Discovery of IACS-9779 and IACS-70465 as Potent Inhibitors Targeting Indoleamine 2,3-Dioxygenase 1 (IDO1) Apoenzyme

Indoleamine 2,3-dioxygenase 1 (IDO1), a heme-containing enzyme that mediates the rate-limiting step in the metabolism of l-tryptophan to kynurenine, has been widely explored as a potential immunotherapeutic target in oncology. We developed a class of inhibitors with a conformationally constrained bicyclo[3.1.0]hexane core. These potently inhibited IDO1 in a cellular context by binding to the apoenzyme, as elucidated by biochemical characterization and X-ray crystallography. A SKOV3 tumor model was instrumental in differentiating compounds, leading to the identification of IACS-9779 (62) and IACS-70465 (71). IACS-70465 has excellent cellular potency, a robust pharmacodynamic response, and in a human whole blood assay was more potent than linrodostat (BMS-986205). IACS-9779 with a predicted human efficacious once daily dose below 1 mg/kg to sustain >90% inhibition of IDO1 displayed an acceptable safety margin in rodent toxicology and dog cardiovascular studies to support advancement into preclinical safety evaluation for human development.

Abstract 3277: IACS-9779, a development candidate that inhibits 2,3-dioxygenase (IDO) activity by blocking heme incorporation into IDO apoenzyme

Increased expression of IDO1 is believed to create a tumor microenvironment that is immunosuppressive. In the course of our research directed at identifying potent and selective inhibitors of IDO1, we identified a class of compounds that inhibited IDO1 activity in a cellular context, but not in isolated enzymatic assays. We have conducted detailed mechanistic studies and shown that these molecules inhibit IDO1 by binding to the apo-enzyme, thus preventing the incorporation of the heme-cofactor into the active site of the holo-enzyme.

Through an extensive medicinal chemistry campaign, we optimized a series of orally bioavailable, highly potent and selective inhibitors of IDO1 that possess excellent pharmacological properties. For several lead molecules, pharmacokinetic (PK) - pharmacodynamic (PD) relationships were established in whole blood and SKOV3 xenograft assays. The inhibition of IDO1 in a human whole-blood assay correlated well with the suppression of tumor kynurenine (KYN) that was observed in SKOV3 xenografts. At plasma concentrations of 3 µM, IACS-9779 supressed tumor KYN levels by 90%. IACS-9779 was well tolerated with excellent in vivo PK properties across multiple preclinical species, and a human PK prediction consistent with a low daily dose needed for full suppression of KYN production via IDO1.

Note: This abstract was not presented at the meeting.

Citation Format: Faika Mseeh, Matthew M. Hamilton, Joseph R. Marszalek, Norma E. Rogers, Connor A. Parker, Simon S. Yu, Zhen Liu, Naphtali J. Reyna, Timothy McAfoos, Brett W. Virgin-Downey, Paul G. Leonard, Jason B. Cross, Ningping Feng, Angela L. Harris, Andy M. Zuniga, Keith Mikule, Martin Tremblay, Yongying Jiang, Mikhila Mahendra, Jihai Pang, Qi Wu, Quanyun Xu, Timothy P. Heffernan, Philip Jones, Richard T. Lewis. IACS-9779, a development candidate that inhibits 2,3-dioxygenase (IDO) activity by blocking heme incorporation into IDO apoenzyme [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3277.

Antisense knockdown of inducible nitric oxide synthase inhibits induction of experimental autoimmune encephalomyelitis in SJL/J mice

We used an antisense oligodeoxynucleotide (ODN) complementary to inducible nitric oxide synthase (iNOS) to inhibit experimental autoimmune encephalomyelitis (EAE) in female SJL/J mice, an animal model for multiple sclerosis. The antisense ODN was administered intraventricularly to mice daily for 10 days beginning at the time of adoptive transfer of myelin basic protein-specific T lymphocytes. The antisense ODN treatment significantly reduced the clinical score of EAE and blocked iNOS mRNA and protein synthesis, as well as iNOS enzyme activity within the central nervous system. The levels of nitric oxide and cyclic guanosine monophosphate were also significantly reduced by the antisense ODN treatment. Neither sense nor random ODN affected clinical EAE or iNOS expression. Moreover, the protein and enzyme activity level of constitutive neuronal nitric oxide synthase was not affected by the antisense ODN. Thus, we have shown that the iNOS antisense ODN specifically blocked iNOS expression and ameliorated the induction of EAE.

