Mevalonate Pathway Provides Ubiquinone to Maintain Pyrimidine

Home , COQ9, Dolichol, Heme a, Mevalonate pathway

Author Manuscript Published OnlineFirst on November 19, 2019; DOI: 10.1158/0008-5472.CAN-19-0650 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

1 Mevalonate pathway provides ubiquinone to maintain pyrimidine 2 synthesis and survival in p53-deficient cancer cells exposed to metabolic 3 stress 4 5 Irem, Kaymak1, Carina, R., Maier1, Werner, Schmitz1, Andrew, D., Campbell2, 6 Beatrice, Dankworth1, Carsten, P., Ade1, Susanne, Walz3, Madelon, Paauwe2, 7 Charis, Kalogirou4, Hecham, Marouf1, Mathias, T., Rosenfeldt5,6, David, M., 8 Gay2,7, Grace, H., McGregor2,7, Owen, J., Sansom2 and Almut, Schulze1,6$# 9 10 1 Theodor-Boveri-Institute, Biocenter, Am Hubland, 97074 Würzburg, Germany 11 2 Cancer Research UK Beatson Institute, Garscube Estate Switchback Road 12 Bearsden Glasgow, G61 1BD 13 3 Comprehensive Cancer Center Mainfranken, Core Unit Bioinformatics, 14 Biocenter, University of Würzburg, Am Hubland, 97074 Würzburg, Germany 15 4 Department of Urology, University Hospital Würzburg, Josef-Schneider-Str. 2, 16 97080 Würzburg 17 5 Department of Pathology, University Hospital Würzburg, Josef-Schneider-Str. 18 2, 97080 Würzburg 19 6 Comprehensive Cancer Center Mainfranken, Josef-Schneider-Str.6, 97080 20 Würzburg, Germany 21 7Institute of Cancer Sciences, University of Glasgow, Garscube Estate, 22 Switchback Road, Bearsden, Glasgow, G61 1QH 23 24 #Corresponding author 25 email: [email protected] 26 $Current address: Division of Tumor Metabolism and Microenvironment, 27 German Cancer Research Center, Im Neuenheimer Feld 281, 69120 28 Heidelberg, Germany ([email protected]) 29 Phone: +49 6221 42 3423 30 31 Running Title: Mevalonate pathway supports ubiquinone synthesis in cancer 32 Conflict of interest: The authors declare no competing financial interests. 33 Keywords: cancer metabolism; colon cancer; p53; mevalonate pathway; 34 SREBP2; ubiquinone; pyrimidine synthesis

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 2019 American Association for Cancer Research. Author Manuscript Published OnlineFirst on November 19, 2019; DOI: 10.1158/0008-5472.CAN-19-0650 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

1 Abstract 2 Oncogene activation and loss of tumor suppressor function changes the 3 metabolic activity of cancer cells to drive unrestricted proliferation. Moreover, 4 cancer cells adapt their metabolism to sustain growth and survival when access 5 to oxygen and nutrients is restricted, such as in poorly vascularized tumor 6 areas. We show here that p53-deficient colon cancer cells exposed to tumor- 7 like metabolic stress in spheroid culture activated the mevalonate pathway to 8 promote the synthesis of ubiquinone. This was essential to maintain 9 mitochondrial electron transport for respiration and pyrimidine synthesis in 10 metabolically compromised environments. Induction of mevalonate pathway 11 enzyme expression in the absence of p53 was mediated by accumulation and 12 stabilization of mature SREBP2. Mevalonate pathway inhibition by statins 13 blocked pyrimidine nucleotide biosynthesis and induced oxidative stress and 14 apoptosis in p53-deficient cancer cells in spheroid culture. Moreover, 15 ubiquinone produced by the mevalonate pathway was essential for the growth 16 of p53-deficient tumor organoids. In contrast, inhibition of intestinal 17 hyperproliferation by statins in an Apc/KrasG12D mutant mouse model was 18 independent of de novo pyrimidine synthesis. Our results highlight the 19 importance of the mevalonate pathway for maintaining mitochondrial electron 20 transfer and biosynthetic activity in cancer cells exposed to metabolic stress. 21 They also demonstrate that the metabolic output of this pathway depends on 22 both genetic and environmental context. 23 24 Significance: 25 p53-deficient cancer cells activate the mevalonate pathway via SREBP2 26 and promote the synthesis of ubiquinone that plays an essential role in reducing 27 oxidative stress and supports the synthesis of pyrimidine nucleotide

1 Introduction 2 The metabolic activity of cancer cells is controlled by genetic alterations 3 and by the tumor microenvironment. Under metabolic stress, defined by 4 reduced access to nutrients and oxygen present in poorly vascularized solid 5 tumors, cancer cells need to adapt their metabolic activity to maintain cell 6 proliferation and survival. One important factor in the adaptation to metabolic 7 stress is the hypoxia inducible factor (HIF), which is stabilized and activated in 8 the absence of oxygen, and promotes the uptake of glucose and its 9 fermentation to lactate while reducing oxidative metabolism (1). However, poor 10 access to the vascular network not only reduces oxygen tension but also lowers 11 the availability of serum-derived nutrients. Therefore, cancer cells need to 12 undergo global rewiring of their metabolic activity to be able to adapt to these 13 conditions. 14 The p53 tumor suppressor is a master regulator of cellular metabolism 15 (2,3). It reduces glucose uptake (4) and alters glycolysis and modulates the flux 16 of metabolites into the pentose phosphate pathway (5-8). Conversely, p53 17 enhances mitochondrial metabolism by promoting the assembly of cytochrome 18 C oxidase (complex IV) and increasing respiration (9). It has been shown that 19 p53 allows cancer cells to adapt to nutrient deprivation, in particular the 20 absence of the amino acid serine and glutamine (10,11). Thus, loss of p53 21 function can increase the sensitivity of cancer cells towards metabolic stress, 22 resulting in a selective vulnerability that could be exploited therapeutically. 23 In this study, we have investigated the role of p53 in the regulation of 24 metabolic processes in colon cancer cells exposed to metabolic stress. In order 25 to recreate the simultaneous reduction in oxygen and nutrient availability found 26 in tumors, we cultured cancer cells as multicellular tumor spheroids. Under 27 these conditions, we find that p53-deficient cancer cells activate the expression 28 of enzymes of the mevalonate pathway via the sterol regulatory element 29 binding protein 2 (SREBP2). Moreover, inhibition of mevalonate pathway 30 activity with statins selectively induced apoptosis in p53-deficient cancer cells 31 exposed to metabolic stress. This effect was mediated by reduced generation 32 of ubiquinone (CoQ10), which p53-deficient cells require to maintain TCA cycle 33 activity, respiration and the synthesis of pyrimidine nucleotides. Our study thus

1 reveals a novel link between the regulation of isoprenoid synthesis and the 2 modulation of electron transfer mediated by ubiquinone in cancer cells. 3 Mevalonate pathway activity is essential for p53-deficient cancer cells to 4 proliferate and survive under the metabolic constraints of the tumor 5 microenvironment. 6 7 8 Materials and Methods 9 Tissue culture and reagents 10 HCT116 p53-isogenic cells were obtained from B. Vogelstein (Johns 11 Hopkins University, Baltimore) and HCT116 p21-isogenic cells from M. 12 Dobbelstein (Georg-August University, Göttingen). RKO p53-isogenic lines 13 were a gift from K.Vousden (Beatson Institute, Glasgow). All other cell lines 14 were from CRUK LRI Research Services, authenticated by STR profiling and 15 used at low passage. Unless stated otherwise, cells were cultured in DMEM 16 with 10% fetal bovine serum (FBS, Gibco), 4 mM L-glutamine and 1% penicillin-

17 streptomycin, at 37°C in a humidified incubator at 5% CO2 and regularly tested 18 for absence of mycoplasma. Etoposide, (R)-Mevalonic acid lithium salt, 19 SB216732, CHIR99021, simvastatin, zoledronic acid monohydrate, coenzyme 20 Q10, NAC, water-soluble cholesterol, uridine and 5-FU were all from Sigma. 21 MG132 and MK2206 were from Bertin Pharma, rapamycin from Cayman 22 Chemicals, mevastatin and YM-53601 from Biomol and nucleosides 23 (EmbryoMax 100x) from Merck-Milipore. 24 25 Spheroid formation, flow cytometry and histology 26 For spheroid formation, 10,000 cells/well were placed in 96-well ultralow 27 attachment plates (Corning® CORN7007) followed by centrifugation at 850g 28 for 10 min. Spheroids were cultured for 12-14 days, during which medium was 29 replaced every three days. 30 Monolayer and spheroid cells were incubated with 20 µM BrdU (Sigma) 31 for 24 hrs, trypsinized and fixed in 80% EtOH. Cells incubated in 2 M HCl with

32 0.5% Triton X-100 for 30 min at room temperature, neutralized with Na2B4O7. 33 and incubated with anti-BrdU-FITC antibodies (Biozol). Cells were washed,

1 treated with RNAse A (24 µg/ml) and propidium iodide (54 µM) for 30 min. 2 Analysis was performed on a BD FACSCanto II using FACSDIVATM software. 3 Spheroids were fixed with 3.7% paraformaldehyde, incubated in 70% 4 ethanol for 16 hrs, mixed with low-melting agarose and paraffin embedded. 5 4μm sections were deparaffinized and rehydrated. Antigen retrieval was 6 performed with citrate buffer (pH 6.0) in a microwave oven for 6 min. Sections 7 were stained with anti-Ki67 (SP6, Thermo Fischer) and anti-Cleaved Caspase 8 3 (Cell Signaling) in PBS/1% BSA at 4°C and biotinylated secondary antibody 9 (Vector Laboratories). Slides were developed with 3,3'-diaminobenzidine (Cell 10 Signaling) and counterstained with Gilmore 3 hematoxylin. For TUNEL staining, 11 sections were heated in citrate buffer (pH 6.0) for 2 min. TUNEL reactions were 12 developed for 1 hour (In Situ Cell Death Detection Kit, Sigma) and 13 counterstained with Hoechst (Sigma). Archival tumor tissue (8) was stained 14 with anti-Ki67 as above. 15 16 RNA sequencing 17 RNA was extracted using RNeasy columns (Qiagen) including DNase I 18 digestion. mRNA was isolated using NEBNext® Poly(A) mRNA Magnetic 19 Isolation Module and library preparation was performed with NEBNext® Ultra™ 20 RNA Library Prep Kit for Illumina following the manufacturer’s instructions. 21 Libraries were size-selected using Agencourt AMPure XP Beads (Beckman 22 Coulter) followed by amplification with 12 PCR cycles. Library quantification 23 and size determination was performed with an Experion system (Bio-Rad) and 24 libraries were sequenced with NextSeq500 (Illumina). 25 RNAseq data were analyzes as described in the supplementary 26 information and are available at GEO (GSE124189). 27 28 RNA extraction and RT-qPCR 29 Total RNA was isolated using PeqGOLD Trifast followed by reverse 30 transcription into cDNA using M-MLV Reverse Transcriptase (Promega) and 31 random hexamer primers. Real-time PCR was performed using Power-up 32 SYBR Green Master Mix (Thermo Fisher Scientific) using Quantitect primers 33 (Qiagen) or custom primers as followed: human ACTB forward 5’- 34 GCCTCGCCTTTGCCGAT-3’ and reverse 5’-CGCGGCGATATCATCATCC-3’;

1 and human CDKN1A forward 5’-TCACTGTCTTGTACCCTTGTGC-3’ and 2 reverse 5’-CGTTTGGAGTGGTAGAAA-3’ (Sigma). All qPCR reactions were 3 performed in technical duplicate on three biologically independent replicate

4 samples. Relative mRNA amounts were calculated using the comparative CT 5 method after normalization to actin B (ACTB). 6 7 Western blotting 8 Cells were lysed in lysis buffer (1% Triton X100, 50 mM Tris pH 7.5, 300

9 mM NaCl, 1 mM EGTA, 1 mM DTT, 1 mM NaVO4 with protease inhibitors for 10 30 minutes and cleared by centrifugation and quantified using BCA assay 11 (Biovision). Nuclear extraction of SREBP2 was performed as previously 12 described (12). Proteins were separated on SDS-PAGE and blotted onto PVDF 13 membrane (Immobilon), blocked with blocking solution (LI-COR) and incubated 14 with primary and secondary antibodies. Signals were detected on an Odyssey 15 scanner and quantified using ImageJ. Antibodies used: p53 (DO-1), p21 (C- 16 19), CCND1 (DSC-6) (from Santa Cruz), HMGCS-1 (#ab155787), histone-3 17 (#ab1791) (from Abcam), SREBP-2 (1D2), ABCA1 (from Novus), SREBP-2 18 (R&D Systems), GSK3 (4G-1E) (from Milipore), PDK1, ACSS2, p-GSK3a/b 19 (Ser21/9), S6 (5G10), p-S6 (Ser240/244), AKT, p-AKT (Ser473), PARP (from 20 Cell Signaling), beta-actin (AC-15), FDFT1, vinculin (from Sigma). Secondary 21 antibodies were from LI-COR Biosciences. 22 23 Stable isotope labelling and mass spectrometry 24 Monolayer cells or spheroids were washed with PBS and medium was 25 replaced with either complete medium or glucose-free medium with 25 mM 26 [U13C]-glucose (Cambridge Isotope Laboratories). Cells were incubated for the 27 indicated times, washed with cold 154 mM ammonium acetate and snap frozen.

