From the Department of Oncology-Pathology Karolinska Institutet, Stockholm, Sweden EXPLORING NOVEL ROLES OF METABOLIC ENZYMES MTHFD2 AND PFKFB3 IN CANCER GENOME STABILITY AND THEIR POTENTIAL AS ANTICANCER THERAPEUTIC TARGETS Nadilly Y. Bonagas Villarreal Stockholm 2021 All previously published papers were reproduced with permission from the publisher. Published by Karolinska Institutet. Printed by Universitetsservice US-AB, 2021 © Nadilly Y. Bonagas Villarreal, 2021 ISBN 978-91-8016-203-6 Cover illustration by Sophia Ceder (http://ceder.graphics/) Exploring novel roles of metabolic enzymes MTHFD2 and PFKFB3 in cancer genome stability and their potential as anticancer therapeutic targets THESIS FOR DOCTORAL DEGREE (Ph.D.) By Nadilly Y. Bonagas Villarreal To be publicly defended at Air & Fire, Science for Life Laboratory, Tomtebodavägen 23a Friday, 4th of June 2021 at 9:30 am Principal Supervisor: Opponent: Professor Thomas Helleday Professor Chris Lord Karolinska Institutet University of London Department of Oncology-Pathology Institute of Cancer Research Division of Breast Cancer Research Co-supervisor(s): Dr. Oliver Mortusewicz Examination Board: Karolinska Institutet Professor Sonia Lain Department of Oncology-Pathology Karolinska Institutet Department of Microbiology, Tumor and Cell Dr. Nina Gustafsson Biology Karolinska Institutet Department of Oncology-Pathology Professor Klas Wiman Karolinska Institutet Department of Oncology-Pathology Professor Bo Stenerlöw Uppsala University Department of Immunology, Genetics and Pathology “No digáis que, agotado su tesoro, de asuntos falta, enmudeció la lira; podrá no haber poetas; pero siempre habrá poesía. … Mientras la ciencia a descubrir no alcance las fuentes de la vida, y en el mar o en el cielo haya un abismo que al cálculo resista, mientras la humanidad siempre avanzando no sepa a dó camina, mientras haya un misterio para el hombre, ¡habrá poesía!” – Gustavo Adolfo Bécquer POPULAR SCIENCE SUMMARY OF THE THESIS According to the World Health Organization (WHO), cancer remains the second leading cause of deaths around the world. They estimate that about one in three people will develop cancer during their lives. Despite our best efforts to develop better treatments, our current ones still cause great discomfort and suffering to the patients and their families. Approximately 90% of all cancer patients receive radio- and/or chemotherapy, which have considerable side effects. These harmful effects are usually caused by the treatment not being able to tell the difference between cancer and normal cells, which results in healthy tissue being affected as well. This highlights the need for more precise therapies with fewer side effects. Previous studies show that certain proteins, such as PFKFB3 and MTHFD2, are commonly overproduced in cancer cells. While these proteins are normally involved in breaking down nutrients like sugars and folic acid to produce energy and cell components, we have discovered that in cancer they can have other functions like protecting the cell’s DNA, i.e., the master code with instructions to how the cells should work and how to make new cells. Under normal circumstances, MTHFD2 is almost non-existent in adult healthy tissue; it is present only in embryos before cells mature into specialized organs. In embryos, MTHFD2 is important for fast growth because it provides the building blocks for all the DNA being assembled in the new cells. These DNA building blocks, called nucleotides, are part of the reason pregnant women are encouraged to take more folic acid – the embryo’s MTHFD2 requires folic acid to make nucleotides. Cancer cells, like embryonic cells, divide extremely fast and find it convenient to reactivate MTHFD2 in order to make more DNA building blocks to support their fast growth. In the case of PFKFB3, we found that when DNA is damaged by radiotherapy for example, PFKFB3 helps recruit the repair proteins needed to make new DNA building blocks and fix the damage. If left unrepaired, the DNA damage would cause cell death, a mechanism put in place to prevent corrupted DNA code from being passed on to new cells. Therefore, tumors benefit from overproducing PFKFB3 to avoid death when their DNA is damaged. Our group has collaborated with the pharmaceutical company Kancera to test their recently developed anti-PFKFB3 drug, while we have generated several drugs of our own that inhibit MTHFD2. These drugs kill the cancer cells which rely on PFKFB3 or MTHFD2, while largely sparing the healthy cells which do not have as much of these proteins. Our hope is that we can use these new drugs to treat many types of cancer more precisely, alone or in combination with radiotherapy and other drugs, to reduce the side effects for patients and even sensitize