Proteasome inhibitors suppress expression of NPM and ARF proteins

Abstract

Proteasome inhibitors stabilize numerous proteins by inhibiting their degradation. Previously we have demonstrated that proteasome inhibitors thiostrepton, MG132 and bortezomib paradoxically inhibit transcriptional activity and mRNA/protein expression of FOXM1. Here we demonstrate that, in addition to FOXM1, the same proteasome inhibitors also decrease mRNA and protein expression of NPM and ARF genes. These data suggest that proteasome inhibitors may suppress gene expression by stabilizing their transcriptional inhibitors.

The 26S proteasome is a multisubunit protease that functions as the major regulator of all short-lived proteins in cells.1 It comprises two 19S regulatory complexes and a 20S core complex. The 20S core constitutes the major proteolytic activity of the proteasome and comprises α- and β-subunits, which are organized into four rings of seven subunits each. The β-subunits (β1, β2 and β5) possess distinct protease activities based upon the cleavage site in the protein substrate.2 Ubiquitin (Ub), a major element of the ubiquitin-proteasome system, is covalently attached to cellular proteins and responsible for their degradation by the proteasome.3,4 Inhibition of proteasome activity affects a wide variety of physiological and pathological processes and has been found to be particularly effective for cancer therapy. The use of proteasome inhibitors (PIs) as drug candidates has emerged from the observation that at specific concentrations, PIs can induce apoptosis in certain leukemia- and lymphoma-derived cells without affecting the normal cells.5 Bortezomib (velcade) is the first FDA-approved proteasome inhibitor that predominantly inhibits the chymotrypsin-like and caspase-like proteasome activity.6,7 It is currently in clinical use and shows efficacy in the treatment of multiple myeloma and mantle cell lymphoma as a single agent,6,7 and it shows synergy in combination with other drugs in vitro.8–10 Since proteasome targets ubiquitin-tagged proteins for degradation and is responsible for proteolytic degradation of the majority of cellular proteins, stabilization of protein expression is a hallmark property of PIs. Contrary to this effect, we found, before, that proteasome inhibitors inhibit the transcriptional activity and expression of an oncogenic transcription factor FOXM1.11,12 In this report, we demonstrate that treatment of cancer cells with proteasome inhibitors, thiostrepton, MG132 and bortezomib led to inhibition of protein expression of ARF and NPM proteins.

HeLa cells were treated with PIs thiostrepton, MG132 and bortezomib for 24 h, and the levels of ARF, NPM and FOXM1 (positive control) were measured by qRT-PCR and immunoblotting. First, we looked at the effect of PIs treatment on mRNA levels of these genes. We found that the downregulation of ARF, NPM and FOXM1 mRNA occurs following treatment with the PIs in HeLa cells (Fig. 1A). Following treatment with thiostrepton, MG132 and bortezomib, the levels of ARF mRNA were suppressed by 65–80%, and NPM mRNA levels were suppressed by 32–46% (Fig. 1A). The levels of FOXM mRNA (used as a positive control) were inhibited by 73–80% by the same PIs (Fig. 1A). The suppression of protein expression occurred very similarly (somewhat weaker) to downregulation of mRNA levels (Fig. 1B and C). While treatment with 5 µM thiostrepton inhibited the expression of ARF and FOXM1 proteins by almost 80%, a 50% reduction in NPM protein level was observed. Five µM MG132 treated HeLa cells demonstrate an inhibition of 20–40% of ARF, NPM and FOXM1 proteins. Finally treatment with 1 µM bortezomib led to 40–60% decrease in ARF, NPM and FOXM1 proteins. Downregulation of protein levels following PIs treatment was quantified using ImageJ. β-actin was used as a loading control (Fig. 1C). On the other hand, proteins like Mcl-1 and p21 get stabilized following treatment with thiostrepton, MG132 and bortezomib11,12 and (data not shown).

In this report we demonstrate that PIs inhibit mRNA and protein expression of two proteins, NPM and p14^ARF. Interestingly, it has been suggested that p14-ARF protein turnover depends, in part, on proteasome function, and following treatment with proteasome inhibitor MG132 a slight increase in ARF protein level occurs.13 Contrary to that, we demonstrated that following treatment with different PIs, mRNA and protein level of ARF are inhibited in HeLa cells. NPM and ARF interact with each other,14 and it is not clear if their interaction linked to suppression of both by PIs. Since PIs stabilize majority of proteins, inhibition of FOXM1, ARF and NPM, protein expression following treatment with PIs seems paradoxical. FOXM1, but not ARF and NPM is a transcription factor. We have postulated previously that following treatment with PIs, a negative regulator of FOXM1 (NRFM) gets stabilized and interacts with FOXM1, leading to inhibition of its transcriptional activity.15 Because of FOXM1 auto-regulation loop the expression of FOXM1 mRNA and protein also get suppressed.16 However, the mechanisms behind inhibition of ARF and NPM by PIs are not well-understood. It is plausible to propose that PIs stabilize negative regulators of transcription of these genes (Fig. 1A), which leads to inhibition of ARF and NPM mRNA expression and overrides the stabilization of these proteins by PIs, leading to suppression of NPM and ARF protein expression. Therefore, if these proteins would be expressed from exogenous promoters, the prediction will be that they will be stabilized by PIs, because they are still targets of ubiquitin-proteasome system. Further experiments are needed to explain the mechanism behind inhibition of ARF and NPM by PIs and to identify potential transcriptional inhibitors of these genes that are stabilized by PIs.

Figures and Tables

Figure 1 Proteasome inhibitors suppress mRNA and protein expression of ARF and NPM. (A) HeLa cells were treated either with DMSO or 5 µM thiostrepton, 5 µM MG132 and 0.5 µM bortezomib for 24 h. Levels of mRNA NPM, p14 ARF and FOXM1 (positive control) were measured using real-time quantitative RT-PCR and normalized with cyclophilin mRNA. The sense/antisense primer sequences were as described, human p14 ARF: 5′-ATG GTG CGC AGG TTC TTG G-3′ and 5′-TGC GGG CAT GGT TAC TGC CTC-3′; FOXM1: 5′-GGA GGA AAT GCC ACA TTA GCG-3′ and 5′-AGG ACT TCT TGG GTC TTG GGG TG-3′; cyclophilin: 5′-GCA GAC AAG GTC CCA AAG ACA G-3′ and 5′-CAC CCT GAC ACA TAA AA C CCT GGG; NPM: 5′-CCA GTG GTC TTA AGG TTG AAG TGT GG-3′ and 5′-TCC AGA TAT ACT TAA GAG TTT CAC ATC-3′. (B) HeLa cells were treated with proteasome inhibitors; 5 µM thiostrepton, 5 µM MG132 and 0.5 µM bortezomib for 24 h as indicated, harvested, and the cell lysates were analyzed for the levels of NPM, p14 ARF and FOXM1 (positive control) with antibodies from Santa Cruz Biotechnology by immunoblotting. β-actin (Sigma) was used as the loading control. (C) Densitometry was performed on scanned immunoblot images using the ImageJ gel analysis tool. The gel analysis tool was used to obtain the absolute intensity (AI) for each experimental band of FoxM1, NPM and ARF. Relative intensity (RI) for each experimental band was calculated by normalizing the experimental AI to the corresponding loading control (β-actin) AI. Columns, mean of three separate experiments; bars, SD.

Acknowledgments

NIH grants 1RO1CA1294414 and 1R21CA134615 (A.L. Gartel).