Skp2 dictates cell cycle-dependent metabolic oscillation between glycolysis and TCA cycle

Liu, Jing; Peng, Yunhua; Shi, Le; Wan, Lixin; Inuzuka, Hiroyuki; Long, Jiangang; Guo, Jianping; Zhang, Jinfang; Yuan, Min; Zhang, Shuangxi; Wang, Xun; Gao, Jing; Dai, Xiangpeng; Furumoto, Shozo; Jia, Lijun; Pandolfi, Pier Paolo; Asara, John M.; Kaelin, William G.; Liu, Jiankang; Wei, Wenyi

doi:10.1038/s41422-020-0372-z

Article
Published: 15 July 2020

Skp2 dictates cell cycle-dependent metabolic oscillation between glycolysis and TCA cycle

Cell Research volume 31, pages 80–93 (2021)Cite this article

7237 Accesses
49 Citations
2 Altmetric
Metrics details

Subjects

A Correction to this article was published on 28 August 2020

This article has been updated

Abstract

Whether glucose is predominantly metabolized via oxidative phosphorylation or glycolysis differs between quiescent versus proliferating cells, including tumor cells. However, how glucose metabolism is coordinated with cell cycle in mammalian cells remains elusive. Here, we report that mammalian cells predominantly utilize the tricarboxylic acid (TCA) cycle in G1 phase, but prefer glycolysis in S phase. Mechanistically, coupling cell cycle with metabolism is largely achieved by timely destruction of IDH1/2, key TCA cycle enzymes, in a Skp2-dependent manner. As such, depleting SKP2 abolishes cell cycle-dependent fluctuation of IDH1 protein abundance, leading to reduced glycolysis in S phase. Furthermore, elevated Skp2 abundance in prostate cancer cells destabilizes IDH1 to favor glycolysis and subsequent tumorigenesis. Therefore, our study reveals a mechanistic link between two cancer hallmarks, aberrant cell cycle and addiction to glycolysis, and provides the underlying mechanism for the coupling of metabolic fluctuation with periodic cell cycle in mammalian cells.

You have full access to this article via your institution.

Download PDF

Glutamine synthetase licenses APC/C-mediated mitotic progression to drive cell growth

Article 10 February 2022

Jiang-Sha Zhao, Shuo Shi, … Yu-Xiong Feng

PER1 suppresses glycolysis and cell proliferation in oral squamous cell carcinoma via the PER1/RACK1/PI3K signaling complex

Article Open access 15 March 2021

Xiaobao Gong, Hong Tang & Kai Yang

NIK promotes metabolic adaptation of glioblastoma cells to bioenergetic stress

Article Open access 15 March 2021

Michael L. Kamradt, Ji-Ung Jung, … Raquel Sitcheran

Introduction

Normally differentiated cells largely rely on tricarboxylic acid (TCA) cycle for converting glucose to energy, whereas proliferating cancer cells are addicted to glycolysis even when oxygen is present, a phenotype termed aerobic glycolysis or “Warburg Effect”, which has been shown to benefit cancer cell growth and tumorigenesis.^1,2 Cancer cells are constantly dividing and require more metabolic intermediates, such us acetyl-CoA, to build macromolecules for their daughter cells.^3,4,5,6,7,8 Extensive investigations have revealed that many oncogenic signaling pathways, such as PKM2, HIF, AKT, Ras and Myc are critical regulatory components in the “Warburg Effect”,^4,5,6 but the underlying mechanism(s) for the link between cell cycle and cancer cell metabolism is largely elusive. A cell cycle-coupled metabolic cycle has been reported in yeasts, where genes with metabolic functions are expressed with robust periodicity, and cells tend not to utilize oxygen for metabolism when dividing event happens,⁹ so as to avoid damage to newly duplicated DNA.¹⁰

Ubiquitination and subsequent degradation of key cyclins dictate the normal transition of cell cycle, where APC/C (anaphase-promoting complex/cyclosome) and SCF (Skp1–Cul1–F-box protein) E3 complex ligases play a key role in this process. Several previous studies indicate a crosstalk between cell cycle regulators and glycolysis,^11,12,13 and several glycolytic enzymes have been reported to be regulated by cell cycle components, such as PFKFB3 (6-phosphofructo-2-kinase/fructose-2, 6-bisphosphatase-3) by SCF^GRR1, SCF^β-TRCP and APC^Cdh1,^12,14,15,16 HK2 (hexokinase 2) by cyclin D1,¹⁷ PFKP (phosphofructokinase, platelet type) and PKM2 (pyruvate kinase M2) by cyclin D3/CDK6.⁷ However, in mammalian cells, there is a lack of a systemic study about the link between cell cycle and metabolic dependence, especially TCA cycle enzymes. We previously found that SCF^β-TRCP E3 ligase degrades lipin1 to promote hepatic lipogenesis in liver,¹⁸ but little is known about whether SCF E3 ligase regulates glucose metabolism in a cell cycle-dependent manner.

In the present study, we attempted to address the mechanism of how the cell cycle machinery remodels cell metabolism, and to explore its role in tumorigenesis. Using Seahorse extracellular flux analyzer and ¹³C-labeled metabolic flux assays, we measured the metabolic dependence in multiple cell lines after cell cycle synchronization with different methods, including nocodazole block, double thymidine block, as well as serum starvation. We found that mammalian cells predominantly utilize the TCA cycle in G1 phase, but prefer glycolysis in S phase. Importantly, coupling cell cycle with metabolism is largely achieved by timely destruction of IDH1/2, key TCA cycle enzymes, in a Skp2- and cyclin E/CDK2-dependent manner. Furthermore, elevated Skp2 abundance in prostate cancer cells destabilizes IDH1 to favor glycolysis and subsequent tumorigenesis. Therefore, our study reveals a mechanistic link between two cancer hallmarks, aberrant cell cycle and addiction to glycolysis.

Results

Mammalian cells in distinct cell cycle phases prefer different glucose metabolism status

To understand the link between cell cycle and metabolism, we firstly synchronized HeLa cells using nocodazole and released cells into different cell cycle phases and measured the rates of glycolysis (indicated by extracellular acidification rate (ECAR)) and TCA cycle (indicated by oxygen consumption rate (OCR)) (Fig. 1a; Supplementary information, Fig. S1a). We observed that glycolysis peaked in early S phase (Fig. 1b), accompanied by a relatively lower rate of TCA cycle (Fig. 1c). In keeping with this, a relatively higher rate of glycolysis in S phase and a relatively higher rate of TCA cycle in G1 phase were also detected in synchronized HeLa cells with double thymidine block and in synchronized mouse embryonic fibroblasts (MEFs) with serum starvation (Supplementary information, Fig. S1b, c). These results suggest that glucose metabolism is likely regulated in a cell cycle-dependent manner in mammalian cells. This may be in part due to ATP needs, as cells rely mostly on the TCA cycle during G1 phase while switching to glycolysis, a less economic form of ATP production, to accumulate intermediate metabolites that are used as building blocks to synthesize biomacromolecules for subsequent DNA replication and cytokinesis.¹⁹

**Fig. 1: Mammalian cells adopt different glucose metabolism pathways in different cell cycle stages, primarily utilizing TCA cycle in G1 phase, but relying on glycolysis in S phase.**

To further explore this cell cycle-dependent metabolic shift between glycolysis and TCA cycle, cells were synchronized using nocodazole and released into either G1 or S phase (Supplementary information, Fig. S1d), followed by labeling with ¹³C6-glucose or ¹³C5-glutamine to profile metabolic intermediates with liquid chromatography-mass spectrometry (LC-MS)²⁰ (Fig. 1d). Notably, cells in S phase exhibited a relatively higher glycolytic flux rate than those in G1 phase (Fig. 1e; Supplementary information, Fig. S1e), which could not be explained by differences of glucose uptake in these two cell cycle phases (Supplementary information, Fig. S1f). In contrast to the relatively fast glycolysis flux in minutes, it took ~2 h for the TCA cycle flux to reach a steady state (Supplementary information, Fig. S1g). In keeping with the metabolic switch from TCA cycle to glycolysis in S phase, we observed a reduction of TCA cycle flux for cells in S phase compared to those in G1 phase (Fig. 1f), which appeared to be independent of glutamine uptake changes (Supplementary information, Fig. S1h). Furthermore, metabolic flux through the pentose phosphate pathway (PPP) revealed by ¹³C6-glucose labeling was also relatively higher for cells in S phase than in G1 phase, consistent with elevated synthesis of fatty acids, aromatic amino acids, and nucleic acids that is coupled with DNA duplication events in S phase (Supplementary information, Fig. S1i–k). Together, these results support a model of cell cycle-dependent metabolic switch from TCA cycle to glycolysis in S phase to facilitate DNA duplication and cell growth.

