br Screening the chemical library of NPDepo br To
3.2. Screening the chemical library of NPDepo
To discover inhibitors of cancer-specific metabolism, we have used the phenotypic screening system described above to assay 559 com-pounds in the chemical library of NPDepo, and obtained two candidate chemotypes; neoantimycin class of compounds (unantimycin A and SW-163A) and a thiourea containing compound (NPL40330). Unantimycin A and SW-163A remarkably decrease OCR and increase ECAR (Fig. 4A and B, Fig. S4), indicating their inhibitory eﬀect on OXPHOS. We have previously tested the cytotoxic eﬀects as well as the antimicrobial and antiprotozoal activities of unantimycin A and SW-163A. As expected, the two inhibitors of respiration showed moderate cytotoxic activity (half-maximal inhibitory concentration (IC50) values against various cancer cell lines: 5.4–39 μM) and relatively strong growth inhibitory activities against pathogenic fungi such as Pyricularia oryzae and Can-dida albicans (IC50 values of SW-163A against the fungi: 0.034–0.12 μM) r> Fig. 4. Unantimycin A, SW-163A, and NPL40330 act as inhibitors of mitochondrial respiration. (A) The chemical structures of unantimycin A, SW-163A, and NPL40330 A. (B) Percentage changes in OCR and ECAR (relative to their respective baseline values) induced by unantimycin A, SW-163A, and NPL40330 were calculated from Fig. S4. Each data point represents the mean ± standard deviation value at 6 min after treatment. Data are mean ± s.d. (n = 3 technical replicates) from one representative experiment out of two independent experiments. (C) Proteomic changes induced by unantimycin A (48 μM, 18 h) and NPL40330 (5 μM, 18 h).
Table 1 Synergistic eﬀects of antimycin/neoantimycins with 2-DG.
Alone With 2-DG
*Values are IC50s against the growth of HeLa cells.
*2-DG alone did not show any significant growth inhibition (< 20%) at test concentrations.
. The results also indicated that, related to the estimated inhibition of OXPHOS, the HeLa 121207-31-6 became more sensitive to these drugs when they were co-treated with 2-DG (Table 1). Furthermore, the ChemPro-teoBase profiling of unantimycin A revealed that it upregulates glyco-lytic enzymes, which suggests that it is a mitochondrial inhibitor (Fig. 4C).
As a representative example of thiourea containing compounds, NPL40330 (Fig. 4A) was noted to induce phenotype similar to those induced by inhibitors of mitochondrial respiration, which are decreased OCR, increased ECAR (Fig. 4B and Fig. S4), and upregulation of gly-colytic proteins (Fig. 4C).
3.3. Determination of molecular targets of hit compounds
We set up an in vitro reconstitution assay for mitochondrial re-spiratory activity in semi-intact cells to elucidate the mode of action of hit compounds (Fig. S5). Digitonin was used as a membrane permea-bilizer.
First, a dose-response study was conducted to determine the op-timum concentration of digitonin. In the presence of succinate, rote-none, and ADP, a concentration-dependent increase in state 3 OCR (an OCR value in the presence of ADP and substrates) was observed, which tended to plateau after the addition of 20 μg/ml digitonin (Fig. 5A, B). In addition, when succinate/rotenone was replaced with other en-dogenous/synthetic substrates and cofactors such as malate/pyruvate, duroquinol, and TMPD/ascorbate, the respiratory activities driven by complexes I, III, and IV, could also be measured (Fig. 5C).
Next, we examined the eﬀects of typical inhibitors of respiration on the activity of each complex in semi-intact cells. In this experiment, test compounds were injected simultaneously with digitonin and specific respiratory substrates for each complex, after which changes in OCR were recorded (Fig. S6). Rotenone is a complex I inhibitor. The results show that it obstructed only the respiratory activity driven by malate/ pyruvate. Given that complex II is generally not involved in respiratory chain reaction when complex I is workable, the complex II inhibitor malonate did not prevent the respiratory activity driven by malate/
Fig. 5. In vitro reconstitution assay for mitochondrial complex activities in digitonin-permeabilized HeLa cells. (A) Dose-response curve for digitonin. The cells were treated with ADP and succinate plus rotenone to stimulate state 3 respiration. (B) State 3 (an OCR value in the presence of ADP and substrates) and state 4O (an OCR value in the presence of ADP, substrates, and oligomycin) OCR were quantified from Fig. 5A using the following equations: state 3 OCR = (measurement #4) – (measurement #8) and state 4O OCR = (measurement #6) – (measurement #8). (C) Comprehensive assessment of mitochondrial complex activities. After three baseline measurements were taken, digitonin was co-injected with ADP and the specific substrates indicated in the panel. Data are mean ± s.d. (n = 3 technical replicates) from one representative experiment out of two independent experiments.