Full Text (PDF)

Published OnlineFirst October 23, 2018; DOI: 10.1158/1535-7163.MCT-18-0863

Small Molecule Therapeutics Molecular Cancer Therapeutics Targeting of Hematologic Malignancies with PTC299, A Novel Potent Inhibitor of Dihydroorotate Dehydrogenase with Favorable Pharmaceutical Properties Liangxian Cao1, Marla Weetall1, Christopher Trotta1, Katherine Cintron1, Jiyuan Ma1, Min Jung Kim1, Bansri Furia1, Charles Romfo1, Jason D. Graci1, Wencheng Li1, Joshua Du1, Josephine Sheedy1, Jean Hedrick1, Nicole Risher1, Shirley Yeh1, Hongyan Qi1, Tamil Arasu1, Seongwoo Hwang1,William Lennox1, Ronald Kong1, Janet Petruska1,Young-Choon Moon1, John Babiak1, Thomas W. Davis1, Allan Jacobson2, Neil G. Almstead1, Art Branstrom1, Joseph M. Colacino1, and Stuart W. Peltz1

Abstract

PTC299 was identified as an inhibitor of VEGFA mRNA engagement in the DHODH conformation in situ is translation in a phenotypic screen and evaluated in the required for its optimal activity. PTC299 has broad and clinic for treatment of solid tumors. To guide precision potent activity against hematologic cancer cells in preclin- cancer treatment, we performed extensive biological char- ical models, reflecting a reduced pyrimidine nucleotide acterization of the activity of PTC299 and demonstrated salvage pathway in leukemia cells. Archived serum samples that inhibition of VEGF production and cell proliferation from patients treated with PTC299 demonstrated increased by PTC299 is linked to a decrease in uridine nucleotides levels of dihydroorotate, the substrate of DHODH, indi- by targeting dihydroorotate dehydrogenase (DHODH), a cating target engagement in patients. PTC299 has advan- rate-limiting enzyme for de novo pyrimidine nucleotide tages over previously reported DHODH inhibitors, includ- synthesis. Unlike previously reported DHODH inhibitors ing greater potency, good oral bioavailability, and lack of that were identified using in vitro enzyme assays, PTC299 off-target kinase inhibition and myelosuppression, and is a more potent inhibitor of DHODH in isolated mito- thus may be useful for the targeted treatment of hemato- chondria suggesting that mitochondrial membrane lipid logic malignancies.

Introduction fourth enzymatic step, the ubiquinone-mediated oxidation of dihydroorotate to orotate in the de novo pyrimidine synthesis De novo pyrimidine nucleotide biosynthesis is activated in pathway (6). DHODH is a validated therapeutic target for the proliferating cancer cells in response to an increased demand for treatment of autoimmune diseases such as rheumatoid arthritis synthesis of DNA, RNA, and phospholipids by activation of (7, 8). Inhibition of DHODH was shown in preclinical studies to oncogenic pathways such as Myc, PI3K, PTEN, and mTOR (1– prevent growth of many types of cancers (9–13). Moreover, it was 5). Dihydroorotate dehydrogenase (DHODH), located on the demonstrated that inhibition of DHODH induces differentiation surface of the inner mitochondrial membrane, catalyzes the of acute myeloid leukemia (AML) cells (14–16). Pharmacologic inhibition of de novo pyrimidine synthesis sensitizes triple-negative breast cancer cells to genotoxic chemotherapy agents and 1PTC Therapeutics, Inc., South Plainfield, New Jersey. 2Department of Microbi- reduces chemotherapy resistance (2). DHODH inhibitors such as ology and Physiological Systems, University of Massachusetts Medical School, teriflunomide, brequinar, and vidofludimus have been consid- Worcester, Massachusetts. ered for use in oncology (17–20), but their use may be limited due Note: Supplementary data for this article are available at Molecular Cancer to a narrow therapeutic window as a result of lesser potency and/ Therapeutics Online (http://mct.aacrjournals.org/). or off-target activities such as their inhibition of kinases (21, 22). Current address for T. Arasu: US Food and Drug Administration, 629 Cranbury There is great interest in the discovery and development of new Road, 2nd Floor, East Brunswick, NJ 08816; and current address for T.W. Davis: DHODH inhibitors for therapeutic uses (23, 24). These efforts PMV Pharmaceuticals Inc., Cedar Brook Corporate Center, 8 Clarke Drive, have mainly relied on in vitro enzyme assays to define and Cranbury Township, NJ 08512. optimize activity, in which the purified enzyme conformation Corresponding Author: Liangxian Cao, PTC Therapeutics, Inc., 100 Corporate may differ from that found in the cells due to lack of membrane fi CT, South Plain eld, NJ 07080. Phone: 908-912-9132; Fax: 908-222-0567; lipids in the assay (25). Recently, it was demonstrated that E-mail: [email protected] DHODH attaches to the inner mitochondrial membrane by doi: 10.1158/1535-7163.MCT-18-0863 binding charged phospholipids which stabilize the flexible sub- 2018 American Association for Cancer Research. strate- and drug-binding site, suggesting that the design and

www.aacrjournals.org 3

Cao et al.

development of novel inhibitors that additionally engage in lipid (Crown Bioscience); otherwise we did not perform any additional interactions may represent a promising approach to targeting cell line authentication in house. The information on the source of DHODH in vivo (26). cells and when they were obtained can be found in Supplemen- PTC299 was identified as an inhibitor of the translation of tary Table S1. VEGFA mRNA using PTC Therapeutics' proprietary GEMS tech- Stable cell lines were generated as reported previously (27). nology phenotypic screening platform (27). PTC299 was in Briefly, plasmid DNA was transfected into HT1080 cells using clinical trials for the treatment of solid tumors (28, 29) and is Fugene-6 transfection reagent (Roche Diagnostics GmbH), fol- currently being reevaluated for use in hematologic cancers. To lowed by selection with 200 mg/mL hygromycin for luciferase- guide safe and effective precision cancer treatment, we performed stable cell lines or 200 mg/mL Zeocin (Invitrogen) for V5 tag stable extensive biochemical characterization of PTC299 and demon- cells in the culture medium. strated that it interacts with and inhibits DHODH, a rate-limiting enzyme in the de novo biosynthesis of pyrimidine nucleotides (5). Western blot analysis Inhibition of VEGFA production by PTC299 is a downstream Western blot was performed as reported previously (27). Anti- effect of inhibiting de novo pyrimidine synthesis as it can be body (C1) specific for VEGFA was purchased from Santa Cruz completely rescued by exogenously added uridine but not by Biotechnologies (1:200 dilution); antibody specific for human other nucleosides. Collectively, PTC299 represents a novel class of b-actin was purchased from Abcam (1:10,000 dilution); antibody potent DHODH inhibitors, and our data demonstrate the poten- specific for V5 tag was purchased from Invitrogen (1:5,000 tial of PTC299 as a treatment option to address unmet needs in dilution). Antibody specific for DHODH was purchased from cancers, such as leukemia and lymphoma, that have lower levels Proteintech (Cat#: 14877-1-AP, 1:500 dilution); and antibody of uridine salvage activity and thus rely on the de novo biosynthesis specific for human prohibitin was purchased from Thermo of pyrimidine nucleotides for survival and rapid proliferation. Scientific (Cat#: PA5-12274 and PA5-14133, 1:1,000 dilution). Immunodetection was done using the corresponding secondary Materials and Methods antibodies conjugated with infrared dyes or horseradish peroxide. The expression levels of proteins were detected with Odyssey General methods (LI-COR) or using enhanced chemiluminescence (Pierce). DNA constructs were generated using standard procedures as reported previously (27). VEGFA 50- and 30-untranslated regions Determination of ELISA EC and cytotoxicity CC values (UTR) were amplified from human genomic DNA (Clontech), 50 50 All ELISAs were performed using commercially available ELISA while the VEGFA165 open reading frame from RNA derived from kits (R&D Systems) according to the manufacturer's instructions. HeLa cells incubated under hypoxia (1% oxygen) was amplified The screen of 240 cell lines was performed at Crown Bioscience. by RT-PCR. DNAs generated with PCR were confirmed by DNA Cells were treated with PTC299 at a series of doses for 72 hours, and sequencing. the inhibition of cell proliferation was determined using a standard PTC299, PTC-371, GSK983, and vidofludimus (4SC-101) were assay, CellTiter-Glo Luminescent Cell Viability Assay (Promega), synthesized at PTC Therapeutics; for details, see the Supplemen- that measures total cellular adenosine triphosphate (ATP) con- tary Materials and Methods in the Supplementary Data File. centrations as an indicator of cell viability. Data generated from Brequinar (SML0113) and teriflunomide (SML0936) were pur- ELISA or cytotoxicity studies were plotted with Prism software. A chased from Sigma, whereas pyrazofurin-MP (P61-B01A) was sigmoidal dose response with a variable slope regression curve was purchased from TriLink Biotechnologies. generated for each compound. Maximal and minimal inhibitions were set at 100% and 0%, respectively, and CC50 values were Cell culture calculated after curve fitting using the Prism software. Tumor cell lines, except for MOLM-13 which were obtained from Dr. Kensuke Kojima in July 2015 (Saga University, Japan) 35S-labeling and immunoprecipitation 0 and Huh-7 which were obtained in February 2001 from the Stuart HT1080 cells harboring either the VEGFA 5 -UTR-VEGFA165-V5 Peltz laboratory at University of Medicine and Dentistry of New construct or the control VEGFA165-V5 were treated for 8 hours with Jersey (now Rutgers University, New Jersey), were purchased from vehicle alone (0.5% DMSO) or vehicle with 100 nmol/L PTC299, the ATCC and were maintained in DMEM (adherent cells) or and then incubated with 35S-labeled methionine and cysteine for RPMI-1640 (suspension cells) supplemented with 10% FBS, approximately 4 hours in the presence or absence of the protein penicillin (50 IU/mL), and streptomycin (50 mg/mL). Culture transporter inhibitor brefeldin A or the proteasome inhibitor media and supplement agents were purchased from Gibco BRL, MG-132. Equal amounts of supernatant or the cell extracts were Invitrogen. Cells bought from the ATCC were expanded and immunoprecipitated with the anti-V5 antibody (Invitrogen). frozen at early passages. Cells were typically used for studies The immunoprecipitated products were resolved by 4% to 20% within 2 to 5 weeks after thawed from frozen stocks. Cells that gradient SDS-PAGE and exposed to Kodak Biomax MR film. were cultured for more than 2 months (about 20 passages) were discarded. Patient-derived AML cells, purchased from AllCells Human tumor xenograft studies (January 2017), Inc. and Champions Oncology, Inc. (April All studies involving animals were performed in accordance with 2017), were cultured in StemSpan medium (Stemcell Technolo- guidelines promulgated by the American Association for Accredi- gies). Cell lines in culture were routinely verified once per month tation of Laboratory Animal Care with the oversight of the animal to be mycoplasma free assayed by using the MycoAlert Myco- use and care committees atRutgersUniversity.Subcutaneous tumor plasma Detection Kit (Cat#: LT07-318, Lonza). The 240 cell lines xenograft studies were performed as reported previously (27). For used for screening of PTC299 activity were authenticated with systemic leukemia lethality model, male NOD-SCID mice were short tandem repeat profiling and mycoplasma tested by the CRO inoculated with MOLT-4 human ALL tumor cells (1 107 cells in