The protein levels of IDH1/2 fluctuate during cell cycle

Glycolytic and TCA cycle fluxes are both governed by a cohort of essential enzymes²¹ (Supplementary information, Fig. S2a). To investigate whether the cell cycle-dependent metabolic switch from TCA cycle to glycolysis in S phase is due to the fluctuation of those enzymes, we further synchronized HeLa cells with either nocodazole or double thymidine block and monitored the protein levels of those enzymes (Supplementary information, Fig. S2b). Among those enzymes, we found that only the protein abundances of IDH1 and IDH2, but not other TCA cycle or glycolytic enzymes, fluctuated during the cell cycle in HeLa cells synchronized by either nocodazole (Fig. 2a; Supplementary information, Fig. S2c) or double thymidine block (Supplementary information, Fig. S2d), featuring relatively lower abundance of both proteins in S phase. Moreover, the fluctuation of IDH1 and IDH2 proteins could be also detected in multiple cell lines that were synchronized by different methods, including MEFs and WPMY1 cells synchronized with serum starvation and HCT116 cells synchronized with nocodazole block (Supplementary information, Fig. S2e–h). Notably, IDH1/2 mRNA levels did not fluctuate (Supplementary information, Fig. S2i, j), indicating that IDH1/2 protein stability might be regulated during the cell cycle.

**Fig. 2: The fluctuation of IDH1/2 protein abundances during cell cycle.**

Three IDH isoenzymes exist in mammalian cells: the mitochondrial NADP⁺-dependent IDH2 and NAD⁺-dependent IDH3 catalyze the conversion of isocitrate to α-ketoglutarate, while the cytosolic NADP⁺-dependent IDH1 catalyzes the same reaction using cytosolic citrate.^22,23 Notably, both Idh1- and Idh2-knockout mice are viable and fertile, with noticeable mitochondrial dysfunction and increased oxidative stress, suggesting that IDH1 and IDH2 partially compensate for each other in vivo.^22,23 To understand the importance of fluctuations of IDH1 and IDH2 during the cell cycle, we measured the metabolic phenotype of IDH1- and IDH2-knockout haploid HAP1 cells generated by CRISPR/Cas9 (Fig. 2b). Compared to wild-type (WT) cells, the TCA cycle flux in IDH2^–/–, and to a lesser extent, IDH1^–/– cells were compromised (Fig. 2c, d; Supplementary information, Fig. S2k). Consistent with this finding, loss of either IDH1 or IDH2 led to growth arrest in galactose-rich media (“metabolic state-dependent lethality”),²⁴ indicating compromised mitochondrial respiration in both IDH1^–/– and IDH2^–/– cells (Fig. 2e). Moreover, compared to WT cells, CRISPR/Cas9-mediated depletion of endogenous IDH1 in HeLa cells also increased glycolysis (Fig. 2f, g), reduced TCA cycle metabolism (Fig. 2h), and increased lactate production (Supplementary information, Fig. S2l). These data suggest that cytosolic IDH1, together with mitochondrial IDH2, likely play essential roles in governing TCA cycle metabolism.

SCF^Skp2 is the E3 ubiquitin ligase for IDH1/2 degradation

We next set out to identify the mechanism(s) responsible for S phase-specific degradation of IDH1/2. We found that endogenous IDH1 and IDH2 protein abundances, but not the mRNA levels, were markedly elevated in cells treated with either the proteasome inhibitor MG132 or the Cullin neddylation inhibitor MLN4924 (Fig. 3a; Supplementary information, Fig. S3a), implicating a Cullin-based E3 ligase in the control of IDH1/2 degradation. Among various Cullins that we examined, IDH1 and IDH2 preferentially interacted with Cullin 1, and to a lesser extent, Cullin 3 (Fig. 3b; Supplementary information, Fig. S3b). Furthermore, depleting Cullin1, but not Cullin 3, Cullin 4A, nor Cullin 4B, led to IDH1 and IDH2 accumulation (Fig. 3c; Supplementary information, Fig. S3c). Further supporting that a Cullin 1-containing E3 ligase(s) regulates IDH1 and IDH2 stability, two other essential components of the canonical SCF ubiquitin ligase complex, Skp1 and Rbx1, also interacted with IDH1 and IDH2 (Supplementary information, Fig. S3d–g). Notably, Flag-tagged IDH1 coimmunoprecipitated with GST-tagged Fbw4 and Skp2 in cells, but not other F-box proteins we examined, under ectopic overexpression conditions (Fig. 3d). However, Skp2, but not Fbw4 promoted IDH1 poly-ubiquitination in cells (Fig. 3e; Supplementary information, Fig. S3h). On the other hand, depleting Skp2, but not Fbw4 dramatically elevated IDH1 in multiple cell lines, and extended half-life of IDH1 (Fig. 3f; Supplementary information, Fig. S3i–l). Moreover, endogenous Skp2 bound to IDH1 (Supplementary information, Fig. S3m). In keeping with this observation, we found that IDH1 and IDH2 were elevated in Skp2^–/– MEFs compared to their WT counterparts, further implicating Skp2 as a physiological negative regulator of IDH protein stability in cells (Fig. 3g).

**Fig. 3: SCF^S^kp2 promotes IDH1/2 ubiquitination and subsequent degradation, thus dictating cell cycle-dependent metabolic shift.**

To further determine whether genetic manipulation of Skp2 affects cell cycle-dependent metabolic shift, we depleted Skp2 in HeLa cells and measured the OCR and ECAR during the cell cycle progression of those cells. Notably, depleting Skp2 largely abolished the cell cycle-dependent fluctuation of IDH1/2 protein abundance (Fig. 3h), which was correlated with reduced glycolytic (Fig. 3i) and OCR oscillations (Fig. 3j) in S phase. To exclude the possibility that these metabolic changes were an indirect consequence of a change in cell cycle distribution due to Skp2 deficiency, we first synchronized cells in G1 or S phase and then performed metabolic studies (Supplementary information, Fig. S3n, o). We found that depleting Skp2 markedly abolished cell cycle-dependent flux changes in glycolytic and TCA cycle intermediates (Fig. 3k, l). Moreover, Skp2 depletion dramatically decreased extracellular lactate levels during S phase, providing further support for a pivotal role of Skp2 in governing the cell cycle-dependent switch to glycolytic metabolism when cells enter S phase (Supplementary information, Fig. S3p, q).

