NAD+ Kinase As a Therapeutic Target in Cancer Philip M

Home , Shikimic acid

Author Manuscript Published OnlineFirst on August 31, 2016; DOI: 10.1158/1078-0432.CCR-16-1129 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

NAD+ Kinase as a Therapeutic Target in Cancer Philip M. Tedeschi, Nitu Bansal, John E. Kerrigan, Emine E. Abali, Kathleen W. Scotto, and Joseph R. Bertino

Departments of Medicine and Pharmacology, Rutgers Robert Wood Johnson Medical School the Rutgers Cancer Institute of New Jersey and the Rutgers Graduate School of Biomedical Sciences New Brunswick, New Jersey

Corresponding Authors: Joseph R. Bertino, The Cancer Institute of New Jersey, 195 Little Albany Street, Room 3033, New Brunswick, NJ 08903. Phone: 732-235-5810; Fax: 732-235-8181; E-mail: [email protected], and Kathleen W. Scotto, The Cancer Institute of New Jersey, 195 Little Albany Street, New Brunswick, NJ 08903. Phone: 732-235-6245; E-mail: scottoka@ cinj.rutgers. edu

Running Title: NAD+ Kinase as a Therapeutic Target in Cancer

Disclosure of Potential Conflicts of Interest: No potential conflicts of interest were disclosed.

Downloaded from clincancerres.aacrjournals.org on September 28, 2021. © 2016 American Association for Cancer Research. Author Manuscript Published OnlineFirst on August 31, 2016; DOI: 10.1158/1078-0432.CCR-16-1129 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Abstract NAD+ kinase (NADK) catalyzes the phosphorylation of nicotinamide adenine dinucleotide (NAD+) to nicotinamide adenine dinucleotide phosphate (NADP+) using ATP as the phosphate donor. NADP+ is then reduced to NADPH by dehydrogenases, in particular glucose-6-phosphate dehydrogenase and the malic enzymes. NADPH functions as an important cofactor in a variety of metabolic and biosynthetic pathways. The demand for NADPH is particularly high in proliferating cancer cells where it acts as a cofactor for the synthesis of nucleotides, proteins and fatty acids. Moreover, NADPH is essential for the neutralization of the dangerously high levels of reactive oxygen species (ROS) generated by increased metabolic activity. Given its key role in metabolism and regulation of ROS, it is not surprising that several recent studies, including in vitro and in vivo assays of tumor growth and querying of patient samples have identified NADK as a potential therapeutic target for the treatment of cancer In this review, we will discuss the experimental evidence justifying further exploration of NADK as a clinically-relevant drug target, and describe our studies with a lead compound, thionicotinamide (TN), an NADK inhibitor prodrug.

Introduction Highly proliferating cancer cells require sufficient amounts of NADH and NADPH to act as reducing agents for reductive synthesis of nucleic acids, protein and lipid biosynthesis (Figure 1A). A lack of these precursors can lead to a halt in cell growth and eventual cell death. NADPH is also key to the maintenance of a healthy redox status in cells (1, 2), since it is involved in neutralizing the reactive oxygen species (ROS) associated with rapid growth. To achieve the metabolic and ROS-mediating requirements associated with rapid proliferation, cancer cells can alter the expression and regulation of metabolic genes. Indeed, several of these genes are under the regulation of oncogenes and tumor suppressor proteins. For example, the tumor suppressor p53 exerts control of NADPH levels by inhibiting glucose-6 phosphate dehydrogenase (G6PD) activity, and by repressing activity of the malate dehydrogenase enzymes M1 and M2; both enzymes contribute to the cellular NADPH pool. (3, 4) (Figure 1A). When p53 is mutated or rendered non-functional through interaction with inhibitory proteins this control is lost; as a result, cancer cells generate more NADPH for protection from ROS and macromolecular synthesis to allow rapid proliferation. Another example is the M2 splice variant of pyruvate kinase (PKM2), which is subjected to complex regulation by both oncogenes and tumor suppressors; overexpression of PKM2 leads to the increased production of NADPH by diverting glucose metabolism into the pentose phosphate pathway (5). All cells harbor some level of ROS. Indeed, moderate levels of ROS are beneficial for cancer cells as they can contribute to increased cell proliferation and genomic instability (6-8). However, excessive ROS can cause oxidative damage to proteins, lipids and DNA, leading to cell death. This fine line between ROS-enhanced and decreased proliferation makes cancer cells vulnerable to stressors that alter this balance, including many chemotherapeutic drugs as well as radiation (9-13). The primary defense against ROS in cancer cells is the upregulation of superoxide dismutase and the NADPH- dependent glutathione (GSH) and thioredoxin redox cycles (14). The GSH system functions via glutathione peroxidase enzymes that inactivate H2O2 and other peroxides by conversion of GSH to reduced glutathione disulfide (GSSG). GSSG is then cycled back to GSH by glutathione reductase using NADPH. Myc, by increasing glutamine uptake, glutaminolysis and therefore GSH production, as well as by promoting the expression of PKM2, also contributes to redox control in cancer cells (15). Thus, the same proteins that when mutated drive the cancer phenotype, are also responsible for the altered metabolism and redox status that allows cancer cells to survive and proliferate. Decreased levels of NADPH are predicted to have a selective and two- pronged effect on tumor survival: inhibition of critical biosynthetic pathways and reduction in the ability of cancer cells to neutralize high ROS levels. A key player in the regulation of NADPH is NAD kinase (NADK), which generates NADP that is converted to NADPH by dehydrogenases. NADK was first identified by Arthur Kornberg in 1950 following isolation from Saccharomyces cerevisiae (16), and has since been purified from a

