Cyclin F-Dependent Degradation of RBPJ Inhibits IDH1R132H-Mediated Tumorigenesis Ruhi S

Published OnlineFirst September 25, 2018; DOI: 10.1158/0008-5472.CAN-18-1772 Cancer Molecular Cell Biology Research

Cyclin F-Dependent Degradation of RBPJ Inhibits IDH1R132H-Mediated Tumorigenesis Ruhi S. Deshmukh, Shalakha Sharma, and Sanjeev Das

Abstract

Cyclin F is a substrate recognition subunit of Skp1-Cul1-F- an RBPJ-dependent manner. Studies in mouse tumor models box protein (SCF) E3 ubiquitin ligase complex. Although there indicated that abrogation of cyclin F expression facilitates have been reports describing the role of cyclin F in the IDH1R132H-mediated tumorigenesis and metastasis. In addi- genotoxic stress response, its function under conditions of tion, increased IDH1R132H levels correlated with reduced altered metabolic homeostasis remain unexplored. Here we cyclin F levels in increasing grades of glioma. These findings report that cyclin F is induced upon metabolic stress in a highlight a novel aspect of cyclin F functions in inhibiting FOXO1-dependent manner. Under metabolic stress condi- tumorigenesis and provide mechanistic insights into regula- R132H tions, cyclin F mediated polyubiquitylation of RBPJ at Lys315, tion of IDH1 . leading to its proteasomal degradation. RBPJ regulated the expression of IDH1, which is often mutated to an oncogenic Significance: These findings reveal mechanistic insights into form IDH1R132H in cancers. Thus, metabolic stress–induced the key role of the cyclin F-RBPJ axis in response to metabolic cyclin F attenuated the oncogenic functions of IDH1R132H in stress in cancer cells. Cancer Res; 78(22); 1–13. 2018 AACR.

Introduction arginine (R132) at the active site. The mutant enzyme produces high levels of the metabolite 2-hydroxyglutarate (2-HG; ref. 9). Cyclin F is the founding member of the F-box family of proteins Furthermore, it has been reported that 2-HG is a competitive and is a substrate recognition component of the SCF E3 ubiquitin inhibitor of 2-OG, and accumulation of 2-HG results in the ligase complex. It has been reported to mediate the ubiquitylation inhibition of the dioxygenase family of enzymes including his- and proteasomal degradation of several proteins to maintain tone demethylases and DNA demethylases that require 2-OG as a genomic stability (1). However, the functions and regulation of cofactor (10). This results in epigenetic alterations that promote cyclin F under conditions of altered metabolic homeostasis tumorigenesis. Thus, it is critical to delineate mechanisms under- remain unexplored. lying IDH1 regulation. Recombination signal binding protein for immunoglobulin In this study, we examined the regulation and functions of kappa J region (RBPJ) is an effector of the Notch pathway (2). cyclin F under metabolic stress conditions. We observed that RBPJ has been linked to regulation of cellular metabolic metabolic stress–induced cyclin F brings about the ubiquitylation processes. A previous study suggests that RBPJ regulates glu- and proteasomal degradation of RBPJ. In addition, IDH1 was cose uptake in memory T cells. Thus, RBPJ is critical for memory identified as a novel RBPJ transcriptional target. Thus, cyclin F T-cell maintenance and secondary immune responses (3). RBPJ inhibits mutant IDH1R132H–promoted tumorigenesis in a RBPJ- has also been reported to promote mTOR complex stability dependent manner. thereby augmenting lipogenesis in a Srebp1c (sterol regulatory element binding transcription factor 1c)-dependent manner (4). Materials and Methods However, the role of RBPJ under metabolic stress conditions is Cell lines and culture conditions yet unclear. U251 and 293A cells were cultured in DMEM containing FBS Isocitrate dehydrogenase 1 (IDH1) is a key enzyme that cata- (Invitrogen), 100 U/mL penicillin, and 100 mg/mL streptomycin lyzes the interconversion of isocitrate and 2-oxoglutarate (2-OG). at 37C. BT142 cells were cultured as described previously (11). IDH1 has been observed to be mutated in diverse cancers (5–8). The cell lines were obtained from ATCC, authenticated, and Among the cancer-associated mutations, the most recurrent one is checked for Mycoplasma contamination. Recombinant adeno- the missense mutation that results in a histidine residue instead of viruses were amplified and titrated as reported previously (12). Cells were grown to approximately 50%–70% confluency and were infected with recombinant adenovirus at a multiplicity of Molecular Oncology Laboratory, National Institute of Immunology, New Delhi, infection (MOI) of 10–20 for the indicated time. GFP-expressing India. adenovirus (Ad-GFP) was used as a negative control. To induce Note: Supplementary data for this article are available at Cancer Research glucose starvation, cells were grown to approximately 50% con- Online (http://cancerres.aacrjournals.org/). fluency and were then cultured in glucose-free DMEM (Invitro- Corresponding Author: Sanjeev Das, Molecular Oncology Laboratory, National gen) containing dialyzed FBS (Invitrogen) and 0.5 mmol/L glu- Institute of Immunology, Aruna Asaf Ali Marg, New Delhi 110067, India. Phone: cose (Sigma-Aldrich). Early passage MEFs were cultured and 911-1267-03702; Fax: 911-1267-42125; E-mail: [email protected] retroviral transduction was performed as described previously doi: 10.1158/0008-5472.CAN-18-1772 (13). Reagents including metformin and MG132 were purchased 2018 American Association for Cancer Research. from Sigma-Aldrich. Palbociclib and MK2206 were purchased

www.aacrjournals.org OF1

Deshmukh et al.