28 For tissue extraction, 150 mg of frozen tissue was homogenized in 3 ml of H2O 29 using an UltraTurrax. 30 For water soluble metabolites, samples were extracted with ice-cold

31 MeOH/H2O (80/20, v/v) containing 0.1 µM lamivudine (Sigma) and separated 32 by centrifugation. Supernatants were transferred to a Strata® C18-E column

33 (Phenomenex) which has been conditioned with 1 ml of CH3CN and 1 ml of

1 MeOH/H2O (80/20, v/v). The eluate was dried and dissolved in 50 μl of a

2 mixture of CH3CN and 5 mM NH4OAc (25/75, v/v). 3 For cholesterol and ubiquinone, samples were extracted with ice-cold

4 MeOH/H2O (80/20, v/v) containing 1 µM CoQ9 (Sigma) and separated by 5 centrifugation. Supernatants were extracted twice with 0.4 ml of hexane, 6 collected and taken to dryness under nitrogen at 35°C. Samples were dissolved 7 in 150 µl of hexane and transferred to Strata® SI-1 columns (Phenomenex), 8 washed with 750 µl hexane and 500 ml hexane/acetic acid ethyl ether (18/1 9 v/v). Ubiquinone was eluted with 0.5 ml hexane/acetic acid ethylester (9/1, v/v). 10 Cholesterol was fully eluted with 0.5 ml hexane/acetic acid ethylester (9/1, v/v). 11 Eluates were dried under nitrogen at 35°C and dissolved in 50 µl iPrOH. 12 Metabolites were analyzed by LC-MS using setting provided in 13 supplementary methods. 14 15 Seahorse Assays 16 Spheroids were washed twice and transferred to XFe96 Spheroid 17 Microplates containing 160 µL of Seahorse XF Assay Medium supplemented 18 with 25 mM D-glucose and 10 mM sodium pyruvate at pH 7.4. Oxygen 19 consumption rates (OCR) were determined using an XF96e Extracellular Flux 20 Analyzer (Software Version 1.4) (Agilent) following manufacturer protocol. 21 During the experiment, 2 μM Oligomycin (Merck-Milipore), 0.5 µM FCCP 22 (Sigma) and 1 µM Rotenone/Antimycin A (Sigma) were injected. OCR of eight 23 biologically independent samples was normalized to spheroid area. 24 25 Organoid Culture 26 Mouse small intestines were isolated from wild-type, VillinCreERApcfl/fl or 27 VillinCreERApcfl/flKrasG12D/+ mice sacrificed three days post-induction with 28 tamoxifen, opened longitudinally and washed with PBS. Crypts were isolated 29 as described (13), mixed with 20 µl Matrigel (BD Bioscience) and plated in 24- 30 well plates in Advanced DMEM/F12 (Thermo Fisher) supplemented with 1% 31 penicillin-streptomycin, 10 mM HEPES, 2 mM glutamine, N2 (Thermo Fisher), 32 B27 (Thermo Fisher), 100 ng/ml Noggin and 50 ng/ml EGF (both from 33 Peprotech). Growth factors were added every two days. Experiments were 34 performed on two biologically independent samples. Genotyping was

1 performed using the following primers: p53fl/fl 5’- 2 GAAGACAGAAAAGGGGAGGG-3’ and 5’-AAGGGGTATGAGGGACAAGG- 3 3’; KrasG12D 5’-GTCTTTCCCCAGCACAGTGC-3’, 5’- 4 CTCTTGCCTACGCCACCAGCTC-3’ and 5’- 5 AGCTAGCCACCATGGCTTGAGTAAGT CTGCA-3’. 6 7 Mice 8 All animal experiments were performed under UK Home Office 9 guidelines using project licences 70-8645 or 70-8646. Experimental protocols 10 were subject to the University of Glasgow animal welfare and ethical review 11 board approval. VillinCreER Apcfl/fl and VillinCreER Apcfl/fl KrasG12D/+ mice have 12 been described previously (14). For induction of intestinal hyperproliferation, 13 mice were given a single intraperitoneal injection of 80 mg/kg tamoxifen on one 14 occasion (VillinCreERApcfl/fl KrasG12D/+), or on two consecutive days 15 (VillinCreERApcfl/fl). Mice were treated with a daily dose of 50 mg/kg simvastatin 16 in 0.5% methylcellulose/5% DMSO or vehicle or a daily dose of 35 mg/kg 17 leflunomide in 100 µl 0.15% carboxymethylcellulose via oral gavage from one 2 18 day post initial tamoxifen injection. For H2O tracing, mice were exposed to 8% 2 19 H2O in their drinking water for 4 days. Mice were given an intraperitoneal 20 injection of 250 µl of cell proliferation reagent (RPN201, GE 21 Healthcare/Amersham) 2 hrs prior to sacrifice and tissue sections were stained 22 for BrdU as described in supplementary methods. 23 24 Statistical analysis 25 Statistical details for each experiment are stated in the figure legends. 26 Graphs were generated using GraphPad Prism 6.0 (GraphPad software). 27 Unless otherwise indicated, statistical significance was calculated using the 28 unpaired two-tailed Student t-test. 29 30 31 Results 32 Spheroid culture induces tumor-like transcriptional signatures and leads 33 to complex metabolic reprogramming

1 To determine the influence of conditions of the tumor microenvironment 2 on colon cancer cells, we used isogenic HCT116 lines that are either wild type 3 (wt) for p53 or an isogenic derivative in which the TP53 gene had been deleted 4 by homologous recombination (15). These cells do not express detectable 5 levels of p53 and fail to induce p21 upon treatment with the DNA damaging 6 agent etoposide (Supplementary Fig. S1A). Both cell lines were cultured either 7 as subconfluent monolayers for 48 hours (MLC), or as large 3-dimensional 8 tumor spheroids (diameter >600 µm), thereby exposing cancer cells to 9 gradients of oxygen and nutrient depletion (16). Spheroid cultures (SPC) 10 showed an overall reduction in proliferation compared to MLC, which was 11 similar in both genotypes (Fig. 1A). However, staining for the proliferation 12 marker Ki67 revealed that p53 wt SPC show proliferation only in the outer 13 regions (Fig. 1B), while SPC of p53-deficient cells present Ki67 positivity 14 throughout their cross-sections (Fig. 1B). Similarly, subcutaneous xenograft 15 colon tumors formed by p53 wt HCT116 cells displayed more heterogenous 16 proliferation patterns compared to their p53-deficient counterparts (Fig. 1C). 17 This suggests that p53 is required for cell cycle arrest induced by the nutrient 18 and oxygen-depleted conditions found in SPC and tumors. 19 We next performed transcriptome analysis of p53 wt and deficient cells 20 cultured as SPC, MLC or xenograft tumors. Principal component analysis 21 (PCA) showed that global gene expression in SPC is more similar to those in 22 tumors rather than MLC (mainly in PC1 accounting for 80% of variance (Fig. 23 1D). Gene set enrichment analysis (GSEA) revealed reduced proliferation 24 (Hallmark_E2F_targets) and induction of interferon response and hypoxia 25 signatures as major transcriptional phenotypes in both SPC and tumors 26 compared to MLC (Fig. 1E and 1F). Moreover, analysis of the cell cycle 27 regulator cyclin D1 (CCND1) and the HIF target pyruvate dehydrogenase 28 kinase (PDK1) confirmed reduced proliferation and induction of hypoxia in SPC 29 compared to MLC (Fig. 1G). 30 We next performed metabolomic analysis to determine differences in 31 metabolism between genotypes and culture conditions (Supplementary Fig. 32 S1B). Stable isotope tracing using [U13C]-glucose showed that SPC increases 33 glucose-dependent lactate synthesis in both p53 wt and deficient cells (Fig. 34 1H). Time course experiments revealed that the labelling of TCA cycle

1 metabolites as well as alanine, glutamate and aspartate reached steady state 2 more rapidly in SPC compared to MLC, as maximal labelling was reached much 3 earlier (Supplementary Fig. S1C, D and E). Moreover, fractions of labelled 4 metabolites were reduced, indicating that the contribution of precursors other 5 that glucose, most likely glutamine, to the TCA cycle is higher in SPC compared 6 to MLC. We also found evidence for pyruvate-dependent anaplerosis, as M+3 7 isotopologues for succinate, fumarate and malate were formed more rapidly in 8 SPC compared to MLC (Supplementary Fig. S1D). This pyruvate-dependent 9 anaplerosis supported the production of aspartate, indicated by the high M+3 10 to M+2 ratio for aspartate in SPC (Fig. 1I). 11 While most metabolic differences between MLC and SPC were found in 12 both genotypes, the total levels of aspartate were higher in SPC from p53 wt 13 cells compared to p53-deficient SPC and also compared to MLC (Fig. 1J). 14 Aspartate is a precursor for pyrimidine nucleotide synthesis and thus essential 15 for proliferation (17). Consistently, glucose-derived labelling of uridine 16 monophosphate (UMP), a central metabolite in pyrimidine biosynthesis, while 17 overall reduced compared to MLC, was higher in p53-deficient SPC compared 18 to their wt counterparts (Supplementary Fig. S1F), potentially reflecting higher 19 demand of nucleotides for proliferation. 20 Together, transcriptomic and metabolic analyses demonstrated that 21 SPC induces hypoxic reprogramming of cellular metabolism in cancer cells. 22 However, oxidative reactions required to generate substrates for anabolic 23 reactions (i.e. aspartate) are still supported through anaplerosis. 24 25 Loss of p53 activates the mevalonate pathway via SREBP2 26 We next compared gene expression signatures between p53 wt and 27 deficient cells under the different culture conditions. The major signatures 28 associated with wt p53 status in all conditions were inflammation and interferon- 29 a response (Supplementary Fig. S2A). Signatures associated with p53 30 deficiency in MLC mapped to TGF-b signaling, spermatogenesis and cell cycle 31 (Supplementary Fig. S2A, left part). In contrast, loss of p53 in SPC and 32 xenograft tumors resulted in the induction of genes associated with cholesterol 33 homeostasis (Supplementary Fig. S2A, middle and right part), many of which