4 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

PTC299 Is a Novel Potent DHODH Inhibitor

200 mL of PBS) by i.v. injection. Eight days after tumor inoculation, m/z 158.0 ! m/z 114.0 for 15N-dihydrorotic acid, and m/z 324.0 mice were randomized into two groups (10 mice per groups) ! m/z 79.0 for 15N-monophosphate uridine (UMP). and treated with compounds as indicated in the results. The mice were observed until moribund, at which time they were euthanized. Conjugation of PTC299 or PTC-371 to Sephorose-6B beads At 2- and 4-week postinoculation, whole blood was obtained from PTC299 or its inactive enantiomer PTC-371 was conjugated to 5 mice pergroup byretro-orbital bleeding, stained with anantibody the epoxy-activated Sepharose 6B according to the manufacturer's against human CD45, and the number of CD45-positive cells was instructions (GE healthcare; Cat. # 17-0480-01). Briefly, 1 g of determined by FACS analysis. epoxy-activated Sepharose 6B was stirred in water (20 mL) for 2 hours at room temperature and then filtered. The wet Sepharose 6B Immunohistochemistry was diluted with pH 11 phosphate/NaOH buffer (5 mL). To a Tumor samples were cut into pieces of less than 3 mm in solution of 1 mmol/L 4-chlorophenyl (S)-6-chloro-1-(4-hydroxy- fi fi phenyl)-1,3,4,9-tetrahydro-2H-pyrido[3,4-b] indole-2-carboxylate thickness and placed in a zinc xative for 24 hours. Paraf n sections were prepared by American HistoLab, Inc. using standard in DMF (12 mL) at 40 C was added the Sepharose 6B in phosphate/ procedures. Mouse endothelial cells were stained with rat anti- NaOH buffer. The reaction mixture was stirred further for 24 hours fi mouse CD31 monoclonal antibody MEC13.3 (BD Pharmingen) at 40 C. The Sepharose 6B-PTC299 product was ltered and and detected with a rabbit anti-rat IgG HRP detection kit accord- washed with DMF (100 mL), water (100 mL), pH 4 buffer (100 ing to the manufacturer's instructions (BD Pharmingen). mL), pH 11 buffer (100 mL), water (100 mL), and 0.5 mol/L NaCl/ water. Sepharose 6B-PTC-371 was prepared with 4-chlorophenyl (R)-6-chloro-1-(4-hydroxyphenyl)-1,3,4,9-tetrahydro-2H-pyrido- Transcriptome profiling with microarrays [3,4-b] indole-2-carboxylate using above reaction procedures. PTC299-resistant and parent HT1080 cells were seeded in 6- well plates (5 105 cells/well) and incubated overnight. Four Pull-down of proteins that are specifically bound to PTC299- replicates of cells for each cell type were performed. Total RNA was conjugated beads purified with RNeasy Mini Kit (Qiagen) according to the manu- Cells in log phase growth were washed with cold PBS once and facturer's instructions. cRNA was prepared according to the stan- then collected by trypsin digestion. After centrifugation at 1,200 dard Affymetrix protocol. cRNA was hybridized at 45C for 16 rpm for 5 minutes, the cells were resuspended in lysis buffer [PBS hours on a GeneChip Human Genome U133 plus 2.0 Array. þ 0.5% Triton X100 þ 1x proteinase inhibitors (Halt, Roche; GeneChips were analyzed using the Affymetrix scanner. Data Cat#: 78440)] and incubated on ice for 10 minutes; supernatants analysis was performed using Transcriptome Analysis Console were collected after passing cells through a 22-gauge needle 10 Software (Thermo Fisher Scientific). The MAS5 normalization times and centrifuging at 13,000 rpm for 15 minutes to remove method was used. ANOVA was used for statistical evaluation. cell debris. Subsequently, 1 mL of cell lysate solution (2 mg/mL protein) was added to each reaction tube containing 25 mL fi Quanti cation of pyrimidine nucleotides in PTC299-treated (packed volume) PTC299- or the R-enantiomer control (PTC- cells 371) beads and incubated for 2 hours at 4C with gentle agitation Cells in log phase growth were cultured in 10 cm dishes (4 on a rotator; after three washes with 1 mL washing Buffer, 5 min/ 6 10 cells/dish) with glutamine-free DMEM containing 1 g/L each, the protein bound with beads was eluted with indicated m glucose, 10% FBS, penicillin (50 IU/mL), streptomycin (50 g/ buffer as shown in the Results section. Proteins in the eluted 15 mL), and 1 mmol/L N-glutamine (Sigma, Cat#: 490024). After samples were verified by Western blot analysis. culturing for the indicated time in the presence of compounds or vehicle control (0.5% DMSO), the cells were washed once with 5 Mitochondria isolated from cultured K562 cells as source of mL cold PBS, harvested, and lysed in 0.5 mL cold distilled H2O DHODH for in vitro enzyme inhibition studies plus 1 mL 80C methanol using a plastic cell scraper. The cell Mitochondria were isolated from K562 cells in log phase growth lysates were then centrifuged at 10,000 x g for 15 minutes at 4 C, using a Dounce homogenizer as reported (30). The isolated mito- and the supernatant was collected for LC-MS/MS analysis of the chondria in the supernatant were pelleted at 12,000 g for 15 15 N-labeled de novo pyrimidine nucleotide metabolites. minutes and resuspended in 1 mL 1x MS homogenization buffer for analysis of downstream DHODH activity as described below. LC-MS/MS detection of metabolites Quantification of 15N-labeled de novo–synthesized pyrimidine In vitro DHODH activity assays nucleotides was carried out on an Accela pump and a PAL The chromogen reduction assay was carried out as previously autosampler coupled to a TSQ Quantum Ultra mass spectrometer. reported (31). DHODH activity was determined in the presence of The mass spectrometer was equipped with a heated electrospray compounds or vehicle control using the standard colorimetric ionization source operated in negative-ion mode (Thermo Fisher DHODH continuous assay in which the oxidation of dihydrooro- Scientific). The ion spray voltage was set at 2500 V, capillary tic acid (DHO) and the subsequent coupled reduction of ubiqui- temperature at 350C, vaporizer temperature at 300C, sheath gas none is measured by monitoring the reduction of 2,6-dichlor- pressure at 50 units, and auxiliary gas pressure at 5 units. A ophenolindophenol (DCIP) on a BioTek Power Wave XS2. The Thermo Fisher Scientific Hypercab column (3 mm, 50 2.1 mm) velocity for each reaction was derived using the GEN5 software. was used and maintained at 50C for metabolite separation. Mobile phase A was 10 mmol/L NH4HCO3 in water, pH 9.4, Docking analysis and mobile phase B was 10 mmol/L NH4HCO3 in ACN-water Software Sybyl X 2.1.1 was used to prepare ligand structures. (9:1). The flow rate was set at 0.50 mL/min. Ion transitions Ligands used in docking procedures were imported as SDF 3D monitored were at m/z 156.0 ! m/z 112.0 for 15N-orotic acid, format which were processed using Pipeline Pilot. Protein crystal

www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 5

Cao et al.