from Selleckchem. AGI5198 was purchased from MedChem were stably transfected (pooled zeomycin-resistant population) Express. AGI5198 (1 mmol/L) was added to the culture medium with control (scrambled), cyclin F, RBPJ, or cyclin F along with Luc2 6 hours prior to the start of starvation time period. RBPJ shRNA. BT142 RBPJ knockdown cells were stably transfected (pooled G418-resistant population) with FLAG-tagged Western blot analysis and immunoprecipitation RBPJK315R. The cell suspension was injected orthotopically into Cells were lysed in lysis buffer [20 mmol/L TrisHCl (pH 7.4), the brain of female nude mice. The mice were randomly assigned 5 mmol/L EDTA, 10 mmol/L Na4P2O7, 100 mmol/L NaF, to groups for injecting different cell lines. The tumor size was 2 mmol/L Na3VO4, 1% (v/v) Nonidet P-40, 1 mmol/L phenyl- assessed for 4 weeks. No blinding was used. For weekly in vivo methylsufonylfluoride, 1 Protease inhibitor mixture (Roche)]. bioluminescence imaging, mice were anesthetized using keta- Equal amounts of proteins per sample were subjected to SDS/ mine (80 mg/kg) and xylazine (10 mg/kg) by intraperitoneal PAGE and transferred to a polyvinylidene difluoride membrane injection. Anesthetized animals were intraperitoneally adminis- (Millipore). The antibodies used are described in Supplementary tered 150 mg/kg D-luciferin in PBS. Imaging was performed in a Table S1. Imaging of all Western blots was performed using a UVP Kodak FX Pro imaging system, and images were analyzed using ChemiDoc-it imager equipped with VisionWorksLS software Carestream imaging software. An X-ray image was taken followed (v7.1; UVP). by acquisition and overlay of the pseudocolor image representing Immunoprecipitations were performed using 500 mg cell the spatial distribution of detected photon counts emerging from extracts pretreated with MNase as described previously (14) and active luciferase within the animal. Camera settings were kept incubated with antibodies as indicated. Subsequent immuno- constant during all measurements. At the end of the experiment, blots were performed as described above. the mice were euthanized, the lungs and liver were harvested, and ex vivo imaging was performed. RT-qPCR Genomic DNA was isolated from blood obtained from eutha- Total RNA was extracted using TRIzol (Invitrogen). nized mice for measurement of circulating tumor cells. qPCR was cDNA synthesis was performed using iScript cDNA synthesis kit performed using primers specific to the human Alu repeat (Bio-Rad) following the manufacturer's protocol. qPCR was per- sequence (forward: 50-ACGCCTGTAATCCCAGCACTT-30; reverse: formed using Maxima SYBR Green Mastermix (Fermentas) in an 50-TCGCCCAGGCTGGAGTGC-30), while mouse actin served as Eppendorf real-time PCR machine. 18S rRNA was used as internal the control (forward: 50-GCTTCTTTGCAGCTCCTTCGTTG-30; 0 0 control for all samples. DDCt method was used to analyze reverse: 5 -TTTGCACATGCCGGAGCCGTTGT-3 ). RT-qPCR data. Error bars are mean SD of three independent experiments with triplicate samples. Primer sequences are listed in Statistical analysis Supplementary Table S2. All experiments were conducted independently at least three times. Results were expressed as mean SD. Sample size was Xenograft studies chosen to give sufficient power for calling significance with All animal protocols were approved by an Institutional standard statistical tests. Statistical analyses were performed by Animal Care and Use Committee. Female nude (nu/nu) mice a standard two-tailed Student t test or one-way ANOVA. P < 0.05 were subcutaneously injected in the right flank with 4 106 was considered significant. cells/0.15 mL of PBS. The mice were randomly assigned to groups for injecting different cell lines. The tumor size was assessed for 4 weeks. No blinding was used. Tumor volume was Results determined by direct measurement with a caliper and was Cyclin F is induced upon metabolic stress in a FOXO1- calculated using the formula: (widest diameter smallest dependent manner diameter2)/2. At the end of the experiment, the mice were To investigate the function of cyclin F upon metabolic stress, we euthanized, tumors were harvested and weighed, and metab- first examined its effect on cyclin F levels. Our results indicated olite and Western blot analyses were performed. For IHC that, unlike cyclins D, E, A, and B, cyclin F levels were induced analysis, tumors were harvested, fixed in formalin, and embed- upon metabolic stress (Fig. 1A; Supplementary Fig. S1A). We also ded in paraffin for sectioning. Five-micron–thick paraffinsec- observed a congruent increase in cyclin F transcript levels (Fig. 1B; tions were dewaxed and immunostained with 1:100 Ki-67 Supplementary Fig. S1B). Similar results were obtained in other antibody (Santa Cruz Biotechnology) or 1:100 CD 31 antibody cell types (Supplementary Fig. S1C–S1F). We did not observe any (Santa Cruz Biotechnology). Secondary detection was per- change in the levels of cyclins D, E, A, B, and F in unstressed formed using anti-mouse Alexa Fluor 488 (Molecular Probes) conditions (25 mmol/L glucose; Supplementary Fig. S1G). We or anti-rabbit Alexa Fluor 555 antibody (Molecular Probes). further examined the half-life of cyclin F protein (Supplementary Counterstaining was done with DAPI. Slides were imaged using Fig. S1H). Our results indicate that there is no significant change in a Zeiss ApoTome microscope, and images were analyzed with the half-life of cyclin F protein under metabolic stress conditions. the Zeiss Zen Blue software. Quantitative analysis of immuno- We also observed that metformin treatment-induced metabolic fluorescence staining was performed using ImageJ software. The stress results in cell-cycle arrest, as has been reported previously average signal intensity from four random fields was used for (15). However, induction of cell-cycle arrest alone by treatment the analysis. with palbociclib (a selective inhibitor of the cyclin-dependent kinases CDK4 and CDK6) does not result in upregulation of cyclin In vivo metastasis assay F levels (Supplementary Fig. S1I and S1J). These results suggest Luc2 BT142 cells were generated by stably transfecting (pooled that the upregulation in cyclin F levels is dependent on alteration neomycin-resistant population) BT142 cells with the in metabolic homeostasis. To delineate the mechanism of cyclin F Luc2 pGL4.51[luc2/CMV/Neo] plasmid (Promega). BT142 cells induction upon metabolic stress, we analyzed its promoter region

OF2 Cancer Res; 78(22) November 15, 2018 Cancer Research

Cyclin F Inhibits IDH1R132H-Mediated Tumorigenesis

Figure 1. Metabolic stress results in cyclin F induction. A and B, U251 cells were subjected to glucose starvation for indicated time points. Western blotting (A) and RT-qPCR (B) were then performed. C, Schematic representation of the FOXO1 consensus binding site at the cyclin F (CCNF) promoter. D, Left, U251 control (control) and FOXO1 knockdown [FOXO1kd (shRNA#1)] cells were subjected to glucose starvation for indicated time points. RT-qPCR was then performed. Right, U251 control (control) and U251 FOXO1 knockdown [FOXO1kd (shRNA#1)] cells were harvested and Western blotting was performed. E, U251 control (control) and U251 FOXO1 knockdown [FOXO1kd (shRNA#1)] cells were subjected to glucose starvation for indicated time points. Western blotting was then performed. F, U251 cells were subjected to glucose starvation for indicated time points. Chromatin immunoprecipitation assay was then performed. G, U251 cells were subjected to glucose starvation for indicated time points. Western blotting was then performed from the cytoplasmic and nuclear fractions. H and I, U251 control (CCNFwt)and U251 CCNFmut cells were subjected to glucose starvation for the indicated time points. Chromatin immunoprecipitation assay (H) and RT-qPCR (I) were performed. Error bars are mean SD of three independent experiments with triplicate samples. , P < 0.05; , P < 0.01; , P < 0.001. and identiﬁed a putative binding site for FOXO1, a key member of lated by FOXO1 but not by other family members like FOXO3a, the FOX (Forkhead box) family of transcription factors involved FOXO4, or FOXO6 (Supplementary Fig. S2D). We further exam- in metabolic stress response (16, 17) (Fig. 1C). We next examined ined the binding of FOXO1 to cyclin F promoter. Upon metabolic cyclin F transcript levels upon metabolic stress, in the presence or stress, increasing levels of FOXO1 were detected at cyclin F pro- absence of FOXO1. We observed that in absence of FOXO1, there moter (Fig. 1F). FOXO1 is phosphorylated by Akt at Ser256 and was no induction in cyclin F upon metabolic stress (Fig. 1D and E). by Cdk1/2 at Ser249, which determine its cellular distribution and Similar results were obtained when other shRNAs were used for hence its activity (18–20). We observed that levels of phosphor- FOXO1 depletion to exclude off-target effects of shRNA (Supple- ylated FOXO1 (Ser256), which is cytoplasmic, declined upon mentary Fig. S2A and S2B). Furthermore, FOXO1 levels remain metabolic stress concomitant to an increase in nuclear FOXO1 unaltered under unstressed and starvation conditions (Fig. 1E; levels (Fig. 1G). However, no phosphorylation at Ser249 site was Supplementary Fig. S2C). Moreover, cyclin F is speciﬁcally regu- detected. Thus, our results suggest that Akt plays a key role in

www.aacrjournals.org Cancer Res; 78(22) November 15, 2018 OF3

Deshmukh et al.