1 are regulated by the SREBP transcription factors (18). Moreover, SREBP target 2 genes (Horton_SREBF_targets) (19) showed strong enrichment in p53- 3 deficient cells in SPC and xenograft tumors but not in MLC (Fig. 2A), suggesting 4 that the combined effect of environment and loss of p53 leads to the 5 upregulation of these genes. As cholesterol homeostasis is preferentially 6 regulated by SREBP2, rather than the closely related SREBP1a or SREBP1c 7 isoforms (19), we next investigated expression of canonical SREBP2 target 8 genes. This showed increased expression of HMGCS1, MVD, HMGCR, 9 DHCR7 and FDFT1 mRNA in HCT116 SPC compared to ML, which was further 10 increased upon loss of p53 (Fig. 2B). Moreover, HMGCS1, FDFT1 and ACSS2 11 showed increased protein levels in p53-deficient SPC from a second isogenic 12 colon cancer cell line, RKO (Fig. 2C). 13 We also investigated whether expression of SREBP2 target genes is 14 associated with TP53 mutation in human colorectal adenocarcinoma (CRC). 15 Analysis of a TCGA dataset (20) revealed higher expression of canonical 16 SREBP2 targets in TP53 mutant tumors (Fig. 2D) and increased expression of 17 HMGCS1 in high grade CRC (Fig. 2E). Moreover, two colon cancer cell lines 18 expressing mutant TP53 (HT29 and DLD1), displayed stronger induction of 19 HMGCS1 expression upon SPC compared to p53 wt cell lines (LS174T and 20 LOVO) (Fig. 2F), corroborating that loss of normal p53 function either through 21 mutation or deletion increases the expression of mevalonate pathway genes. 22 Wild type p53 was shown to inhibit mevalonate pathway genes through 23 induction of the cholesterol transporter ABCA1 (21). In agreement with this 24 study, we found that ABCA1 mRNA expression was strongly reduced in p53- 25 deficient cells both in MLC and SPC (Supplementary Fig. S2B). However, 26 levels of ABCA1 protein were higher in p53-deficient MLC and completely 27 absent in SPC (Supplementary Fig. S2C). Interestingly, ABCA1 is a target for 28 miRNA-33, which is encoded by an intron within the SREBF2 gene (22,23). It 29 is therefore possible that ABCA1 is repressed in SPC via a miRNA-dependent 30 mechanism. 31 To address the mechanism of mevalonate pathway regulation in our 32 system, we first confirmed that increased expression of HMGCS1 protein in 33 p53-deficient SPC is abolished upon shRNA-mediated silencing of SREBP2 34 (Supplementary Fig. S2D and E). We also established that p53-deficient SPC

1 contain high levels of the 55 kDa mature form of SREBP2 (Fig. 2G), which 2 represents the active transcription factor. Accumulation of mature SREBP2 and 3 enhanced target expression was also observed in MLC of p53-deficient 4 HCT116 cells cultured in lipid-reduced medium (Supplementary Fig. S2F and 5 G), a condition that induces SREBP processing (18). Nuclear accumulation of 6 mature SREBP2 is mediated by increased processing of the precursor or by 7 stabilization of the mature protein. As SREBP processing is induced by 8 mTORC1 (24,25), we first investigated the activity of this pathway. We found 9 that phosphorylation of the mTORC1 substrate p70S6K (indicated by the higher 10 relative abundance of the upper band) and its downstream target S6 ribosomal 11 protein (S6RB), is strongly increased in SPC compared to MLC (Fig. 2H and 12 Supplementary Fig. S2H). This was surprising as hypoxia, a major feature of 13 SPC, inhibits the mTORC1 pathway (26,27). Indeed, exposure of HCT116 MLC 14 to hypoxia decreased S6RB phosphorylation and slightly reduced HMGCS1 15 expression (Supplementary Fig. S2I). 16 As increased mTORC1 activity was observed in SPC from both 17 genotypes, we also addressed whether loss of p53 alters protein stability of 18 mature SREBP2. Treatment with the proteasome inhibitor MG132 only 19 increased mature SREBP2 in p53 wt SPC, confirming that mature SREBP2 is 20 more stable when p53 is absent (Fig. 2I). Mature SREBP2 is phosphorylated 21 by glycogen synthase kinase 3 (GSK3), leading to its ubiquitination and 22 degradation (28). We found an overall increase in GSK3 phosphorylation on 23 serine 21 (GSK3a) and serine 9 (GSK3b) in SPC compared to MLC, with a 24 further increase in p53-deficient cells (Fig. 2J), indicating reduced activity of the 25 kinase upon p53 loss. Consistently, treatment of p53 wt SPC with GSK3 26 inhibitors increased levels of mature SREBP2 and restored expression of 27 HMGCS1 mRNA to the same level found in p53-deficient cells (Fig. 2K and L). 28 Treatment with the mTORC1 inhibitor rapamycin reduced mature SREBP2 and 29 HMGCS1 mRNA in p53-deficient SPC (Fig. 2K and M). However, this was 30 independent of AKT, as treatment with MK2206 did not affect GSK3 or S6RB 31 phosphorylation (Supplementary Fig. S2J).

1 Together, this suggests that loss of p53 in SPC induces nuclear 2 accumulation of mature SREBP2 through activation of mTORC1 and inhibition 3 of GSK3. 4 5 Inhibition of mevalonate synthesis induces apoptosis in p53-deficient 6 spheroids 7 We next tested whether the mevalonate pathway contributes to cancer 8 cell survival in the metabolically compromised environment of SPC. We used 9 statins, a class of lipid-lowering drugs that inhibit the activity of HMGCR, the 10 rate limiting enzyme of the pathway (Fig. 3A). Statin treatment increased the 11 expression of SREBF target genes, due to inactivation of the negative feedback 12 loop (29), and resulted in global downregulation of cell cycle and epithelial to 13 mesenchymal transition (EMT) expression signatures regardless of genotype 14 (Supplementary Fig. S3A). Protein levels of the S-phase proteins cyclin A 15 (CCNA1) and aurora kinase A (AURKA) were also reduced (Supplementary 16 Fig. S3B), confirming that the mevalonate pathway contributes to proliferation 17 (30) and disruption of tissue architecture (31). 18 When investigating the effect of mevalonate pathway inhibitors on cell 19 viability, we found strong inhibition of proliferation in MLC, irrespective of 20 genotype (Supplementary Fig. S3C). In contrast, in SPC only p53-deficient cells 21 were sensitive to mevastatin treatment, indicated by TUNEL staining, while p53 22 wt cells were largely resistant to this treatment (Fig. 3B and C). Mevastatin- 23 induced apoptosis in p53-deficient cells was blocked by addition of mevalonate, 24 the product of the HMGCR reaction (Fig. 3B and C), confirming the specificity 25 of the inhibitor. Apoptotic cells positive for TUNEL and cleaved caspase 3 were 26 mainly found in the core regions, where cells are experiencing the most severe 27 oxygen and nutrient depletion (Fig. 3B and Supplementary Fig. S3D), 28 suggesting that the mevalonate pathway supports cell viability under metabolic 29 stress. 30 As it has been shown that induction of p21 (CDKN1A) is required for the 31 p53-dependent remodeling of metabolism in response to serine deprivation in 32 colon cancer (10), we asked whether failure to induce p21 could be responsible 33 for the induction of apoptosis by statins in p53-deficient cells. However, while 34 short-term simvastatin treatment (24h) induced CDKN1A mRNA only in p53 wt

1 cells, both genotypes increased CDKN1A mRNA and protein expression after 2 longer statin exposure (72h) (Fig. 3D and E). This indicates that mevalonate 3 pathway inhibition induces p21 through a p53-independent mechanism. 4 Furthermore, statin treatment of SPC of p21-deficient HCT116 cells did not 5 induce apoptosis (Fig. 3F and G), confirming that p21 is dispensable for statin- 6 resistance of p53 wt cells. 7 8 Mevalonate pathway inhibition blocks the production of ubiquinone 9 The mevalonate pathway facilitates the synthesis of isoprenoids, which 10 are substrates for cholesterol synthesis, protein prenylation as well as the 11 synthesis of dolichol, heme A and ubiquinone (Fig. 4A) (32). Ubiquinone 12 consists of a benzoquinone ring derived from tyrosine linked to a tail comprising 13 10 (human) or 9 (mouse) isoprenoid units and functions as electron transfer 14 molecule between the complexes of the respiratory chain (33). To determine 15 mevalonate pathway activity, we treated SPC of p53 wt and deficient cells with 16 [U13C]-glucose and determined label incorporation into different metabolites. 17 SPC of p53-deficient cells increase the incorporation of labelled carbons into 18 mevalonate, resulting in an overall increased abundance of this metabolite (Fig. 19 4B and C). In addition, p53-deficient SPC displayed overall higher levels of 20 acetyl-CoA, the substrate of the mevalonate pathway, without major difference 21 in the proportional labelling of the M+2 fraction, which is generated from citrate 22 (Supplementary Fig. S4A and B). However, despite the observed increase in 23 mevalonate synthesis, the amount of total and labelled cholesterol was much 24 lower in p53-deficient SPC compared to their wt counterparts (Fig. 4D and E). 25 In contrast, the amount of total and labelled ubiquinone was higher in p53- 26 deficient cells, demonstrating a re-routing of metabolites into the ubiquinone 27 synthesis pathway (Fig. 4F and G). Treatment with simvastatin decreased the 28 amount of both metabolites and completely abolished their glucose-derived 29 labelling (Fig. 4D-F and Supplementary Fig. S4C-D). Importantly, isotopologue 30 peak distribution for cholesterol and ubiquinone were similar in both genotypes 31 (Fig. 4E and G), indicating comparable contribution of glucose to the acetyl- 32 CoA pool (34). 33 To investigate whether a p53-dependent switch in the routing of 34 metabolites in the mevalonate pathway can also be observed in tumors, we

1 determined the abundance of cholesterol, 7-dihydroxy cholesterol (7-DHC) and 2 ubiquinone in xenograft colon tumors of p53 wt and deficient HCT116 cells. In 3 support of the results obtained in spheroid cultures, p53 wt colon tumors 4 displayed somewhat higher levels of cholesterol and 7-DHC compared to p53- 5 deficient colon tumors. In contrast, levels of ubiquinone were overall higher in 6 p53-deficient colon tumors, although this difference failed to reach significance 7 (Fig. 4H). Together, these results suggest that loss of p53 alters mevalonate 8 pathway flux to support the production of ubiquinone. 9 10 Inhibition of ubiquinone synthesis impairs TCA cycle and respiration and 11 results in oxidative stress 12 Ubiquinone is an essential component of the mitochondrial electron 13 transport chain (ETC) where it shuttles electrons between NADH-CoQ

14 reductase (complex I) or succinate dehydrogenase (complex II) and CoQH2- 15 cytochrome c reductase (complex III) (Fig. 5A). Using stable isotope tracing 16 with [U13C]-glucose, we found that simvastatin reduced labelled and unlabelled 17 fractions of aspartate and most TCA cycle metabolites in SPC from both 18 genotypes (Fig. 5B and Supplementary Fig. S5A). However, simvastatin 19 reduced basal and maximal oxygen consumption rates (OCR) in SPC of p53- 20 deficient cells, which was rescued by mevalonate addition. In contrast, SPC 21 from p53 wt cells only displayed a small reduction in maximal respiration upon 22 statin treatment (Fig. 5C). 23 Reduced availability of oxygen as final electron acceptor of the electron 24 transport chain can lead to electron leakage and the formation of reactive 25 oxygen species (35). We therefore reasoned that inhibition of ubiquinone 26 synthesis could cause oxidative stress under the hypoxic conditions in SPC, 27 which could lead to the induction of apoptosis. Indeed, replenishing spheroid 28 cultures either with ubiquinone or the antioxidant N-acetyl-cysteine (NAC) was 29 as effective as mevalonate in preventing statin-induced apoptosis in SPC (Fig. 30 5D and E), while cell-permeable cholesterol had no effect (Supplementary Fig. 31 S5B). In addition, the viability of statin-treated MLC was not restored by the 32 addition of ubiquinone (Supplementary Fig. S5C), indicating that multiple 33 products of this pathway are needed to support the rapid proliferation of cancer 34 cells observed in MLC.