variety of organisms (17-20). Human cytoplasmic NADK (cNADK) is an obligate dimer, with a monomer size of about 50 kilodalton. Unlike several prokaryotic NADKs, human cNADK is unable to utilize inorganic phosphate; instead, the phosphorylation of NAD to NADP requires ATP and magnesium ion as a co-factor (21). Of note, the active site is composed of portions of both monomers, a distinctive feature of this enzyme (Figure 1B). While the structure and function of the human cNADK have been known for over a decade, recently a mitochondrial NADK (mNADK) has been discovered, and may play a predominant role in controlling the ROS generated in the mitochondria (22, 23)(vide infra). Most of the studies of NADK have focused on the enzyme in plants and bacteria. For example, NADK has been shown to play diverse roles in Arabidopsis thaliana, participating in nitrogen/carbon metabolism, photosynthetic efficiency and protection from oxidative stress (24-26). The key role of NADK in increasing NADPH levels and macromolecular synthesis has been exploited by overexpressing this enzyme to increase the output of a desired product (27-30) with less attention given to approaches to inhibit NADK activity. To date, most efforts to develop NADK inhibitors have primarily focused on pathogenic bacteria, due to differences in active site structure and substrates between prokaryotic and mammalian NADK (27); in this regard, a special effort has been made to identify NADK inhibitors for the treatment of multidrug resistant tuberculosis and malaria (31, 32). Inhibition of human NADK for treatment of diseases such as inflammation and cancer has only recently been proposed (33).

Mitochondrion-Localized NAD Kinase Despite the key role of NADPH, and therefore NADK in modulating cellular ROS, our studies showed that knockdown of cNADK had only a modest effect on ROS levels (34) and others have shown that overexpression of cNADK (cytoplasmic NADK) only resulted in minimal protection from ROS (35). Pollack et al suggested that human cells may depend more on the NADP+ dependent dehydrogenases to reduce NADP+, rather than levels of NADP+ for oxidative defense (35). These results may also be explained by the recent identification of a mitochondrial NADK (mNADK). As NADP is membrane impermeable, the existence of a mitochondrial NADK was long postulated but only recently characterized (21, 22). Mitochondrial NADH and NADPH are required for protection from oxidative stress as well as for the biosynthesis of macromolecules. Similar to cNADK, mNADK may play a major role in generating NADH and NADPH for protection of mitochondria from ROS. As tumor cells have increased levels of ROS, inhibition of mNADK might at first glance seem an effective strategy. However, this will have to be carefully tested, as the likely critical role of this enzyme for protection against ROS in the heart and brain could prove this approach too toxic for clinical use. Supporting this concern, a pediatric patient was recently identified with a deficiency in mNADK due to a homozygous nonsense mutation in exon 10 leading to a premature stop codon (36). By the time he was 4.5 years old, the patient developed severe encephalopathy, spastic quadriplegia, blindness and epilepsy, and died soon thereafter. Thus, cNADK may be a more feasible target for inhibitor development.