Cutting Edge: Antisense Knockdown of Inducible Nitric Oxide Synthase Inhibits Induction of Experimental Autoimmune Encephalomyelitis in SJL/J Mice

We used an antisense oligodeoxynucleotide (ODN) complementary to inducible nitric oxide synthase (iNOS) to inhibit experimental autoimmune encephalomyelitis (EAE) in female SJL/J mice, an animal model for multiple sclerosis. The antisense ODN was administered intraventricularly to mice daily for 10 days beginning at the time of adoptive transfer of myelin basic protein-specific T lymphocytes. The antisense ODN treatment significantly reduced the clinical score of EAE and blocked iNOS mRNA and protein synthesis, as well as iNOS enzyme activity within the central nervous system. The levels of nitric oxide and cyclic guanosine monophosphate were also significantly reduced by the antisense ODN treatment. Neither sense nor random ODN affected clinical EAE or iNOS expression. Moreover, the protein and enzyme activity level of constitutive neuronal nitric oxide synthase was not affected by the antisense ODN. Thus, we have shown that the iNOS antisense ODN specifically blocked iNOS expression and ameliorated the induction of EAE.

Gender differences of inducible nitric oxide production in SJL/J mice with experimental autoimmune encephalomyelitis

We identified gender related differences of inducible nitric oxide synthase (iNOS) expression and NO production in mice with experimental autoimmune encephalomyelitis (EAE). When myelin basic protein-specific T-lymphocytes derived from female mice were transferred, the female recipients developed more severe EAE and expressed higher levels of iNOS and NO than male recipients. When the T-lymphocytes derived from males were transferred, severe EAE was induced in neither female or male recipients and neither iNOS nor NO were detectable. These data show an association between No production and EAE severity, suggesting a possible role of NO in the pathogenesis of EAE.

Inducible nitric-oxide synthase and nitric oxide production in human fetal astrocytes and microglia. A kinetic analysis

The understanding of the induction and regulation of inducible nitric-oxide synthase (iNOS) in human cells may be important in developing therapeutic interventions for inflammatory diseases. In the present study, we not only demonstrated that human fetal mixed glial cultures, as well as enriched microglial cultures, synthesize iNOS and nitric oxide (NO) in response to cytokine stimulation, but also assessed the kinetics of iNOS and NO synthesis in human fetal mixed glial cultures. The iNOS mRNA was expressed within 2 h after stimulation and decreased to base line by 2 days. Significant levels of iNOS protein appeared within 24 h after stimulation and remained elevated during the culture period. A dramatic increase in NO production and NO-mediated events, such as the induction of cyclic guanosine monophosphate (cGMP), NADPH diaphorase activity, and nitrotyrosine occurred 3 days after stimulation, a delay of 48 h from the time of the first expression of iNOS enzyme. This delay of NO production was altered by the addition of tetrahydrobiopterin, but not by the addition of L-arginine, heme, flavin adenine dinucleotide (FAD), flavin mononucleotide (FMN), or NADPH. These findings suggest that a post-translational regulatory event might be involved in iNOS-mediated NO production in human glia.