Cyclin E/CDK2 phosphorylates IDH1 at Thr157 residue

SCF^Skp2 typically binds and ubiquitinates its downstream substrates in a phosphorylation-dependent manner.²⁵ We therefore examined a panel of modifying kinase(s) potentially involved in Skp2-mediated degradation of IDH1/2 in cells. Notably, cyclin E/CDK2, and to a lesser extent, cyclin A/CDK2, promoted Skp2-mediated degradation of IDH1 and IDH2 in cells (Fig. 4a; Supplementary information, Fig. S4a). In support of a physiological role for cyclin E1 and cyclin A2 as negative regulators of IDH1/2, IDH1 and IDH2 proteins were accumulated in Ccne1^–/– and Ccna2^–/– MEFs, but not in Ccne2^–/–, Ccnd1^–/–, Ccnd2^–/– nor Ccnd3^–/– MEFs, accompanied by relatively higher oxidative phosphorylation rate in Ccne1^–/– MEFs (Fig. 4b; Supplementary information, Fig. S4b–d). Furthermore, CDK2 inhibition led to reduced IDH1 ubiquitination, resulting in extended half-life of the IDH1 protein (Supplementary information, Fig. S4e, f). In keeping with this observation, we found that Ccne1^–/– MEFs had higher OCR level (Fig. 4c). Notably, cyclin D3/CDK6 has been recently reported to inhibit glycolysis via directly phosphorylating PFKP and PKM2,⁷ whereas here we reveal a crucial role for cyclin E1/CDK2 and cyclin A2/CDK2 in suppressing TCA cycle largely by promoting the degradation of the TCA cycle enzymes, IDH1/2. Thus, these two mechanisms might represent complementary and synergistic molecular switches for tightly controlling the metabolic cycle in a cell cycle-dependent manner.

Furthermore, we revealed that cyclin E/CDK2 phosphorylated IDH1 in vitro (Fig. 4d). To further determine the phosphorylation site in IDH1, we aligned the amino acid sequences of IDH1 and IDH2 across species (Fig. 4e); there are three potential phosphorylation sites fitting the canonical CDK2 phosphorylation consensus motif.²⁶ By mutating each possible CDK2 phosphorylation residue to analine (T77A, S94A, and T157A), we found that cyclin E/CDK2 primarily phosphorylated the evolutionarily conserved T157 site of IDH1 (Fig. 4f). As CDK2 exerts its kinase activity through binding either cyclin E or cyclin A,^27,28 in the latter part of this study, we primarily focused on the molecular mechanism underlying cyclin E1/CDK2-mediated degradation of IDH1/2. Importantly, the phosphorylation on T157 of exogenous IDH1 could be detected using mass spectroscopy (Supplementary information, Fig. S4g). The T157 site is also conserved in mitochondrial IDH2 (T197), suggesting that the Skp2/cyclin E/CDK2 signaling axis also negatively regulates IDH2 through this site, presumably before newly synthesized IDH2 enters the mitochondria (Supplementary information, Fig. S7b).

In keeping with a critical role of T157 in Skp2-mediated degradation of IDH1, we found that mutating T157, but not the other two SP/TP motif residues T77 or S94, to alanine abolished cyclin E/CDK2-induced Skp2 interaction with recombinant IDH1 in vitro (Fig. 4g). Moreover, synthetic peptides with amino acid sequence derived from the putative phospho-degron region in IDH1 (T157) and IDH2 (T197) bound to recombinant Skp2, but not Fbw4 in vitro, only when T157 in IDH1 or T197 in IDH2 was phosphorylated (Supplementary information, Fig. S4h–j). As a result, IDH1-T157A mutant was neither ubiquitinated by Skp2 in cells (Fig. 4h) and in vitro (Fig. 4i, j), nor degraded by Skp2 in cells (Fig. 4k; Supplementary information, Fig. S4k–m).

Non-degradable IDH1-T157A mutant abolishes metabolic shift during cell cycle and compromises cell proliferation

To explore the function of cyclin E/CDK2-dependent phosphorylation of IDH1 in metabolic oscillation during cell cycle, we constructed stable cell lines that ectopically expressed either IDH1-WT or IDH1-T157A mutant in HeLa cells. After synchronization by nocodazole block and release, unlike IDH1-WT, the protein abundance of IDH1-T157A mutant did not fluctuate in HeLa cells during the cell cycle (Fig. 5a). In keeping with the protein abundance of IDH1, the metabolic shift from TCA to glycolysis during S phase was also compromised in IDH1-T157A mutant cells (Fig. 5b, c). Compared with the cells expressing IDH1-WT, the cells expressing IDH1-T157A mutant had similar levels of glycolytic intermediates in G1 and S phases (Fig. 5d). These results indicated that the non-degradable IDH1 mutant abolished the fluctuation of IDH1 levels, leading to a lack of metabolic shift during cell cycle progression. In addition, non-degradable IDH1-T157A mutation resulted in a modest increase in G1 cells (Supplementary information, Fig. S5a).

**Fig. 5: Non-degradable IDH1 is resistant to cell cycle-dependent fluctuation in abundance, and restricts cell proliferation.**

To further understand how cell cycle-dependent metabolic shift regulates tumorigenesis, we generated several stable cell lines, including HeLa, A375, U2OS, DU145 and PC3 cells, that either express IDH-WT or IDH1-T157A mutant (Fig. 5e; Supplementary information, Fig. S5b, f, j, k). As a result of inefficiency of metabolic shift during cell cycle progression due to the abolishment of fluctuation of the IDH1-T157A mutant protein abundance, those cells expressing non-degradable IDH1-T157A mutant had compromised proliferation (Fig. 5f; Supplementary information, Fig. S5c, g). In keeping with that, those cells expressing the IDH1-T157A mutant also had compromised anchorage-independent growth indicated by crystal violet and soft agar colony formation assays (Fig. 5g–i; Supplementary information, Fig. S5d, e, h–n). Moreover, the tumors derived from PC3 cells expressing the non-degradable IDH1-T157A mutant were significantly smaller than those expressing IDH1-WT or GFP in the xenograft nude mouse model with subcutaneous injection of cancer cells (Fig. 5j, k). These data indicated that the non-degradable IDH1-T157A mutant abolished metabolic shift during cell cycle, leading to compromised tumorigenesis both in vitro and in vivo, possibly due to impaired delivery of glycolytic intermediates needed for the robust assembly of biomass during S phase to support rapid cell growth and xenograft tumorigenesis.

Reverse correlation between Skp2 and IDH1 in prostate cancer

Skp2 plays an important role in prostate tumorigenesis.²⁹ To further determine whether Skp2 regulates cancer cell metabolism in prostate cancer setting, we adopted a panel of prostate cancer (PrCa) cell lines, including C4-2, DU145, LNCaP, PC3, 22Rv1 and VCaP, for subsequent metabolic assays (Fig. 6a). Notably, we observed an inverse correlation between Skp2 and IDH1 protein abundances, but not IDH1 mRNA levels in the panel of PrCa cell lines (Fig. 6a; Supplementary information, Fig. S6a, b). Compared to four PrCa cell lines characterized by a Skp2^low and IDH1^high expression pattern (C4-2, LNCaP, VCaP, and 22Rv1), two PrCa cells characterized by a distinct Skp2^high and IDH1^low expression pattern (DU145 and PC3) displayed elevated rate of glycolysis, and reduced rate of oxidative phosphorylation (Fig. 6b, c; Supplementary information, Fig. S6c). Notably, the fluctuation of IDH1/2 protein abundance during cell cycle can also be detected in Skp2^low cell line 22Rv1, but not in the two Skp2^high cell lines, likely due to stabilization of Skp2 in Skp2^high cell lines that leads to less dramatic fluctuation of Skp2 itself (Supplementary information, Fig. S6d, f, g). In keeping with this notion, 22Rv1 cells also displayed cell cycle-dependent metabolic shift (Supplementary information, Fig. S6e).