Cytoplasmic NADK expression in cancer. Using a gain-of-function screening platform, a novel cNADK mutant, NADK-I90F, was isolated from a panel of mutants originating in pancreatic ductal adenocarcinoma cancer patients (37). Our laboratory characterized the molecular structure and enzymatic activity of this mutant cNADK and found that, as compared to the wild type enzyme, cNADK-I90F had a higher Vmax and lower Km for NAD+ and ATP, indicating an increase in activity (37). Consistent with this, cells expressing this mutant had low ROS levels and elevated levels of NADPH (Figure 2). When the effect of NADK-I90F on cell growth and metabolism was evaluated in vitro and in vivo, it was found to drive transformation of normal pancreatic ductal cells due to its increased enzymatic activity. This further supports that cNADK, both wild type and certain mutant forms, are clinically relevant enzyme targets for cancer therapy. Notably, both over expression and several additional NADK mutants have been identified in a small percentage (<5%) in multiple tumor types, including pancreatic cancer in the The Cancer Genome Atlas (TCGA) database, but more information is required to determine the significance of the mutations reported.

The Search for Potent Inhibitors of Human NADK—Preclinical and Clinical Studies of NAD Analogs One of the first attempts to develop NADK inhibitors was made by Pankiewicz et al (32, 38), using a novel approach. Although over 30 clinically-approved kinase inhibitors have been developed that bind to the ATP binding site of various tyrosine protein kinases, these investigators reasoned that, since the ATP binding site in NADK is solvent- exposed and chelation with Mg++ is necessary for ATP binding, targeting the binding of ATP to NADK would not be an effective approach. Rather, they focused their efforts on inhibitors that targeted the NAD binding site in the enzyme (33). Although NAD kinases belong to a large family of NAD-utilizing enzymes with conserved NAD binding sites, the domain in human NADK was considered sufficiently different from that of most NAD- dependent enzymes to allow the identification of selective inhibitors. For example, the NAD binding site has been successfully targeted in other enzymes: tiazofurin, a potent inhibitor of inosine monophosphate dehydrogenase (IMPDH), is used to treat chronic myelogenous leukemia (CML), and mycophenolic acid (MPA), another IMPDH inhibitor, is used as an immunosuppressant (33). In collaboration with Pankowicz et al (32) we examined several NAD analogs. Of particular interest was benzamide riboside, which was anabolized to benzamide adenine nucleoside (BAD) in cells. We found that in addition to its known potent inhibition of inosine monophosphate dehydrogenase (IMPDH), it also inhibited NADK, and as a result markedly lowered NADPH levels resulting in a loss of dihydrofolate reductase (DHFR) levels, as this coenzyme protects DHFR from proteolysis (39) BAD inhibited growth of CEM lymphocytic cells at a low micromolar concentration. Unfortunately, although the precursor drug, benzamide riboside had encouraging preclinical anticancer activity, animal toxicity studies showed that treatment with BAD caused skeletal muscle loss and hepatotoxicity, precluding further drug development

