Apoptosis-inducing factor (AIF) exhibits reactive oxygen species (ROS)-generating NADH oxidase activity of unknown significance, which is dispensable for apoptosis. We knocked out the aif gene in two human colon carcinoma cell lines that displayed lower mitochondrial complex I oxidoreductase activity and produced less ROS, but showed increased sensitivity to peroxide-or drug-induced apoptosis. AIF knockout cells failed to form tumors in athymic mice or grow in soft agar. Only AIF with intact NADH oxidase activity restored complex I activity and anchorage-independent growth of aif knockout cells, and induced aif-transfected mouse NIH3T3 cells to form foci. AIF knockdown in different carcinoma cell types resulted in lower superoxide levels, enhanced apoptosis sensitivity and loss of tumorigenicity. Antioxidants sensitized AIF-expressing cells to apoptosis, but had no effect on tumorigenicity. In summary, AIF-mediated resistance to chemical stress involves ROS and probably also mitochondrial complex I. AIF maintains the transformed state of colon cancer cells through its NADH oxidase activity, by mechanisms that involve complex I function. On both counts, AIF represents a novel type of cancer drug target.

A novel type of antibacterial screening method, a target mechanism-based whole-cell screening method, was developed to combine the advantages of target mechanism- and whole-cell-based approaches. A mycobacterial reporter strain with a synthetic phenotype for caseinolytic protease (ClpP1P2) activity was engineered, allowing the detection of inhibitors of this enzyme inside intact bacilli. A high-throughput screening method identified bortezomib, a human 26S proteasome drug, as a potent inhibitor of ClpP1P2 activity and bacterial growth. A battery of secondary assays was employed to demonstrate that bortezomib indeed exerts its antimicrobial activity via inhibition of ClpP1P2: Down- or upmodulation of the intracellular protease level resulted in hyper- or hyposensitivity of the bacteria, the drug showed specific potentiation of translation error-inducing aminoglycosides, ClpP1P2-specific substrate WhiB1 accumulated upon exposure, and growth inhibition potencies of bortezomib derivatives correlated with ClpP1P2 inhibition potencies. Furthermore, molecular modeling showed that the drug can bind to the catalytic sites of ClpP1P2. This work demonstrates the feasibility of target mechanism-based whole-cell screening, provides chemical validation of ClpP1P2 as a target, and identifies a drug in clinical use as a new lead compound for tuberculosis therapy.

Human caspase-4 does not have a corresponding mouse ortholog. Caspase-4 falls within the class of “inflammatory caspases,” being homologous with human caspases 1 and 5 and mouse caspases 1, 11, and 12. To address the function of caspase-4, we generated caspase-4-deficient human THP1 monocytic cell lines which exhibited substantially reduced LPS-induced secretion of several chemokines and cytokines, including IL-8 (CXCL8), CCL4 (macrophage-inflammatory protein-1β), CCL20 (macrophage-inflammatory protein-3α), and IL-1β. The LPS-induced expression of the mRNAs encoding these cytokines was correspondingly reduced in the caspase-4-deficient clones. Because a specific NF-κB inhibitor blocked LPS-induced IL-8 and CCL4 mRNA expression as well as IL-8 and CCL4 secretion in THP1 cells, we investigated the role of caspase-4 in NF-κB signaling. LPS-induced NF-κB nuclear translocation and activation were inhibited in all caspase-4-deficient clones. LPS stimulation led to the interaction of endogenous caspase-4 and TNFR-associated factor 6 (TRAF6) via a TRAF6-binding motif (PPESGE), which we identified in caspase-4. Mutation of this site in caspase-4 resulted in the loss of the TRAF6-caspase-4 interaction. Similar TRAF6-binding motifs are known to be functionally important for TRAF6 interactions with other molecules including caspase-8, and for mediating NF-κB activation in various immune and nonimmune cell types. Our data suggest that the TRAF6-caspase-4 interaction, triggered by LPS, leads to NF-κB-dependent transcriptional up-regulation and secretion of important cytokines and chemokines in innate immune signaling in human monocytic cells.

To identify novel inhibitors of Mycobacterium tuberculosis cell envelope biosynthesis, we employed a two-step approach. First, we screened the diverse synthetic small molecule 71,544-compound Enamine library for growth inhibitors using the non-pathogenic surrogate Mycobacterium bovis BCG as screening strain and turbidity as readout. Second, 16 confirmed hits were tested for their ability to induce the cell envelope stress responsive promoter piniBAC controlling expression of red fluorescent protein in an M. bovis BCG reporter strain. Using a fluorescence readout, the acetamide E11 was identified. Resistant mutant selection and whole genome sequencing revealed the mycolic acid transporter Mmpl3 as a candidate target of E11. Biochemical analysis using mycobacterial spheroplasts and various membrane assays suggest that E11 indirectly inhibits MmpL3-facilitated translocation of trehalose monomycolates by proton motive force disruption. E11 showed potent bactericidal activity against growing and non-growing M. tuberculosis, low cytotoxic, and hemolytic activity and a dynamic structure activity relationship. In addition to activity against M. tuberculosis, E11 was active against the non-tuberculous mycobacterium M. abscessus, an emerging opportunistic pathogen. In conclusion, we identified a novel bactericidal anti-mycobacterial lead compound targeting MmpL3 providing an attractive starting point for optimization.