Evidence for the presence of an unusual nitric oxide- and citrulline-producing enzyme in rat kidney

We have found an enzymatic activity obtained from rat kidney capable of producing citrulline and NOx. (nitrate and nitrite) which was resistant to inhibition by conventional arginine analogues. This enzyme activity does not require any calcium or calmodulin and was found to be induced during pregnancy. This unique enzyme was found to be tissue and species specific. Another unique feature of this enzyme is that it did not bind to 2'5'-ADP-sepharose under standard conditions. Western blot analysis of the 100,000 g kidney supernatant using monoclonal antibody for macrophage inducible nitric oxide synthase failed to produce a band for inducible nitric oxide synthase. HPLC and capillary ion analysis for nitrate and nitrite (NOx) showed clear peaks for [3H] L-citrulline and NOx, respectively, which were not changed either in the absence of calcium and calmodulin or in the presence of 300 microM S-ethylisothiourea, which has been shown to be a very potent and selective inhibitor of inducible nitric oxide synthase with a Ki of about 14.7 nM. These results suggest the possible existence of another isoform of nitric oxide synthase with very distinct properties from the known isoforms.

Arginase activity in endothelial cells: inhibition by NG-hydroxy-L-arginine during high-output NO production

Rat aortic endothelial cells were found to contain both constitutive and lipopolysaccharide (LPS)-inducible arginase activity. Studies were performed to determine whether induction of nitric oxide synthase (NOS) by LPS and cytokines is accompanied by sufficient arginase induction to render arginine concentrations rate limiting for high-output NO production. Unactivated cells contained abundant arginase activity accompanied by continuous urea formation. LPS induced the formation of both inducible NOS (iNOS) and arginase, and this was accompanied by increased production of NO, citrulline, and urea. Immunoprecipitation experiments revealed the constitutive presence of arginase-I in both unactivated and LPS-activated cells and arginase-II induction by LPS. Arginase-I and iNOS were verified by reverse transcriptase-polymerase chain reaction. Induction of large amounts of iNOS by LPS plus several cytokines resulted in large quantities of NO, citrulline, and NG-hydroxy-L-arginine (NOHA), but urea production was markedly diminished. Decreased urea production was attributed to increased formation of NOHA, the precursor to NO and citrulline and a potent inhibitor of arginase-I activity with an inhibitory constant of 10-12 microM. Inhibition of iNOS activity by NG-methyl-L-arginine decreased NO and NOHA production and increased urea production. This study reveals for the first time that substantial arginase activity is present constitutively in rat aortic endothelial cells, a different isoform of arginase is induced by LPS, and intracellular arginase activity can be markedly inhibited during cytokine induction of iNOS because of NOHA formation. The inhibition of arginase activity that occurs by NOHA during marked iNOS induction may be a mechanism to ensure sufficient arginine availability for high-output production of NO.

The cloned neurotensin receptor mediates cyclic GMP formation when coexpressed with nitric oxide synthase cDNA

Rat neurotensin (NT) receptor (NTR) cDNA was subcloned into the pRC-CMV expression vector and transfected into 293 cells, and cellular clones that stably expressed the NTR were isolated and characterized. [3H]NT binding to membranes prepared from the NTR cDNA-transfected cells displayed specificity and saturability, with an apparent Kd of 1.25 nM and a Bmax of 43.4 pmol/mg of protein (approximately 3.5 x 10(6) binding sites/cell). NT stimulated an increase in [3H]inositol phosphate levels in the NTR-expressing cells up to 2500% of basal levels. The response was time and dose dependent, with an EC50 of 10.4 nM. NT also stimulated cAMP formation in these cells, with an EC50 of 27.0 nM. In addition, NT evoked an increase in the level of intracellular calcium. Approximately 60% of the calcium rise was attributable to the release of intracellular stores and 40% was attributable to calcium influx. Although NTR occupancy has been shown to stimulate cGMP formation in several brain preparations and cell lines, NT was unable to mediate cGMP synthesis in the NTR-expressing 293 cells. We found that 293 cells have guanylate cyclase activity but have undetectable levels of nitric oxide synthase (NOS) activity. Because it was possible that the production of nitric oxide is required as the mediator of NT-induced cGMP synthesis, we subcloned NOS cDNA into the pCEP4 expression vector and transiently expressed it in the NTR cells. We report that NT increased cGMP levels up to 375% of basal levels when NOS cDNA was coexpressed and that the increase was completely inhibited by the NOS inhibitor N omega-nitro-L-arginine. NT-induced cGMP accumulation was time and dose dependent, with an EC50 of 1.7 nM. To our knowledge, this is the first report of NT mediating cGMP formation with a cloned receptor and the first evidence that NT-induced cGMP accumulation requires the production of nitric oxide.