cancers which have become resistant to therapy. ABSTRACT Altered tumor metabolism has been described as early as the 1920s, but it was only in recent decades that proteomic and metabolomic studies revealed that the ways in which tumors rewire their nutrient and energy pathways are more diverse and have more implications for treatment outcome than previously thought. There is now a great interest in characterizing promising metabolic targets and identifying novel ways by which to exploit them for cancer treatment. This thesis work is part of an ongoing effort to elucidate the molecular mechanisms behind metabolic cancer targets specifically at the interface of genome stability, their role in the pathogenesis of different tumor types and genetic contexts, and their suitability as drug targets for clinically relevant treatment strategies. In Paper I, we present a new role for the glycolysis enzyme 6-phosphofructo 2-kinase/fructose 2,6-bisphosphatase 3 (PFKFB3) in homologous recombination (HR). We used gene silencing and pharmacological inhibitors to investigate the role of PFKFB3 in the response to DNA damage induced by ionizing radiation (IR). We found that PFKFB3 promotes the recruitment of DNA repair factors and supplies nucleotides for DNA synthesis through its interaction with ribonucleotide reductase (RNR). We also validated the antitumor preclinical potential of PFKFB3 inhibitor KAN0438757 and showed it specifically sensitized cancer cells to IR. In Paper II, we solve the first crystal structure of human one-carbon metabolism enzyme methylenetetrahydrofolate dehydrogenase 2, methenyltetrahydrofolate cyclohydrolase (MTHFD2) in complex with its cofactors and a weak inhibitor, LY345899. We developed biochemical activity and target engagement assays to evaluate the binding and inhibition of MTHFD2 by LY345899 in cancer cell models. With the newfound structural insights to determine key residues important for substrate and cofactor binding, we were able to undertake a structure-based drug discovery program targeting MTHFD2 detailed in Paper III. Paper III expands on the groundwork laid out in Paper II to develop first-in-class, highly potent and cell active inhibitors of MTHFD2 (MTHFD2i). Again, using gene silencing techniques, we identified a novel role for MTHFD2 in genome maintenance, which we confirmed with our small molecule inhibitors. We show that MTHFD2i induce replication stress and apoptosis selectively in transformed cells as a result of impaired de novo thymidylate synthesis and genomic uracil misincorporation. We established an in vivo model of acute myeloid leukemia (AML) and showed that MTHFD2i significantly prolonged survival and outperformed the standard of care compound cytarabine (AraC), providing proof-of-concept for the translational potential of MTHFD2i as anticancer drugs. In Paper IV, we further elaborate on the role of MTHFD2 in genome maintenance in response to DNA damage. We found that MTHFD2 accumulates and associates to chromatin upon DNA double strand breaks (DSBs) and promotes DNA repair through HR. Loss of MTHFD2 significantly impairs HR activity, with MTHFD2i specifically sensitizing cancer cells to PARP inhibitors in vitro and delaying tumor growth when combined with a PARP inhibitor in vivo. Taken together, these studies showcase these two metabolic enzymes, PFKFB3 and MTHFD2, in a new light as novel DNA damage response (DDR) targets. Our findings provide compelling evidence to propose the intersection of cancer metabolism and genome stability as an untapped source of novel anticancer targets warranting more mechanistic and drug development efforts. LIST OF SCIENTIFIC PAPERS I. Gustafsson, N. M. S., Färnegårdh, K., Bonagas, N., Huguet Ninou, A., Groth, P., Wiita, E., Jönsson, M., Hallberg, K., Lehto, J., Pennisi, R., Martinsson, J., Norström, C., Hollers, J., Schultz, J., Andersson, M., Markova, N., Marttila, P., Kim, B., Norin, M., Olin, T., and Helleday, T. Targeting PFKFB3 radiosensitizes cancer cells and suppresses homologous recombination. Nat Commun (2018) 9, 3872 II. Gustafsson, R., Jemth, A. S., Gustafsson, N.M.S., Färnegårdh, K., Loseva, O., Wiita, E., Bonagas, N., Dahllund, L., Llona-Minguez, S., Häggblad, M., Henriksson, M., Andersson, Y., Homan, E., Helleday, T., and Stenmark P. Crystal structure of the emerging cancer target MTHFD2 in complex with a substrate-based inhibitor. Cancer Res (2017) 77, 937-948 III. Gustafsson, N. M. S.*, Bonagas, N.*, Henriksson, M.*, Wiita, E., Gustafsson, R., Marttila, P., Borhade, S., Vallin, K., Sarno, A., Svensson, R., Göktürk, C., Pham, T., Jemth, A. S., Loseva, O., Green, A. C., Cookson, V., Sandberg, L., Rasti, A., Unterlass, J. E., Haraldsson, M., Andersson, Y., Scaletti, E. R., Bengtsson, C., Paulin, C. B. J., Sanjiv, K., Abdurakhmanov,