Strategic Incorporation of Polarity in Heme-Displacing Inhibitors of Indoleamine-2,3-dioxygenase-1 (IDO1).

pubs.acs.org/acsmedchemlett Letter Strategic Incorporation of Polarity in Heme-Displacing Inhibitors of Indoleamine-2,3-dioxygenase‑1 (IDO1) Catherine White,* Meredeth A. McGowan,* Hua Zhou, Nunzio Sciammetta, Xavier Fradera, Jongwon Lim, Elizabeth M. Joshi, Christine Andrews, Elliott B. Nickbarg, Phillip Cowley, Sarah Trewick, Martin Augustin, Konstanze von Köenig, Charles A. Lesburg, Karin Otte, Ian Knemeyer, Hyun Woo, Wensheng Yu, Mangeng Cheng, Peter Spacciapoli, Prasanthi Geda, Xuelei Song, Nadya Smotrov, Patrick Curran, Mee Ra Heo, Pravien Abeywickrema, J. Richard Miller, David Jonathan Bennett, and Yongxin Han Cite This: ACS Med. Chem. Lett. 2020, 11, 550−557 ACCESS Metrics & More Read Online Article Recommendations sı Supporting Information * ABSTRACT: Indoleamine-2,3-dioxygenase-1 (IDO1) has emerged as a target of significant interest to the field of cancer immunotherapy, as the upregulation of IDO1 in certain cancers has been linked to host immune evasion and poor prognosis for patients. In particular, IDO1 inhibition is of interest as a combination therapy with immune checkpoint inhibition. Through an Automated Ligand Identification System (ALIS) screen, a diamide class of compounds was identified as a promising lead for the inhibition of IDO1. While hit 1 possessed attractive cell-based potency, it suffered from a significant right-shift in a whole blood assay, poor solubility, and poor pharmacokinetic properties. Through a physicochemical property-based approach, including a focus on lowering AlogP98 via the strategic introduction of polar substitution, compound 13 was identified bearing a pyridyl oxetane core. Compound 13 demonstrated improved whole blood potency and solubility, and an improved pharmacokinetic profile resulting in a low predicted human dose. KEYWORDS: IDO1, heme displacer, oxetane, AlogP98 I established with epacadostat12,13 in combination with anti-PD1 (programmed cell death protein 1) antibodies.14 The recent failure of this combination to show improved efficacy over antiPD-1 monotherapy in a phase 3 study (ECHO-301/KEYNOTE-252) has prompted speculation that epocadostat may have failed to achieve sufficient target engagement in the tumor microenvironment.15 Thus, the investigation of IDO1 ndoleamine-2,3-dioxygenase-1 (IDO1) is a heme-containing enzyme which catalyzes the oxidation of essential amino acid tryptophan (Trp) as the first and rate-limiting step of the kynurenine pathway. Both depletion of Trp and the production of kynurenine and other pathway metabolites contribute to local immunosuppression, manipulating multiple components of the innate and adaptive immune system including CD8+ T cells, effector T (Teff) cells, and natural killer (NK) cells.1−4 Many human tumors have been shown to exploit this pathway by upregulating the expression of IDO1,5−7 and an increased level of IDO1 expression in tumor cells is correlated with poor prognosis in several tumor types.8,9 Given its role in contributing to immune escape, IDO1 has emerged as an important therapeutic target in cancer immunotherapy.10,11 Initial proof of concept for IDO1 inhibition in the clinic was © 2020 American Chemical Society Received: January 6, 2020 Accepted: March 10, 2020 Published: March 10, 2020 550 https://dx.doi.org/10.1021/acsmedchemlett.0c00010 ACS Med. Chem. Lett. 2020, 11, 550−557 ACS Medicinal Chemistry Letters pubs.acs.org/acsmedchemlett inhibition continues in the clinic,16 and it remains highly desirable to discover new highly potent and selective IDO1 inhibitors which allow for achievement of higher levels of target engagement. Over the past several years, there have been multiple reports describing alternative binding modes and mechanisms of action for IDO1-selective inhibitory small molecules. Epacadostat inhibits IDO1 via a competitive binding mechanism, occupying the Trp binding site and coordinating to the hemebound iron, while an uncompetitive inhibitor such as mitomycin C occupies an allosteric site located on the opposite side of the heme in the protein.17 Additionally, it has recently been demonstrated that the heme cofactor is labile and can be displaced entirely, allowing for inhibitors to bind to the apoenzyme and thus prevent heme from rebinding.18 In fact, it has been shown that a majority of IDO1 present in the cell is in its apo form. This class of inhibitors are therefore particularly effective in a cellular setting, but require extended incubation times and elevated temperatures to effect inhibition in a functional enzymatic assay where the heme is preincorporated. As part of our efforts toward identifying a small molecule inhibitor of IDO1, we performed an affinity-based primary screen of 260,000 compounds against IDO1 using Automated Ligand Identification System (ALIS) technology. This screening method relies on identification of small molecules which bind to a target of interest via a mass-spectrometry-based detection using pooled libraries of compounds with known and distinct masses.19−22 Significantly, the protein employed for this screen was found to have achieved between 80% and 90% heme incorporation during its biosynthesis, and thus both apo and heme-bound forms of the protein were present (see Supporting Information for protein preparation protocol). From this screening campaign, diamide 1 was identified as a promising hit, displaying potent activity in a HeLa cell-based assay measuring production of the product of IDO1-mediated tryptophan oxidation, N-formylkynurenine (Hela IC50 = 9 nM, see Supporting Information for assay details). Interestingly, 1 was inactive in a biochemical IDO1 inhibition assay, prompting speculation that 1 acts as a heme-displacing IDO1 inhibitor as described above.23 A high-resolution cocrystal structure was obtained with apo IDO1 and compound 1 (Figure 1; PDB [protein database] entry 6 V52), which confirmed that 1 binds in such a manner that it occupies the binding site of the absent heme cofactor. This binding mode is further illustrated by an overlay24 of the Letter crystal structure of 1 with a recently disclosed heme-displacing inhibitor (compound 2, Figure 2)18,25 and heme (PDB entry Figure 2. Overlay of compounds 1 (teal) and 2 (gray) with the heme binding location (yellow). The A- and C- pockets are indicated. Residues omitted for clarity. 6e40).26 The structure also reveals that 1 is anchored to the protein via a network of hydrogen bonds, extending from the propyl-substituted amide directly to Ser167 and (via water) His346, as well as two water-mediated hydrogen bonds from the aryl amide to Ser267 and Arg343. The aryl amide extends below the plane of the absent heme to occupy a largely hydrophobic pocket (which we termed the “C” pocket). The npropyl amide substituent extends into and partially filled a hydrophobic pocket above the absent heme (termed the “A” pocket),17 which in the case of compound 2 is occupied by an aryl substituent. As an initial lead, compound 1 displayed potent suppr (...truncated)