Inducible nitric oxide synthase from a rat alveolar macrophage cell line is inhibited by nitric oxide

The objective of this study was to determine whether inducible nitric oxide (NO) synthase from a rat alveolar macrophage cell line (NR8383) activated by LPS plus IFN-gamma could be regulated by NO, one of the two products of the enzymatic reaction. This study was based on previous observations in this laboratory that NO is a negative feedback modulator of constitutive NO synthase from rat cerebellum. NO synthase activity was determined by monitoring the formation of 3H-L-citrulline from 3H-L-arginine in the presence of added cofactors. NO synthase catalyzed the conversion of L-arginine to equimolar quantities of NO and L-citrulline. NO and S-nitrosothiols inhibited NO synthase activity and this effect was enhanced by superoxide dismutase and attenuated by oxyhemoglobin. Nitrite and nitrate, the oxidation products of NO, as well as L-citrulline, the amino acid end-product, produced no significant effects on NO synthase activity. The inhibitory effect of NO on NO synthase appeared to be partially reversible upon addition of oxyhemoglobin. The inhibitory effect of NO was mimicked by other heme ligands including carbon monoxide, cyanide, and manganese-protoporphyrin IX. These observations indicate that (1) enzyme-bound heme plays a mechanistic role in the catalytic conversion of L-arginine to NO plus L-citrulline; (2) NO may function as a negative feedback modulator of inducible NO synthase by interacting with enzyme-bound heme; and (3) negative feedback modulation by NO may represent a mechanism by which the potentially toxic L-arginine-NO pathway in activated alveolar macrophages is turned off.

Inducible nitric oxide synthase from a rat alveolar macrophage cell line is inhibited by nitric oxide

The objective of this study was to determine whether inducible nitric oxide (NO) synthase from a rat alveolar macrophage cell line (NR8383) activated by LPS plus IFN-gamma could be regulated by NO, one of the two products of the enzymatic reaction. This study was based on previous observations in this laboratory that NO is a negative feedback modulator of constitutive NO synthase from rat cerebellum. NO synthase activity was determined by monitoring the formation of 3H-L-citrulline from 3H-L-arginine in the presence of added cofactors. NO synthase catalyzed the conversion of L-arginine to equimolar quantities of NO and L-citrulline. NO and S-nitrosothiols inhibited NO synthase activity and this effect was enhanced by superoxide dismutase and attenuated by oxyhemoglobin. Nitrite and nitrate, the oxidation products of NO, as well as L-citrulline, the amino acid end-product, produced no significant effects on NO synthase activity. The inhibitory effect of NO on NO synthase appeared to be partially reversible upon addition of oxyhemoglobin. The inhibitory effect of NO was mimicked by other heme ligands including carbon monoxide, cyanide, and manganese-protoporphyrin IX. These observations indicate that (1) enzyme-bound heme plays a mechanistic role in the catalytic conversion of L-arginine to NO plus L-citrulline; (2) NO may function as a negative feedback modulator of inducible NO synthase by interacting with enzyme-bound heme; and (3) negative feedback modulation by NO may represent a mechanism by which the potentially toxic L-arginine-NO pathway in activated alveolar macrophages is turned off.