Thionicotinamide. During a search of a small compound library for novel NAD analogs we identified nicotinamide adenine dinucleotide (NADS) and nicotinamide adenine dinucleotide phosphate (NADPS) and showed that the cytotoxic effect of NADS/NADPS was largely due to inhibition of NADK activity (40). As thionicotinamide (TN) is converted to NADPS upon entering the cell, we tested the possible role of TN as a prodrug inhibitor of NADK, and found that it was as effective as NADPS. Since TN is more readily available than NADS/NADPS, we pursued our analyses using TN as our lead compound. We examined several cell lines for sensitivity to TN. The T-cell lymphoma cell lines CEM-CCRF and MOLT-4, the diffuse large B cell lymphoma cell line RL, as well as the C85 colon cancer cell line were all sensitive, with IC50’s ~10uM. Inhibition of NADK activity by TN or shRNA-mediated knockdown decreased colony formation and the growth of C85 cells to a similar degree (notably, the colonies were small, suggesting effects on early precursors), supporting the hypothesis that it is the effect of TN on NADK as well as inhibition of glucose-6-phosphate dehydrogenase (34), that underlies the cytotoxic effects observed (Figure 3). TN decreased the levels of NADPH in C85 cells, resulting in a modest but reproducible increase in steady-state ROS levels, particularly when the oxidative stressor hydrogen peroxide was present (~3-fold increase). Importantly, other NADPH-reliant pathways, such as fatty acid synthesis and nucleotide synthesis, were also inhibited by TN (34). TN was also effective in sensitizing cells to several chemotherapeutic agents including methotrexate, gemcitabine, docetaxel and irinotecan (34). As there was an increase in the level of ROS as well as reduction in NADPH levels in cells treated with both TN and chemotherapy, the TN-induced loss of NADK activity/NADPH contributed to the synergistic effect. Thus, both inhibition of macromolecular synthesis and an increase in oxidative stress may play an important role in the toxicity of these combinations. The effectiveness of TN as a single agent was examined in a pilot study of two human xenograft models in mice. TN caused tumor regression and a protracted halt in tumor growth for the duration of observation (Figure 4). The human C85 colon cancer xenograft was treated with the same dose and similar schedule of TN as the RL lymphoma xenograft, and tumor response was compared to C85 xenografts in which NADK was downregulated by shRNA (~50%). Importantly, both TN and shRNA inhibited the growth of this tumor without toxicity, validating NADK as a key anti-cancer target (34). These promising results justify further development of NAD analogs for clinical use. The only NADK inhibitor that has been tested in the clinic is 6-amino nicotinamide, which had shown promising preclinical anticancer activity, although animal studies showed that neurologic toxicity was dose limiting (41). Unfortunately, neurotoxicity prevented dose escalation and minimal activity was seen at the maximum tolerated dose in clinical trials (42). The mechanism underlying the antitumor and neurotoxic effects of this drug are not clear, and have been ascribed to its ability to inhibit phosphogluconate dehydrogenase and hence the pentose phosphate pathway (43). However, an earlier report showed that it potently inhibited the phosphorylation of NAD in mitochondria (44), a more compelling explanation for the neurotoxic effects. , Nicotinamide was found to reverse these side effects, as well as its antitumor activity,

presumably by inhibiting uptake and or conversion of 6-amino nicotinamide to 6-amino NAD, its active form. Notably, in contrast, no neurotoxicity was observed with TN, indicating that, whatever the basis for this toxic effect with 6-amino nicotinamide, it is not a characteristic of all nicotinamide analogs.

Will Inhibitors of NADK Have Selective Anti-Cancer Activity? The most compelling evidence that NADK is a novel target for anticancer drug development comes from our studies shown above that knockdown of this enzyme in the human colon carcinoma cells inhibited clonogenic tumor cell proliferation in vitro and xenograft growth in vivo. Additionally, knockdown studies of cytoplasmic NADPH and inhibition of cNADK and G6PD by TN have not demonstrated neurotoxicity in mice at doses that inhibited tumor growth, suggesting that drugs that specifically target the cytoplasmic enzyme may have a safe therapeutic index. We conclude that inhibition of cNADK will have selective anticancer activity by reducing the NADPH pool available for biosynthetic pathways and protection from ROS. TN by decreasing NADPH levels also had the additional effect of destabilizing and thereby decreasing levels of dihydrofolate reductase (DHFR). As this decrease in DHFR contributed to the toxicity of TN, a combination of a DHFR inhibitor and inhibition of NADK by TN had synergistic anticancer activity in vitro (35). Moreover, synergistic cell kill was also observed when TN was used in combination with other drugs that inhibit DNA synthesis, likely due to the fact that both treatments lead to enhanced ROS and decreased macromolecular synthesis (3). The role of mitochondrial NADK in protecting cell from ROS and mitochondrial metabolism still needs to be clarified, as well as studies that compare the effect of NADK inhibitors on the cytoplasmic and mitochondrial enzyme. To identify inhibitors that select between cNADK and mNADK, as well as mutant forms of cNADK, two approaches are being used: a library screen, and rational design using a crystal structure of the enzyme (44). While the key binding site residues are conserved between the mitochondrial and cytoplasmic enzymes (21, 22), there are significant structural differences between the two enzymes that may be exploited (Figure 5). The conserved residues are D184, D383, L188, N284 and E285; Apart from the potential for difference in binding site shape and dynamics (induced fit), there are proximal differences where selectivity may be achieved. These include: the R385 residue in cNADK (glycine in mNADK); the D412 residue in mNADK (while this is D383 in cNADK, the residue side chain points away from the binding site); F211 provides a pi stack interaction for the adenine ring of NAD in cNADK (replaced by L192 in the mitochondrial enzyme) and the more subtle Y327 residue (tryptophan in mNADK not shown in figure), giving a larger pi stacking group for interaction with the nicotinamide ring of NAD. New leads obtained from the library screen will be modified to enhance selectivity and efficacy based upon what we learn from in silico docking to the crystal structure.