1 2 Production of ubiquinone by the mevalonate pathway supports 3 pyrimidine nucleotide biosynthesis 4 Ubiquinone also functions as an electron acceptor for dihydroorotate 5 dehydrogenase (DHODH), an essential enzyme for the generation of 6 pyrimidine nucleotides for DNA and RNA synthesis (Fig. 6A). Stable isotope 7 tracing showed higher incorporation of glucose-derived carbons into UMP in 8 SPC of p53-deficient cells (Fig. 6B). This was detected in the M+5 fraction, 9 representing labelling via ribose, but also in the M+7/M+8 fractions, 10 representing ribose plus either two or three labelled carbons derived from 11 aspartate (Fig. 6B). Treatment with statins significantly lowered labelling and 12 overall levels of UMP in SPC from both genotypes (Fig. 6B and C). This was 13 restored by supplementing statin-treated SPC with either mevalonate or 14 ubiquinone (Fig. 6C), confirming that ubiquinone is rate-limiting for pyrimidine 15 synthesis under these conditions. Moreover, addition of nucleosides or uridine, 16 which can readily be taken up by cells and used to replenish the nucleotide pool 17 via the salvage pathway, was sufficient to block the induction of apoptosis by 18 statins in p53-deficient SPC (Fig. 6D and Supplementary Fig. S6A). 19 The antimetabolite drug 5-fluoro-uracil (5-FU), which is standard-of-care 20 for advanced CRC, exerts its effect mostly through inhibition of thymidylate 21 synthase (TYMS) (36). TYMS converts dUMP to dTMP for DNA synthesis, and 22 5-FU treatment leads to DNA damage and cell death. We therefore investigated 23 whether statins alter 5-FU sensitivity of cancer cells under the metabolic 24 constraints of SPC. Interestingly, while p53 wt HCT116 cells showed 25 remarkable resistance towards 5-FU, most likely due to the low proliferation of 26 these cells in this condition, the drug sensitized the cells to simvastatin 27 treatment (Fig. 6E). In contrast, p53-deficient cells already showed induction of 28 apoptosis in response to statin alone, which was not further increased by 5-FU 29 (Fig. 6E). 30 Collectively, these results demonstrate that ubiquinone production by 31 the mevalonate pathway is essential for pyrimidine biosynthesis in cancer cells. 32 Inhibition of ubiquinone synthesis blocks the viability of p53-deficient cells 33 under the metabolic constraints of SPC. In contrast, p53 wt cells are initially

1 resistant to statin treatment but can be sensitized by the anti-metabolite 5-FU, 2 which blocks dTMP synthesis and causes DNA and RNA damage. 3 4 The metabolic output of the mevalonate pathway depends on 5 environmental context 6 To investigate the role of the mevalonate pathway under different 7 conditions resembling the tumor microenvironment, we used organoid cultures 8 of intestinal epithelial cells from mice carrying conditional alleles of Apc, Trp53 9 or KrasG12D together with VillinCreERT2. Efficient recombination of the Trp53 and 10 Kras locus were confirmed by PCR (Supplementary Fig. S7A). Placed in 11 organoid culture medium, these cells grow as large cysts without any signs of 12 differentiation (13). Interestingly, while simvastatin only had a minor effect on 13 the growth of Apc-deficient organoids, Apc/p53 double deficient cells showed a 14 severe reduction in organoid growth, which was fully restored by mevalonate 15 supplementation (Fig. 7A and B). Reduced organoid growth was accompanied 16 by induction PARP cleavage, a marker of apoptotic cell death (Supplementary 17 Fig. S7B). Similar results were also obtained for Apcfl/fl/KrasG12D and 18 Apcfl/fl/p53fl/fl/KrasG12 organoids (Fig. 7A and B), demonstrating that the deletion 19 of Trp53 sensitizes the organoids towards mevalonate pathway inhibition. 20 Moreover, inhibition of organoid growth in Apc/p53 deficient and Apcfl/fl/ 21 p53fl/fl/KrasG12D/+ cells was robustly restored by addition of either ubiquinone or 22 nucleosides (Fig. 7C and D), confirming that the provision of ubiquinone for 23 nucleotide biosynthesis is an essential function of the mevalonate pathway in 24 CRC tumor organoids. 25 We next assessed the ability of simvastatin to suppress intestinal 26 hyperproliferation induced by acute deletion of Apc and activation of Kras in 27 vivo. This was achieved by crossing mice carrying conditional alleles of Apc or 28 an activated allele of Kras (Apcfl/fl or Apcfl/fl;KrasG12D/+) to mice bearing the 29 VillinCreERT2 transgene (37). After induction of CRE-dependent recombination,

30 mice were treated for 4 days with simvastatin or vehicle and with D2O to assess 31 cholesterol and ubiquinone synthesis in vivo (38). Histological analysis of BrdU 32 positive cells demonstrated that simvastatin had no effect on proliferation in 33 Apc-deficient intestinal crypts, but blocked hyperproliferation induced by Kras 34 activation (Fig. 7E and F). Deuterium tracing revealed that the highly

1 proliferating intestinal crypts in VillinCREERT2Apcfl/fl;KrasG12D/+ mice exhibit high 2 levels of cholesterol synthesis, which was blocked by simvastatin (Fig. 7G). In 3 contrast, ubiquinone synthesis was already lower in intestines from double 4 mutant mice and not further reduced by simvastatin (Fig. 7G), indicating that 5 cholesterol rather than ubiquinone is the limiting metabolite produced by the 6 mevalonate pathway in this system. This was further corroborated by the 7 finding that hyperproliferation of intestinal crypts in 8 VillinCREERT2Apcfl/fl;KrasG12D/+ mice was insensitive to the DHODH inhibitor 9 leflunomide (Supplementary Fig. S7C and D), which has recently been shown 10 to block growth of breast cancer cells (39). Together, these results indicate that 11 de novo pyrimidine synthesis is dispensable for KRAS-induced intestinal 12 hyperproliferation and that the metabolic output of the mevalonate pathway 13 depends on genetic factors and microenvironmental context. 14 15 16 Discussion 17 Metabolic gradients in tumors are likely to simultaneously limit access to 18 oxygen and nutrients, making adaptation by metabolic compensation 19 challenging (40). One potential response of cancer cells to nutrient deprivation 20 is cell cycle arrest, which alleviates the metabolic demand of nucleotide 21 biosynthesis for DNA replication, allowing cancer cells to survive until nutrients 22 become available, for example after formation of new blood vessels or 23 engagement of metabolic symbiosis (41,42). Using spheroid cultures (SPC) as 24 model, we show here that p53 wt colon cancer cells respond to metabolic 25 deprivation by reducing proliferation. In contrast, p53-deficient CRC cells are 26 able to maintain proliferation in the spheroid center, where nutrient and oxygen 27 supply is restricted. Contrary to monolayer cultures, gene expression 28 signatures in SPC are characteristic of cell cycle arrest and induction of 29 hypoxia, similar to those found in tumor tissue. Moreover, stable isotope tracing 30 showed that SPC engage in hypoxic remodeling of their metabolism, with 31 reduced glucose oxidation, enhanced lactate production and increased TCA 32 cycle anaplerosis from pyruvate. Pyruvate anaplerosis promotes glutamine- 33 independent growth of cancer cells (43) and supports aspartate synthesis in

1 succinate dehydrogenase (SDH) deficient cancer cells (44). We found that SPC 2 of p53-deficient colon cancer cells showed reduced aspartate levels, indicating 3 its enhanced usage for pyrimidine biosynthesis. 4 Importantly, p53-deficient CRC cells in SPC or grown as xenograft 5 tumors increase expression of mevalonate pathway enzymes and upregulation 6 of SREBP2 targets was observed in p53-mutant CRC patient samples and cell 7 lines. Previous studies have shown that mutant p53 can bind to SREBP2 and 8 promoting its transcriptional activity during disruption of mammary tissue 9 architecture (31), and that wt p53 represses the mevalonate pathway through 10 ABCA1-dependent inhibition of SREBP2 processing (21). We demonstrate 11 here that loss of p53 in SPC promotes expression of SREBP2 target genes by 12 activating mTORC1 signaling, which drives the processing of SREBP2 (24), 13 and by limiting the activity of GSK3, which controls the phosphorylation- 14 dependent degradation of mature SREBP2 (28). The combination of mTORC1 15 activation and inhibition of GSK3 results in the accumulation of mature SREBP2 16 and increases the expression of its target genes. 17 Our study also shows that tumor-like metabolic stress alters the 18 sensitivity of cancer cells towards mevalonate pathway inhibition. In monolayer, 19 both genotypes were highly sensitive to mevalonate pathway inhibitors, most 20 likely because cells require cholesterol for rapid proliferation (32). However, 21 when exposed to metabolic stress, p53-proficient cells were largely resistant to 22 statin treatment, while p53-deficient cancer cells showed induction of 23 apoptosis. Cell death was restricted to the nutrient- and oxygen-deprived center 24 of the spheroids, indicating that the mevalonate pathway provides essential 25 metabolic functions under these conditions. Surprisingly, sensitivity towards 26 mevalonate pathway inhibition was independent of p21, suggesting that the 27 protective effect of wt p53 is independent of its role as transcriptional inducer 28 of this target. Indeed, it has been shown that an acetylation-deficient form of 29 p53 that is unable to induce p21 retains important tumor suppressive functions 30 (45). 31 We also demonstrate that p53-dependent metabolic rewiring of the 32 mevalonate pathway supports the synthesis of ubiquinone, an important 33 electron transport molecule of the ETC (46). Previous studies indicate that 34 nutrient deprivation increases dependency of cancer cells on ETC activity

1 (47,48), particularly for the generation of aspartate as precursor for pyrimidine 2 synthesis (49). DHODH, the enzyme converting dihydroorotate to orotate 3 during UMP synthesis, requires electron transfer via ubiquinone and has been 4 shown to support the growth of respiration-deficient tumors (39). This suggests 5 that ubiquinone synthesis by the mevalonate pathway supports pyrimidine 6 synthesis, particularly when efficient electron transport is hampered by low 7 oxygen availability. Ubiquinone deprivation also induces oxidative stress, 8 especially when demand for biosynthetic reactions that deliver electrons to the 9 ETC is high. We found that statin-induced cell death was prevented by 10 antioxidants or by the addition of nucleosides or uridine, which allow cells to 11 switch to the salvage pathway, suggesting that reducing de novo pyrimidine 12 synthesis prevents ROS formation and cell death. Furthermore, the anti- 13 metabolite 5-FU, which blocks dTMP synthesis and induces DNA and RNA 14 damage, sensitized p53 wt SPC to statin treatment. 5-FU may impose 15 additional strain on pyrimidine biosynthesis and/or increase oxidative stress, 16 both of which would enhance the dependency of cancer cells on ubiquinone. 17 While clinical trials combining statins with 5-FU in CRC have produced some 18 promising results (50), our study suggests that p53 status could determine the 19 outcome of mevalonate pathway inhibition in CRC. 20 The dependence of p53-deficient cancer cells on mevalonate pathway 21 activity was also confirmed in apc-/- intestinal tumor organoids, where deletion 22 of p53, either alone or in combination with Kras activation, induced sensitivity 23 towards statin treatment. Addition of ubiquinone or nucleosides restored growth 24 of p53-deficient tumor organoids in the presence of statins, suggesting that 25 cells require mevalonate pathway-derived ubiquinone to counteract oxidative 26 stress and support biosynthetic reactions. Indeed, LGR5+ intestinal stem cells 27 are enriched for gene expression signatures linked to purine and pyrimidine 28 metabolism (51) and are highly dependent on mitochondrial metabolism (52). 29 We also found that statins block Kras-dependent hyperproliferation in 30 Apc-deficient intestinal crypts. However, in contrast to our findings in SPC and 31 organoids, this was associated with reduced cholesterol rather than ubiquinone 32 synthesis. Cholesterol is required for membrane synthesis (32) and could be 33 the major metabolic output of the mevalonate pathway in rapidly proliferating 34 tissues. Moreover, cells within the intestinal mucosa may not be exposed to

