Discovery of novel proteasome inhibitors using a high-content cell-based screening system

Abstract

The regulated degradation of damaged or misfolded proteins, as well as down-regulation of key signaling proteins, within eukaryotic and bacterial cells is catalyzed primarily by large, ATP-dependent multimeric proteolytic complexes, termed proteasomes. Inhibition of proteasomal activity affects a wide variety of physiological and pathological processes, and was found to be particularly effective for cancer therapy. We report here on the development of a novel high throughput assay for proteasome inhibition using a unique, highly sensitive live-cell screening, based on the cytoplasm-to-nucleus translocation of a fluorescent proteasome inhibition reporter (PIR) protein, consisting of nuclear localization signal-deficient p53 derivative. We further show here that mdm2, a key negative regulator of p53 plays a key role in the accumulation of PIR in the nucleus upon proteasome inhibition. Using this assay, we have screened the NCI Diversity Set library, containing 1,992 low molecular weight synthetic compounds, and identified four proteasome inhibitors. The special features of the current screen, compared to those of other approaches are discussed.

Figure 1. Engineering and validation of the PIR reporter system.

(A) Schematic representation of the PIR protein. PIR consists of yellow fluorescent protein (YFP) fused to the C-terminus of the human p53 mutant R175H, carrying a triple mutation in the bipartite NLS in which three consecutive lysine residues were replaced with alanines K319A, K320A, and K321A. (B) Nuclear accumulation of the PIR protein upon treatment with proteasome inhibitors. PIR cells were exposed to MG132, Bortezomib, and ALLN at the indicated concentrations for 6 h. PIR-associated YFP-fluorescence (upper panel) and immunostaining with anti-β-catenin antibody (lower panel) are seen. (C) PIR cells were incubated for 6 h without proteasome inhibitor (control), or with 10 µM MG132. Fractionated cytoplasm and nuclear lysates were analyzed by Western blot with anti-p53 antibody. A band corresponding to PIR was detected in the nuclear fraction only following treatment with MG132.

Figure 2. MDM2 promotes PIR nuclear translocation.

(A) Overexpression of MDM2 results on the PIR nuclear localization without additional stimuli. PIR cells were transfected with wild-type MDM2, MDM2 mutant deficient on p53 binding (Δ 9-58), or MDM2 mutant with abolished E3 ligase site (Ser 440). Cells expressing both p53 and MDM2 were visualized by immunofuorescence staining with the anti-MDM2 monoclonal antibodies. PIR has a nuclear localization in the cells expressing wt MDM2 and MDM2 (Ser 440), and remains cytoplasmic in the cells transfected with MDM2 (Δ 9-58). (B) Mdm2 siRNA prevents bortezomide–induced translocation of PIR to the nucleus. PIR cells were transiently transfected with 200 pmol control-siRNA or Mdm2-siRNA. Forty-eight hours after transfection, bortezomide (0.1 µM) was added for an additional 6 h, and immunofuorescence staining for MDM2 was performed as described in Materials and Methods.

Figure 3. A flow chart of the screening procedure.

For the screening assay, H1299-PIR reporter cells were plated in 384-well plates for 24 h and treated with compounds of the NCI Diversity Set library at two concentrations (1 and 10 µM), one compound per well. Following 12 hours of incubation, cells were fixed in 3% paraformaldehyde and screened for PIR cellular localization with WiScan™ automated microscope system. Acquired images were analyzed for PIR nuclear translocation, and selected hits were confirmed by microscopy-based and biochemical methods, and tested for compound cytotoxicity.

Figure 4. Positive hit compounds inhibit 26S proteasome.

(A) PIR cells were treated with hit compounds and MG132 for 6 hr at the following concentrations: NSC3907 – 20 µM; NSC99671 – 50 µM; NSC310551 – 0.3 µM; NSC321206 – 0.15 µM; and MG132 - 5 µM). Whole cell lysates were immunoblotted for ubiquitin (upper panel) and β-catenin (middle panel). Tubulin (lower panel) was used as an internal loading control. (B) Quantitation of the amounts of ubiquitinated proteins (blue) and β-catenin (red) in the absence of and upon treatment with the different inhibitors. (C) Positive hit compounds inhibit purified proteasome in vitro. Purified 26S proteasomes from rabbit muscle were incubated for the indicated times in the presence of 30 µM of our positive hits (100 µM for NSC3907). MG-132 at a 5 µM concentration served as a positive control. NSC3907 (blue), NSC99671 (green), NSC310551 (magenta), NSC321206 (orange), without treatment (red) and MG-132 (black).

Figure 5. Effect of the hit compounds on the viability of PIR cells.

PIR cells were treated with the active compounds for 48 hr at 11 concentrations ranging from 0.1 to 100 µM and the cells' viability was assessed by AlamarBlue assay. Results are expressed as the viability ratio of treated to untreated cells and represent the mean ± SD of 6 repeats.