**Fig. 6: Skp2 dedicates the metabolic phenotypes of prostate cancer cell lines in part by promoting IDH1 degradation.**

To further determine the causal role of Skp2 in IDH1 protein abundance and cell metabolism, we further depleted Skp2 in the two Skp2^high cells and ectopically expressed Skp2 in three Skp2^low cells (Fig. 6d, g). Importantly, depletion of endogenous Skp2 in the two Skp2^high cells increased p27 and IDH1/2 protein levels (Fig. 6d), resulting in reduced glycolysis (Fig. 6e) and increased oxidative phosphorylation (Fig. 6f). By contrast, ectopic expression of Skp2 in Skp2^low cells, such as LNCaP, C4-2 and 22Rv1, reduced p27 and IDH1/2 protein levels (Fig. 6g), leading to increased glycolysis (Fig. 6h) and reduced oxidative phosphorylation (Fig. 6i). These results provide further support for a critical role of Skp2 in destabilizing IDH1/2, thereby coupling metabolism to cell cycle progression.

Skp2 inhibition leads to IDH1/2 accumulation and renders metabolic shift from glycolysis to TCA cycle

To further determine whether targeting the Skp2–IDH1 axis provides therapeutic possibility for prostate cancer, we treated 22Rv1 and LNCaP cells with the Skp2 inhibitor, SKPin C1.³⁰ Notably, SKPin C1 significantly stabilized both IDH1 and IDH2 in 22Rv1 and LNCaP cells in a dose-dependent manner (Fig. 7a; Supplementary information, Fig. S7a). As expected, IDH1 was cytosolic regardless of SKPin C1 treatment (Supplementary information, Fig. S7b). However, IDH2, which is normally mitochondrial, was detected in the cytoplasm after SKPin C1 treatment, suggesting that SCF^Skp2-mediated degradation of IDH2 possibly occurs before its translocation into the mitochondria (Supplementary information, Fig. S7c). Moreover, SKPin C1 treatment phenocopied the effects of expressing the non-degradable T157A-IDH1 mutant with respect to cellular proliferation (Fig. 7b). These effects were on target because they were largely abolished in cells lacking Skp2 (Fig. 7c). Furthermore, SKPin C1 treatment led to a shift from glycolysis to TCA cycle, similar to genetic depletion of Skp2 (Fig. 7d, e); this effect was IDH1/2 dependent because it was largely abolished in cells lacking IDH1 or IDH2 (Fig. 7d–f).

**Fig. 7: Skp2 inhibitor represses prostate cancer via shifting cell metabolism from glycolysis to TCA cycle.**

p27, which arrests cell cycle in G1 phase by inhibiting CDK kinase activities, is one of the best-characterized Skp2 ubiquitin substrates.³¹ Importantly, depleting P27 in multiple cell lines did not phenocopy the effects of increasing Skp2 with respect to IDH1/2 abundance (Supplementary information, Fig. S7c, d) or cell metabolism (Supplementary information, Fig. S7e, f), thus arguing against the possibility that the effects of Skp2 on IDH1/2 stability and the shift to glycolysis in S phase were indirectly mediated by fluctuations in p27 levels (Supplementary information, Fig. S7g). Moreover, depleting IDH1 in Skp2-depleted cells partially rescued their metabolic phenotype (redirecting cell metabolism in favor of glycolysis), even in the G1 phase (Fig. 7g, h; Supplementary information, Fig. S7h), but did not rescue their colony growth ability (Supplementary information, Fig. S7i, j). It is therefore likely that degradation of both IDH and p27 contributes to the oncogenic role of Skp2 (Fig. 7i).

Discussion

Cancer cells are quickly divided and are in high demand of biomacromolecules, including lipids, nucleotides and amino acids to prepare for the DNA replication in S phase and subsequent cell division.^3,4,5,6,7,8 In yeast, the cell metabolism is coordinated with cell division; the gene expression of metabolic enzymes fluctuate with robust periodicity, particularly, the expression of genes related to oxygen consumption is repressed during division period.⁹ To date, several lines of evidence advocate a bi-directional interplay between the cell cycle and metabolic machineries. On one hand, key glycolytic enzymes, including HK2, PFKFB3, PFKP, and PKM2, are directly regulated in a cell cycle-dependent manner.^{7,12,14,15,16,17} Mechanistically, the HK2 gene expression is regulated in a cyclin D1-dependent manner,¹⁷ while the enzyme activities of PFKP and PKM2 are regulated by cyclin D3/CDK6-dependent phosphorylation events during the cell cycle.⁷ Moreover, the glycolysis-promoting enzyme PFKFB3 is ubiquitinated by several E3 ligases, including SCF^GRR1, SCF^β-TRCP and APC^Cdh1, during the cell cycle.^{11,12,14,15,16} Besides, the regulation of cellular metabolism dependence is also critical for maintaining stemness of stem cells. For example, Fbxo15, another F-box protein, has been reported to control the mitochondrial biogenesis and metabolism in embryonic stem cells via degrading KBP (KIF1-binding protein).³² On the other hand, a perturbed metabolic state can also compromise cell cycle progression.³³

To systematically analyze the metabolic fluctuation during cell cycle, we synchronized several cells with various methods and traced cell metabolism by using Seahorse extracellular analyzer and ¹³C-glucose/glutamine labeling flux assays. The metabolic results indicate that in S phase, cells shift to high rates of glycolysis and low rates of TCA cycle to enable more flux of intermediates into the biomass synthesis pathways (Fig. 1; Supplementary information, Fig. S1). By monitoring the protein expressions of all enzymes in glycolysis and TCA cycle for the synchronized cells in different cell cycle phases, we found that only the abundance of the rate-limiting enzymes in TCA cycle, namely IDH1/2, fluctuated during cell cycle (Fig. 2; Supplementary information, Fig. S2). We further found that the protein stability of IDH1/2 was regulated by SCF^Skp2 E3 ubiquitin ligase (Fig. 3; Supplementary information, Fig. S3). Specifically, during the G1/S transition, accumulated cyclin E activates CDK2,²⁷ which in turn phosphorylates IDH1 at the conserved Thr157 residue, leading to its recognition and subsequent ubiquitination by SCF^Skp2 E3 ligase (Figs. 4, 5; Supplementary information, Figs. S4, S5).

Given that Skp2 plays a pivotal oncogenic role in prostate cancer,²⁹ we further explored the pathological effect of Skp2–IDH1 axis in PrCa setting. Notably, we found that the protein abundance of IDH1 was inversely correlated with that of Skp2, and the Skp2–IDH1 signaling axis contributed to the dictating of the distinctive cancer “Warburg” metabolic phenotype (Fig. 6; Supplementary information, Fig. S6). This study therefore reveals a novel oncogenic role of Skp2 independent of its other biological substrates such as p27 in cell cycle regulation, by promoting the metabolic switch from utilization of TCA cycle to glycolysis, landing further support for the notion that targeting Skp2 may provide a potent anti-cancer therapy in part by suppressing cancer metabolism (Fig. 7; Supplementary information, Fig. S7).