1 metabolic deprivation, as they have access to the nutrient-rich contents of the 2 intestinal lumen, including nucleosides generated by the degradation of diet- 3 derived nucleic acids. 4 5 Together, our findings reveal a novel function of the mevalonate pathway 6 in supporting the synthesis of ubiquinone for electron transfer and pyrimidine 7 biosynthesis in p53-deficient cancer cells exposed to environmental stress. 8 However, our results also show that the dependence on mevalonate pathway- 9 derived metabolites is determined by environmental context. Mevalonate 10 pathway inhibition may therefore be most effective under conditions of nutrient 11 and oxygen deprivation. Beneficial effects of mevalonate pathway inhibitors 12 have already been demonstrated in several cancer entities, including CRC 13 (53,54). The results of this study indicate that mevalonate pathway inhibitors 14 may need to be combined with treatments that induce metabolic stress, such 15 as anti-angiogenic therapy. 16 17 Acknowledgements 18 We thank B.Vogelstein (Johns Hopkins University, Baltimore), K.Vousden (The 19 Francis Crick Institute, London) and M.Dobbelstein (University Göttingen) for 20 providing p53 and p21 isogenic colon cancer cell lines, respectively. We thank 21 C.Schülein-Völk and U.Eilers for help with automated cell counting. We also 22 thank S.Janaki Raman and M.T.Snaebjörnsson for critically reading the 23 manuscript. This study was funded by grants from the German Research 24 Foundation FOR2314 and SCHU2670-1 (A.Schulze), the Graduate School of 25 Life Sciences Würzburg (I.Kaymak), GRK2243 (C.R.Maier), the Rosetrees 26 Trust (G.McGregor), the CRUK Grand Challenge (M.Paauwe and D.M.Gay) 27 and core funding to the Beatson Institute from Cancer Research UK (A17196) 28 (O.J.Sansom). 29 30 31 References: 32 1. Denko NC. Hypoxia, HIF1 and glucose metabolism in the solid tumour. 33 Nat Rev Cancer 2008;8:705-13 34 2. Floter J, Kaymak I, Schulze A. Regulation of Metabolic Activity by p53. 35 Metabolites 2017;7

1 3. Kruiswijk F, Labuschagne CF, Vousden KH. p53 in survival, death and 2 metabolic health: a lifeguard with a licence to kill. Nat Rev Mol Cell Biol 3 2015;16:393-405 4 4. Kawauchi K, Araki K, Tobiume K, Tanaka N. p53 regulates glucose 5 metabolism through an IKK-NF-kappaB pathway and inhibits cell 6 transformation. Nat Cell Biol 2008;10:611-8 7 5. Hitosugi T, Zhou L, Elf S, Fan J, Kang HB, Seo JH, et al. 8 Phosphoglycerate mutase 1 coordinates glycolysis and biosynthesis to 9 promote tumor growth. Cancer Cell 2012;22:585-600 10 6. Bensaad K, Tsuruta A, Selak MA, Vidal MN, Nakano K, Bartrons R, et 11 al. TIGAR, a p53-inducible regulator of glycolysis and apoptosis. Cell 12 2006;126:107-20 13 7. Jiang P, Du W, Wang X, Mancuso A, Gao X, Wu M, et al. p53 regulates 14 biosynthesis through direct inactivation of glucose-6-phosphate 15 dehydrogenase. Nat Cell Biol 2011;13:310-6 16 8. Ros S, Floter J, Kaymak I, Da Costa C, Houddane A, Dubuis S, et al. 6- 17 Phosphofructo-2-kinase/fructose-2,6-biphosphatase 4 is essential for 18 p53-null cancer cells. Oncogene 2017;36:3287-99 19 9. Matoba S, Kang JG, Patino WD, Wragg A, Boehm M, Gavrilova O, et al. 20 p53 Regulates Mitochondrial Respiration. Science 2006 21 10. Maddocks OD, Berkers CR, Mason SM, Zheng L, Blyth K, Gottlieb E, et 22 al. Serine starvation induces stress and p53-dependent metabolic 23 remodelling in cancer cells. Nature 2013;493:542-6 24 11. Tajan M, Hock AK, Blagih J, Robertson NA, Labuschagne CF, Kruiswijk 25 F, et al. A Role for p53 in the Adaptation to Glutamine Starvation through 26 the Expression of SLC1A3. Cell Metab 2018 27 12. Wang X, Sato R, Brown MS, Hua X, Goldstein JL. SREBP-1, a 28 membrane-bound transcription factor released by sterol-regulated 29 proteolysis. Cell 1994;77:53-62 30 13. Sato T, Stange DE, Ferrante M, Vries RG, Van Es JH, Van den Brink S, 31 et al. Long-term expansion of epithelial organoids from human colon, 32 adenoma, adenocarcinoma, and Barrett's epithelium. Gastroenterology 33 2011;141:1762-72 34 14. Cammareri P, Vincent DF, Hodder MC, Ridgway RA, Murgia C, Nobis 35 M, et al. TGFbeta pathway limits dedifferentiation following WNT and 36 MAPK pathway activation to suppress intestinal tumourigenesis. Cell 37 Death Differ 2017;24:1681-93 38 15. Bunz F, Dutriaux A, Lengauer C, Waldman T, Zhou S, Brown JP, et al. 39 Requirement for p53 and p21 to sustain G2 arrest after DNA damage. 40 Science 1998;282:1497-501 41 16. Sutherland R, Freyer J, Mueller-Klieser W, Wilson R, Heacock C, 42 Sciandra J, et al. Cellular growth and metabolic adaptations to nutrient 43 stress environments in tumor microregions. Int J Radiat Oncol Biol Phys 44 1986;12:611-5 45 17. Sullivan LB, Gui DY, Hosios AM, Bush LN, Freinkman E, Vander Heiden 46 MG. Supporting Aspartate Biosynthesis Is an Essential Function of 47 Respiration in Proliferating Cells. Cell 2015;162:552-63 48 18. Horton JD, Goldstein JL, Brown MS. SREBPs: activators of the complete 49 program of cholesterol and fatty acid synthesis in the liver. The Journal 50 of clinical investigation 2002;109:1125-31

1 19. Horton JD, Shah NA, Warrington JA, Anderson NN, Park SW, Brown 2 MS, et al. Combined analysis of oligonucleotide microarray data from 3 transgenic and knockout mice identifies direct SREBP target genes. 4 Proceedings of the National Academy of Sciences of the United States 5 of America 2003;100:12027-32 6 20. Cancer Genome Atlas N. Comprehensive molecular characterization of 7 human colon and rectal cancer. Nature 2012;487:330-7 8 21. Moon SH, Huang CH, Houlihan SL, Regunath K, Freed-Pastor WA, 9 Morris JPt, et al. p53 Represses the Mevalonate Pathway to Mediate 10 Tumor Suppression. Cell 2018 11 22. Gerin I, Clerbaux LA, Haumont O, Lanthier N, Das AK, Burant CF, et al. 12 Expression of miR-33 from an SREBP2 intron inhibits cholesterol export 13 and fatty acid oxidation. J Biol Chem 2010;285:33652-61 14 23. Najafi-Shoushtari SH, Kristo F, Li Y, Shioda T, Cohen DE, Gerszten RE, 15 et al. MicroRNA-33 and the SREBP host genes cooperate to control 16 cholesterol homeostasis. Science 2010;328:1566-9 17 24. Porstmann T, Santos CR, Griffiths B, Cully M, Wu M, Leevers S, et al. 18 SREBP activity is regulated by mTORC1 and contributes to Akt- 19 dependent cell growth. Cell Metab 2008;8:224-36 20 25. Porstmann T, Griffiths B, Chung YL, Delpuech O, Griffiths JR, 21 Downward J, et al. PKB/Akt induces transcription of enzymes involved 22 in cholesterol and fatty acid biosynthesis via activation of SREBP. 23 Oncogene 2005;24:6465-81 24 26. Arsham AM, Howell JJ, Simon MC. A novel hypoxia-inducible factor- 25 independent hypoxic response regulating mammalian target of 26 rapamycin and its targets. J Biol Chem 2003;278:29655-60 27 27. Liu L, Cash TP, Jones RG, Keith B, Thompson CB, Simon MC. Hypoxia- 28 induced energy stress regulates mRNA translation and cell growth. Mol 29 Cell 2006;21:521-31 30 28. Sundqvist A, Bengoechea-Alonso MT, Ye X, Lukiyanchuk V, Jin J, 31 Harper JW, et al. Control of lipid metabolism by phosphorylation- 32 dependent degradation of the SREBP family of transcription factors by 33 SCF(Fbw7). Cell Metab 2005;1:379-91 34 29. DeBose-Boyd RA, Ye J. SREBPs in Lipid Metabolism, Insulin Signaling, 35 and Beyond. Trends Biochem Sci 2018;43:358-68 36 30. Shamma A, Takegami Y, Miki T, Kitajima S, Noda M, Obara T, et al. Rb 37 Regulates DNA damage response and cellular senescence through 38 E2F-dependent suppression of N-ras isoprenylation. Cancer Cell 39 2009;15:255-69 40 31. Freed-Pastor WA, Mizuno H, Zhao X, Langerod A, Moon SH, Rodriguez- 41 Barrueco R, et al. Mutant p53 Disrupts Mammary Tissue Architecture via 42 the Mevalonate Pathway. Cell 2012;148:244-58 43 32. Mullen PJ, Yu R, Longo J, Archer MC, Penn LZ. The interplay between 44 cell signalling and the mevalonate pathway in cancer. Nat Rev Cancer 45 2016;16:718-31 46 33. Dallner G, Sindelar PJ. Regulation of ubiquinone metabolism. Free 47 Radic Biol Med 2000;29:285-94 48 34. Kamphorst JJ, Chung MK, Fan J, Rabinowitz JD. Quantitative analysis 49 of acetyl-CoA production in hypoxic cancer cells reveals substantial 50 contribution from acetate. Cancer Metab 2014;2:23

1 35. Brunelle JK, Bell EL, Quesada NM, Vercauteren K, Tiranti V, Zeviani M, 2 et al. Oxygen sensing requires mitochondrial ROS but not oxidative 3 phosphorylation. Cell Metab 2005;1:409-14 4 36. Longley DB, Harkin DP, Johnston PG. 5-fluorouracil: mechanisms of 5 action and clinical strategies. Nat Rev Cancer 2003;3:330-8 6 37. el Marjou F, Janssen KP, Chang BH, Li M, Hindie V, Chan L, et al. 7 Tissue-specific and inducible Cre-mediated recombination in the gut 8 epithelium. Genesis 2004;39:186-93 9 38. McGregor G, Campbell A, Fey S, Tumanov S, Sumpton D, Rodriguez 10 Blanco G, et al. Targeting the metabolic response to statin-mediated 11 oxidative stress produces a synergistic anti-tumor response. Cancer Res 12 in press 13 39. Bajzikova M, Kovarova J, Coelho AR, Boukalova S, Oh S, Rohlenova K, 14 et al. Reactivation of Dihydroorotate Dehydrogenase-Driven Pyrimidine 15 Biosynthesis Restores Tumor Growth of Respiration-Deficient Cancer 16 Cells. Cell Metab 2018 17 40. Eales KL, Hollinshead KE, Tennant DA. Hypoxia and metabolic 18 adaptation of cancer cells. Oncogenesis 2016;5:e190 19 41. Pisarsky L, Bill R, Fagiani E, Dimeloe S, Goosen RW, Hagmann J, et al. 20 Targeting Metabolic Symbiosis to Overcome Resistance to Anti- 21 angiogenic Therapy. Cell reports 2016;15:1161-74 22 42. Allen E, Mieville P, Warren CM, Saghafinia S, Li L, Peng MW, et al. 23 Metabolic Symbiosis Enables Adaptive Resistance to Anti-angiogenic 24 Therapy that Is Dependent on mTOR Signaling. Cell reports 25 2016;15:1144-60 26 43. Cheng T, Sudderth J, Yang C, Mullen AR, Jin ES, Mates JM, et al. 27 Pyruvate carboxylase is required for glutamine-independent growth of 28 tumor cells. Proceedings of the National Academy of Sciences of the 29 United States of America 2011;108:8674-9 30 44. Cardaci S, Zheng L, MacKay G, van den Broek NJ, MacKenzie ED, 31 Nixon C, et al. Pyruvate carboxylation enables growth of SDH-deficient 32 cells by supporting aspartate biosynthesis. Nat Cell Biol 2015;17:1317- 33 26 34 45. Li T, Kon N, Jiang L, Tan M, Ludwig T, Zhao Y, et al. Tumor Suppression 35 in the Absence of p53-Mediated Cell-Cycle Arrest, Apoptosis, and 36 Senescence. Cell 2012;149:1269-83 37 46. Wang Y, Hekimi S. Understanding Ubiquinone. Trends Cell Biol 38 2016;26:367-78 39 47. Birsoy K, Possemato R, Lorbeer FK, Bayraktar EC, Thiru P, Yucel B, et 40 al. Metabolic determinants of cancer cell sensitivity to glucose limitation 41 and biguanides. Nature 2014 42 48. Zhang B, Wang J, Wang X, Zhu J, Liu Q, Shi Z, et al. Proteogenomic 43 characterization of human colon and rectal cancer. Nature 44 2014;513:382-7 45 49. Gui DY, Sullivan LB, Luengo A, Hosios AM, Bush LN, Gitego N, et al. 46 Environment Dictates Dependence on Mitochondrial Complex I for 47 NAD+ and Aspartate Production and Determines Cancer Cell Sensitivity 48 to Metformin. Cell Metab 2016;24:716-27 49 50. Lee J, Jung KH, Park YS, Ahn JB, Shin SJ, Im SA, et al. Simvastatin 50 plus irinotecan, 5-fluorouracil, and leucovorin (FOLFIRI) as first-line

