Regulation of energy homeostasis by the ubiquitin-independent REGγ proteasome

Introduction

Maintenance of energy homeostasis is essential for survival and proper function of all cells. Intracellular energy homeostasis is closely related to protein degradation and synthesis. Cells mainly use the ubiquitin (Ub)-dependent proteasome system (UPS) and autophagy-lysosome system for protein degradation and the ribosomes for protein synthesis1. Interestingly, autophagy serves as an energy-saving process2, whereas both the protein synthesis and the Ub-dependent protein degradation are high energy-consuming processes3,4. Therefore, the exquisite balance between these protein degradation and synthesis systems is required to maintain proper protein and energy homeostasis. Indeed, ribosomal subunits can be targeted for degradation by both UPS5 and autophagy6. Notably, growing numbers of proteasomal substrates have been identified to be degraded by Ub-independent proteasome pathway (UIPP), and importantly, the UIPP provides cells a shortcut to degrade proteins without ATP consumption, suggesting that it serves as an energy-saving protein degradation pathway7. However, the functions of UIPP have not got enough attention7. The proteasome is a large protein complex consisting of a 20S proteolytic core and three different proteasomal activators including 19S (or PA700), 11S (or PA28, REG) and PA200. Differently, the 19S activator binds to the 20S core and mediates protein turnover in an Ub- and ATP-dependent manner, whereas the 11S proteasome mainly promotes Ub-independent protein degradation. Previous studies revealed that REGγ (or PA28γ), one of the 11S proteasomal activators8,9, promotes Ub- and ATP-independent proteasomal degradation of steroid receptor coactivator-3 and the cell cycle inhibitor p21 (refs 10,11). Our previous study demonstrated that REGγ deficiency induces autophagy-dependent lipid degradation, indicating a role for UIPP in lipid metabolism12. Interestingly, starvation can increase proteasome activity with no upregulation of UPS13, suggesting that cell may activate UIPP to achieve energy-saving protein turnover under low energy status. However, the effectiveness of UIPP in energy homeostasis and cell fate decision under starvation remains unknown.

Limiting energy consumption in disadvantageous circumstances is critical for cell survival. Transcription of ribosomal RNA (rRNA), the first step in ribosome synthesis, is a highly energy-consuming process14,15. The TBP-TAFI complex SL1, transcription activator UBF and the RNA polymerase I (Pol I) enzyme with associated factors such as TIF1A and TIF-IC form the minimal complex required for rDNA transcription16,17,18,19.The synthesis of rRNA is tuned to match environmental nutrition conditions. Nutrients and growth factors positively regulate rRNA synthesis to adapt to cell proliferation through ERK- and mTOR-dependent TIF-IA phosphorylation15, whereas glucose starvation downregulates rRNA synthesis to limit energy consumption by activating AMPK-dependent phosphorylation of TIF1A20. Of note, during the past decade, the silent information regulator (Sir2)-like family deacetylases (also known as sirtuins) have emerged as important regulators in cell stress resistance and energy metabolism21,22,23,24. In mammals, seven sirtuins (SirT1-SirT7) have been identified. Interestingly, SirT1 forms an energy-dependent nucleolar silencing complex (eNoSC) with NML and SUV39H1 and acts as an energy-dependent repressor of rDNA transcription4, whereas SirT7, the only sirtuin enriched in nucleoli, associates with Pol I and UBF and positively regulates rDNA transcription25,26,27. Clearly, multiple signalling pathways are involved in dynamic regulation of rDNA transcription, but how these different, sometimes even antagonistic, pathways are coordinated to fine-tune rRNA synthesis to maintain energy homeostasis and cell survival under stress conditions remains to be clarified.

In this study, we reveal that REGγ-deficient cells exhibit high energy consumption and are sensitive to energy stress through increasing SirT7-directed rDNA transcription. Moreover, AMPK also plays a key role in the REGγ-SirT7 pathway in turning off rDNA transcription under energy stress conditions. Furthermore, REGγ reduction sensitizes tumours to 2DG (a competitive glycolysis inhibitor) treatment in vivo. Our findings disclose a role of the UIPP in maintaining cellular energy homeostasis, suggesting that REGγ is a potential therapeutic target for tumour-starving treatment.

Results

REGγ deficiency promotes energy consumption

Although the UIPP provides cells an energy-saving protein turnover shortcut, the contribution of this process in energy balance is unknown. Previous studies reported that the REGγ knockout (KO) mice displayed reduced body weight and retarded growth28,29. Our recent study showed that REGγ-KO mice exhibited over-consumption of food12. These observations prompted us to test the role of REGγ-proteasome in energy metabolism. Interestingly, we observed that REGγ knockout (KO) led to a significant downregulation of intracellular ATP level accompanied by an upregulation of ADP-to-ATP ratio in MEF cells, and REGγ recomplementation reversed these changes (Fig. 1a,b). Decreased level of ATP was also observed in REGγ stable knockdown human cancer cell lines (Fig. 1c). To further assess the role of REGγ in energy homeostasis, we treated cells with glucose deprivation (GD). Results showed that REGγ-KO led to a rapid and severe decrease of intracellular ATP level under GD, REGγ reconstitution significantly retarded the reduction rate of intracellular ATP level in GD-treated REGγ-KO cells, and cellular ATP level was significantly restored through glucose resupplementation in GD-treated REGγ-KO cells (Fig. 1d). Similar results were also obtained in cancer cell lines with stable REGγ knockdown (Fig. 1e). The above data indicate that REGγ plays an essential role in maintaining intracellular energy homeostasis.

Figure 1: REGγ deficiency promotes energy consumption and starvation-induced cell death.

(a–c) REGγ regulates cellular energy homeostasis under normal growth conditions. (A,B) MEF cells from wild-type (+/+, WT) and REGγ knockout (−/−, KO) mice were cultured in DMEM-high glucose medium. The relative intracellular ATP levels (a) and the cellular ADP/ATP ratios (b) were detected. To determine the specific effect of REGγ, REGγ-KO MEF cells were infected with lentiviral vectors expressing REGγ. Western blots show REGγ expression. (c) The relative intracellular ATP levels in HeLa, HCT116 and HCT116−/− (p53 null) cancer cells with stable knockdown of REGγ (ShR1 or ShR2) or a vector control (ShN) cultured in DMEM-high glucose medium. Western blots show the knockdown efficiency. (d,e) REGγ deficiency promotes energy consumption. Indicated cell lines were cultured in glucose-free DMEM (glucose deprivation, GD) for the indicated time periods, or re-supplemented with glucose (+gluc.) for 4 or 6 h after GD. The relative intracellular ATP levels were analysed. (f-l) REGγ deficiency promotes energy-dependent cell death. (f) REGγ-WT and -KO MEF cells were treated with GD for 16 h and apoptosis was analysed by FACS. Quantitative data show percentage of apoptosis. (g) REGγ-WT and -KO MEF cells were treated with GD for 16 h and analysed for activated caspase-3 and poly (ADP-ribose) polymerase cleavage by western blotting. (h) REGγ-WT and -KO MEF cells were treated with GD for the indicated times and the large-scale DNA fragmentation was determined by agarose gel electrophoresis. (i,j) Indicated cell lines were treated with GD, or re-supplemented with glucose (+ gluc.) for the indicated time periods after GD. Cell viability was determined using MTT assay. (k,l) Indicated cells were treated with GD in the presence or absence of methyl pyruvate (MP, 10 μM) for the indicated time periods, and cell viability was analysed using MTT assay. All experiments were repeated three times, data represent mean±s.d., *P<0.05, **P<0.01; Student’s t-test is used throughout. See also Supplementary Fig. 1.