structures were downloaded from the Protein Data Bank (PDB; In a dose-dependent and stereo-selective manner, PTC299 http://www.rcsb.org/). Protein structure preparations were per- inhibited hypoxia-induced VEGFA protein production in HeLa formed using the standard protocol of Sybyl X 2.1.1, and the cells with an EC50 of 1.64 0.83 nmol/L (Fig. 1B). Similar activity resulting structures were saved as SYBYL ''.mol2'' files. The pockets was also observed in HT1080 cells which constitutively produce protocols were generated using standard fully automated Surflex- high levels of VEGFA (Supplementary Fig. S1E). The effect of Dock procedures (32). Sybyl provides a docking score for each PTC299 is not limited to secreted isoforms of the protein as the docked conformation, which includes two parts: an affinity score levels of high molecular weight matrix-bound variants of (-logKd) and a crash score (also pKd units). The crash score is the VEGFA189/206 were similarly decreased (Supplementary Fig. degree of inappropriate penetration into the protein by the ligand S1F). PTC-371, the R-enantiomer of PTC299, was inactive in as well as the degree of internal self-clashing that the ligand is inhibiting VEGFA synthesis. experiencing. The top docking scores for each ligand were used for To verify the role of the VEGFA 50-UTR for PTC299 activity, a V5 comparison. epitope–tagged plasmid containing the VEGFA165 open reading The Docking Mode was set as Surflex_Dock GeomX(SFXC). Input frame was constructed (Fig. 1C, top plot). PTC299 dose-depen- options were set to allow 3D Coordinate Generation, if necessary. dently inhibited VEGF 50-UTR–mediated production of the Protein movement was set to allow for the flexibility of protein VEGFA165-V5 protein as determined by Western blot analysis atoms whose van der Waals surface distances from ligand atoms are (Fig. 1C) and by ELISA (Supplementary Fig. S1G). Pulse-chase < 4 Å and for the adaptation of the active site conformation to the experiments using 35S-radiolabled methionine and cysteine dem- docked ligand. The covalent force field weighting of the ligand was onstrated that PTC299 decreased the levels of newly synthesized set to 1.0, whereas the covalent force field weighting of the protein VEGFA protein (Fig. 1D) while not decreasing total protein was set to 0.6 to allow heavy atoms to move. The maximum number synthesis (Supplementary Fig. S1H). This effect could not be of poses was set to 10 to optimize and rescore each ligand after a reversed by treatment with the protein transport blocker brefeldin docking run that included protein movement. Additional Starting A (BFA) or with the proteasome inhibitor MG132 (Supplemen- Conformations per Molecule were set to 6 (GeomX). The maximum tary Fig. S1I and S1J). These results suggest that PTC299 treatment number of conformations per fragment to be submitted to the results in the inhibition of VEGFA translation rather than the docking procedure was set to 20. The maximum number of poses inhibition of VEGFA secretion or promotion of VEGFA protein per ligand was set to 20. The minimum root-mean-square distance degradation and support the conclusion that the VEGFA 50-UTR is between final poses was set to 0.50. The Sybyl version of the required for the suppression of VEGFA mRNA translation by AMBER7 FF99 Force Field was used for energy calculations (33). PTC299. In a mouse tumor xenograft model, PTC299 dose-dependently Statistical analysis inhibited the growth of HT1080 fibrosarcoma (Fig. 1E), without Data are the mean SD or SEM as indicated for quantitative overt toxicity or significant body weight changes at any dose when experiments. For statistical analysis, P values were derived using compared with treatment with the vehicle alone (Supplementary unpaired Student t tests for any study with only two groups Fig. S1K). In parallel, treatment of mice with PTC299 resulted in a presented. Otherwise, comparisons of groups were performed on dose-dependent reduction in the concentrations of intratumor log-transformed data using a one-way ANOVA test. All analyses and circulating plasma human VEGFA relative to those in control were made using GraphPad Prism Software. mice (Fig. 1F) but had no significant impact on the levels of other cytokines such as PDGF and FGF-2 in tumors as assessed in a Data availability separate study (Supplementary Fig. S1L). At the suboptimal dose All data generated or analyzed during this study are included in of 0.3 mg/kg, treatment of mice with PTC299 resulted in partial this article and its Supplementary Information Files. Additional reductions in the levels of both tumor and plasma human VEGFA, information is available from the corresponding author upon indicating that inhibition of intratumor human VEGFA correlates request. with tumor growth control. When stained with anti-murine CD31 antibody, the diameters Results of tumor vessels were visibly reduced, and their distribution was PTC299 was identified from a phenotypic screen and chemical more uniform in tumors from PTC299-treated mice than in optimization as an inhibitor of VEGFA protein synthesis vehicle-treated mice (Fig. 1G), consistent with results in preclin- Using the contextual regulation of the production of VEGFA in ical models using other VEGF-targeted therapies (34, 35). the tumor cell imparted by critical elements in the 50- and 30-UTRs of the VEGFA mRNA, we initiated a drug discovery program using Inhibition of VEGFA production by PTC299 is a downstream PTC Therapeutics' proprietary GEMS technology phenotypic effect of inhibiting de novo pyrimidine synthesis screening platform (27). Subsequently, structure–activity–rela- To gain insight into the mechanism of action of PTC299, we tionship (SAR) studies led to the identification of PTC299 (Fig. generated drug-resistant cells by culturing HT1080 cells in esca- 1A) as a clinical development candidate for cancer therapy. lating concentrations (1 nmol/L to 1,000 nmol/L) of PTC299, PTC299 is the pharmacologically active S-enantiomer of 4-chlor- doubling the PTC299 concentration weekly. PTC299-resistant ophenyl (1S)-6-chloro-1-(4-methoxyphenyl)-1,3,4,9-tetrahydro- HT1080 cells proliferate at a rate similar to that of wild-type 2H-beta-carboline-2-carboxylate [synthesis of PTC299 (5) and HT1080 cells (Supplementary Fig. S2A), but the IC50 for the intermediates 1–4 shown in Supplementary Fig. S1A]. Similar to inhibition of VEGFA production by PTC299 in resistant cells is the compounds identified in a high-throughput screen as previ- shifted dramatically, from low nmol/L to single-digit mmol/L ously described (27), biological characterization suggested that values (Supplementary Fig. S2B). Transcriptome profiling with PTC299 inhibited VEGFA-50UTR–mediated mRNA translation microarrays to compare the gene expression in PTC299-resistant rather than transcription (Supplementary Fig. S1B–S1D). and in parent HT1080 cells demonstrated that the gene with the

6 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

PTC299 Is a Novel Potent DHODH Inhibitor

Figure 1. Structure, biological activity, and stereoselectivity of PTC299. A, Chemical structure of PTC299, the 1S conﬁguration, and the inactive R enantiomer, PTC-371. B, PTC299 inhibits hypoxia-induced VEGFA expression in HeLa cells. Results are expressed as the percent inhibition of hypoxia-induced VEGFA production relative to vehicle-treated controls. Hypoxia (1% oxygen)-induced VEGFA production is measured by subtracting the VEGFA produced under normoxic conditions (21% oxygen) from the levels of VEGFA produced under hypoxia. Data are mean SD from a representative study (n ¼ 3). C, The VEGFA 50-UTR is required for

PTC299 inhibition of a V5-tagged VEGFA165 protein production. HT1080 cells were stably transfected with the constructs shown in the diagrams on the top of the graph. The culture supernatant and cell lysates were collected for Western blot after treatment with PTC299 or 0.5% DMSO control for 36 hours. D, PTC299 selectively inhibits production of pulse 35S-labeled VEGFA. The VEGFA-V5 stable cell lines were treated with PTC299 (100 nmol/L) for 8 hours and then pulsed/chased with 35S-radiolabled methionine and cysteine for 4 hours prior to immunoprecipitation with V5 antibody. E, PTC299 dose-dependently inhibits HT1080 tumor growth in mice. Values represent the average SEM of 10 mice per group. Compound or vehicle was orally administered b.i.d. as indicated in the graph. , P < 0.001, one-way ANOVA test. F, PTC299 reduces intratumor and serum levels of human VEGFA. Values represent the average SEM of 10 mice. One- way ANOVA test compared with vehicle; , P < 0.05 and , P < 0.001. G, PTC299 inhibits tumor angiogenesis, and tumor sections were stained with anti-CD31 antibody. www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 7

Cao et al.