determining FOXO1 subcellular localization under metabolic transcript levels (Fig. 3B). Moreover, no decline in interaction with stress conditions. To further corroborate the role of FOXO1 in cyclin F was observed upon metabolic stress in presence of MG132 cyclin F transcriptional regulation, we performed reporter assays. (Supplementary Fig. S3E). We next examined RBPJ protein levels Our results indicated that cyclin F promoter was robustly trans- in presence or absence of cyclin F. We observed that cyclin F activated over the time course of starvation period but abrogation depletion abrogates RBPJ downregulation upon metabolic stress of FOXO1 expression inhibited transactivation of cyclin F pro- (Fig. 3C; Supplementary Fig. S4A). Furthermore, MG132 treat- moter (Supplementary Fig. S2E). Moreover, deletion of the ment inhibited the downregulation of RBPJ levels upon metabolic FOXO1-binding site in the reporter assay abolished FOXO1- stress, while in the absence of cyclin F, MG132 treatment had no mediated transactivation. Furthermore, CRISPR-Cas9–mediated significant effect on RBPJ levels. Similar results were obtained in disruption of the FOXO1-binding site in the cyclin F promoter other cell types (Supplementary Fig. S4B and S4C). We also region abrogates FOXO1-mediated regulation of cyclin F expres- examined whether cyclin F has any effect on the half-life of RBPJ sion (Supplementary Fig. S2F; Fig. 1H and I). Taken together, protein (Supplementary Fig. S4D). In the presence of cyclohex- these results suggest that FOXO1 induces cyclin F upon metabolic imide, there was a significant cyclin F–dependent decline in RBPJ stress. protein levels under starvation conditions with a distinct reduction in half-life. We also observed that unlike wild-type RBPJ, Cyclin F interacts with RBPJ cyclin F noninteracting mutant RBPJRxI/AxA levels are not down- To gain insights into its functions, we employed a biochemical regulated under metabolic stress conditions (Supplementary Fig. approach for identifying cyclin F–interacting proteins (Fig. 2A). S4E). We next investigated whether cyclin F mediates RBPJ ubi- Many novel as well as known interactors of cyclin F were identified quitylation. We observed that RBPJ was polyubiquitylated upon including CP110 and NUSAP1 (21, 22). Among the novel inter- metabolic stress that was inhibited upon abrogation of cyclin F actors, RBPJ was of particular interest as little is understood about expression (Fig. 3D). Moreover, RBPJ polyubiquitylation was its regulation. Thus, we further examined cyclin F–RBPJ interac- found to be K48-linked (Fig. 3E; Supplementary Fig. S4F). Similar tion. We observed that RBPJ coimmunoprecipitated with cyclin F results were obtained in other cell types (Supplementary Fig. S4G (Fig. 2B and C). To confirm whether cyclin F directly interacts with and S4H). Similar observations were made when other shRNAs RBPJ, GST pull-down experiments were performed (Supplemen- were used for cyclin F depletion to exclude off-target effects of tary Fig. S3A). Our results indicated that cyclin F directly interacts shRNA (Supplementary Fig. S4I and S4J). Furthermore, polyubi- with RBPJ. We further observed that RBPJ specifically interacts quitylation of RBPJ was abrogated in the presence of K48R with cyclin F but not with other F-box proteins (Supplementary ubiquitin mutant (Supplementary Fig. S4K). Fig. S3B). We then mapped the domain of RBPJ, to which cyclin F To identify the site of polyubiquitylation, we performed GST binds (Fig. 2D). We observed that cyclin F immunoprecipitated pull-down assays under glucose starvation conditions. The GST- with FLAG-RBPJ segment (RBPJ-3D) encompassing the C-termi- RBPJ fusion protein comprising the amino acids 301–500 was nal RHR domain. Previous reports have suggested that cyclin F observed to be polyubiquitylated (Fig. 3F). We further mutated binds to the Cy motif (RxL/RxI) in its substrates (21). The C- each of the three lysine residues within this region. Our results terminal region of RBPJ has one Cy motif located between indicated that the mutation of K315 residue specifically abrogated residues 304 and 306. Disruption of this motif led to loss of K48-linked polyubiquitylation (Fig. 3G). Moreover, protein cyclin F–RBPJ interaction. Using FLAG-tagged cyclin F constructs sequence analysis of RBPJ region encompassing amino acids we also mapped the domain of cyclin F that binds to RBPJ (Fig. 305–318 suggests that K315 site is conserved across diverse 2E). Our data indicate that RBPJ binds to FLAG–cyclin F segment organisms (Fig. 3H). Furthermore unlike wild-type RBPJ, (cyclin F–2D) containing the cyclin domain. Earlier reports indi- RBPJK315R levels are not downregulated under glucose starvation cate that the hydrophobic patch in the cyclin domain is involved conditions (Supplementary Fig. S4L). These results indicate that in substrate binding (23). We also observed that cyclin F mutated metabolic stress–induced cyclin F promotes K48-linked polyubi- in this hydrophobic patch, cyclin FM309A, cannot interact with quitylation of RBPJ at K315, resulting in its proteasomal RBPJ. Next, we examined the interaction between endogenous degradation. cyclin F and RBPJ. Our results suggested that although RBPJ coimmunoprecipitated with cyclin F, the interaction was attenu- Cyclin F regulates IDH1 levels in a RBPJ-dependent manner ated over the time course of metabolic stress as RBPJ protein levels To determine the effect of RBPJ downregulation on cellular were downregulated (Fig. 2F and G; Supplementary Fig. S3C and metabolic processes, we examined a list of genes involved in S3D). Thus, our results indicate that cyclin F binds to RBPJ, but energy metabolism in cyclin F knockdown and cyclin F/RBPJ upon metabolic stress, the interaction was lost due to down- double knockdown cells by RT-qPCR. We observed that in regulation of RBPJ. U251 control cells, IDH1 expression was downregulated upon metabolic stress concomitant to decline in RBPJ levels. However, Cyclin F targets RBPJ for proteasomal degradation in U251 cyclin F knockdown cells, which constitutively express We further investigated the mechanism of RBPJ downregula- high levels of RBPJ, IDH1 levels were elevated and there was no tion under metabolic stress conditions. Our results indicated that significant change upon metabolic stress. Upon depletion of both there was no significant change in RBPJ transcript levels upon cyclin F and RBPJ, IDH1 expression was constitutively down- metabolic stress (Fig. 3A). Because ubiquitin-mediated protea- regulated (Fig. 4A and B; Supplementary Fig. S5A). Similar results some degradation pathway is frequently involved in regulation of were obtained in other cell types (Supplementary Fig. S5B–S5E). protein levels, we determined the effects of the proteasome Similar observations were made when other sets of shRNAs were inhibitor, MG132, on RBPJ protein levels. We observed that used for cyclin F and RBPJ depletion to exclude off-target effects of MG132 treatment attenuated the decline in RBPJ protein levels shRNA (Supplementary Fig. S5F and S5G). Furthermore, upon upon metabolic stress while there was no significant change in its ectopic expression of ubiquitylation-resistant RBPJ mutant