Materials and methods

Cell culture

Human embryonic kidney 293 (HEK293) cells, HEK293FT, HeLa, DLD1, HCT116, U2OS, A375, VCaP, HAP1 cells and MEFs were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM) containing 10% fetal bovine serum (FBS), 100 units of penicillin and 100 mg/mL streptomycin. PC3, DU145, 22Rv1, LNCaP and C4-2 cells were cultured in RPMI1640 containing 10% FBS, 100 units of penicillin and 100 mg/mL streptomycin. RWPE cells were maintained in keratinocyte serum free medium (K-SFM, Invitrogen, 44019). Skp2^+/+ and Skp2^–/– MEFs were described previously.³⁴ Ccna2^f/f, Ccne1^–/–Ccne2^–/–, Ccne1^–/–, Ccne2^–/–, Ccnd1^–/–, Ccnd2^–/– and Ccnd3^–/– MEFs were gifts from Dr. Piotr Sicinski. HAP1-IDH1^–/– (HZGHC003323c006) and HAP1-IDH2^–/– (HZGHC000919c010) cells were purchased from Horizon Discovery. HAP1 is a near-haploid human cell line that was derived from KBM-7, a chronic myelogenous leukemia (CML) cell line.³⁵ HeLa-IDH1^–/– cells were generated using CRISPR/Cas9 with a guide sequence of 5′-TACGAAATATTCTGGGTGGC-3′.³⁶ Cell culture transfection, lentiviral packaging and subsequent infection of various cell lines were performed according to the protocol described previously.³⁷ To determine the proliferation ability of HAP1 after depletion of IDH1 or IDH2, cells were cultured in H-DMEM, then transferred into DMEM without glucose (Thermo Fisher, 11966025) after adding either 25 mM of D-glucose or D-galactose.

HeLa, HCT116, WPMY1, 22Rv1, DU145, PC3 cells and MEFs were used for synchronization. HeLa cells, which have low endogenous Skp2 activity, were used for ectopic expression-based degradation assays. HEK293 cell line was used for ubiquitination assays and co-IP assays to define the interaction between two ectopically expressed proteins, which is the most frequently used cell line for this type of experiment. Human prostate cancer cells, DU145, PC3, LNCaP, VCaP, 22Rv1 and C4-2, were used for evaluating endogenous Skp2 and IDH1 levels, as well as Skp2 knockdown and Skp2 overexpression. HAP1, LNCaP, and 22Rv1 cells were also used for treatments with Skp2 inhibitor, SKPin C1.

Plasmid construction

Skp2 cDNA was subcloned into CMV-GST, pcDNA3-HA and Lenti-puro-HA vectors via BamHI and XhoI sites. IDH1-WT cDNA was subcloned into pET28a-His, pGEX-GST, Flag-CMV and Lenti-hygro-HA vectors via BamHI and XhoI sites. Site-directed mutagenesis to generate various IDH1 degron mutants was performed using the QuikChange XL Site-Directed Mutagenesis Kit (Stratagene) according to the manufacturer’s instructions. HA-cyclin A, HA-cyclin E, HA-CDK2, HA-ERK1, HA-GSK3β and HA-Rbx1 were generated by cloning the corresponding cDNAs into pcDNA3-HA vector via BamHI and XhoI sites. CMV-GST-Fbl3a, CMV-GST-Fbl13, CMV-GST-Fbl18, CMV-GST-Fbo16, CMV-GST-β-TRCP1, CMV-GST-Fbw4, CMV-GST-Fbw6, CMV-GST-Fbw7 and CMV-GST-Skp2 were gifted from Dr. Wade Harper. Myc-cullin 1, Myc-cullin 2, Myc-cullin 3, Myc-cullin 4A, Myc-cullin 4B and Myc-cullin 5 were gifted from Dr. James DeCaprio. The lentiviral vectors containing Skp2 and p27 shRNAs were described before.³⁴ The lentiviral vectors containing cullin 1, cullin 3 and Fbw4 shRNAs were purchased from Open biosystem.

Cell cycle synchronization

Cell synchronization with nocodazole arrest, double thymidine block or serum starvation was described previously.^38,39 Briefly, HeLa cells or HCT116 cells were incubated with 10 μg/mL of nocodazole for 20 h, followed by knocking of dish on hard surface to dislodge mitotic cells and washing with PBS for 3 times. The cells were released for the indicated time periods before harvest. For double thymidine blocking, HeLa, 22Rv1, PC3 or DU145 cells were incubated with 2 mM thymidine for 16 h, then the thymidine was washed away with PBS and cells were grown in fresh media for 9 h, followed by the second round of 2 mM thymidine treatment for 15 h. After that, the cells were washed with PBS for 3 times and released for the indicated time periods before harvest. For serum starvation, MEFs and WPMY1 cells were cultured in serum-free media for 48 h, and then transferred into H-DMEM with 10% FBS for the indicated time periods before harvest.

Seahorse XF24 extracellular bioenergetics analysis

OCR and ECAR were measured using Seahorse XF24 analyzer (Boston, MA, USA). The cell numbers and FCCP concentration were optimized based on previous reports and titration experiments. OCR assays used Seahorse XF basal media containing 25 mM glucose, 1 mM sodium pyruvate and 2 mM glutamine, while ECAR assays used Seahorse XF basal media containing no glucose, no pyruvate and 2 mM glutamine. For OCR assays, the final concentrations of oligomycin, FCCP and antimycin A were 1, 0.3 and 1 μM, respectively, unless indicated otherwise. For ECAR assays, the final concentrations of glucose, oligomycin and 2-DG were 10 mM, 1 μM and 50 mM, respectively, unless indicated otherwise. After the measurement, cells were trypsined and counted, and all data were normalized with cell number.

Immunoblot (IB) and immunoprecipitation (IP) analyses

Cells were lysed in EBC buffer (50 mM Tris, pH 7.5, 120 mM NaCl, 0.5% NP-40) supplemented with protease inhibitors (cOmplete Mini, Roche) and phosphatase inhibitors (phosphatase inhibitor cocktail set I and II, Calbiochem). The protein concentrations of the lysates were measured using the Bio-Rad protein assay on a Beckman Coulter DU-800 spectrophotometer. The lysates were then resolved by SDS-PAGE and immunoblotted with the indicated antibodies. For immunoprecipitation, 1 mg lysates were incubated with the appropriate sepharose beads for 4 h at 4 °C. Immuno-complexes were washed four times with NETN buffer (20 mM Tris, pH 8.0, 100 mM NaCl, 1 mM EDTA and 0.5% NP-40) before being resolved by SDS-PAGE and immunoblotted with the indicated antibodies.

In vitro kinase assay

IDH1 in vitro kinase assays were performed as previously reported.⁴⁰ Briefly, His-IDH1 was expressed in BL21 E. coli and purified using Ni-NTA (Ni-nitrilotriacetic acid) agarose according to the manufacturer’s instructions. One microgram of His-tagged IDH1 WT or mutant protein was incubated in the absence or presence of cyclin E/Cdk2 kinase in kinase assay buffer (10 mM HEPES, pH 8.0, 10 mM MgCl₂, 1 mM dithiothreitol, 0.1 mM ATP). The reaction was initiated by the addition of 10× kinase assay buffer in a volume of 30 μL for 45 min at 30 °C followed by the addition of SDS-PAGE sample buffer to stop the reaction. The reaction products were resolved by SDS-PAGE.

In vitro pull-down assay

His-Skp2 and GST-IDH1 were expressed in BL21 E. coli and purified using Ni-NTA agarose or Glutathione Sepharose 4B according to the manufacturer’s instructions. The GST-IDH1 proteins (2 μg) were eluted using elution buffer (50 mM Tris-HCl, pH 8.0, 10 mM reduced glutathione) and incubated with or without cyclin E/Cdk2 in kinase assay buffer for 1 h. Then, the reaction solution was added with His-Skp2 beads (1 μg) and incubated for 3 h at 4 °C followed by the addition of SDS-PAGE sample buffer to stop the reaction. The reaction products were resolved by SDS-PAGE.