8 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

PTC299 Is a Novel Potent DHODH Inhibitor

greatest increase in expression in PTC299-resistant cells is cytidine By utilizing this methoxy group, we linked either PTC299, or its deaminase (CDA), whereas the gene with greatest decrease is inactive R-enantiomer PTC-371, to epoxy-activated Sepharose 6B Eukaryotic Translation Initiation Factor 1A, Y-Linked (EIF1AY) (Fig. 3A). In pilot studies, we focused on those highly enriched [Fig. 2A, the microarray data can be found in Supplementary Excel proteins in PTC299-conjugated bead pull-down samples by cut- File S1 or downloaded from GEO database (Accession number: ting out the most enriched protein bands for the proteomic GSE121266)]. These changes in expression were verified by quan- studies and demonstrated that these proteins are Prohibitin-1/2 titative PCR as shown in Supplementary Fig. S2C. Because CDA is and voltage-dependent anion channel (VDAC). After demons- an enzyme in the uridine salvage pathway, we evaluated the effect trating that PTC299 inhibits de novo pyrimidine synthesis and of PTC299 treatment on intracellular nucleotide levels. PTC299 has an activity profile similar to that of the DHODH inhibitor selectively decreased the total levels of the deoxypyrimidine brequinar (Fig. 2; Supplementary Table S2), we performed nucleotides dCTP and dTTP, but not those of the deoxypurine another pull-down study and demonstrated that PTC299 beads nucleotides dATP and dGTP (Fig. 2B). Metabolic labeling studies enriched not only for Prohibitin-1/2 but also for DHODH in the using 15N-glutamine demonstrated that PTC299 inhibited de novo pull-down samples (Fig. 3B). To determine if binding to these pyrimidine synthesis, whereas its inactive R-enantiomer PTC-371 proteins was PTC299 specific, we first performed "compete-on" had no effect (Fig. 2C). Further studies demonstrated that inhi- studies and demonstrated that preincubation with 100 mmol/L bition was dose-dependent and measurable after only 30 minutes PTC299 selectively blocked the binding of DHODH to the of treatment (Supplementary Fig. S3A and S3B). The inhibition PTC299-conjugated beads while it had no effect on the binding profile of PTC299 in these assays is similar to that of the known of Prohibitin-1/2 (Fig. 3C). The inactive enantiomer PTC-371 had DHODH inhibitor brequinar rather than that of the UMP no effect on the binding of either DHODH or Prohibitin-1/2. synthase (UMPS) inhibitor pyrazofurin-monophosphate Similar results were observed when isolated mitochondrial lysates (refs. 18, 36; Supplementary Table S2). As shown in Fig. 2D, were used as the protein source (Supplementary Fig. S3C). We treatment with pyrazofurin-monophosphate resulted in an over next demonstrated that both free PTC299 and brequinar effec- 40-fold increase of 15N-orotate, the substrate of UMPS. However, tively blocked the binding of DHODH to the PTC299-conjugated cotreatment with PTC299 blocked the pyrazofurin-MP–induced beads while having no effect on prohibitin binding (Fig. 3D), increase of 15N-orotate, consistent with inhibition of DHODH suggesting that PTC299 and brequinar bind to the same site in that acts upstream of UMPS. DHODH. Subsequently, we performed a "competing off" study We performed nucleoside rescue studies and demonstrated that and demonstrated both PTC299 and brequinar effectively eluted the PTC299-mediated inhibition of VEGFA protein synthesis was out the DHODH from the PTC299-conjugated beads while the dose-dependently blocked by the addition of uridine, but not by inactive isomer PTC-371 did not (Fig. 3E), further supporting the the addition of adenosine, cytidine, or guanosine, to the culture notion that PTC299 and brequinar bind to the same pocket in medium (Fig. 2E and F). As a consequence of DHODH inhibition DHODH. Proteomic study of those elution samples verified that and subsequent pyrimidine nucleotide depletion, PTC299 also one of the most selectively enriched proteins was DHODH induced an S phase cell-cycle arrest in HT1080 cells (Fig. 2G, 2nd (Supplementary Table S3). Because PTC299 is species selective plot). This cell-cycle arrest was prevented by the addition of the and does not have activity in mouse and rat cell lines tested in vitro pyrimidine nucleosides uridine or cytidine (Fig. 2G, bottom two (Supplementary Table S4), we performed a side-by-side pull- plots), but not the purine nucleosides adenosine or guanosine (L. down study with human and rat cell lysates. As shown in Fig. Cao; unpublished data). These data suggest that inhibition of 3F, PTC299-conjugated beads enriched for both human DHODH VEGFA translation and cell-cycle arrest by PTC299 are two sep- and Prohibitin-1/2. However, these beads only bound to rat arate downstream effects of inhibiting de novo pyrimidine nucle- prohibitin rather than rat DHODH when compared with the otide synthesis (Fig. 2H). blank beads or the PTC-371–conjugated beads. These data suggest that PTC299 specifically binds to human DHODH, and that other PTC299 interact with and inhibits the activity of DHODH enriched proteins such as Prohibitin-1/2 are PTC299-bead selec- Extensive SAR studies demonstrated that a methoxy group on tive rather than PTC299 specific. PTC299 can tolerate modification and is not critical for the Further studies demonstrated PTC299 inhibited the activity of biological activity of PTC299 (Y.C. Moon; unpublished data). recombinant DHODH using in vitro enzyme assays, though not as

Figure 2. Inhibition of VEGFA translation by PTC299 is coupled with inhibition of de novo pyrimidine synthesis. A, Transcriptome profiling of PTC299-resistant HT1080 cells using microarrays. Highlighted on the left volcano plot are those genes with greater than 2-fold changes in expression with a FDR less than 0.05; the heat maps of the expression of these genes are shown on the right. B, PTC299 selectively decreases pyrimidine nucleotides in HT1080 cells. HT1080 cells were treated with PTC299 (100 nmol/L) or 0.5% DMSO control for 8 hours. Data are representative of two independent studies. C, PTC299 inhibits de novo pyrimidine nucleotide synthesis. HT1080 cells in log phase growth were treated with compounds (100 nmol/L) or vehicle control (0.5% DMSO) in the presence of 1 mmol/L 15N-glutamine for 8 hours. 15N-labeled CTP, UTP, and UMP were quantified with LC/MS. Data are representative of two independent studies. D, PTC299 acts upstream of UMPS. Cells were treated with 1 mmol/L PTC299 or MP-pyrazofurin or combination of the two for 8 hours. Peak area ratio of orotic acid (normalized to an internal control) for each treatment was determined, and average SD was shown in the picture (n ¼ 2). MP-pyrazofurin is a UMPS inhibitor. E, Inhibition of VEGFA production in HT1080 cells by PTC299 is completely blocked by addition of exogenous uridine into the cell culture. Cells were treated with PTC299 (100 nmol/L) or vehicle (0.5% DMSO) in the presence of 100 mmol/L various nucleosides (indicated in the figure) for 48 hours. Data are mean SD, n ¼ 2. F, Inhibition of VEGFA production in HT1080 cells by PTC299 is dose-dependently blocked by addition of exogenous uridine. Assays were done in triplicate, and data represent average inhibition against DMSO control after VEGFA levels normalized to viable cell number measured by Celltiter-Glo. Data represent the mean SD, n ¼ 2. G, Cell-cycle arrest in S-phase induced by PTC299 is completely blocked by addition of exogenous uridine or cytidine into the cell culture. HT1080 cells were treated with PTC299 (100 nmol/L) or vehicle (0.5% DMSO) in the presence of 100 mmol/L of the indicated nucleosides for 24 hours. H, A scheme showing de novo and salvage pathways of pyrimidine nucleotide synthesis. CAD, Carbamoyl-phosphate synthetase 2, aspartate transcarbamylase, and dihydroorotase; UCK, uridine cytidine kinase.

www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 9

Cao et al.

Figure 3. PTC299 binds to and inhibits the activity of DHODH. A, Schematic drawing of PTC299, the inactive isomer linked to the agarose beads and the outline of chemical pull- down study. B, PTC299 beads bind to DHODH and Prohibitin-1/2. Pull-down samples from Blank-, PTC-371–, and PTC299-conjugated beads were resolved on 4%–20% SDS-PAGE. Left plot shows the silver staining, whereas the right plot shows Westerns blotted with antibodies against human DHODH, PHB1 (Prohibitin-1), PHB2 (Prohibitin-2), and VDAC respectively. C, PTC299 selectively inhibits DHODH binding to the PTC299 beads. HT1080 cell lysates were preincubated with 100 mmol/L free PTC299 or PTC-371 as indicated in the figure for 2 hours at 4C prior to the pull-down study described in the methods. D, Western blot analysis demonstrated that free PTC299 and brequinar competes DHODH binding to the PTC299 beads. The experiment was done in the same way as for Fig. 3C. E, Western blot analysis demonstrated that free PTC299 selectively elutes DHODH from the PTC299 beads. F, PTC299-conjugated beads selective bind to human rather than rat DHODH. G, PTC299 inhibits DHODH activity in isolated mitochondria, data are representative of three independent mitochondria preparations, and orotate production was quantified by LC-MS/MS analysis of the end product after 30 minutes of enzyme reaction. H, Computer modeling using crystal structures of the human DHODH protein. Both the brequinar analog C44 (Purple) and PTC299 (cyan) bind deep in the pocket, but the inactive enantiomer PTC-371 (orange) does not, which is reflected by the docking scores on the right. The cofactor flavin mononucleotide (FMN) and the product orotate are shown in yellow and blue, respectively. Right plot shows a localized view that PTC299 (cyan) binds to the pocket of human DHODH crystal at 4IGH. Amino acids within 5 angstroms of PTC299 are shown as a ribbon. Amino acid residues within 3 angstroms were labeled (hydrophobic, red, charged, purple; polar, blue). Relevant side chains that make up the pocket are displayed explicitly.