OF4 Cancer Res; 78(22) November 15, 2018 Cancer Research

Cyclin F Inhibits IDH1R132H-Mediated Tumorigenesis

Figure 2. RBPJ is a cyclin F–interacting protein. A, U251 cells were infected with adenovirus expressing GFP (Ad-GFP) or cyclin F tagged with HA and FLAG epitopes (Ad-Cyclin F HF) for 24 hours. The cells were treated with MG132 (10 mmol/L) for the last 6 hours of the infection period. The cyclin F–associated proteins were detected by immunoprecipitation, followed by SDS-PAGE and silver staining. B, U251 cells were infected with the indicated adenoviruses for 24 hours. The cells were treated with MG132 (10 mmol/L) for the last 6 hours of the infection period. Immunoprecipitation using anti-FLAG antibody followed by Western blotting was then performed. C, U251 cells were infected with the indicated adenoviruses for 24 hours. The cells were treated with MG132 (10 mmol/L) for the last 6 hours of the infection period. Immunoprecipitation using anti-RBPJ antibody followed by Western blotting was then performed. D, U251 cells were transfected with an empty vector (EV) or the indicated FLAG-tagged RBPJ constructs. Twenty-four hours posttransfection, immunoprecipitation followed by Western blotting was then performed. E, U251 cells were transfected with an empty vector (EV) or the indicated FLAG-tagged cyclin F constructs. Twenty-four hours posttransfection, immunoprecipitation followed by Western blotting was then performed. F, U251 cells were subjected to glucose starvation for indicated time points. Immunoprecipitation using anti-cyclin F antibody followed by Western blotting was then performed. G, U251 cells were subjected to glucose starvation for indicated time points. Immunoprecipitation using anti-RBPJ antibody followed by Western blotting was then performed.

(RBPJK315R), there was no decline in IDH1 levels upon metabolic transcription start site (Fig. 4D; ref. 24). We investigated the stress in contrast to the downregulation observed in case of wild- binding of RBPJ to IDH1 promoter. In U251 control cells, RBPJ type RBPJ (Fig. 4C; Supplementary Fig. S5H). Similar results were bound to the IDH1 promoter but the levels detected at the obtained in BT142 cells (Supplementary Fig. S5I). promoter declined upon metabolic stress (Fig. 4E). However, We next analyzed the IDH1 promoter region and identiﬁed a upon abrogation of cyclin F expression, the decline in binding of putative RBPJ binding site approximately 5.8 kb upstream of the RBPJ to IDH1 promoter under starvation conditions was not

www.aacrjournals.org Cancer Res; 78(22) November 15, 2018 OF5

Deshmukh et al.

Figure 3. Cyclin F mediates RBPJ ubiquitylation at K315 residue. A, U251 cells were subjected to glucose starvation for indicated time points. RT-qPCR was then performed. B, U251 cells were subjected to glucose starvation for indicated time points. The cells were treated with MG132 for the last 6 hours of the indicated time points. Western blotting (left) and RT-qPCR (right) were performed. C, U251 control (control) and cyclin F knockdown [CCNFkd (shRNA#1)] cells were subjected to glucose starvation for indicated time points. Cells were treated with MG132 for the last 6 hours of the indicated time points. Western blotting was then performed. D, The indicated cells were subjected to glucose starvation for indicated time points and treated with MG132 (10 mmol/L) for the last 6 hours. Immunoprecipitation using anti- RBPJ antibody followed by Western blotting was then performed. E, The indicated cells were subjected to glucose starvation for indicated time points and treated with MG132 (10 mmol/L) for the last 6 hours. Immunoprecipitation using anti-RBPJ antibody followed by immunoblotting was then performed. F, U251 cells were transfected with constructs expressing full-length or different domains of RBPJ (1D: 1–157 aa, 2D: 158–300 aa, 3D: 301–500 aa) as GST fusion proteins. Twenty-four hours posttransfection, the cells were subjected to glucose starvation for indicated time points and treated with MG132 (10 mmol/L) for the last 6 hours of starvation period. GST pull-down followed by immunoblotting was then performed. G, Left, U251 control (control) and RBPJ knockdown [RBPJkd (shRNA#1)] cells were harvested and Western blotting was performed. Right, U251 RBPJ knockdown cells were transfected with FLAG-tagged wild-type RBPJ, RBPJK315R, RBPJK319R, or RBPJK335R. Twenty-four hours posttransfection, the cells were subjected to glucose starvation for indicated time points and treated with MG132 (10 mmol/L) for the last 6 hours of starvation period. Immunoprecipitation using anti-FLAG antibody followed by immunoblotting was then performed. H, The amino acid sequence around human RBPJ K315 was aligned with those from other species. Error bars are mean SD of three independent experiments with triplicate samples.

OF6 Cancer Res; 78(22) November 15, 2018 Cancer Research

Cyclin F Inhibits IDH1R132H-Mediated Tumorigenesis

Figure 4. Cyclin F inhibits RBPJ-dependent IDH1 induction. A, U251 control (control), cyclin F knockdown [CCNFkd (shRNA#1)], and cyclin F/RBPJ double knockdown [CCNFkd (shRNA#1)/RBPJkd (shRNA#1)] cells were subjected to glucose starvation for indicated time points. RT-qPCR was then performed for the indicated genes. Heatmap comparing relative mRNA levels is shown. Red and green indicate upregulation or downregulation, respectively. B, The indicated cells were subjected to glucose starvation. Western blotting was then performed. C, U251 RBPJ knockdown cells were transfected with FLAG-tagged wild-type RBPJ and RBPJK315R. Twenty-four hours posttransfection, the cells were subjected to glucose starvation for indicated time points. Western blotting was then performed. D, Schematic representation of the RBPJ-binding site at the IDH1 promoter. E, U251 RBPJ knockdown cells were stably transfected (pooled neomycin-resistant population) with FLAG-tagged RBPJK315R. U251 control (control), cyclin F knockdown [CCNFkd (shRNA#1)], cyclin F/RBPJ double knockdown [CCNFkd (shRNA#1)/RBPJkd (shRNA#1)] cells, as well as U251 RBPJ knockdown cells expressing RBPJK315R [RBPJkd (shRNA#1), RBPJK315R] were subjected to glucose starvation for indicated time points. Chromatin immunoprecipitation assay was then performed. Error bars are mean SD of three independent experiments with triplicate samples. , P < 0.01.

www.aacrjournals.org Cancer Res; 78(22) November 15, 2018 OF7

Deshmukh et al.