1 chemotherapy in metastatic colorectal patients: a multicenter phase II 2 study. Cancer Chemother Pharmacol 2009;64:657-63 3 51. Lindeboom RG, van Voorthuijsen L, Oost KC, Rodriguez-Colman MJ, 4 Luna-Velez MV, Furlan C, et al. Integrative multi-omics analysis of 5 intestinal organoid differentiation. Mol Syst Biol 2018;14:e8227 6 52. Rodriguez-Colman MJ, Schewe M, Meerlo M, Stigter E, Gerrits J, Pras- 7 Raves M, et al. Interplay between metabolic identities in the intestinal 8 crypt supports stem cell function. Nature 2017;543:424-7 9 53. Liu JC, Hao WR, Hsu YP, Sung LC, Kao PF, Lin CF, et al. Statins dose- 10 dependently exert a significant chemopreventive effect on colon cancer 11 in patients with chronic obstructive pulmonary disease: A population- 12 based cohort study. Oncotarget 2016;7:65270-83 13 54. Sehdev A, Shih YC, Huo D, Vekhter B, Lyttle C, Polite B. The role of 14 statins for primary prevention in non-elderly colorectal cancer patients. 15 Anticancer Res 2014;34:5043-50 16 17

1 Figure Legends: 2 Figure 1: Spheroid cultures replicate tumor-like transcriptional profiles 3 and show pyruvate-dependent anaplerosis 4 A) HCT116 p53+/+ and p53-/- cells were cultured as subconfluent monolayer 5 cultures (MLC) for 48 hrs or as multi-layered tumor spheroid cultures (SPC) for 6 14 days. Cells were incubated with BrdU for 24 hrs and analyzed by FACS. 7 B) HCT116 p53+/+ and p53-/- cells were cultured as SPC for 14 days, fixed and 8 embedded in paraffin. Histological sections were analyzed for expression of the 9 proliferation marker Ki67. Representative images of three spheroids analyzed 10 per condition are shown. 11 C) Analysis of proliferation in HCT116 p53+/+ and p53-/- xenograft tumor tissue 12 from (8) using Ki67. Representative images of tumors from six animals per 13 group are shown. 14 D) HCT116 p53+/+ and p53-/- cells were cultured as MLC for 48 hrs or as SPC 15 for 14 days. RNA was analyzed by RNA-SEQ and compared to RNA-SEQ data 16 from HCT116 p53+/+ and p53-/- cells grown as xenograft tumors (8). Principal 17 component analysis (PCA) shows overall higher similarity in gene expression 18 signatures between tumors (T) and SPC compared to MLC. 19 E) Gene set enrichment analysis (GSEA) comparing MLC and SPC cultures of 20 HCT116 p53+/+ and p53-/- cells. Enrichment plots for 21 HALLMARK_E2F_TARGETS, 22 BROWNE_INTERFERON_RESPONSE_GENES and 23 MANALO_HYPOXIA_UP are shown. 24 F) Enrichment plots for the same gene sets as in E comparing MLC and tumors 25 (T) of HCT116 p53+/+ and p53-/- cells. 26 G) Western blots (WB) showing levels of cyclin D1 (CCND1) and pyruvate 27 dehydrogenase kinase 1 (PDK1) in HCT116 p53+/+ or p53-/- cells grown as MLC 28 or SPC. Vinculin is shown as loading control. 29 H-J) HCT116 p53+/+ and p53-/- cells were cultured as MLC or SPC and labelled 30 for 16 hours with [U13C]-glucose. Cells were extracted and metabolites were 31 analyzed by LC-MS. Data show mean ±SEM of three independent biological 32 replicates. Results from time-resolved experiments are provided in Figure S1. 33 H) Relative peak intensities for lactate.

1 I) Ratios of M+3 and M+2 isotopologues for aspartate. 2 J) Relative peak intensities of aspartate. 3 4 Figure 2: Loss of p53 induces enzymes of the mevalonate pathway via 5 activation of SREBP2 6 A) Enrichment plots for HORTON_SREBF_TARGETS (19) for HCT116 p53+/+ 7 and p53-/- cells cultured as MLC, SPC or xenograft colon tumors. 8 B) Expression of canonical mevalonate pathway genes and SREBF2 in 9 HCT116 p53+/+ or p53-/- cells grown as MLC or SPC. Data show mean ±SEM 10 of three independent biological replicates. (*p<0.05; **p<0.01; ***p<0.001; 11 ****p<0.0001, unpaired two-tailed Student’s t test). 12 C) Western blots showing levels of HMGCS1, FDFT1, ACSS2 and p53 in RKO 13 p53+/+ and RKO p53-/- cells grown as SPC. Vinculin is shown as loading control. 14 D) Combined z-score for the expression of canonical mevalonate pathway 15 genes (HMGCS1, HMGCR, MVD, DHCR7, ACSS2, FDFT1 and SREBF2) was 16 calculated for tumors from the TCGA colorectal adenocarcinoma dataset. 17 Mevalonate pathway signature values were compared between all p53 wt (n = 18 94) and p53 mutant (n = 88) tumors. p = 0.0051 was determined using an 19 unpaired two-tailed Student’s t test. 20 E) Expression of HMGCS1 in colorectal adenocarcinoma tumors from the 21 TCGA dataset according to stage (PT1-4). 22 F) Expression of HMGCS1 in TPP53 wt and mutant colon cancer cell lines 23 grown as MLC or SPC. 24 G) WB showing expression of HMGCS1 and mature SREBP2 in HCT116 p53+/+ 25 or p53-/- cells grown as MLC or SPC. Actin is shown as loading control. 26 H) WB showing levels of phosphorylated ribosomal protein S6 (P-S6RB) and 27 total ribosomal protein S6 expression (S6RB) in HCT116 p53+/+ or p53-/- cells 28 grown as MLC or SPC. Actin is shown as loading control. 29 I) SPC of HCT116 p53+/+ or p53-/- cells were treated with 20µM MG132 or 30 solvent for 1h. Levels of mature SREBP2 were detected by WB. Vinculin is 31 shown as loading control. 32 J) Western blots showing phosphorylation on GSK3a/b (serine 21/9) and total 33 GSK3a/b protein in HCT116 p53+/+ or p53-/- cells grown as MLC or SPC.

1 Numbers display P-GSK3/GSK3 signal ratio. Actin is shown as loading control. 2 Graph shows mean ±SEM ratios of biologically independent replicate SPC 3 samples (n=4, *p≤0.05 was determined using a paired two-tailed Student’s t 4 test). 5 K) HCT116 p53+/+ or p53-/- cells grown as SPC were treated with 20nM of 6 rapamycin (RAPA), 30 µM of SB216763 (SB) or 10 µM of CHIR99021 (CHIR) 7 for 24 hours and mature SREBP2 was analyzed by WB. Vinculin is shown as 8 loading control. Numbers show signal intensity for mSREBP normalized to 9 vinculin. 10 L) Expression of HMGCS1 mRNA in HCT116 p53+/+ or p53-/- cells grown as 11 spheroids and treated with 10 µM of CHIR99021 (CHIR) for 72 hours. Data 12 show mean ±SEM of three independent biological replicates. (*p<0.05, 13 unpaired two-tailed Student’s t test). 14 M) Effect of rapamycin on HMGCS1 expression in spheroid cultures of HCT116 15 p53+/+ or p53-/- cells. Data are presented as mean of duplicate samples. 16 17 Figure 3: The mevalonate pathway is essential for the survival of p53- 18 deficient colon cancer cells 19 A) Diagram showing selected metabolites of the mevalonate pathway and 20 HMGCR, the molecular target of statins. 21 B) HCT116 p53+/+ or p53-/- cells were grown as SPC and treated with 10 µM 22 mevastatin (MST) or solvent (DMSO) either alone or in combination with 0.5 23 mM mevalonate (MVL) for 72 hours. Spheroids were fixed and histological 24 sections were analyzed for the presence of apoptotic cells by TUNEL staining. 25 Images show representative results of three spheroids analyzed per condition. 26 C) Quantitation of data shown in B. Data are presented as mean ±SEM of at 27 least 3 spheroids analyzed per condition. (**p<0.01, unpaired two-tailed 28 Student’s t test). 29 D) HCT116 p53+/+ and p53-/- cells grown as SPC were treated with 10µM 30 simvastatin (SIM) or solvent (DMSO) for 24 or 72 hours. Expression of p21 31 (CDKN1A) mRNA was determined by qPCR. Data show mean ±SEM of three 32 independent biological replicates. (*p<0.05; ***p<0.001, unpaired two-tailed 33 Student’s t test).

1 E) HCT116 p53+/+ and p53-/- cells were grown as SPC and treated with 10µM 2 simvastatin (SIM) or solvent (DMSO) either alone or in combination with 0.5 3 mM mevalonate for 72 hours. Expression of p21 protein was determined by 4 western blotting. Vinculin is shown as loading control. 5 F) HCT116 p21-/- cells were grown as SPC and treated with 10 µM simvastatin 6 (SIM) or solvent (DMSO) either alone or in combination with 0.5 mM 7 mevalonate (MVL) for 72 hours. Spheroids were fixed and histological sections 8 were analyzed for the presence of apoptotic cells by TUNEL staining. Images 9 show representative results of three spheroids analyzed per condition. 10 G) HCT116 p21+/+ or p21-/- cells were grown as SPC and treated with 10 µM 11 simvastatin (SIM) or solvent (DMSO) either alone or in combination with 0.5mM 12 mevalonate (MVL) for 72 hours. Expression of p21 protein was determined by 13 western blotting. Vinculin is shown as loading control. 14 15 Figure 4: Inhibition of mevalonate synthesis blocks the production of 16 ubiquinone in colon cancer cells 17 A) Schematic showing the branching of the mevalonate pathway into 18 cholesterol biosynthesis, the generation of isoprenoids for protein prenylation 19 and the synthesis of dolichol, heme A and ubiquinone (CoQ10). 20 B-C) HCT116 p53+/+ or p53-/- cells were grown as SPC and labelled with [U13C]- 21 glucose for 16 hrs before extraction and analysis of mevalonate isotopologues. 22 Data show mean ±SEM of three independent biological replicates. 23 B) Relative peak intensities of labelled and unlabelled fractions for mevalonate. 24 C) Relative peak intensities of individual labelled fractions for mevalonate. 25 D-G) HCT116 p53+/+ or p53-/- cells were grown as SPC and treated with 10 µM 26 simvastatin (SIM) or solvent (DMSO) for 72 hrs. For the last 16 hours, cells 27 were labelled with [U13C]-glucose before cells were extracted and metabolites 28 were analyzed by LC-MS. Data show mean ±SEM of three independent 29 biological replicates. 30 D) Relative peak intensities of labelled and unlabelled fractions for cholesterol. 31 E) Relative peak intensities of individual isotopologues for cholesterol. 32 F) Relative peak intensities of labelled and unlabelled fractions for ubiquinone 33 (CoQ10). 34 G) Relative peak intensities of individual isotopologues for ubiquinone.