10 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

PTC299 Is a Novel Potent DHODH Inhibitor

Figure 4. PTC299 inhibits DHODH activity and reduces elevated VEGFA levels in patients with NF2. A, Reduction in mean levels of serum VEGFA after 8 weeks of PTC299 treatment. Data are represented as mean SEM, n ¼ 11. B, VEGFA reduction for individual patients over the course of the PTC299 treatment. C, Serum levels of DHO are correlated with PTC299 concentration in patient-002. D, Serum levels of DHO are inversely correlated with serum VEGFA levels in patient-002. E, PTC299 increases serum DHO in 3 additional patients whose serum samples were still available at the time of analysis. BLQ: below the level of quantiﬁcation.

potently as do brequinar or teriflunomide (Supplementary Fig. pull-down studies, the inactive enantiomer, PTC-371, bound less S3D). However, when purified mitochondria were used as the efficiently in the pocket with a final docking score of 7.05, source of DHODH, PTC299 inhibited DHODH activity more probably because of the unique nonsymmetrical shape of the potently than did teriflunomide and with potency similar to that cavity. Although PTC299 is a less potent inhibitor of rhDHODH of brequinar (Fig. 3G; Supplementary Fig. S3E). Along with the in vitro, it more potently inhibits DHODH from isolated mito- result that uridine can completely block the inhibition of VEGFA chondria (Fig. 3G; Supplementary Fig. S3E). Because PTC299 is production and cell-cycle arrest by PTC299, these data indicate lipophilic, its entry into the binding pocket of the enzyme is likely that DHODH is the primary target responsible for the observed facilitated when the enzyme is embedded in the mitochondrial activities of PTC299. membrane with engagement of membrane phospholipids. In studies using purified recombinant protein, there is no membrane Computer modeling shows that PTC299 binds to the same site or lipid layer that may hinder the entry of PTC299 into the binding in DHODH as does brequinar and with similar strength pocket. Efforts to cocrystallize PTC299 and purified recombinant Docking studies on available human DHODH protein (PDB DHODH have not been successful to date (Nonato MC, Uni- ID: 4IGH, http://www.rcsb.org/) revealed that PTC299 and the versidade de S~ao Paulo, Brazil; personal communication), pos- potent brequinar analog C44 bind to the same site with similar sibly because of the requirement for the lipid engagement with the strength (Fig. 3H), consistent with the pull-down results that enzyme (26). We are currently working to optimize crystallo- demonstrated brequinar competes with DHODH binding to graphic conditions to determine the structure of lipid-engaged PTC299 beads. The Sybyl Surflex docking score, calculated based DHODH interaction with PTC299. on the binding affinity and crash score (37), is 9.55 for C44 versus 9.27 for PTC299. The interaction of PTC299 with the binding PTC299 inhibits DHODH activity in patients with pocket is mainly a "hand-in-glove" type nonpolar hydrophobic neurofibromatosis type 2 interaction. Unlike with C44, there are no H-bond interactions Comprehensive preclinical IND-enabling safety pharmacolo- between PTC299 and Arg136 in DHODH. As anticipated from the gy, toxicology, and genotoxicity studies (Supplementary Materials

www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 11

Cao et al.

and Methods in the Supplementary Data File) demonstrated that for CDA and 4.6-fold for UPP1) in the hematopoietic tumor lines PTC299 is safe and well tolerated in preclinical studies (Supple- than in the solid tumor lines (Fig. 5A; Supplementary Table S7). mentary Table S5). Subsequently, several clinical trials in healthy We then performed LC-MS/MS analysis of UMP pools from volunteers and in patients with solid tumors were conducted to salvage and de novo pathways in a panel of 8 tumor cell lines assess PTC299 (28, 29), including a prospective phase IIa study in and demonstrated that PTC299-sensitive cells in general have less 2009–2010 in patients with neurofibromatosis type 2 (NF2), an salvage UMP production relative to de novo UMP production autosomal-dominant tumor-suppressor syndrome characterized when compared with the insensitive tumor cells (Fig. 5B). These by bilateral vestibular Schwannomas. Eleven patients were data suggest that cells with reduced uridine salvage activity are enrolled including 4 males and 7 females with a median age of more sensitive to PTC299. 25 (range, 19–44) at the time of enrollment. The demographic We then compared the potency of PTC299 in leukemia cells information on these patients is summarized in Supplementary with that of several well-known DHODH inhibitors (17–19). Table S6. After 8 weeks of PTC299 treatment, levels of serum PTC299 was the most potent inhibitor with an IC50 of about 1 VEGFA were significantly decreased (Fig. 4A, P < 0.004, paired nmol/L, over 10- to 1000-fold more potent than brequinar, Student t test). Baseline levels in healthy volunteers were reported vidofludimus, or teriflunomide (Fig. 5C), consistent with its to be approximately 140 pg/mL (38), indicating that serum superior potency in the inhibition of VEGFA production in tumor VEGFA levels are elevated in NF2 patients and were reduced by cells (Supplementary Fig. S6). We also tested PTC299 against AML treatment with PTC299 to levels found in healthy volunteers. patient-derived cells. As shown in Fig. 5D, PTC299 inhibited the Circulating VEGFA was decreased in 11 of 11 patients as shown proliferation of all five AML patient-derived cell lines tested with in Fig. 4B. IC50 values ranging from 2 to 60 nmol/L. In addition, we have We next examined archived serum samples from some patients shown that PTC299 induced differentiation in AML MOLM-13 in the NF2 clinical study for DHO levels, hypothesizing that cells (Supplementary Fig. S5B), consistent with the activity of inhibition of DHODH would result in higher levels of this various DHODH inhibitors as recently reported (14). substrate as seen in patients with genetic deficiency in the After confirming in vitro activity in leukemia cells, we assessed DHODH enzyme (39). As shown in Fig. 4C and D, serum DHO the efficacy of PTC299 in two mouse models of leukemia. As levels in Patient 010-002 who had been on study for 64 weeks shown in Fig. 1E, the maximal effect in control of tumor growth were initially below the lower limit of quantification. After treat- and VEGFA production was achieved at a dose of 3 mg/kg b.i.d. ment with PTC299, plasma concentrations of PTC299 and serum Later studies found that qd dosing 10 mg/kg was equivalently levels of DHO were significantly increased (Fig. 4C), whereas effective (M. Weetall; unpublished data) and thus this experi- serum VEGFA protein levels were continuously decreased (Fig. mentally more feasible dosing regimen was used in the leukemia 4D). For another 3 patients whose sera were also available at the animal models described below. time the analyses were performed, DHO levels were also below In the first model, MOLT-4 human acute lymphoblastic leu- the lower limit of quantification prior to treatment but were kemia (ALL) cells were injected i.v. into NOD-SCID mice which significantly increased with PTC299 treatment (Fig. 4E). The were then treated with PTC299 or with doxorubicin at the indi- increased levels of DHO in these samples indicate that PTC299 cated doses. Doxorubicin was tested at the highest dose tolerated inhibited DHODH in the previous clinical studies. by NOD-SCID mice. As shown in Fig. 5E, vehicle-dosed mice had a median survival time (MST) of 46 days and did not survive PTC299 demonstrates broad and potent inhibition of beyond 71 days. Doxorubicin did not prolong survival time. The hematologic cancer cell proliferation MST for the mice dosed with PTC299 was 136 days (P < 0.05, one- PTC299 inhibits de novo pyrimidine nucleotide synthesis in all way ANOVA, multiple comparisons vs. vehicle). The number of cancer cells analyzed whether the cells are responsive to inhibition circulating MOLT-4 cells was significantly reduced in mice treated of VEGFA production or not (Supplementary Fig. S4). Because with PTC299 but not in mice treated with doxorubicin (Supple- pyrimidine nucleotides can be synthesized by the de novo or the mentary Fig. S5C). salvage pathway (40), some cancer cells may be insensitive to In a second study, MOLM-13 AML cells were injected into the PTC299 due to increased pyrimidine nucleotide salvage activity. flank of nude mice to generate a solid tumor. Cytarabine (AraC) is Thus, we determined the IC50 (concentration to inhibit cell a deoxycytidine analog used as a standard chemotherapeutic proliferation by 50%) of PTC299 against a panel of 240 tumor agent to treat AML (42). As shown in Fig. 5F, PTC299 treatment cell lines. We classified 68 cells lines as responders and 172 cells alone resulted in a significant delay of tumor growth when lines as nonresponders based on an arbitrary IC50 cutoff of 1 compared with vehicle or AraC treatment; the median time to mmol/L. Based on these results, sensitivity to PTC299 was reach a tumor volume of 1,000 mm3 was 23 days for PTC299 observed across different tumor types (Supplementary Fig. versus 7 and 13 days for vehicle- and AraC-treated mice, respec- S5A). When grouped as hematopoietic versus solid tumor cells, tively. The combination of PTC299 with AraC further delayed 57% (32/56) of hematopoietic lines were sensitive to PTC299, tumor growth for 39 days. The increased activity of the combi- whereas only 18% (33/184) in the solid tumor lines were sensitive nation of AraC with PTC299 may reflect a more efficient incor- to PTC299. poration of AraC into DNA due to a reduction in the levels of Gene expression of pyrimidine nucleotide synthesis enzymes competing endogenous pyrimidine nucleotides. was analyzed using published microarray data in the Cancer Cell Line Encyclopedia (CCLE) database (41). Of 240 cell lines tested, 232 were found in the CCLE database including 54 hematopoietic Discussion and 178 solid tumor cell lines. Two salvage enzymes, CDA and Tumor cells require the upregulation and integration of mul- uridine phosphorylase 1 (UPP1) that convert cytidine and uracil, tiple biosynthetic processes including synthesis of nucleotides respectively, to uridine, were significantly less expressed (3.2-fold and proteins to drive uncontrolled cell proliferation (43). Here,