observed. Upon abrogation of expression of both cyclin F and knockdown cells. In presence of RBPJK315R, metabolic stress did RBPJ, no significant levels of RBPJ could be detected at the IDH1 not trigger significant alterations in 5hmC levels, but AGI5198 promoter. Furthermore, in case of ubiquitylation-resistant RBPJ treatment resulted in an increase in 5hmC levels. Similar observa- mutant (RBPJK315R), metabolic stress had no effect on its ability to tions were made in IDH1R132H-expressing U251 cells (Supple- bind IDH1 promoter. We further performed reporter assays. We mentary Fig. S6A–S6C). Similar results were obtained when other observed that transactivation of IDH1 promoter was downregu- sets of shRNAs were used for cyclin F and RBPJ depletion to lated upon metabolic stress concomitant to decline in RBPJ levels. exclude off-target effects of shRNA (Supplementary Fig. S6D and Moreover, abrogation of RBPJ expression as well as deletion of the S6E). We next investigated the effects of cyclin F on the methyl- RBPJ-binding site inhibited IDH1 promoter transactivation (Sup- ation status of H3K4, H3K9, and H3K27. In BT142 control cells, plementary Fig. S5J). In addition, presence or absence of Notch1 metabolic stress led to a decline in H3K9me3 and H3K27me3 did not have any effect on IDH1 levels (Supplementary Fig. S5K). levels, concomitant to RBPJ and IDH1 downregulation. But These data suggest that IDH1 is a Notch-independent RBPJ similar decrease was not observed in BT142 cyclin F knockdown transcriptional target. cells (Fig. 5C). In BT142 cyclin F/RBPJ double knockdown cells, We further analyzed the effects of Akt signaling on RBPJ and the H3K9me3 and H3K27me3 levels were lower as compared IDH1 levels under metabolic stress conditions (Supplementary with the control cells, and there was no discernable change upon Fig. S5L). Our results indicate that under metabolic stress condi- metabolic stress. Upon ectopic expression of RBPJK315R, tions, Akt-dependent FOXO1 phosphorylation is downregulated, H3K9me3 and H3K27me3 levels were not significantly altered which promotes cyclin F induction, leading to RBPJ degradation upon metabolic stress. But when ectopic expression of mutant and IDH1 downregulation. Thus, inhibition of Akt abrogates RBPJ (RBPJK315R) was combined with AGI5198 treatment, it FOXO1 phosphorylation, resulting in constitutively elevated resulted in a decline in H3K9me3 and H3K27me3 levels, as levels of cyclin F and reduced levels of RBPJ and IDH1. These compared with the control cells. H3K4me3 levels remained data suggest that Akt signaling modulates cyclin F–RBPJ–IDH1 unaltered in these conditions. These results indicate that cyclin axis. Previous reports suggest that metabolic stress results in p53- F-dependent RBPJ degradation inhibits IDH1R132H-mediated epi- dependent cell-cycle arrest (13). Hence, we further examined the genetic reprogramming. effect of metabolic stress on the cell-cycle progression as well as To explore the physiological significance of cyclin F–mediated levels of cyclin F, RBPJ, and IDH1 in U251 control and p53 regulation of IDH1R132H epigenetic program, we analyzed the knockdown cells (Supplementary Fig. S5M and S5N). As reported transcript levels of key tumor suppressor genes (TSG) reported to previously, we also observed that metabolic stress triggers cell- be epigenetically silenced in gliomas (25–28). Our data indicate cycle arrest in presence of p53, but not in absence of p53. that the levels of these TSGs are upregulated upon metabolic However, metabolic stress resulted in induction of cyclin F and stress, which is inhibited upon abrogation of cyclin F expression consequent downregulation of RBPJ and IDH1 levels in both (Fig. 5D). In BT142 cyclin F/RBPJ double knockdown cells, the U251 control and p53 knockdown cells. Moreover, refeeding with expression of the TSGs was elevated as compared with control glucose resulted in a reversal of these effects in both U251 control cells, and there was no significant alteration in the levels upon and p53 knockdown cells. Taken together, these results suggest metabolic stress. Ectopic expression of RBPJK315R inhibited the that alteration in metabolic homeostasis determines cyclin F– upregulation of the TSGs upon metabolic stress while simulta- mediated regulation of RBPJ and consequently IDH1 levels. neous treatment with AGI5198 abrogated the inhibition. We further analyzed the promoter regions of these genes by MeDIP Cyclin F inhibits IDH1R132H-mediated epigenetic silencing of assays (Supplementary Fig. S6F). Our data indicate that in BT142 key tumor suppressor genes control cells, metabolic stress–induced cyclin F promotes demeth- Because our results indicated that mutant IDH1 (IDH1R132H)in ylation of the CpG islands in the promoters of the TSGs and BT142 cells is regulated by RBPJ in a cyclin F–dependent manner, depletion of cyclin F resulted in a loss of this effect. However, in we investigated the effect of cyclin F on IDH1R132H activity. BT142 cyclin F/RBPJ double knockdown cells, the extent of IDH1R132H-generated 2-HG levels were downregulated upon promoter methylation was lower as compared with control cells, metabolic stress, but upon abrogation of cyclin F expression, no and there was no significant alteration upon metabolic stress. In decline in 2-HG levels was observed (Fig. 5A). Upon simulta- the presence of RBPJK315R, promoter demethylation upon meta- neous abrogation of cyclin F and RBPJ expression, 2-HG levels bolic stress was attenuated, whereas simultaneous treatment were downregulated, irrespective of metabolic stress, as compared with AGI5198 resulted in reduced promoter methylation, with with control cells. On the other hand, upon ectopic expression of no discernable change under metabolic stress conditions. These RBPJK315R, there was no downregulation of 2-HG levels upon results indicate that cyclin F–dependent RBPJ degradation inhi- metabolic stress, but treatment with AGI5198, an inhibitor of bits IDH1R132H-mediated epigenetic silencing of TSGs. IDH1R132H, resulted in decline of 2-HG levels irrespective of metabolic stress. 2-HG inhibits cellular dioxygenases including Cyclin F–mediated RBPJ downregulation represses IDH1R132H DNA and histone demethylases (10). Hence, we analyzed the oncogenic effects effect of cyclin F on 5-hydroxymethyl cytosine (5hmC) levels. In We next investigated the effect of cyclin F on the malignant BT142 control cells, 5hmC levels were upregulated upon meta- phenotype of IDH1R132H-expressing cells. Our results indicated bolic stress. Upon abrogation of cyclin F expression, there was no that metabolic stress downregulated the invasiveness, migration significant change in 5hmC levels upon metabolic stress (Fig. 5B). potential, and anchorage-independent growth (Fig. 6A–C). Abro- On the other hand, upon simultaneous abrogation of cyclin F and gation of cyclin F expression enhanced the invasiveness, migration RBPJ expression, 5hmC levels were elevated as compared with potential, and anchorage-independent growth, with no signifi- control cells. Moreover, there was no significant change in 5hmC cant change upon metabolic stress. However, the invasiveness, levels upon metabolic stress in BT142 cyclin F/RBPJ double migration potential, and anchorage-independent growth of

OF8 Cancer Res; 78(22) November 15, 2018 Cancer Research

Cyclin F Inhibits IDH1R132H-Mediated Tumorigenesis

Figure 5. IDH1R132H-mediated epigenetic silencing of TSGs is inhibited by cyclin F. A–D, BT142 RBPJ knockdown cells were stably transfected (pooled neomycin-resistant population) with FLAG-tagged RBPJK315R. BT142 control (control), cyclin F knockdown [CCNFkd (shRNA#1)], cyclin F/RBPJ double knockdown [CCNFkd (shRNA#1)/RBPJkd (shRNA#1)] cells, as well as BT142 RBPJ knockdown cells expressing RBPJK315R [RBPJkd (shRNA#1), RBPJK315R] were subjected to glucose starvation and AGI5198 treatment, as indicated. The levels of 2-HG and 2-OG were measured and their ratio was plotted (A). The levels of 5hmC were measured (B). Western blotting was performed (C). RT-qPCR was performed (D). Error bars represent mean SD of three independent experiments with triplicate samples. , P < 0.01.