In vivo ubiquitination assay

Denatured in vivo ubiquitination assays were performed as previously described.³⁹ Briefly, HEK293 cells were transfected with Flag-IDH1, His-ubiquitin and HA-Skp2. 48 h after transfection, 30 μM MG132 was added to block proteasome degradation for 6 h and cells were harvested in denatured buffer (6 M guanidine-HCl, pH 8.0, 0.1 M Na₂HPO₄/NaH₂PO₄, 10 mM imidazole). After sonication, the ubiquitinated proteins were purified by incubation with Ni-NTA matrices for 3 h at room temperature. The pull-down products were sequentially washed twice in buffer A, twice in buffer A/TI mixture (buffer A: buffer TI = 1:3) and once in buffer TI (25 mM Tris-HCl, pH 6.8, 20 mM imidazole). The poly-ubiquitinated proteins were separated by SDS-PAGE for IB analyses.

In vitro ubiquitination assay

The in vitro ubiquitination assays were performed as previously described.⁴⁰ To purify the SCF^Skp2 E3 ligase complex, 293 T cells were transfected with vectors encoding Flag-Skp2, HA-Cul1, Myc-Skp1 and HA-Rbx1. The SCF^Skp2 E3 complexes were purified from the WCLs using anti-Flag-M2 agarose. Briefly, GST-IDH1-WT and GST-IDH1-T157A were expressed in BL21 E.coli and purified using Glutathione Sepharose 4B according to the manufacturer’s instructions. Purified, recombinant GST-IDH1-WT and GST-IDH1-T157A proteins were incubated with purified SCF^Skp2 complexes in the presence of purified, recombinant active E1, E2 (UBCH5A and UBCH3), cyclin E/CDK2, cyclin A/CDK2, ATP and ubiquitin. The reactions were stopped by the addition of 2× SDS-PAGE sample buffer, and the reaction products were resolved by SDS-PAGE and probed with the indicated antibodies.

Cycloheximide chasing assay

To determine the half-life of IDH1 protein, cells were treated with 200 μg/mL of cycloheximide and harvested at the indicated time points, and the cell lysate was then subjected to SDS-PAGE and IB of the indicated proteins.

FACS analysis

Cells that were synchronized with nocodazole and released were collected at the indicated time points and stained with propidium iodide (PI) according to the manufacturer’s instructions. Stained cells were analyzed with a Dako-Cytomation MoFlo sorter (Dako).

Peptide-binding assay

The IDH1 peptides with/without phosphorylation modification were synthesized in LifeTein, LLC (Somerset, New Jersey, USA). Each peptide contained an N-terminal biotin and free C-terminus. The peptides were diluted into 1 mg/mL for further biochemical assays. The sequences were listed below:

IDH1 Biotin-TDFVVPGPGKVEITYTPSDGTQKVTYLVHNF.

pIDH1 Biotin-TDFVVPGPGKVEITYT(p)PSDGTQKVTYLVHNF.

IDH2 Biotin-TDFVADRAGTFKMVFTPKDGSGVKEWEVYNF.

pIDH2 Biotin-TDFVADRAGTFKMVFT(p)PKDGSGVKEWEVYNF.

Peptides (2 μg) were incubated with 10 μg of recombinant Skp2 proteins for 4 h at 4 °C, and 10 μL Streptavidin agarose was added in the sample for another 1 h. The agarose was washed four times with NETN buffer. Bound proteins were added in 2× SDS-PAGE loading buffer and resolved by SDS-PAGE for IB analysis.

Mass spectrometry analysis

Mass spectrometry analysis was performed as described previously with minor modifications.²⁶ Briefly, anti-Flag-IDH1 immunoprecipitations were performed with the WCLs derived from three 10-cm dishes of HEK293 cells co-transfected with Flag-IDH1, HA-cyclin E and HA-CDK2. The immunoprecipitated proteins were resolved by SDS-PAGE, and stained by Gelgold staining buffer. The band containing IDH1 was reduced with 10 mM DTT for 30 min, alkylated with 55 mM iodoacetamide for 45 min, and in-gel digested with trypsin. The resulting peptides were extracted from the gel and analyzed by microcapillary reversed phase liquid chromatography-tandem mass spectrometry (LC-MS/MS) using a high resolution Orbitrap Elite (Thermo Fisher Scientific) in positive ion DDA mode via CID, as previously described. MS/MS data were searched against the human protein database using Mascot (Matrix Science) and data analysis was performed using the Scaffold 4 software.

Clonogenic survival and soft agar assay

Cells were cultured in 10% FBS-containing DMEM or RPMI-1640 media before being plated into 6-well plate with 10,000 cells (3000 cells for HeLa) per well. Ten days later, cells were fixed with 10% acetic acid/10% methanol for 10 min, stained with 0.4% crystal violet/20% ethanol, followed by counting of the colony numbers. For soft agar assays, cells were seeded in 0.4% low-melting-point agarose in DMEM or RPMI-1640 containing 10% FBS with 100,000 per well (30,000 cells for HeLa), and layered onto 0.8% agarose in DMEM or RPMI-1640 with 10% FBS. The plates were kept in the cell culture incubator for 3–4 weeks after which the cells were stained with iodonitrotetrazolium chloride and colonies were counted.

Extraction of labeled metabolites

U-¹³C6-glucose-labeled DMEM was prepared with non-glucose, non-glutamine and non-pyruvate DMEM by adding 10 mM of U-¹³C6 D-glucose, 1 mM sodium pyruvate and 2 mM glutamine. U-¹³C5-glutamine-labeled DMEM was prepared with non-glucose, non-glutamine and non-pyruvate DMEM by adding 2 mM of U-¹³C5 glutamine, 1 mM sodium pyruvate and 25 mM glucose. ¹³C flux assay was performed according to the previous report.³⁸ Briefly, medium was refreshed 1 h before cell harvest to remove accumulated metabolic wastes. For metabolite labeling, before harvesting sample, the cells were changed to U-¹³C6-glucose-labeled media for labeling for 30, 60 and 120 s or U-¹³C5-glutamine-labeled media for labeling for 1, 2 and 3 h. The medium was aspirated completely and 4 mL of dry ice cold 80% MeOH was added, followed by placing of the plates at –80 °C for 30 min. Then the metabolites were extracted as previously described and normalized by protein amount. All metabolites were analyzed as previously described.²⁰

Glucose and glutamine uptake

The uptakes of glucose and glutamine for HeLa cells in either G1 phase or S phase were measured using Glucose Uptake Cell-Based Assay Kit and Glutamate Assay kit according to the manufacturer’s protocol. For glucose uptake, cells were stained with 2-NBDG followed by flow cytometry analysis (excitation/emission = 485/538 nm). For glutamine uptake, cells were harvested and analyzed for OD₄₅₀.

Fractionation of cytoplasm, mitochondria and nuclei

Cells were harvested and subjected to fractionation of cytoplasm, mitochondria, and nuclei using Cell Fractionation kit. All fractions and the WCL were subjected to IB analysis of the indicated proteins, with tubulin, citrate synthase, and Histon H3 as markers of cytoplasm, mitochondria, and nuclei, respectively.

qPCR

Cells were harvested and the total RNA was extracted with TRIzol (Thermo Fisher Scienctific). The RNA was reverse-transcribed (PrimeScript RT Master Mix, RR036A, Takara) and target genes were analyzed with qPCR (FastStart Universal SYBR Green Master, 04913914001, Roche) using the CFX realtime PCR system (BioRad). GAPDH were used as internal control.

Xenograft assays in nude mice

22Rv1 cells stably expressing HA-IDH1-WT or HA-IDH1-T157A were inoculated into the flank of male nude mice (5 × 10⁶ cells per injection, 7–8 mice for each group). After 4 weeks, the mice were sacrificed humanely, and the xenograft tumors were dissected and weighed.