12 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

PTC299 Is a Novel Potent DHODH Inhibitor

Figure 5.

PTC299 demonstrates broad activity and potent inhibition of leukemia cell proliferation in vitro and in animal models. A, Box plot of log2 expression level (microarray) for CDA and UPP1, two pyrimidine salvage enzymes for converting cytidine and uracil, respectively, to uridine. The box represents the lower and upper quartile of the data. The white line in the box represents the median value. The whiskers represent the extension of data up to 1.5x of interquartile range (IQR) below the lower quartile or above the upper quartile. P values are calculated with Software R (https://www.r-project.org/, version 3.3.1) based on t test comparing hematopoietic tumors (n ¼ 54) and solid tumors (n ¼ 178); B, PTC299-sensitive cells in general have less salvage UMP production. Four sensitive (HeLa, HT1080, A549, and K562) and four insensitive lines (U87MG, MCF7, PC3, and Huh7) were cultured in the presence of 15N-Glutamine for 8 hours, then the unlabeled and 15N-labeled UMP in the cell lysates were measured by LC-MS/MS analysis. Data are mean SEM, n ¼ 4, P value from t test. C, PTC299 inhibits the proliferation of MOLM13 AML cells more potently than do other known DHODH inhibitors. Cells were treated with indicated compounds for 72 hours, and cell proliferation was measured using Celltiter Glo (Promega). % inhibition was calculated against the vehicle (0.5% DMSO) treatment. Data are mean SD, n ¼ 2. D, PTC299 is active in all ﬁve AML patient-derived cells in vitro. Patient-derived AML cells were treated with PTC299 for 72 hours, and then cell proliferation was assessed using Celltiter Glo (Promega). Assays were done in duplicate (AML-002, 004, and 005) or triplicate (AML-001 and 003). E, PTC299 prolongs survival of mice bearing human ALL after i.v. inoculation of NOD/SCID mice with MOLT-4 cells (10 mice/group). Treatment with PTC299 (10 mg/kg, qd), or with doxorubicin (0.5 mg/kg or 1 mg/kg i.p. once per week), was initiated 7 days after tumor inoculation; F, PTC299 inhibits MOLM-13 AML xenograft tumor growth in nude mice. Values shown represent the mean of 8 mice per group, and error bars show the SEM. www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 13

Cao et al.

we demonstrate that decreased translation of VEGFA mRNA by with side effects believed to be due to their off-target activities such PTC299 is linked to the inhibition of de novo pyrimidine nucle- as inhibition of kinases (21, 22). Recently, the inhibition of otide synthesis as exogenously added uridine, but not other DHODH by GSK983, which is structurally related to PTC299, nucleosides, can completely block the inhibition of VEGFA pro- was reported (51). However, we determined that this compound duction by PTC299. Unlike uridine, cytidine only rescued the cell- has a considerably lower bioavailability in mice than does cycle arrest in S-phase but not the inhibition of VEGFA produc- PTC299 (Supplementary Fig. S7). tion. This discrepancy may be due to the lower levels of CDA PTC299 represents a novel class of DHODH inhibitors with expression (Fig. 2A; Supplementary Fig. S2C), and cytidine could favorable biochemical and pharmacologic properties that may not be efficiently converted to uridine in the HT1080 cells. distinguish it from previously published DHODH inhibitors: (1) Moreover, exogenously added cytidine in cells expressing low Unlike the well-known DHODH inhibitors teriflunomide and levels of CDA could be quickly converted to CTP/dCTP which in brequinar, PTC299 did not show off-target effects including any turn might further inhibit the CDA and other salvage enzymes overt effects on 205 kinases and 62 Novascreen pharmacologic (44, 45). targets tested (Supplementary Table S5) and did not result in Extensive biological characterization demonstrated that myeloid suppression in patients with solid tumors (28, 29); (2) DHODH is the primary target of PTC299 as all the observed PTC299 has been extensively tested and well-tolerated in preclin- activities of PTC299 can be completely rescued by addition of ical studies. PTC299 did not inhibit the proliferation of normal exogenous uridine rather than other nucleosides. The proteomic primary cells including hematopoietic stem cells at concentra- study showed other potential targets may exist, and the selectivity tions 1,000-fold greater than its IC50 in tumor cells (Supplemen- and significance of those proteins identified in the proteomic tary Table S5); (3) DHODH inhibitors were identified by others study warrant further investigation. using in vitro enzymatic assay in which the purified enzyme Similar to PTC299, other DHODH inhibitors such as brequinar conformation may differ from that found in the cells due to lack also inhibit VEGFA production albeit less potently (Supplemen- of lipids in the assay (25, 26). The potent activity of PTC299 in tary Fig. S6B and S6C). The link between the inhibition of VEGFA mitochondrial preparations is likely due to the presence of the mRNA translation and de novo pyrimidine synthesis was unan- lipid bilayer when the enzyme is in the mitochondrial membrane ticipated but may be critical for cancer cells to synchronize rapid that may facilitate the entry and/or docking of the lipophilic proliferation and angiogenesis for tumor progression. Inhibition PTC299 into the binding pocket of the DHODH enzyme. This of VEGFA mRNA translation by PTC299 is mediated by the 50- is consistent with the recent reports that interactions with the UTR of the mRNA prompting multiple hypotheses for this link: small lipophilic pocket in DHODH are critical for potent inhi- (1) Depletion of uridine nucleotide pools may affect certain bition of its enzymatic activity (23, 26); (4) PTC299-resistant translation initiation factor(s) required for the noncanonical HT1080 cells are not cross-resistant to DHODH inhibitors such as protein synthesis of stress-regulated mRNAs such as the VEGFA brequinar or teriflunomide (Supplementary Fig. S6), indicating mRNA. We observed that eIF1AY gene expression decreased over that the mode of action for PTC299 is different from that of other 90% in PTC299-resistant HT1080 cells, when compared with DHODH inhibitors even though they target the same docking site; wild-type HT1080 cells (Fig. 2A; Supplementary Fig. S2). eIF1A (5) It was demonstrated that inhibition of DHODH activates p53 controls translation start codon recognition and has been expression resulting in inhibition of cancer cell proliferation and reported to affect IRES-mediated translation (46). (2) Low levels apoptosis (11). However, we analyzed the p53 expression in the of uridine nucleotides may also affect the mRNA substrate direct- CCLE database and found no correlation of p53 expression with ly. For example, depletion of pyrimidine nucleotides in cancer the activity of PTC299 (IC50) in those cell lines. Dr. Kensuke cells can cause a starvation-like stress response which may lead to Kojima (Kyoto University, Japan) tested PTC299 in a panel of changes in pseudouridylation state of mRNA (47). Reduction of leukemia cell lines and also found that PTC299 activity does not dyskerin expression, a component of pseudouridine synthase, has correlate with the expression of p53 (personal communication); been shown to selectively increase VEGFA-IRES–mediated trans- and finally, (6) a docking model shows that PTC299 binds lation (48). As uridines in RNAs are substrates for dyskerin, DHODH in the same pocket and with similar strength as does depletion of uridine nucleotides by PTC299 may reduce free brequinar analog C44. However, the interaction of PTC299 with substrate competition and thus activate dyskerin to inhibit VEGFA the binding pocket is mainly a "hand-in-glove," nonpolar lipo- mRNA translation. (3) A recent study has demonstrated that the philic interaction (Fig. 3H). These findings demonstrate that activity of mTOR, which controls synthesis of proteins and PTC299 is structurally and functionally different from previously nucleotides (5), can be regulated by nucleotide pools (49). Thus, published DHODH inhibitors, thus it may be a useful new tool it is also possible that inhibition of the translation of regulated for preclinical and clinical investigation of targeting DHODH for mRNAs, such as the VEGFA mRNA, is mediated via disrupting the therapeutic uses in treatment of cancer and immune diseases. production of pyrimidine nucleotides causing imbalances in The elucidation of the mechanism of action of PTC299 will nucleotide pools. Elucidating the molecular mechanisms linking facilitate the optimization of biomarker identification, patient inhibition of DHODH and VEGFA mRNA translation requires selection, and design of combination therapies with other drug(s) further investigation. for clinical development. Here, our data together point to the Two inhibitors of human DHODH have been approved for use potential clinical utility of PTC299 for the treatment of cancers in autoimmune diseases, leflunomide for rheumatoid arthritis, such as leukemia and lymphoma which in general have lower and its active metabolite, teriflunomide for multiple sclerosis pyrimidine salvage activity and thus depend on de novo pyrimi- (50). Brequinar, another potent inhibitor of human DHODH, dine synthesis for uncontrolled proliferation. PTC299 was placed was tested in the 1990s in clinical trials for solid tumors but was on clinical hold following the occurrence of severe hepatotoxicity associated with myeloid suppression and had a limited thera- in 2 patients of 279 individuals administered with PTC299 (28, peutic window (18, 22). These DHODH inhibitors are associated 29). Based on recent preclinical assessments demonstrating