BT142 cyclin F/RBPJ double knockdown cells was lower as com- To further investigate the differences in tumor phenotype, we pared with control cells, and there was no significant alteration checked the 2-HG levels in these tumors. Cyclin F knockdown upon metabolic stress. In presence of RBPJRxI/AxA and RBPJK315R, resulted in elevated 2-HG levels while concomitant knockdown of invasiveness, migration potential and anchorage-independent RBPJ lowered 2-HG levels (Supplementary Fig. S7B, left). In growth was elevated with no significant change under metabolic presence of RBPJK315R, 2-HG levels were elevated while simulta- stress conditions. However, treatment with AGI5198 led to a neous AGI5198 treatment reduced 2-HG levels. Concurring reduction in the invasiveness, migration potential, and anchor- observations were made when 5hmC levels were examined (Sup- age-independent growth of RBPJK315R-expressing cells with no plementary Fig. S7B, right). Furthermore, increased Ki67 and significant change under metabolic stress conditions. Similar CD31 staining, which are indicative of proliferation rate and results were obtained in U251 cells expressing IDH1R132H (Sup- angiogenesis, respectively, was observed upon cyclin F knock- plementary Fig. S6G–S6I). Similar observations were made when down while simultaneous knockdown of RBPJ mitigated these other sets of shRNAs were used for cyclin F and RBPJ depletion to effects. In presence of RBPJK315R, increased proliferation and exclude off-target effects of shRNA (Supplementary Fig. S6J and angiogenesis was observed while simultaneous AGI5198 admin- S6K). These results indicate that cyclin F–mediated RBPJ degra- istration resulted in decreased proliferation and angiogenesis dation inhibits IDH1R132H-dependent malignant transformation. (Supplementary Fig. S7C and S7D). These results suggest that We next examined the effect of cyclin F on tumorigenicity of cyclin F suppresses IDH1R132H-mediated tumorigenesis. IDH1R132H-expressing cells. Under metabolic stress conditions, cyclin F depletion resulted in significantly larger tumors as com- IDH1R132H-promoted aggressive tumor phenotype is inhibited pared with control cells (Fig. 6D and E; Supplementary Fig. S7A). by cyclin F However codepletion of RBPJ resulted in reduced tumor growth. We further investigated the tumor-suppressive effects of cyclin Presence of RBPJK315R promoted tumor growth while simulta- F on the metastatic potential of IDH1R132H-expressing cells. We neous administration of AGI5198 attenuated the tumor growth. observed that under metabolic stress conditions, BT142 cyclin F

www.aacrjournals.org Cancer Res; 78(22) November 15, 2018 OF9

Deshmukh et al.

Figure 6. Cyclin F downregulates IDH1R132H oncogenic functions. A and B, BT142 control (control), cyclin F knockdown [CCNFkd (shRNA#1)], cyclin F/RBPJ double knockdown [CCNFkd (shRNA#1)/RBPJkd (shRNA#1)] cells, as well as BT142 RBPJ knockdown cells expressing RBPJRxI/AxA [RBPJkd (shRNA#1), RBPJRxI/AxA], or RBPJK315R [RBPJkd (shRNA#1), RBPJK315R] were subjected to glucose starvation and AGI5198 treatment, as indicated. In vitro invasion of cells was measured (A). The migration potential of cells was measured (B). Error bars represent mean SD of three independent experiments with triplicate samples. , P < 0.01. C, The indicated cells were seeded in soft agar and maintained in 25 mmol/L or 0.5 mmol/L glucose, and treated with AGI5198 as indicated. Representative images are shown (left). Scale bars, 200 mm. Numbers of colonies with a diameter greater than 20 mm were counted (right). Error bars represent mean SD of three independent experiments. , P < 0.01. D, The indicated cells were injected subcutaneously into nude mice. After subcutaneous tumors were established, the mice were administered metformin (5 mg/mL in drinking water) and AGI5198 (450 mg/kg per day via gavage) daily, as indicated. The data shown are representative of three independent experiments (n ¼ 5 mice per group). Error bars represent mean SD from ﬁve individual mice. , P < 0.01. E, Tumor lysates from D above were prepared, and Western blotting was performed.

knockdown cells formed much larger tumors as compared with istration led to lowered CTC count. Moreover, the levels of the control cells (Fig. 7A and B). Depletion of cyclin F also resulted in epithelial marker E–cadherin were lower, while those of the numerous metastatic nodules in the liver as well as lungs (Fig. 7C mesenchymal markers vimentin and ﬁbronectin were elevated and D). Similar observations were made with cells expressing in tumors arising from cells depleted for cyclin F, or cells expres- RBPJK315R. However, treatment with AGI5198 resulted in reduced sing RBPJK315R, which was reversed upon AGI5198 treatment tumor size and metastasis. Moreover, concomitant depletion of (Supplementary Fig. S7E). Thus, we concluded that cyclin F both cyclin F and RBPJ also resulted in reduced tumor size and suppresses IDH1R132H-mediated tumor growth and metastasis metastasis. We also observed higher number of circulating tumor in a RBPJ-dependent manner. cells (CTC) in mice implanted with cells depleted for cyclin F or Several previous studies have reported increased levels of cells expressing RBPJK315R (Fig. 7E). However, AGI5198 admin- IDH1R132H in gliomas (29). Thus, we next examined whether

OF10 Cancer Res; 78(22) November 15, 2018 Cancer Research

Cyclin F Inhibits IDH1R132H-Mediated Tumorigenesis

Figure 7. Cyclin F represses IDH1R132H-promoted aggressive tumor phenotype. A, BT142Luc2 control (control), cyclin F knockdown [CCNFkd (shRNA#1)], cyclin F/RBPJ double knockdown [CCNFkd (shRNA#1)/RBPJkd (shRNA#1)] cells, as well as BT142Luc2 RBPJ knockdown cells expressing RBPJK315R [RBPJkd (shRNA#1), RBPJK315R] were orthotopically injected into the brain of nude mice. After 1 week, the mice were administered metformin (5 mg/mL in drinking water) and AGI5198 (450 mg/kg per day via gavage) daily, as indicated. Representative images are shown. The data shown are representative of three independent experiments using 5 individual mice per group. B, Bioluminescence quantification of A was performed. The data shown are representative of three independent experiments using five individual mice per group. Error bars represent mean SD from five individual mice (n ¼ 5 mice per group; , P < 0.01). C and D, At the end of 4 weeks, organs from mice orthotopically implanted with BT142Luc2 cyclin F knockdown cells in A were harvested and analyzed by ex vivo imaging (C) as well as hematoxylin and eosin staining of tumor sections (D). Representative images are shown. The data shown are representative of three independent experiments using five individual mice per group. E, At the end of 4 weeks, blood from mice in A was used to isolate genomic DNA for the measurement of circulating tumor cells. The data shown are representative of three independent experiments using five individual mice per group. Error bars represent mean SD from five individual mice (n ¼ 5 mice per group; , P < 0.01). F, Representative images of immunostaining of cyclin F, RBPJ, and IDH1R132H on sections of normal brain tissue and different grades of human gliomas. Scale bar, 10 mm. there is any association between cyclin F and IDH1R132H levels in study suggests that hypoxia-induced miR-210 downregulates different grades of gliomas. We observed that cyclin F levels were several mitosis-related genes including cyclin F by directly binding lower in gliomas as compared with normal brain tissue while to the 30-UTR. Thus, miR-210 perturbs mitotic progression by RBPJ and IDH1R132H levels were elevated (Fig. 7F). Furthermore, inducing chromosome misalignment and mis-segregation (30). with increase in tumor grades, cyclin F levels declined while RBPJ Oct4 has been reported to transactivate cyclin F and NIPP1 and IDH1R132H levels increased (Supplementary Fig. S7F–S7H). (nuclear inhibitor of protein phosphatase 1; ref. 31). The These results indicate that dysregulation of cyclin F expression and Oct4-cyclin F/Nipp1 pathway inhibits the protein phosphatase consequent upregulation of IDH1R132H is critical for glioma 1 complex–mediated dephosphorylation of Rb to promote progression. embryonic stem cell self-renewal (32). Here we report that under metabolic stress conditions cyclin F levels are induced. Our results cyclin F Discussion further indicate that FOXO1 binds to promoter and triggers its expression. Thus, our findings provide mechanistic Cyclin F by virtue of its role in maintaining genomic stability is insights into regulation of cyclin F expression. critical for normal cell physiology as well as transformation. Despite being a key transcription factor regulating diverse However, little is known about cyclin F regulation. A previous physiologic processes, the regulation of RBPJ remains poorly

