondria, where it initiates cell apoptosis. To screen for compounds that disrupt 14-3-3z:BAD interactions, we developed bioluminescence resonance energy transfer (BRET) sensors by fusing luciferase and mCitrine to 14-3-3z and BAD fragments, respectively. Sensors were validated for their performance with well-established 14-3-3 inhibitors (FTY720 and I, 2-5). Treatment of transfected cells that express the BRET sensor with FTY720 and I, 2-5 caused significant dose- and time-dependent reductions in BRET. We next adapted the sensor to a high-throughput format and screened for inhibitors of 14-3-3z:BAD interactions from a 1971 FDA-approved drug library. Expression of a BAD mutant that contained Ser112/136A mutations, which cannot interact with 14-3-3z, reduced BRET by 25% (baseline); thus, any drug that reduced BRET by more than 25% was considered to completely dissociate 14-3-3ζ:BAD complexes. Our screen identified 138 hits, which were further assessed for their capacity to induce cell death and apoptosis by high-content analysis of Hoechst, propidium iodide, and Annexin V incorporation. Hits that induced over 40% of cell death were selected for further testing in CD8+ T cells from mice and humans.

Conclusion: We have effectively developed a novel BRET sensor that can detect 14-3-3z:BAD interactions to permit screening for pro-apoptotic compounds in living cells. The results of our screen may lead to the discovery and re-purposing of drugs that delay the progression of T1D by killing cytotoxic T cells.