14 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

PTC299 Is a Novel Potent DHODH Inhibitor

antitumor activity in AML models at low doses of PTC299, we are C. Romfo, J.D. Graci, J. Sheedy, J. Hedrick, N. Risher, S. Yeh, H. Qi, T. Arasu, reevaluating the potential use of PTC299 in leukemia and lym- R. Kong, A. Jacobson, J.M. Colacino, S.W. Peltz phoma, anticipating that efficacious concentration of PTC299 will Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): L. Cao, M. Weetall, C. Trotta, J. Ma, C. Romfo, be lower than those achieved in previous clinical studies in solid J.D. Graci, W. Li, J. Du, S. Yeh, H. Qi, T. Arasu, R. Kong, J. Petruska, A. Jacobson, tumors. A. Branstrom, J.M. Colacino, S.W. Peltz Writing, review, and/or revision of the manuscript: L. Cao, M. Weetall, Conclusion C. Trotta, J. Ma, M.J. Kim, B. Furia, C. Romfo, J.D. Graci, W. Li, J. Du, J. Petruska, J. Babiak, N.G. Almstead, A. Branstrom, J.M. Colacino, S.W. Peltz PTC299 is a novel potent inhibitor of DHODH that is differ- Administrative, technical, or material support (i.e., reporting or organizing entiated from and has advantages over other DHODH inhibitors. data, constructing databases): L. Cao, M. Weetall, C. Romfo, H. Qi, A. Jacobson, Studies demonstrated the target engagement of PTC299 in N.G. Almstead, A. Branstrom, J.M. Colacino patients, thus supporting its potential for treatment of hemato- Study supervision: L. Cao, M. Weetall, J. Sheedy, R. Kong, Y.-C. Moon, J. Babiak, N.G. Almstead, J.M. Colacino, S.W. Peltz logic malignancies which rely on de novo pyrimidine nucleotide Other (performed laboratory experiments): K. Cintron synthesis for survival and proliferation. Other (medicinal chemist): H. Qi Other (originally designed and synthesized PTC299 molecule): T. Arasu Disclosure of Potential Conflicts of Interest The authors are or were employed by PTC Therapeutic and have received Acknowledgments salary compensation for time, effort, and hold or held financial interest in the We thank all the members of the Discovery group at PTC for the thoughtful company. Travel expenses related to attendance at scientific conferences to discussion and constructive advises throughout the PTC299 program. We thank present data were reimbursed by PTC Therapeutics, Inc. All patents related to the Jana Narasimhan for a critical reading and discussion of this article. This work presented here are held by PTC Therapeutics, Inc. program was partially supported by funding from NCI to J.M. Colacino and L. Cao (2 R44 CA108330-02). Authors' Contributions Conception and design: L. Cao, M. Weetall, C. Trotta, C. Romfo, T. Arasu, S. The costs of publication of this article were defrayed in part by the payment of Hwang, R. Kong, T.W. Davis, J.M. Colacino, S.W. Peltz page charges. This article must therefore be hereby marked advertisement in Development of methodology: L. Cao, M. Weetall, J. Ma, M.J. Kim, B. Furia, C. accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Romfo, T. Arasu, S. Hwang, W. Lennox, R. Kong, S.W. Peltz Acquisition of data (provided animals, acquired and managed patients, Received August 1, 2018; revised August 22, 2018; accepted October 17, 2018; provided facilities, etc.): L. Cao, M. Weetall, J. Ma, M.J. Kim, B. Furia, published first October 23, 2018.

References 1. Martinez-Outschoorn UE, Peiris-Pages M, Pestell RG, Sotgia F, Lisanti MP. 12. Dorasamy MS, Choudhary B, Nellore K, Subramanya H, Wong PF. Cancer metabolism: a therapeutic perspective. Nat Rev Clin Oncol Dihydroorotate dehydrogenase inhibitors target c-Myc and arrest mel- 2017;14:113. anoma, myeloma and lymphoma cells at S-phase. J Cancer 2017;8: 2. Brown KK, Spinelli JB, Asara JM, Toker A. Adaptive reprogramming of de 3086–98. novo pyrimidine synthesis is a metabolic vulnerability in triple-negative 13. Zhang X, Yang J, Chen M, Li L, Huan F, Li A, et al. Metabolomics profiles breast cancer. Cancer Discov 2017;7:391–9. delineate uridine deficiency contributes to mitochondria-mediated apo- 3. McMahon SB. Control of nucleotide biosynthesis by the MYC oncoprotein. ptosis induced by celastrol in human acute promyelocytic leukemia cells. Cell Cycle 2008;7:2275–6. Oncotarget 2016;7:46557–72. 4. Mathur D, Stratikopoulos E, Ozturk S, Steinbach N, Pegno S, Schoenfeld 14. Sykes DB, Kfoury YS, Mercier FE, Wawer MJ, Law JM, Haynes MK, et al. S, et al. PTEN regulates glutamine flux to pyrimidine synthesis and Inhibition of dihydroorotate dehydrogenase overcomes differentiation sensitivity to dihydroorotate dehydrogenase inhibition. Cancer Discov blockade in acute myeloid leukemia. Cell 2016;167:171–86. 2017;7:380–90. 15. Lewis TA, Sykes DB, Law JM, Munoz B, Rustiguel JK, Nonato MC, et al. 5. Robitaille AM, Christen S, Shimobayashi M, Cornu M, Fava LL, Moes S, Development of ML390: a human DHODH inhibitor that induces differ- et al. Quantitative phosphoproteomics reveal mTORC1 activates de novo entiation in acute myeloid leukemia. ACS Med Chem Lett 2016;7:1112–7. pyrimidine synthesis. Science 2013;339:1320–3. 16. Wu D, Wang W, Chen W, Lian F, Lang L, Huang Y, et al. Pharmacologic 6. Rawls J, Knecht W, Diekert K, Lill R, Loffler M. Requirements for the inhibition of dihydroorotate dehydrogenase induces apoptosis and mitochondrial import and localization of dihydroorotate dehydrogenase. differentiation in acute myeloid leukemia cells. Haematologica 2018; Eur J Biochem 2000;267:2079–87. 103:1472–83. 7. Vyas VK, Ghate M. Recent developments in the medicinal chemistry and 17. Baumann P, Mandl-Weber S, Volkl A, Adam C, Bumeder I, Oduncu F, et al. therapeutic potential of dihydroorotate dehydrogenase (DHODH) inhi- Dihydroorotate dehydrogenase inhibitor A771726 (leflunomide) induces bitors. Mini Rev Med Chem 2011;11:1039–55. apoptosis and diminishes proliferation of multiple myeloma cells. Mol 8. Hoffmann HH, Kunz A, Simon VA, Palese P, Shaw ML. Broad-spectrum Cancer Ther 2009;8:366–75. antiviral that interferes with de novo pyrimidine biosynthesis. Proc Natl 18. Cody R, Stewart D, DeForni M, Moore M, Dallaire B, Azarnia N, et al. Acad Sci U S A 2011;108:5777–82. Multicenter phase II study of brequinar sodium in patients with advanced 9. Koundinya M, Sudhalter J, Courjaud A, Lionne B, Touyer G, Bonnet L, et al. breast cancer. Am J Clin Oncol 1993;16:526–8. Dependence on the pyrimidine biosynthetic enzyme DHODH is a syn- 19. Fitzpatrick LR, Small JS, Doblhofer R, Ammendola A. Vidofludimus thetic lethal vulnerability in mutant KRAS-driven cancers. Cell Chem Biol inhibits colonic interleukin-17 and improves hapten-induced colitis in 2018;25:705–717.e11. rats by a unique dual mode of action. J Pharmacol Exp Ther 2012; 10. Mohamad Fairus AK, Choudhary B, Hosahalli S, Kavitha N, Shatrah O. 342:850–60. Dihydroorotate dehydrogenase (DHODH) inhibitors affect ATP deple- 20. Arteaga CL, Brown TD, Kuhn JG, Shen HS, O'Rourke TJ, Beougher K, et al. tion, endogenous ROS and mediate S-phase arrest in breast cancer cells. Phase I clinical and pharmacokinetic trial of Brequinar sodium (DuP 785; Biochimie 2017;135:154–63. NSC 368390). Cancer Res 1989;49:4648–53. 11. Ladds MJGW, van Leeuwen IMM, Drummond CJ, Chu S, Healy AR, 21. Xu X, Williams JW, Gong H, Finnegan A, Chong AS. Two activities of the Popova G, et al. A DHODH inhibitor increases p53 synthesis and immunosuppressive metabolite of leflunomide, A77 1726. Inhibition of enhances tumor cell killing by p53 degradation blockage. Nat Commun pyrimidine nucleotide synthesis and protein tyrosine phosphorylation. 2018;9:1107. Biochem Pharmacol 1996;52:527–34.