Conclusions

The in vitro and in vivo effects of NADK inhibition on cancer cell death, combined with the identification of an activating mutation in NADK that initiates carcinogenesis makes a strong case for developing inhibitors that lower NADPH tumor levels, and raises the possibility of finding specific inhibitors of this or other mutant NADK enzymes. The identification of new, more selective inhibitors will provide a tool to dissect the role of cNADK and mNADK in the cancer phenotype and provide a new class of clinically- relevant, broad-spectrum anti-cancer therapeutics for further development.

References

1. Cairns RA, Harris IS, Mak TW. Regulation of cancer cell metabolism. Nat Rev Cancer 2011;11:85-95. 2. Vander Heiden MG, Cantley LC, Thompson CB. Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 2009;324:1029-33. 3. Jiang P, Du W, Mancuso A, Wellen KE, Yang X. Reciprocal regulation of p53 and malic enzymes modulates metabolism and senescence. Nature 2013;493:689-93. 4. Jiang P, Du W, Wang X, Mancuso A, Gao X, Wu M, et al. p53 regulates biosynthesis through direct inactivation of glucose-6-phosphate dehydrogenase. Nat Cell Biol 2011;13:310-6. 5. Anastasiou D, Poulogiannis G, Asara JM, Boxer MB, Jiang JK, Shen M, et al. Inhibition of pyruvate kinase M2 by reactive oxygen species contributes to cellular antioxidant responses. Science 2011;334:1278-83. 6. Ying W. NAD+/NADH and NADP+/NADPH in cellular functions and cell death: regulation and biological consequences. Antiox Redox Signal 2008;10:179-206. 7. Boonstra J, Post JA. Molecular events associated with reactive oxygen species and cell cycle progression in mammalian cells. Gene 2004;337:1-13. 8. Schafer FQ, Buettner GR. Redox environment of the cell as viewed through the redox state of the glutathione disulfide/glutathione couple. Free Radic Biol Med 2001;30:1191-212. 9. Alexandre J, Nicco C, Chereau C, Laurent A, Weill B, Goldwasser F, et al. Improvement of the therapeutic index of anticancer drugs by the superoxide dismutase mimic mangafodipir. J Natl Cancer Inst 2006;98:236-44. 10. Fei ZH, Wu K, Chen YL, Wang B, Zhang SR, Ma SL. Capilliposide isolated from Lysimachia capillipes Hemsl. induces ROS generation, cell cycle arrest, and apoptosis in human nonsmall cell lung cancer cell lines. Evid Based Complement Alternat Med 2014;2014:497456. 11. Qian C, Wang Y, Zhong Y, Tang J, Zhang J, Li Z, et al. Wogonin-enhanced reactive oxygen species-induced apoptosis and potentiated cytotoxic effects of chemotherapeutic agents by suppression Nrf2-mediated signaling in HepG2 cells. Free Radic Res 2014;48:607-21. 12. Xinyuan Li PF, Jietang Mai, Eric T Choi, Hong Wang, Xiao-feng Yang. Targeting mitochondrial reactive oxygen species as novel therapy for inflammatory diseases and cancers. J Hematol Oncol. 2013;6:19. 13. Wondrak GT. Redox-directed cancer therapeutics: molecular mechanisms and opportunities. Antioxid Redox Signal 2009;11:3013-69.