1 H) Xenograft tumors from HCT116 p53+/+ or p53-/- cells were extracted and 2 levels of cholesterol, 7-dihydroxycholesterol (7-DHC) and ubiquinone (CoQ10) 3 were determined by LC-MS. Data are shown as mean ±SEM of six p53+/+ and 4 five p53-/- colon tumors. (*p<0.05, unpaired two-tailed Student’s t test). 5 6 Figure 5: Inhibition of mevalonate synthesis blocks TCA cycle activity and 7 cellular respiration and induces oxidative stress 8 A) Diagram showing the role of ubiquinone (Q10) in electron transport within 9 the mitochondrial electron transport chain (ETC). 10 B) HCT116 p53+/+ or p53-/- cells were grown as SPC and treated with 10 µM 11 simvastatin (SIM) or solvent (DMSO) for 72 hrs. For the last 16 hours, cells 12 were labelled with [U13C]-glucose before cells were extracted and metabolite 13 levels were analyzed by LC-MS. Relative peak intensities of isotopologues for 14 aspartate are shown. Data show mean ±SEM of three independent biological 15 replicates. 16 C) HCT116 p53+/+ or p53-/- cells were grown as SPC and treated with 10 µM 17 simvastatin (SIM) or solvent (DMSO) either alone or in the presence of 0.5 mM 18 mevalonate for 72 hrs. Oxygen consumption rates (OCR) were determined 19 using the Seahorse Bioanalyzer. Oligomycin (oligo), FCCP and 20 rotenone/antimycin A (R/A) were added to determine ATP-dependent, maximal 21 and basal respiration. Data are presented as mean ± SEM of 12 spheroids 22 analyzed per condition. 23 D) HCT116 p53+/+ or p53-/- cells were grown as SPC and treated with 10 µM 24 simvastatin (SIM) or solvent (DMSO) either alone or in combination with 0.5 25 mM mevalonate (MVL), 10 µM ubiquinone (Q10) or 5 mM N-acetylcysteine 26 (NAC) for 72 hours. Spheroids were fixed and histological sections were 27 analyzed for the presence of apoptotic cells by TUNEL staining. Images show 28 representative results of three spheroids analyzed per condition. 29 E) Quantitation of data shown in D. Data are presented as mean ± SEM of at 30 least 3 spheroids analyzed per condition. (**p<0.01, unpaired two-tailed 31 Student’s t test). 32 33 Figure 6: Mevalonate pathway activity is essential for pyrimidine 34 nucleotide biosynthesis and survival in colon cancer cells

1 A) Diagram showing the role of ubiquinone (Q10) in the conversion of 2 dihydroorotate to orotate during pyrimidine biosynthesis. 3 B) HCT116 p53+/+ or p53-/- cells were grown as SPC and treated with 10 µM 4 simvastatin (SIM) or solvent (DMSO) for 72 hours. For the last 16 hours, cells 5 were labelled with [U13C]-glucose before metabolites were extracted and 6 analyzed by LC-MS. Relative peak intensities (left graph) and total labelled 7 fractions (right graph) for UMP and are shown. Data show mean ±SEM of three 8 independent biological replicates. (*p<0.05; **p<0.01, unpaired two-tailed 9 Student’s t test) 10 C) HCT116 p53+/+ or p53-/- cells were grown as SPC and treated with 10 µM 11 simvastatin (SIM) or solvent (DMSO) for 72 hours either alone or in combination 12 with 0.5 mM mevalonate (MVL) or 10 µM ubiquinone (Q10). For the last 16 13 hours, cells were labelled with [U13C]-glucose before metabolites were 14 extracted and analyzed by LC-MS. Relative peak intensities (left graph) and 15 total labelled fractions (right graph) for UMP and are shown. Data show mean 16 ±SEM of three independent biological replicates. (*p<0.05; **p<0.01, unpaired 17 two-tailed Student’s t test). 18 D) HCT116 p53+/+ or p53-/- cells were grown as SPC and treated with solvent 19 (DMSO) or 10 µM simvastatin (SIM) either alone or in combination with 20 nucleosides (150 µM each of cytidine, guanosine, adenosine, uridine and 50µM 21 of thymidine = NCL) for 72 hours. Spheroids were fixed and histological 22 sections were analyzed for the presence of apoptotic cells by TUNEL staining. 23 Data are presented as mean ±SEM of at least 3 spheroids analyzed per 24 condition. (*p<0.05, ****p<0.0001, unpaired two-tailed Student’s t test). 25 E) HCT116 p53+/+ or p53-/- cells were grown as SPCs and treated with solvent 26 (DMSO), 10 µM simvastatin (SIM), 10 µM 5-fluorouracil (5-FU) or a combination 27 of the two for 72 hours. Spheroids were analyzed by TUNEL staining. Data are 28 presented as mean ±SEM of at least 3 spheroids analyzed per condition. 29 (*p<0.05, ***p<0.001, unpaired two-tailed Student’s t test). 30 31 Figure 7: Simvastatin reduces growth of p53-deficient tumor organoids 32 and blocks proliferation in Apc/p53-deficient Kras-transformed intestinal 33 crypts

1 A) Primary mouse intestinal cells derived from VillinCREERT2Apcfl/fl, 2 VillinCREERT2Apcfl/fl;p53fl/fl, VillinCREERT2Apcfl/fl;KrasG12D or 3 VillinCREERT2Apcfl/fl;p53fl/fl;KrasG12D animals were used to generate organoid 4 cultures. Organoids were treated with 10 µM simvastatin (SIM) either alone or 5 in combinations with 0.5 mM mevalonate (MEV) for 48 hrs. Images show 6 representative microscopic fields from three independent replicate cultures. 7 B) Quantitation of data shown in (A). Data are presented as mean ±SEM of 8 microscopic fields from three independent cultures. (*p<0.05; **p<0.01, 9 ****p<0.0001, unpaired two-tailed Student’s t test). 10 C) Apcfl/fl;p53fl/fl or Apcfl/fl;p53fl/fl;KrasG12D organoids were treated with 10 µM 11 simvastatin either alone or in combination with 10 µM ubiquinone (Q10) or 12 nucleosides (150 µM each of cytidine, guanosine, adenosine, uridine and 50µM 13 of thymidine) for 48 hrs. Images show representative microscopic fields from 14 three independent replicate cultures. 15 D) Quantitation of data shown in (C). Data are presented as mean ±SEM of 16 microscopic fields from three independent cultures. (****p<0.0001, unpaired 17 two-tailed Student’s t test). 18 E) VillinCREERT2;Apcfl/fl and VillinCREERT2;apcfl/fl;KrasG12D/+ mice were treated 19 with a single intraperitoneal injection of 80 mg/kg of tamoxifen on one occasion 20 (VillinCreERT2Apcfl/fl KrasG12D/+), or on two consecutive days 21 (VillinCreERT2Apcfl/fl). From one day post-induction, mice were treated with a 22 daily dose of 50 mg/kg simvastatin (in 0.5% methyl cellulose/ 5% DMSO). After 23 four days, mice were sacrificed and intestinal mucosa was fixed, paraffin 24 embedded and histological sections were stained for BrdU incorporation. 25 Representative images are shown. 26 F) Three intestinal crypts for each genotype and treatment were scored for 27 BrdU positive cells. 28 G) Fraction of cholesterol and ubiquinone (CoQ9) containing deuterated water 29 in intestinal mucosa from the different genotypes 30 F+G (*p<0.05 unpaired two-tailed Student’s t test).

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 2019 American Association for Cancer Research. Author Manuscript Published OnlineFirst on November 19, 2019; DOI: 10.1158/0008-5472.CAN-19-0650Kaymak et al. Figure 1 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

A B spheroids D 100 p53+/+ p53-/- Ki67 Ki67 p53NULL MLC tumors p53WT p53NULL p53WT SPC p53NULL 50 0.2 p53WT T

150μM 150μM

150μm PC2 (18%) 0.0

%BrdU incorporation %BrdU C monolayer xenograft tumors spheroids 0 p53+/+ p53-/- Ki67 Ki67 -0.2

p53+/+ p53-/- p53+/+ p53-/- MLC SPC PC1 (80%)

50μM 50μM

E HALLMARK_E2F_TARGETS F HALLMARK_E2F_TARGETS 0.8 0.8 0.8 0.7 0.7 0.7 0.7 0.6 0.6 0.6 0.6 0.5 0.5 0.5 0.5 0.4 0.4 0.4 0.4 0.3 0.3 0.3 NES 5.67 0.3 NES 5.40 NES 3.61 NES 3.0 0.2 0.2 0.2 0.2 q≤0.05 0.1 q≤0.05 0.1 q≤0.05 0.1 q≤0.05 0.1 Enrichment score (ES) Enrichment score (ES) Enrichment score (ES) 0.0 0.0 0.0 0.0 Enrichment score (ES)

p53+/+ p53-/- MLC p53+/+ SPC MLC p53-/- SPC MLC T MLC T BROWNE_INTERFERON_RESP_GENES BROWNE_INTERFERON_RESP_GENES 0.0 0.0 0.0 0.0 -0.1 -0.1 -0.1 -0.1 -0.2 -0.2 -0.3 -0.2 -0.2 -0.3 -0.4 -0.4 -0.3 -0.3 -0.5 -0.4 -0.4 -0.6 NES -3.22 -0.5 NES -2.28 NES -2.24 -0.7 -0.6 NES -3.47 -0.5 -0.5 -0.7 -0.6 Enrichment score (ES) -0.8 -0.6 q≤0.05

q≤0.05 q≤0.05 Enrichment score (ES) Enrichment score (ES) q≤0.05 -0.8 Enrichment score (ES)

MLC p53+/+ SPC MLC p53-/- SPC MLC p53+/+ T MLC p53-/- T MANALO_HYPOXIA_UP MANALO_HYPOXIA_UP 0.0 0.0 0.0 0.0 -0.1 -0.1 -0.1 -0.1 -0.2 -0.2 -0.3 -0.3 -0.2 -0.2 -0.4 NES -3.12 -0.4 NES -3.32 -0.3 -0.3 -0.5 -0.5 NES -1.80 -0.4 NES -1.93 -0.6 q≤0.05 -0.4

Enrichment score (ES) q≤0.05

Enrichment score (ES) -0.6 Enrichment score (ES) q≤0.05 Enrichment score (ES) q≤0.05 -0.7 -0.5 -0.5

MLC p53+/+ SPC MLC p53-/- SPC MLC p53+/+ T MLC p53-/- T

G MLC SPC H Lactate I Aspartate J Aspartate 800 2.0 p53+/+ p53-/- p53+/+ p53-/- 250 M+4 600 200 124 kDa VCL M+3 1.5 M+3 M+2 M+2 150 400 M+1 1.0 M+1 49 kDa PDK1 M+0 100 M+0 M+3/M+2 200 0.5 50 peak intensity/prot 34 kDa CCND1 peak intensity/prot 0 0.0 0 + +/+ +/+ 53-/-53 p53-/-53 p53-/- p53+/+p p p53-/- p53+/ p53-/-p53+/+p53-/- p53+/+ p MLC SPC MLC SPC SPCMLC

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 2019 American Association for Cancer Research. Author Manuscript Published OnlineFirst on November 19, 2019; DOI: 10.1158/0008-5472.CAN-19-0650Kaymak et al. Figure 2 A Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Monolayer Spheroids Tumors HORTON_SREBF_TARGETS HORTON_SREBF_TARGETS HORTON_SREBF_TARGETS

0.2 0.0 0.0 -0.1 -0.1 0.1 -0.2 -0.2 -0.3 0.0 -0.3 -0.4 -0.1 NES 0.67 -0.5 NES -2.64 -0.4 NES -1.8 -0.5 -0.2 q=0.991 -0.6 q≤0.05 q=0.057 -0.7 -0.6 Enrichment score (ES) Enrichment score (ES) Enrichment score (ES) p53+/+ p53-/- p53+/+ p53-/- p53+/+ p53-/- B HMGCS1 MVD HMGCR DHCR7 FDFT1 SREBF2 10 2.5 6 5 * 4 p53+/+ ** 6 * p53-/- ** n.s. 8 2.0 4 * **** *** ** 3 4 4 6 ** 1.5 3 * 2 2 4 1.0 **** ** 2 *** 2 ** 1 2 0.5 1 Relative mRNA expression Relative mRNA expression Relative mRNA expression Relative mRNA expression Relative Relative mRNA expression 0 Relative mRNA0.0 expression 0 0 0 0 MLC SPC MLC SPC MLC SPC MLC SPC MLC SPC MLC SPC