www.aacrjournals.org Cancer Res; 78(22) November 15, 2018 OF11

Deshmukh et al.

understood. A recent report suggests that miR-133a-3p down- results in downregulation of IDH1R132H levels. Decline in regulates RBPJ expression to inhibit the tumorigenic properties IDH1R132H levels diminishes the malignant phenotype. Thus, R132H (33). RBPJ has also been reported to be induced under hypoxic our findings delineate a mechanism for regulation of IDH1 conditions (34). However, the mechanism of RBPJ induction expression. under hypoxic conditions remains unclear. Furthermore, emerging reports also indicate an important role of RBPJ in controlling Disclosure of Potential Conflicts of Interest cellular metabolism (3, 4). Nonetheless, the regulation of RBPJ in No potential conflicts of interest were disclosed. response to metabolic cues remains unexplored. Our findings establish the role of cyclin F in the posttranslational regulation of Authors' Contributions RBPJ under metabolic stress conditions. Cyclin F mediates RBPJ Conception and design: R.S. Deshmukh, S. Das ubiquitylation at the conserved K315 residue, leading to its Development of methodology: R.S. Deshmukh, S. Das proteasomal degradation. Thus, our findings provide a new Acquisition of data (provided animals, acquired and managed patients, dimension to the repertoire of cyclin F substrates. provided facilities, etc.): R.S. Deshmukh, S. Sharma, S. Das Analysis and interpretation of data (e.g., statistical analysis, biostatistics, IDH1 is a frequently mutated metabolic enzyme in diverse computational analysis): R.S. Deshmukh, S. Sharma, S. Das cancer types (35). However, molecular mechanisms underlying Writing, review, and/or revision of the manuscript: R.S. Deshmukh, S. Sharma, regulation of IDH1 expression are yet unclear. In pancreatic S. Das cancer cells, HuR, a RNA-binding protein binds to and stabi- Administrative, technical, or material support (i.e., reporting or organizing lizes IDH1 transcript. Elevated IDH1 levels promote chemore- data, constructing databases): R.S. Deshmukh, S. Das sistance in these cells by augmenting antioxidant defense Study supervision: S. Das mechanism (36). A recent report suggests that exogenously Acknowledgments expressed FOXO3A transactivates IDH1. FOXO3A depletion We thank the members of the Molecular Oncology Lab for helpful discus- reduces 2-HG levels and downregulates proliferation capacity sions. The authors would also like to acknowledge the support from National (37). Our results suggest that there is no FOXO1-dependent Institute of Immunology. R.S. Deshmukh and S. Sharma were supported by a alteration in IDH1 levels under normal physiologic conditions fellowship from the Department of Biotechnology, Ministry of Science and (Supplementary Fig. S8A and S8B). Under metabolic stress Technology, Government of India. This work was supported by a grant from the conditions, FOXO1 induces cyclin F, which in turn mediates Department of Biotechnology, Government of India [BT/HRD/NBA/37/01/ RBPJ degradation, leading to IDH1 downregulation. Conse- 2015 (i) to S. Das]. quently, depletion of FOXO1 results in elevated levels of RBPJ The costs of publication of this article were defrayed in part by the payment of and IDH1 under metabolic stress conditions, while codepletion page charges. This article must therefore be hereby marked advertisement in of FOXO1 and RBPJ leads to IDH1 downregulation. Thus, our accordance with 18 U.S.C. Section 1734 solely to indicate this fact. data indicate that IDH1 is a direct RBPJ transcriptional target. In R132H cancer cells expressing the oncogenic mutant IDH1 ,cyclin Received June 11, 2018; revised August 22, 2018; accepted September 18, F–mediated RBPJ degradation under low-nutrient conditions 2018; published first September 25, 2018.

References 1. Galper J, Rayner SL, Hogan AL, Fiﬁta JA, Lee A, Chung RS, et al. Cyclin F: a 10. Xu W, Yang H, Liu Y, Yang Y, Wang P, Kim S-H, et al. Oncometabolite 2- component of an E3 ubiquitin ligase complex with roles in neurodegen- hydroxyglutarate is a competitive inhibitor of a-ketoglutarate-dependent eration and cancer. Int J Biochem Cell Biol 2017;89:216–20. dioxygenases. Cancer Cell 2011;19:17–30. 2. Andersson ER, Sandberg R, Lendahl U. Notch signaling: simplicity in 11. Luchman HA, Stechishin OD, Dang NH, Blough MD, Chesnelong C, Kelly design, versatility in function. Development 2011;138:3593–612. JJ, et al. An in vivo patient-derived model of endogenous IDH1-mutant 3. Maekawa Y, Ishifune C, Tsukumo S, Hozumi K, Yagita H, Yasutomo K. glioma. J Neurooncol 2011;14:184–91. þ Notch controls the survival of memory CD4 T cells by regulating glucose 12. Satija YK, Das S. Tyr99 phosphorylation determines the regulatory milieu uptake. Nat Med 2015;21:55–61. of tumor suppressor p73. Oncogene 2016;35:513. 4. Pajvani UB, Qiang L, Kangsamaksin T, Kitajewski J, Ginsberg HN, Accili D. 13. Jones RG, Plas DR, Kubek S, Buzzai M, Mu J, Xu Y, et al. AMP-activated Inhibition of Notch uncouples Akt activation from hepatic lipid accumu- protein kinase induces a p53-dependent metabolic checkpoint. Mol Cell lation by decreasing mTORC1 stability. Nat Med 2013;19:1054–60. 2005;18:283–93. 5. Mardis ER, Ding L, Dooling DJ, Larson DE, McLellan MD, Chen K, et al. 14. Groisman R, Polanowska J, Kuraoka I, Sawada J, Saijo M, Drapkin R, et al. Recurring mutations found by sequencing an acute myeloid leukemia The ubiquitin ligase activity in the DDB2 and CSA complexes is differen- genome. New Engl J Med 2009;361:1058–66. tially regulated by the COP9 signalosome in response to DNA damage. Cell 6. Parsons DW, Jones S, Zhang X, Lin JC-H, Leary RJ, Angenendt P, et al. An 2003;113:357–67. integrated genomic analysis of human glioblastoma multiforme. Science 15. Ben Sahra I, Laurent K, Loubat A, Giorgetti-Peraldi S, Colosetti P, Auberger 2008;321:1807–12. P, et al. The antidiabetic drug metformin exerts an antitumoral effect in 7. Amary MF, Bacsi K, Maggiani F, Damato S, Halai D, Berisha F, et al. IDH1 vitro and in vivo through a decrease of cyclin D1 level. Oncogene 2008; and IDH2 mutations are frequent events in central chondrosarcoma and 27:3576. central and periosteal chondromas but not in other mesenchymal tumours. 16. Greer EL, Oskoui PR, Banko MR, Maniar JM, Gygi MP, Gygi SP, et al. The J Pathol 2011;224:334–43. energy sensor AMP-activated protein kinase directly regulates the mam- 8. Borger DR, Tanabe KK, Fan KC, Lopez HU, Fantin VR, Straley KS, et al. malian FOXO3 transcription factor. J Biol Chem 2007;282:30107–19. Frequent mutation of isocitrate dehydrogenase (IDH)1 and IDH2 in 17. Lin A, Yao J, Zhuang L, Wang D, Han J, Lam EWF, et al. The FoxO-BNIP3 cholangiocarcinoma identiﬁed through broad-based tumor genotyping. axis exerts a unique regulation of mTORC1 and cell survival under energy Oncologist 2011;17:72–9. stress. Oncogene 2014;33:3183. 9. Dang L, White DW, Gross S, Bennett BD, Bittinger MA, Driggers EM, et al. 18. Brunet A, Bonni A, Zigmond MJ, Lin MZ, Juo P, Hu LS, et al. Akt promotes Cancer-associated IDH1 mutations produce 2-hydroxyglutarate. Nature cell survival by phosphorylating and inhibiting a Forkhead transcription 2009;462:739–44. factor. Cell 1999;96:857–68.