Statistical analysis

The statistical significance between experimental groups was determined by Student’s t-test or one-way ANOVA. The threshold for statistical significance was set to P < 0.05.

Change history

28 August 2020
An amendment to this paper has been published and can be accessed via a link at the top of the paper.

References

Warburg, O. On the origin of cancer cells. Science 123, 309–314 (1956).
Article CAS Google Scholar
Vander Heiden, M. G., Cantley, L. C. & Thompson, C. B. Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 324, 1029–1033 (2009).
Article Google Scholar
Bensaad, K. et al. TIGAR, a p53-inducible regulator of glycolysis and apoptosis. Cell 126, 107–120 (2006).
Article CAS Google Scholar
Gordan, J. D., Thompson, C. B. & Simon, M. C. HIF and c-Myc: sibling rivals for control of cancer cell metabolism and proliferation. Cancer Cell 12, 108–113 (2007).
Article CAS Google Scholar
Manning, B. D. & Cantley, L. C. AKT/PKB signaling: navigating downstream. Cell 129, 1261–1274 (2007).
Article CAS Google Scholar
Christofk, H. R. et al. The M2 splice isoform of pyruvate kinase is important for cancer metabolism and tumour growth. Nature 452, 230–233 (2008).
Article CAS Google Scholar
Wang, H. et al. The metabolic function of cyclin D3-CDK6 kinase in cancer cell survival. Nature 546, 426–430 (2017).
Article CAS Google Scholar
Hanahan, D. & Weinberg, R. A. Hallmarks of cancer: the next generation. Cell 144, 646–674 (2011).
Article CAS Google Scholar
Tu, B. P., Kudlicki, A., Rowicka, M. & McKnight, S. L. Logic of the yeast metabolic cycle: temporal compartmentalization of cellular processes. Science 310, 1152–1158 (2005).
Article CAS Google Scholar
Chen, Z., Odstrcil, E. A., Tu, B. P. & McKnight, S. L. Restriction of DNA replication to the reductive phase of the metabolic cycle protects genome integrity. Science 316, 1916–1919 (2007).
Article CAS Google Scholar
Tudzarova, S. et al. Two ubiquitin ligases, APC/C-Cdh1 and SKP1-CUL1-F (SCF)-beta-TrCP, sequentially regulate glycolysis during the cell cycle. Proc. Natl. Acad. Sci. USA 108, 5278–5283 (2011).
Article CAS Google Scholar
Colombo, S. L. et al. Anaphase-promoting complex/cyclosome-Cdh1 coordinates glycolysis and glutaminolysis with transition to S phase in human T lymphocytes. Proc. Natl. Acad. Sci. USA 107, 18868–18873 (2010).
Article CAS Google Scholar
Bao, Y. et al. Energy management by enhanced glycolysis in G1-phase in human colon cancer cells in vitro and in vivo. Mol. Cancer Res. 11, 973–985 (2013).
Article CAS Google Scholar
Benanti, J. A., Cheung, S. K., Brady, M. C. & Toczyski, D. P. A proteomic screen reveals SCFGrr1 targets that regulate the glycolytic–gluconeogenic switch. Nat. Cell Biol. 9, 1184–1191 (2007).
Article CAS Google Scholar
Herrero-Mendez, A. et al. The bioenergetic and antioxidant status of neurons is controlled by continuous degradation of a key glycolytic enzyme by APC/C–Cdh1. Nat. Cell Biol. 11, 747–752 (2009).
Article CAS Google Scholar
Almeida, A., Bolaños, J. P. & Moncada, S. E3 ubiquitin ligase APC/C-Cdh1 accounts for the Warburg effect by linking glycolysis to cell proliferation. Proc. Natl. Acad. Sci. USA 107, 738–741 (2010).
Article CAS Google Scholar
Sakamaki, T. et al. Cyclin D1 determines mitochondrial function in vivo. Mol. Cell Biol. 26, 5449–5469 (2006).
Article CAS Google Scholar
Shimizu, K. et al. The SCFβ-TRCP E3 ubiquitin ligase complex targets Lipin1 for ubiquitination and degradation to promote hepatic lipogenesis. Sci. Signal. 10, eaah4117 (2017).
Article Google Scholar
Pavlova, N. N. & Thompson, C. B. The emerging hallmarks of cancer metabolism. Cell Metab. 23, 27–47 (2016).
Article CAS Google Scholar
Yuan, M., Breitkopf, S. B., Yang, X. & Asara, J. M. A positive/negative ion-switching, targeted mass spectrometry-based metabolomics platform for bodily fluids, cells, and fresh and fixed tissue. Nat. Protoc. 7, 872–881 (2012).
Article CAS Google Scholar
Srere, P. A. Complexes of sequential metabolic enzymes. Annu. Rev. Biochem. 56, 89–124 (1987).
Article CAS Google Scholar
Kim, S. et al. Suppression of tumorigenesis in mitochondrial NADP+-dependent isocitrate dehydrogenase knock-out mice. Biochim. Biophys. Acta 1842, 135–143 (2014).
Article CAS Google Scholar
Itsumi, M. et al. Idh1 protects murine hepatocytes from endotoxin-induced oxidative stress by regulating the intracellular NADP(+)/NADPH ratio. Cell Death Differ. 22, 1837–1845 (2015).
Article CAS Google Scholar
Gohil, V. M. et al. Nutrient-sensitized screening for drugs that shift energy metabolism from mitochondrial respiration to glycolysis. Nat. Biotechnol. 28, 249–255 (2010).
Article CAS Google Scholar
Wang, Z., Liu, P., Inuzuka, H. & Wei, W. Roles of F-box proteins in cancer. Nat. Rev. Cancer 14, 233–247 (2014).
Article CAS Google Scholar
Liu, P. et al. Cell-cycle-regulated activation of Akt kinase by phosphorylation at its carboxyl terminus. Nature 508, 541–545 (2014).
Article CAS Google Scholar
Koff, A. et al. Formation and activation of a cyclin E-cdk2 complex during the G1 phase of the human cell cycle. Science 257, 1689–1694 (1992).
Article CAS Google Scholar
Zhang, H., Kobayashi, R., Galaktionov, K. & Beach, D. pl9skp1 and p45skp2 are essential elements of the cyclin A-CDK2 S phase kinase. Cell 82, 915–925 (1995).
Article CAS Google Scholar
Lin, H. K. et al. Skp2 targeting suppresses tumorigenesis by Arf-p53-independent cellular senescence. Nature 464, 374–379 (2010).
Article CAS Google Scholar
Wu, L. et al. Specific small molecule inhibitors of Skp2-mediated p27 degradation. Chem. Biol. 19, 1515–1524 (2012).
Article CAS Google Scholar
Carrano, A. C., Eytan, E., Hershko, A. & Pagano, M. SKP2 is required for ubiquitin-mediated degradation of the CDK inhibitor p27. Nat. Cell Biol. 1, 193–199 (1999).
Article CAS Google Scholar
Donato, V. et al. The TDH–GCN5L1–Fbxo15–KBP axis limits mitochondrial biogenesis in mouse embryonic stem cells. Nat. Cell Biol. 19, 341–351 (2017).
Article CAS Google Scholar
Owusu-Ansah, E., Yavari, A., Mandal, S. & Banerjee, U. Distinct mitochondrial retrograde signals control the G1-S cell cycle checkpoint. Nat. Genet. 40, 356–361 (2008).
Article CAS Google Scholar
Inuzuka, H. et al. Acetylation-dependent regulation of Skp2 function. Cell 150, 179–193 (2012).
Article CAS Google Scholar
Carette, J. E. et al. Haploid genetic screens in human cells identify host factors used by pathogens. Science 326, 1231–1235 (2009).
Article CAS Google Scholar
Sanjana, N. E., Shalem, O. & Zhang, F. Improved vectors and genome-wide libraries for CRISPR screening. Nat. Methods 11, 783–784 (2014).
Article CAS Google Scholar
Boehm, J. S., Hession, M. T., Bulmer, S. E. & Hahn, W. C. Transformation of human and murine fibroblasts without viral oncoproteins. Mol. Cell Biol. 25, 6464–6474 (2005).
Article CAS Google Scholar
Wan, L. et al. APCCdc20 suppresses apoptosis through targeting Bim for ubiquitination and destruction. Dev. Cell 29, 377–391 (2014).
Article CAS Google Scholar
Wei, W. et al. Degradation of the SCF component Skp2 in cell-cycle phase G1 by the anaphase-promoting complex. Nature 428, 194–198 (2004).
Article CAS Google Scholar
Inuzuka, H. et al. SCF FBW7 regulates cellular apoptosis by targeting MCL1 for ubiquitylation and destruction. Nature 471, 104–109 (2011).
Article CAS Google Scholar