C D E F HMGCS1 RKO SPC ** HMGCS1 10 MLC ** SPC

10 10000 p53+/+ p53-/- 8 124 kDa VCL 6 79 kDa ACSS2 5 4 57 kDa HMGCS1 2 53 kDa p53 z-score RESM / Relative mRNA expression Relative 0

0 2000 4000 6000 8000 )

48 kDa FDFT1 relative expression (RPKM) WT) 0

pT1 pT2 pT3 pT4 LOVO (WT) wt LS174T ( HT29 (R273HDLD1 (S241F) mut Tumor stage TP53 status G H I MLC SPC SPC MLC SPC p53+/+ p53-/- p53+/+ p53-/- p53+/+ p53-/- p53+/+ p53-/- p53+/+ p53-/- 60-78 mSREBP2 -+-+MG132 kDa 42 kDa ACTIN 124 kDa VCL 57 kDa HMGCS1 32 kDa P-S6RB (Ser240/244) 60-78 42 kDa ACTIN kDa mSREBP2 32 kDa S6RB 1.0 2.32 1.0 0.84

SPC J 2.5 MLC SPC * K SPC p53+/+ p53-/- p53+/+ p53-/- 2.0 p53+/+ p53-/- α β 51 kDa P-GSK3 / 1.5 CON RAPA SB CHIR CONRAPA SB CHIR 47 kDa (Ser21/9) 51 kDa 124 kDa VCL 47 kDa GSK3α/β 1.0

P-GSK3/GSK3 0.5 60-78 42 kDa ACTIN kDa mSREBP2 1.0 0.93 3.05 4.35 0.0 1.0 1.06 2.25 2.88 1.0 0.59 0.87 0.86

p53+/+ p53-/-

L HMGCS1 M HMGCS1 4 CON 5 CON * CHIR99021 RAPA 4 3 3 2 2 1 1 Relative mRNA expression

0 mRNA expression Relative 0 p53+/+ p53-/- p53+/+ p53-/-

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 2019 American Association for Cancer Research. Author Manuscript Published OnlineFirst on November 19, 2019; DOI: 10.1158/0008-5472.CAN-19-0650Kaymak et al. Figure 3 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

A B DMSO MST MST + MVL C Acetyl-CoA 400 p53+/+ HMG-CoA ** HMGCR p53-/- statins 300 Mevalonate p53+/+

150µm 150µm 150µm 200 Farnesyl-PP 100 Squalene TUNEL positive cells/area

p53-/- 0 Cholesterol T MS MVL MST MVL 150µm 150µm 150µm DMSO + DMSO +

purple=TUNEL MST MST blue=DAPI

D CDKN1A CDKN1A E spheroid 3 * 24h 6 72h p53+/+ p53-/- DMSO * DMSO DMSOSIM SIM + DMSO SIM SIM + SIM SIM MVL MVL 2 4 124 kDa *** VCL 21 kDa p21

1 2 Relative mRNA expression Relative Relative mRNA expression0 0 p53+/+ p53-/- p53+/+ p53-/-

F G

DMSO SIM SIM + MVL spheroid p21+/+ p21-/- DMSOSIM SIM + DMSO SIM SIM + MVL MVL 124 kDa VCL

p21-/- 21 kDa p21

110µm 110µm 110µm

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 2019 American Association for Cancer Research. Author Manuscript Published OnlineFirst on November 19, 2019; DOI: 10.1158/0008-5472.CAN-19-0650Kaymak et al. Figure 4 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

A B Mevalonate C Mevalonate 1.5 0.6 Glucose labelled p53+/+ labelled unlabelled p53-/- Acetyl-CoA * 1.0 0.4 Mevalonate

Farnesyl-PP Geranyl geranyl-PP 0.5 0.2 Dolichol peak intensity/prot peak intensity/prot Squalene Heme A Ubiquinone (CoQ10) 0.0 0.0 7-DHC /- 0 4 6 M+ M+2 M+ M+ Cholesterol p53+/+ p53- D E Cholesterol Cholesterol 100 150 labelled 50 p53+/+ DMSO unlabelled p53-/- DMSO 100 5 4 50 3 2 peak intensity/prot peak intensity/prot 1 0 0 2 3 5 +0 +1 0 5 6 1 6 M M M+ M+ M+4M+ M+6M+7M+8M+9 +1 M M+11M+1M+132 M+14M+1M+1M+17M+18M+19M+2M+20 M+22M+23M+24M+25M+2M+27

p53-/- SIM p53+/+ SIM p53+/+ DMSOp53-/- DMSO

F G Ubiquinone (CoQ10) Ubiquinone (CoQ10) 50 60 40 p53+/+ DMSO labelled 30 p53-/- DMSO unlabelled 20 10 40 3

2 20 peak intensity/prot peak intensity/prot 1

0 0 2 5 8 9 0 4 6 1 7 0 6 9 M+0M+1M+M+3M+4M+M+6M+7M+M+ M+1M+11M+12M+13M+1M+15M+1M+17M+18M+19M+20M+2M+22M+23M+24M+25M+26M+2M+28M+29M+3M+31M+32M+33M+34M+35M+3M+37M+38M+3M+40M+41M+42M+43M+44M+45M+46M+47M+48M+49M+50

p53-/- SIM p53+/+ SIM p53+/+ DMSOp53-/- DMSO

H Cholesterol 7-DHC Ubiquinone (CoQ10) 120 400 * 0.8 100 300 0.6 80 )/tissue weight )/tissue weight 7 7 60 200 0.4

40 100 0.2 20 peak intensity/tissue weight intensity/tissue peak peak intensity (x10 intensity peak 0 peak(x10 intensity 0 0.0 p53+/+ p53-/- p53+/+ p53-/- p53+/+ p53-/-

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 2019 American Association for Cancer Research. Author Manuscript Published OnlineFirst on November 19, 2019; DOI: 10.1158/0008-5472.CAN-19-0650Kaymak et al. Figure 5 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

A B C Aspartate HCT116 p53+/+ HCT116 p53-/- ATP 500 FCCP R/A 500 oligo FCCP R/A - cyt C Synthase 400 oligo intermembrane e DMSO space 400 400 SIM - M+4 - e - e e 300 SIM+MVL I Q10 III IV M+3 300 300 e- M+2 matrix II 200 200 200 1/2 O2 M+1 NADH + H O Succinate Fumarate +H 2 M+0 NAD+ 100 100 100

α-KG NADH Malate peak intensity/prot

0 OCR (pmol/min/spheoid area) 0

0 area) (pmol/min/spheroid OCR Aconitate Oxaloacetate 050100150200 0 50 100 150 200 Citrate Time (minutes) Time (minutes) Aspartate

p53+/+ SIMp53-/- SIM p53+/+ DMSOp53-/- DMSO D E DMSO SIM SIM + MVL SIM + Q10 SIM + NAC *** 500 p53+/+ *** **** *** p53-/- 400 p53+/+ 300

150µm 150µm 150µm 150µm 150µm 200

100 p53-/-

TUNEL positive cells/area positive TUNEL 0 M 0 SI SIM 150µm 150µm 150µm 150µm 150µm DMSO DMSO SIM + MVL SIM + MVL purple=TUNEL SIM +SIM Q10 + NAC SIM +SIM Q1 + NAC blue=DAPI

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 2019 American Association for Cancer Research. Author Manuscript Published OnlineFirst on November 19, 2019; DOI: 10.1158/0008-5472.CAN-19-0650Kaymak et al. Figure 6 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

A B UMP UMP M+9 30 10 labelled carboamoylaspartate UMP M+8 * M+7 8 Dihydroorotatate Orotate 20 M+6 ATP M+5 6 ** - cyt C intermembrane e Synthase space DHODH M+4 4 - - 10 M+3 e e - e- e peak intensity/prot I III IV peak intensity/prot M+2 2 - Q10 e M+1 matrix II 0 M+0 0 1/2 O2 + +H H2O

p53-/- SIM p53-/- SIM p53+/+ SIM p53+/+p53-/- SIM DMSO p53+/+ DMSOp53-/- DMSO p53+/+ DMSO

C UMP UMP

10 M+9 5 * labelled M+8 * 4 M+7 M+6 3 **** 5 M+5 n.s. *** M+4 2 * M+3 peak intensity/prot M+2 peak intensity/prot 1 M+1 0 M+0 0 0 0 1 O MVL

p53+/+ SIM p53-/- SIM p53+/+ SIM p53-/- SIM p53+/+ DMSO p53-/- DMSO p53+/+ DMSO p53-/- DMS p53-/- SIM + Q10 p53-/- SIM + Q10 p53+/+p53+/+ SIM + SIMMVL + Q1 p53-/- SIM + MVL p53+/+p53+/+ SIM + SIM + Q p53-/- SIM + MVL

D E

400 p53+/+ 400 **** * p53+/+ * p53-/- *** p53-/- 300 300 * * 200 * 200

100 100 TUNEL positive cell/area positive TUNEL

TUNEL positive cells/area positive TUNEL 0 0 U SIM 5-FU SIM 5-F SIM SIM DMSO DMSO DMSO DMSO M + 5-FU M + 5-FU SI SIM + NCL SIM + NCL SI

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 2019 American Association for Cancer Research. Author Manuscript Published OnlineFirst on November 19, 2019; DOI: 10.1158/0008-5472.CAN-19-0650Kaymak et al. Figure 7 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

A Simvastatin control Simvastatin + mevalonate fl/fl

Apc B

200µm 200µm 200µm Apcfl/fl Apcfl/fl p53fl/fl Apcfl/fl, KrasG12D/+ Apcfl/fl p53fl/fl G12D/+

fl/fl 1.5 1.5 Kras 1.5 1.5 * * p53 **** fl/fl 1.0 1.0 1.0 1.0 ERT2 Apc 200µm 200µm 200µm **

0.5 0.5 0.5 0.5 G12D/+ spheroid area per field spheroid area per field VillinCRE spheroid area per field spheroid area per field Kras 0.0 0.0 0.0 0.0 fl/fl M SIM 200µm SIM SIM SI 200µm 200µm DMSO

Apc DMSO DMSO DMSO SIM+MEV SIM+MEV SIM+MEV SIM+MEV fl/fl G12D/+ p53 fl/fl Kras

Apc 200µm 200µm 200µm

C Simvastatin Simvastatin D control Simvastatin + Q10 + nucleosides Apcfl/fl p53fl/fl Apcfl/fl p53fl/fl KrasG12D/+

fl/fl 1500 1500 **** ****

p53 **** fl/fl 1000 **** 1000 **** Apc ERT2 200µm 200µm 200µm 200µm **** 500 500 fl/fl spheroid area per field spheroid area per field VillinCRE G12D/+ p53

fl/fl 0 0 0

Kras 1 SIM SIM Apc 200µm 200µm 200µm 200µm DMSO +NUCL DMSO M+Q10 SIM+Q M SI M+NUCL SI SI

E F G Villin CREER Apcfl/fl Villin CREER Apcfl/fl KrasG12D/+ Cholesterol Ubiquinone (CoQ9) Vehicle Simvastatin Vehicle Simvastatin * * 80 0.65 0.9 * * * * 60 0.60

40 0.55 0.8

20 0.50 Fraction Enrichment Fraction Fraction Enrichment

0 0.45 0.7 No. of BrdU pos cells per half crypt cells half per pos BrdU of No. fl fl/fl fl/fl /+ fl/ /+ D D 2 G12D/+ G12 Apc G1 + SIM + + SIM Apc l + SIM Apc + SIM + SIM fl/fl / fl/f + SIM fl/fl D D/+ Kras 2 Kras 2 Kras fl/fl G1 G1 Apc fl/fl G12D/+ Apc Apc fl/fl Apc Kras Apc Kras Apc Kras fl/fl fl/fl fl/fl

Apc Apc Apc

Mevalonate pathway provides ubiquinone to maintain pyrimidine synthesis and survival in p53-deficient cancer cells exposed to metabolic stress

Irem Kaymak, Carina Ramona Maier, Werner Schmitz, et al.

Cancer Res Published OnlineFirst November 19, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-19-0650

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2019/11/19/0008-5472.CAN-19-0650.DC1

Author Author manuscripts have been peer reviewed and accepted for publication but have not yet Manuscript been edited.

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/2019/11/19/0008-5472.CAN-19-0650. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.