www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 15

Cao et al.

22.XuX,WilliamsJW,ShenJ,GongH,YinDP,BlinderL,etal.In 36. Ohnuma T, Roboz J, Shapiro ML, Holland JF. Pharmacological and vitro and in vivo mechanisms of action of the antiproliferative and biochemical effects of pyrazofurin in humans. Cancer Res 1977;37: immunosuppressive agent, brequinar sodium. J Immunol 1998;160: 2043–9. 846–53. 37. Jain AN. Effects of protein conformation in docking: improved pose 23. Sainas S, Pippione AC, Giorgis M, Lupino E, Goyal P, Ramondetti C, prediction through protein pocket adaptation. J Comput Aided Mol Des et al. Design, synthesis, biological evaluation and X-ray structural 2009;23:355–74. studies of potent human dihydroorotate dehydrogenase inhibitors 38. Di RF, Azzaro MP, Palumbo GA, Bagnato S, Stagno F, Giustolisi GM, et al. based on hydroxylated azole scaffolds. Eur J Med Chem 2017;129: Elevated vascular endothelial growth factor (VEGF) serum levels in idio- 287–302. pathic myelofibrosis. Leukemia 2001;15:976–80. 24. Lucas-Hourani M, Munier-Lehmann H, El MF, Malmquist NA, Harpon J, 39. Duley JA, Henman MG, Carpenter KH, Bamshad MJ, Marshall GA, Ooi CY, Coutant EP, et al. Original 2-(3-Alkoxy-1H-pyrazol-1-yl)azines inhibitors et al. Elevated plasma dihydroorotate in Miller syndrome: Biochemical, of human dihydroorotate dehydrogenase (DHODH). J Med Chem diagnostic and clinical implications, and treatment with uridine. Mol 2015;58:5579–98. Genet Metab 2016;119:83–90. 25. Basso LGM, Mendes LFS, Costa-Filho AJ. The two sides of a lipid-protein 40. Lane AN, Fan TW. Regulation of mammalian nucleotide metabolism and story. Biophys Rev 2016;8:179–91. biosynthesis. Nucleic Acids Res 2015;43:2466–85. 26. Costeira-Paulo J, Gault J, Popova G, Ladds MJGW, van Leeuwen IMM, Sarr 41. Barretina J, Caponigro G, Stransky N, Venkatesan K, Margolin AA, Kim S, M, et al. Lipids shape the electron acceptor-binding site of the peripheral et al. The Cancer Cell Line Encyclopedia enables predictive modelling of membrane protein dihydroorotate dehydrogenase. Cell Chem Biol anticancer drug sensitivity. Nature 2012;483:603–7. 2018;25:309–17. 42. Wang X, Harrison JS, Studzinski GP. Enhancement of arabinocytosine 27. CaoL,WeetallM,BombardJ,QiH,ArasuT,LennoxW,etal.Discovery (AraC) toxicity to AML cells by a differentiation agent combination. of novel small molecule inhibitors of expression in tumor cells using a J Steroid Biochem Mol Biol 2016;164:72–8. cell-based high throughput screening platform. PLoS One 2016;11: 43. Cunningham JT, Moreno MV, Lodi A, Ronen SM, Ruggero D. Protein and e0168366. nucleotide biosynthesis are coupled by a single rate-limiting enzyme, 28. Weetall M, Davis T, Elfring G, Northcutt V, Cao L, Moon YC, et al. Phase 1 PRPS2, to drive cancer. Cell 2014;157:1088–103. study of safety, tolerability, and pharmacokinetics of PTC299, an inhibitor 44. Ipata PL, Falcone G, Serra MC. Some regulatory properties of cytidine of stress-regulated protein translation. Clin Pharmacol Drug Dev 2016; deaminase and uridine phosphorylase of Bacillus cereus. FEBS Lett 1970; 5:296–305. 10:67–70. 29. Packer RJ, Rood BR, Turner DC, Stewart CF, Fisher M, Smith C, et al. Phase I 45. Gemble S, Ahuja A, Buhagiar-Labarchede G, Onclercq-Delic R, Dairou J, and pharmacokinetic trial of PTC299 in pediatric patients with refractory or Biard DS, et al. Pyrimidine pool disequilibrium induced by a cytidine recurrent central nervous system tumors: a PBTC study. J Neurooncol deaminase deficiency inhibits PARP-1 activity, leading to the under rep- 2015;121:217–24. lication of DNA. PLoS Genet 2015;11:e1005384. 30. Clayton DA, Shadel GS. Isolation of mitochondria from cells and tissues. 46. Jaafar ZA, Oguro A, Nakamura Y, Kieft JS. Translation initiation by the Cold Spring Harb Protoc 2014;2014:db. hepatitis C virus IRES requires eIF1A and ribosomal complex remodeling. 31. Chen SF, Perrella FW, Behrens DL, Papp LM. Inhibition of dihydroor- Elife 2016;5. otate dehydrogenase activity by brequinar sodium. Cancer Res 1992; 47. Wu G, Radwan MK, Xiao M, Adachi H, Fan J, Yu YT. The TOR signaling 52:3521–7. pathway regulates starvation-induced pseudouridylation of yeast U2 32. Ruppert J, Welch W, Jain AN. Automatic identification and representa- snRNA. RNA 2016;22:1146–52. tion of protein binding sites for molecular docking. Protein Sci 1997; 48. Rocchi L, Pacilli A, Sethi R, Penzo M, Schneider RJ, Trere D, et al. Dyskerin 6:524–33. depletion increases VEGF mRNA internal ribosome entry site-mediated 33. Cheatham TE III, Cieplak P, Kollman PA. A modified version of the Cornell translation. Nucleic Acids Res 2013;41:8308–18. et al. force field with improved sugar pucker phases and helical repeat. J 49. Emmanuel N, Ragunathan S, Shan Q, Wang F, Giannakou A, Huser N, et al. Biomol Struct Dyn 1999;16:845–62. Purine nucleotide availability regulates mTORC1 activity through the Rheb 34. Prewett M, Huber J, Li Y, Santiago A, O'Connor W, King K, et al. Anti- GTPase. Cell Rep 2017;19:2665–80. vascular endothelial growth factor receptor (fetal liver kinase 1) mono- 50. Schiff MH, Strand V, Oed C, Loew-Friedrich I. Leflunomide: efficacy and clonal antibody inhibits tumor angiogenesis and growth of several mouse safety in clinical trials for the treatment of rheumatoid arthritis. Drugs and human tumors. Cancer Res 1999;59:5209–18. Today (Barc) 2000;36:383–94. 35. Lichtenbeld HC, Ferarra N, Jain RK, Munn LL. Effect of local anti-VEGF 51. Deans RM, Morgens DW, Okesli A, Pillay S, Horlbeck MA, Kampmann M, antibody treatment on tumor microvessel permeability. Microvasc Res et al. Parallel shRNA and CRISPR-Cas9 screens enable antiviral drug target 1999;57:357–62. identification. Nat Chem Biol 2016;12:361–6.

16 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

Targeting of Hematologic Malignancies with PTC299, A Novel Potent Inhibitor of Dihydroorotate Dehydrogenase with Favorable Pharmaceutical Properties

Liangxian Cao, Marla Weetall, Christopher Trotta, et al.

Mol Cancer Ther 2019;18:3-16. Published OnlineFirst October 23, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-18-0863

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2018/10/23/1535-7163.MCT-18-0863.DC1

Cited articles This article cites 49 articles, 12 of which you can access for free at: http://mct.aacrjournals.org/content/18/1/3.full#ref-list-1

Citing articles This article has been cited by 7 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/18/1/3.full#related-urls

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 Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/18/1/3. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.