OF12 Cancer Res; 78(22) November 15, 2018 Cancer Research

Cyclin F Inhibits IDH1R132H-Mediated Tumorigenesis

19. Huang H, Regan KM, Lou Z, Chen J, Tindall DJ. CDK2-dependent phos- 28. Mueller W, Nutt CL, Ehrich M, Riemenschneider MJ, von Deimling A, van phorylation of FOXO1 as an apoptotic response to DNA damage. Science den Boom D, et al. Downregulation of RUNX3 and TES by hypermethyla- 2006;314:294–7. tion in glioblastoma. Oncogene 2007;26:583–93. 20. Yuan Z, Becker EBE, Merlo P, Yamada T, DiBacco S, Konishi Y, et al. 29. Yan H, Parsons DW, Jin G, McLendon R, Rasheed BA, Yuan W, et al. IDH1 Activation of FOXO1 by Cdk1 in cycling cells and postmitotic neurons. and IDH2 mutations in gliomas. New Engl J Med 2009;360:765–73. Science 2008;319:1665–8. 30. He J, Wu JB, Xu NH, Xie WD, Li MN, Li JN, et al. MiR-210 disturbs mitotic 21. D'Angiolella V, Donato V, Vijayakumar S, Saraf A, Florens L, progression through regulating a group of mitosis-related genes. Nucleic Washburn MP, et al. SCFCyclin F controls centrosome homeostasis Acids Res 2013;41:498–508. and mitotic ﬁdelity through CP110 degradation. Nature 2010;466: 31. Campbell PA, Perez-Iratxeta C, Andrade-Navarro MA, Rudnicki MA. Oct4 138–42. targets regulatory nodes to modulate stem cell function. PLoS One 2007;2: 22. Emanuele MJ, Elia AEH, Xu QK, Thoma CR, Izhar L, Leng YM, et al. Global e553. identiﬁcation of modular cullin-RING ligase substrates. Cell 2011;147: 32. Schoeftner S, Scarola M, Comisso E, Schneider C, Benetti R. An Oct4-pRb 459–74. axis, controlled by miR-335, integrates stem cell self-renewal and cell cycle 23. D'Angiolella V, Donato V, Forrester FM, Jeong YT, Pellacani C, control. Stem Cells 2013;31:717–28. Kudo Y, et al. Cyclin F-mediated degradation of ribonucleotide 33. Huang Y, Wu Y, Dong J, Han D, Yang S, Jiang L. MicroRNA-133a-3p exerts reductase M2 controls genome integrity and DNA repair. Cell 2012;149: inhibitory effects on gallbladder carcinoma via targeting RBPJ. Am J Cancer 1023–34. Res 2016;6:2448–62. 24. Tun T, Hamaguchi Y, Matsunami N, Furukawa T, Honjo T, Kawaichi M. 34. Onishi H, Yamasaki A, Kawamoto M, Imaizumi A, Katano M. Hypoxia but Recognition sequence of a highly conserved DNA binding protein RBP-J not normoxia promotes Smoothened transcription through upregulation kappa. Nucleic Acids Res 1994;22:965–71. of RBPJ and Mastermind-like 3 in pancreatic cancer. Cancer Lett 25. D'Urso PI, D'Urso OF, Storelli C, Catapano G, Gianfreda CD, Montinaro A, 2016;371:143–50. et al. Retrospective protein expression and epigenetic inactivation studies 35. Dang L, Yen K, Attar EC. IDH mutations in cancer and progress toward of CDH1 in patients affected by low-grade glioma. J Neurooncol development of targeted therapeutics. Ann Oncol 2016;27:599–608. 2011;104:113–8. 36.ZareiM,LalS,ParkerSJ,NevlerA,Vaziri-GoharA,DukleskaK,etal. 26. Merlo A, Herman JG, Mao L, Lee DJ, Gabrielson E, Burger PC, et al. 50 CpG Posttranscriptional upregulation of IDH1 by HuR establishes a pow- island methylation is associated with transcriptional silencing of the erful survival phenotype in pancreatic cancer cells. Cancer Res 2017; tumour suppressor p16/CDKN2/MTS1 in human cancers. Nat Med 77:4460–71. 1995;7:686–795. 37. Charitou P, Rodriguez-Colman M, Gerrits J, van Triest M, Groot Koerkamp 27. Horiguchi K, Tomizawa Y, Tosaka M, Ishiuchi S, Kurihara H, Mori M, et al. M, Hornsveld M, et al. FOXOs support the metabolic requirements of Epigenetic inactivation of RASSF1A candidate tumor suppressor gene at normal and tumor cells by promoting IDH1 expression. EMBO Rep 3p21.3 in brain tumors. Oncogene 2003;22:7862–5. 2015;16:456–66.

www.aacrjournals.org Cancer Res; 78(22) November 15, 2018 OF13

Cyclin F-Dependent Degradation of RBPJ Inhibits IDH1 R132H-Mediated Tumorigenesis

Ruhi S. Deshmukh, Shalakha Sharma and Sanjeev Das

Cancer Res Published OnlineFirst September 25, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-18-1772

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2018/09/25/0008-5472.CAN-18-1772.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/early/2018/10/29/0008-5472.CAN-18-1772. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.