14. Buldak RJ, Buldak L, Kukla M, Gabriel A, Zwirska-Korczala K. Significance of selected antioxidant enzymes in cancer cell progression. Pol J Pathol 2014;65:167-75. 15. Liu YC, Li F, Handler J, Huang CR, Xiang Y, Neretti N, et al. Global regulation of nucleotide biosynthetic genes by c-Myc. PLoS One 2008;3:e2722. 16. Kornberg A, Pricer WE, Jr. On the structure of triphosphopyridine nucleotide. The J Biol Chem 1950;186:557-67. 17. Apps DK. Pigeon-liver NAD kinase. The structural and kinetic basis of regulation of NADPH. Eur J Biochem 1975;55:475-83. 18. Bulygina ER, Telepneva VI. [Isolation of NAD-kinase from pigeon heart]. Biokhimiia.1980;45:2019-27. 19. Butler JR, McGuinness ET. Candida utilis NAD+ kinase: purification, properties and affinity gel studies. Int J Biochem 1982;14:839-44. 20. Tseng YM, Harris BG, Jacobson MK. Isolation and characterization of yeast nicotinamide adenine dinucleotide kinase. Biochim Biophys Acta1979;568:205-14. 21. Lerner F, Niere M, Ludwig A, Ziegler M. Structural and functional characterization of human NAD kinase. Biochem Biophys Res Commun 2001;288:69-74. 22, Ohashi K, Kawai S, Murata K. Identification and characterization of a human mitochondrial NAD kinase. Nat Commun 2012;3:1248. 23 Zhang R. MNADK, a long-awaited human mitochondrion-localized NAD kinase. J Cell Physiol 2015;230:1697-701. 24. Takahara K, Kasajima I, Takahashi H, Hashida SN, Itami T, Onodera H, et al. Metabolome and photochemical analysis of rice plants overexpressing Arabidopsis NAD kinase gene. Plant Physiol 2010;152:1863-73. 25. Takahashi H, Takahara K, Hashida SN, Hirabayashi T, Fujimori T, Kawai- Yamada M, et al. Pleiotropic modulation of carbon and nitrogen metabolism in Arabidopsis plants overexpressing the NAD kinase2 gene. Plant Physiol 2009;151:100- 13. 26. Waller JC, Dhanoa PK, Schumann U, Mullen RT, Snedden WA. Subcellular and tissue localization of NAD kinases from Arabidopsis: compartmentalization of de novo NADP biosynthesis. Planta 2010;231:305-17. 27. Shi F, Li Y, Li Y, Wang X. Molecular properties, functions, and potential applications of NAD kinases. Acta biochimica et biophysica Sinica. 2009;41:352-61. 28. Wang Y, San KY, Bennett GN. Improvement of NADPH bioavailability in Escherichia coli by replacing NAD(+)-dependent glyceraldehyde-3-phosphate dehydrogenase GapA with NADP (+)-dependent GapB from Bacillus subtilis and addition of NAD kinase. J Ind Microbiol Biotechnol 2013;40:1449-60. 29. Cui YY, Ling C, Zhang YY, Huang J, Liu JZ. Production of shikimic acid from Escherichia coli through chemically inducible chromosomal evolution and cofactor metabolic engineering. Microb Cell Fact 2014;13:21. 30. Li ZJ, Cai L, Wu Q, Chen GQ. Overexpression of NAD kinase in recombinant Escherichia coli harboring the phbCAB operon improves poly(3-hydroxybutyrate) production. Appl Microbiol Biotechnol 2009;83:939-47. 31. Bi J, Wang H, Xie J. Comparative genomics of NAD(P) biosynthesis and novel antibiotic drug targets. J Cell Physiol 2011;226:331-40.