Download references

Acknowledgements

This work was supported in part by the NIH grants (CA229307 and CA200573 to W.W.; CA183914 to L.W.; R01CA068490, P50CA101942 and R35CA210068 to W.G.K.), the National Basic Research Program of China (2015CB553602 to J.K.L.; 2015CB856302 to J.G.L.), the National Natural Science Foundation of China (91649106, 31570777, 31770917 and 31700684 to J.K.L.; 81802787 to Y.P.). Fundamental Research Funds for the Central Universities (08143008 and 08143101 to J.K.L.; zrzd2017013 to J.G.L.) and American Cancer Society (to H.I.). We thank Wangxiao He, Zhanwu Hou and Huadong Liu for their help with the peptide synthesis, thank Evan Chen for his kind help with metabolite labeling, and thank Brian J. North and Wei lab members for critical reading of the manuscript, and members of the Wei, Pandolfi, Kaelin and Liu laboratories for helpful discussions.

Author information

These authors contributed equally: Jing Liu, Yunhua Peng, Le Shi

Authors and Affiliations

Department of Pathology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, 02215, USA
Jing Liu, Lixin Wan, Hiroyuki Inuzuka, Jianping Guo, Jinfang Zhang, Xiangpeng Dai & Wenyi Wei
Center for Mitochondrial Biology and Medicine, The Key Laboratory of Biomedical Information Engineering of Ministry of Education, School of Life Science and Technology and Frontier Institute of Science and Technology, Xi’an Jiaotong University, Xi’an, Shaanxi, 710049, China
Jing Liu, Yunhua Peng, Le Shi, Jiangang Long, Shuangxi Zhang, Xun Wang, Jing Gao & Jiankang Liu
Department of Molecular Oncology, H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL, 33612, USA
Lixin Wan
Institute of Precision Medicine, the First Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, 510275, China
Jianping Guo
Medical Research Institute, School of Medicine, Wuhan University, Wuhan, Hubei, 430071, China
Jinfang Zhang
Cancer Research Institute, Beth Israel Deaconess Cancer Center, Department of Medicine and Pathology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, 02215, USA
Min Yuan & John M. Asara
Children’s Medical Center Research Institute, University of Texas Southwestern Medical Center, Dallas, TX, 75390, USA
Xun Wang
Division of Radiopharmaceutical Chemistry, Cyclotron and Radioisotope Center, Tohoku University, Sendai, 980-8578, Japan
Shozo Furumoto
Cancer Institute, Longhua Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai, 200032, China
Lijun Jia
Division of Signal Transduction, Beth Israel Deaconess Medical Center and Department of Medicine, Harvard Medical School, Boston, MA, 02215, USA
Pier Paolo Pandolfi
Howard Hughes Medical Institute, Chevy Chase, MD, USA
William G. Kaelin Jr.
Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA, USA
William G. Kaelin Jr.

Authors

Jing Liu
View author publications
You can also search for this author in PubMed Google Scholar
Yunhua Peng
View author publications
You can also search for this author in PubMed Google Scholar
Le Shi
View author publications
You can also search for this author in PubMed Google Scholar
Lixin Wan
View author publications
You can also search for this author in PubMed Google Scholar
Hiroyuki Inuzuka
View author publications
You can also search for this author in PubMed Google Scholar
Jiangang Long
View author publications
You can also search for this author in PubMed Google Scholar
Jianping Guo
View author publications
You can also search for this author in PubMed Google Scholar
Jinfang Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Min Yuan
View author publications
You can also search for this author in PubMed Google Scholar
Shuangxi Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Xun Wang
View author publications
You can also search for this author in PubMed Google Scholar
Jing Gao
View author publications
You can also search for this author in PubMed Google Scholar
Xiangpeng Dai
View author publications
You can also search for this author in PubMed Google Scholar
Shozo Furumoto
View author publications
You can also search for this author in PubMed Google Scholar
Lijun Jia
View author publications
You can also search for this author in PubMed Google Scholar
Pier Paolo Pandolfi
View author publications
You can also search for this author in PubMed Google Scholar
John M. Asara
View author publications
You can also search for this author in PubMed Google Scholar
William G. Kaelin Jr.
View author publications
You can also search for this author in PubMed Google Scholar
Jiankang Liu
View author publications
You can also search for this author in PubMed Google Scholar
Wenyi Wei
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

J.L. and Y.P. designed and performed most of the experiments with assistance from L.S., L.W., H.I., J.G.L., J.P.G., J.Z., S.Z., X.W., J.G., X.D., S.F. and L.J., M.Y. and J.M.A. performed the LC-MS/MS metabolomic profiling and mass spectrometry analysis of IDH1 T157 phosphorylation. W.W., J.K.L., P.P.P., and W.G.K. supervised the study. J.L. and W.W. wrote the manuscript. All authors commented on the manuscript.

Corresponding authors

Correspondence to Jiankang Liu or Wenyi Wei.

Ethics declarations

Competing interests

W.W. is a co-founder and consultant of the ReKindle Therapeutics. All other authors declare no competing interests.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Liu, J., Peng, Y., Shi, L. et al. Skp2 dictates cell cycle-dependent metabolic oscillation between glycolysis and TCA cycle. Cell Res 31, 80–93 (2021). https://doi.org/10.1038/s41422-020-0372-z

Download citation

Received: 03 March 2020
Accepted: 23 June 2020
Published: 15 July 2020
Issue Date: January 2021
DOI: https://doi.org/10.1038/s41422-020-0372-z

This article is cited by

Cancer immunometabolism: advent, challenges, and perspective
- Qin Dang
- Borui Li
- Xiaowu Xu
Molecular Cancer (2024)
Transcriptomic analysis identifies CYP27A1 as a diagnostic marker for the prognosis and immunity in lung adenocarcinoma
- Yi Yin
- Muqun He
- Xianhe Xie
BMC Immunology (2023)
Metabolic reprogramming in colorectal cancer: regulatory networks and therapy
- Jieping Zhang
- Shaomin Zou
- Lekun Fang
Cell & Bioscience (2023)
Small extracellular vesicles delivering lncRNA WAC-AS1 aggravate renal allograft ischemia‒reperfusion injury by inducing ferroptosis propagation
- Xinyuan Li
- Xiang Peng
- Xin Gou
Cell Death & Differentiation (2023)
Rare sugar l-sorbose exerts antitumor activity by impairing glucose metabolism
- Hui-Lin Xu
- Xiaoman Zhou
- Xiao-Dong Gao
Communications Biology (2023)