Negative feedback regulation of endothelial cell function by nitric oxide

The objective of this study was to determine whether nitric oxide (NO) could function as a negative feedback modulator of endothelial cell function by inhibiting NO synthase in vascular endothelial cells. The rationale for this approach was a previous study from this laboratory, which revealed that NO inhibits neuronal NO synthase from rat cerebellum. In the present study, NO and NO-donor agents noncompetitively inhibited NO synthase derived from bovine aortic endothelial cells. Oxyhemoglobin blocked the inhibitory action of NO and by itself increased NO synthase activity. This finding suggests that NO acts as a negative feedback modulator of NO synthase. In intact aortic endothelial cells grown on microcarrier beads and perfused in a bioassay cascade system, pretreatment of cells with NO-donor agents caused a marked inhibition of endothelial NO biosynthesis in response to bradykinin and increased fluid shear or flow. When isolated bovine pulmonary arterial rings precontracted by phenylephrine were used, pretreatment of arterial rings with NO-donor agents diminished endothelium-dependent arterial relaxation involving the L-arginine-NO pathway without altering endothelium-independent relaxation to NO itself. On the basis of these studies, NO is suggested to play an important negative feedback regulatory role on endothelial NO synthase and, therefore, vascular endothelial cell function.

Oxidation of nitric oxide in aqueous solution to nitrite but not nitrate: comparison with enzymatically formed nitric oxide from L-arginine

Nitric oxide (NO) in oxygen-containing aqueous solution has a short half-life that is often attributed to a rapid oxidation to both NO2- and NO3-. The chemical fate of NO in aqueous solution is often assumed to be the same as that in air, where NO is oxidized to NO2 followed by dimerization to N2O4. Water then reacts with N2O4 to form both NO2- and NO3-. We report here that NO in aqueous solution containing oxygen is oxidized primarily to NO2- with little or no formation of NO3-. In the presence of oxyhemoglobin or oxymyoglobin, however, NO and NO2- were oxidized completely to NO3-. Methemoglobin was inactive in this regard. The unpurified cytosolic fraction from rat cerebellum, which contains constitutive NO synthase activity, catalyzed the conversion of L-arginine primarily to NO3- (NO2-/NO3- ratio = 0.25). After chromatography on DEAE-Sephacel or affinity chromatography using 2',5'-ADP-Sepharose 4B, active fractions containing NO synthase activity catalyzed the conversion of L-arginine primarily to NO2- (NO2-/NO3- ratio = 5.6) or only to NO2-, respectively. Unpurified cytosol from activated rat alveolar macrophages catalyzed the conversion of L-arginine to NO2- without formation of NO3-. Addition of 30 microM oxyhemoglobin to all enzyme reaction mixtures resulted in the formation primarily of NO3- (NO2-/NO3- ratio = 0.09 to 0.20). Cyanide ion, which displaces NO2- from its binding sites on oxyhemoglobin, inhibited the formation of NO3-, thereby allowing NO2- to accumulate. These observations indicate clearly that the primary decomposition product of NO in aerobic aqueous solution is NO2- and that further oxidation to NO3- requires the presence of additional oxidizing species such as oxyhemoproteins.

Constitutive nitric oxide synthase from cerebellum is reversibly inhibited by nitric oxide formed from L-arginine

The objective of this study was to determine whether constitutive nitric oxide (NO) synthase from rat cerebellum could be regulated by the two products of the reaction, NO and L-citrulline, utilizing L-arginine as substrate. NO synthase activity was determined by monitoring the formation of 3H-citrulline from 3H-L-arginine in the presence of added cofactors. The rate of citrulline formation in enzyme reaction mixtures was non-linear. Addition of superoxide dismutase (SOD; 100 units) inhibited NO synthase activity and made the rate of product formation more non-linear, whereas addition of oxyhemoglobin (HbO2; 30 microM) increased NO synthase activity, made the rate of product formation linear and also abolished the effect of SOD. Added NO (10 microM) inhibited NO synthase activity and this inhibition was potentiated by SOD and abolished by HbO2. Added L-citrulline (1 mM) did not alter NO synthase activity. The two NO donors, S-nitroso-N-acetylpenicillamine (200 microM) and N-methyl-N'-nitro-N-nitrosoguanidine (200 microM) mimicked the inhibitory effect of NO and inhibition of NO synthase activity by NO was reversible. These observations indicate clearly that NO formed during the NO synthase reaction or added to the enzyme reaction mixture causes a reversible inhibition of NO synthase activity. Thus, NO may function as a negative feedback modulator of its own synthesis.