32. Petrelli R, Sham YY, Chen L, Felczak K, Bennett E, Wilson D, et al. Selective inhibition of nicotinamide adenine dinucleotide kinases by dinucleoside disulfide mimics of nicotinamide adenine dinucleotide analogues. Bioorg Med Chem 2009;17:5656-64. 33. Petrelli R, Felczak K, Cappellacci L. NMN/NaMN adenylyltransferase (NMNAT) and NAD kinase (NADK) inhibitors: chemistry and potential therapeutic applications. Curr Med Chem 2011;18:1973-92. 34. Tedeschi PM, Lin H, Gounder M, Kerrigan JE, Abali EE, Scotto K, et al. Suppression of cytosolic NADPH pool by thionicotinamide increases oxidative stress and synergizes with chemotherapy. Mol Pharmacol 2015;88:720-7. 35. Pollak N, Niere M, Ziegler M. NAD kinase levels control the NADPH concentration in human cells. J Biol Chem 2007;282:33562-71. 36. Houten SM, Denis S, Te Brinke H, Jongejan A, van Kampen AH, Bradley EJ, et al. Mitochondrial NADP(H) deficiency due to a mutation in NADK2 causes dienoyl-CoA reductase deficiency with hyperlysinemia. Hum Mol Genet 2014;23:5009-16. 37. Tsang YH, Dogruluk T, Tedeschi PM, Wardwell-Ozgo J, Lu H, Espitia M, et al. Funtional annotation of rare gene aberration drivers of pancreatic cancer. Nat Commun 2016;7:10500. 38. Pankiewicz KW, Watanabe KA, Lesiak-Watanabe K, Goldstein BM, Jayaram HN. The chemistry of nicotinamide adenine dinucleotide (NAD) analogues containing C- nucleosides related to nicotinamide riboside. Curr Med Chem 2002;9:733-41. 39. Roussel B, Johnson-Farley N, Kerrigan JE, Scotto KW, Banerjee D, Felczak K, et al. A second target of benzamide riboside: dihydrofolate reductase. Cancer Biol Ther 2012;13:1290-8. 40. Hsieh YC, Tedeschi P, Adebisi Lawal R, Banerjee D, Scotto K, Kerrigan JE, et al. Enhanced degradation of dihydrofolate reductase through inhibition of NAD kinase by nicotinamide analogs. Mol Pharmacol 2013;83:339-53. 41. Herken H, Meyer-Estorf G, Halbhubner K, Loos D. Spastic paresis after 6- aminonicotinamide: metabolic disorders in the spinal cord and electromyographically recorded changes in the hind limbs of rats. Naunyn Schmiedebergs Arch Pharmacol 1976;293:245-55. 42. Herter FP, Weissman SG, Thompson HG, Jr., Hyman G, Martin DS. Clinical experience with 6-aminonicotinamide. Cancer Res 1961;21:31-7. 43. Koutcher JA, Alfieri AA, Matei C, Meyer KL, Street JC, Martin DS. Effect of 6- aminonicotinamide on the pentose phosphate pathway: 31P NMR and tumor growth delay studies. Magn Reson Med 1996;36:887-92. 44. RCSB Protein Data Bank [database on the Internet]. Piscataway (NJ): Rutgers, The State University of New Jersey; La Jolla (CA): San Diego Supercomputer Center (SDSC) and Skaggs School of Pharmacy and Pharmaceutical Science; 1971 [cited 2016 Jun 6]. Wang H, Tempel W, Wernimont AK, Tong Y, Guan X, Shen Y, et al. Crystal structure of human NAD kinase; Accession 3PFN [about 2 p.]. Available from: http://www.rcsb.org/pdb/explore/explore.do?structureId=3PFN. Files updated weekly.

Figure 1. (A) The role of NAD kinase (NADK) and dehydrogenases (glucose -6-phosphate dehydrogenase (G6PD) and the malic enzymes, M1,M2 in generating NADPH. NADPH is required for reductive synthesis (nucleic acids, proteins and fatty acids) and protects cells from ROS by generating reduced forms of glutathione (GSH) and thioredoxin (Trx). (B) Model of NAD+ (space filling model) bound to human NADK with each monomer colored green and cyan created using Pymol.

Figure 2. Cells expressing mutant NADK-I90F exhibit low ROS and elevated NADP and NADPH levels. (A) Total NADP and NADPH (NADpt) and NADPH levels were determined in normal pancreatic ductal cells (HPDE), pancreatic ductal adenocarcinoma cells (BxPC-3 and Panc-1), or cell lines expressing either the vector control, NADK or NADK-I90F using the NADP/NADPH-glo assay (Promega). (B) ROS levels were assayed in normal pancreatic ductal cells (HPDE) or pancreatic cancer cells (BxPC-3 and Panc-1). (C) Expression levels of NADK and NADK-I90F were determined. Adapted from Tsang et al. (37).

Figure 3. Exogenous nicotinamide in culture media can abrogate thionicotinamide (ThioNa) toxicity. (A) ThioNa toxicity is inversely correlated with nicotinamide levels. Untreated C85 cells and C85 cells in which cNADK has been stably knocked down are unaffected. Representative wells are shown for each condition. (B) Average colony increases in ThioNa-treated cells as nicotinamide levels increase. (C) Average colony number increases as media nicotinamide levels increase. (D) The proposed intracellular biosynthetic pathway from ThioNa to nicotinamide adenine dinucleotide phosphate, reduced (NADPSH). Nam, nicotinamide; n.s., not significant. Reprinted with permission from Tedeschi et al. (34): Suppression of cytosolic NADPH pool by thionicotinamide increases oxidative stress and synergizes with chemotherapy. Molecular Pharmacology 2015;88:720-7.

Figure 4. (A) ThioNa induces tumor regression in RL lymphoma xenografts. 6 to 8 week old NOD/SCID-γ male mice were inoculated s.c with RL cells. When the tumors were palpable, animals were treated with 100 mg/kg TN on days indicated by arrows. Tumor volume was determined using calipers (B) Inhibition of NADK by ThioNa or siRNA-mediated knockdown of NADK reduces tumor growth in C85 xenografts. NOD/SCID-γ male mice were inoculated s.c with one million C85 cells or C85 cells expressing shRNA directed against NADK. When tumors were palpable (Day 3), animals bearing normal C85 xenografts were dosed with 100mg/kg ThioNa on days indicated with arrows. Tumor volume was measured as above. This figure shows days 1-12 of the study, for clarity. In both studies, there was no evidence of toxicity (weight loss, neurotoxicity). Reprinted with permission from Tedeschi et al. (34): Suppression of cytosolic NADPH pool by thionicotinamide increases oxidative stress and synergizes with chemotherapy. Molecular Pharmacology 2015;88:720-7.

Figure 5. Structural alignment of the cytoplasmic NADK (cNADK) crystal structure (enzyme depicted as ribbons colored blue with key residues depicted as CPK colored ball-stick models and NAD+ depicted as a ball-stick model colored magenta) with the homology model of the mitochondrial enzyme mNADK (depicted as ribbons colored red). The mNADK residues are ball- stick models colored green with labels in italics. (Illustration prepared using VMD).

Figure 1:

A Nad+ KINASE NAD+ NADP+

G6PD

MALIC enzymes (M1/M2)

NADPH

Reductive Defense against ROS biosynthesis (GSH, Trx)

Figure 2:

AC 16 WT Vec I90F cells

4 12 52 NADK NADPt 8 NADPH 38 GAPDH

4 BxPC-3

pmol per 2.5 × 10 0 WT Vec Vec WT I90F Vec WT I90F Vec WT I90F I90F HPDE BxPC-3 Panc-1 52 NADK B

) 38 GAPDH 2 60 12 6 Panc-1 40 8 4 Vec WT 20 4 2 I90F

0 0 0 ROS (ﬂuorescence × 10 HPDE BxPC-3 Panc-1

Figure 3:

A Nicotinamide B P < 0.001 32.8 μM08.2 μM μM P < 0.001 25

Untreated 20

10 shNADK 5

Average colony size (pixel) 0

75 μM ThioNa

Control, 0 μM Nam ThioNa, 0 μM Nam shRNA, 0 μM Nam ThioNa, 8.2 μM Nam shRNA, 8.2 μM Nam Control, 8.2 μM Nam shRNA, 32.8 μM Nam CDP < 0.01 Control, 32.8 μM Nam ThioNa, 32.8 μM Nam P < 0.001 50 n.s. NAD+ 40 NADK G6PD nucleosidase 30

20 ThioNa NADS NADPS NADPSH No. of colonies 10

0 Nicotinamide

Control, 0 μM Nam ThioNa, 0 μM Nam shRNA, 0 μM Nam Control, 8.2 μM Nam ThioNa, 8.2 μM Nam shRNA, 8.2 μM Nam Control, 32.8 μM Nam ThioNa, 32.8 μM Nam shRNA, 32.8 μM Nam

Figure 4:

1,500 ) 3

1,000

500

Tumor volume (mm Untreated ThioNa

0 0 10 20 30 Day B 175

P value 150 Untreated <0.001 *** ThioNa <0.01 ) ** 3 125 shNADK <0.05 *

100

75 *** 50 ** *** Tumor volume (mm 25 * ** ** 0 024681012 Day

Figure 5:

L192 F211

D184

L188 N284

D383 D412

R385 Y327 E285

NAD+ Kinase as a Therapeutic Target in Cancer

Philip M. Tedeschi, Nitu Bansal, John E. Kerrigan, et al.

Clin Cancer Res Published OnlineFirst August 31, 2016.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-16-1129

Author Author manuscripts have been peer reviewed and accepted for publication but have not yet been Manuscript 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://clincancerres.aacrjournals.org/content/early/2016/08/30/1078-0432.CCR-16-1129. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.