Subasumstat | Fao Signal

Discovery of TAK-981, a First-in-Class Inhibitor of SUMO-Activating Enzyme for the Treatment of Cancer

Steven P. Langston,* Stephen Grossman,* Dylan England, Roushan Afroze, Neil Bence,
Douglas Bowman, Nancy Bump, Ryan Chau, Bei-Ching Chuang, Christopher Claiborne, Larry Cohen, Kelly Connolly, Matthew Duﬀey, Nitya Durvasula, Scott Freeze, Melissa Gallery, Katherine Galvin, Jeﬀrey Gaulin, Rachel Gershman, Paul Greenspan, Jessica Grieves, Jianping Guo, Nanda Gulavita, Shumet Hailu, Xingyue He, Kara Hoar, Yongbo Hu, Zhigen Hu, Mitsuhiro Ito, Mi-Sook Kim,
Scott Weston Lane, David Lok, Anya Lublinsky, William Mallender, Charles McIntyre, James Minissale, Hirotake Mizutani, Miho Mizutani, Nina Molchinova, Koji Ono, Ashok Patil, Mark Qian,
ImageJessica Riceberg, Vaishali Shindi, Michael D. Sintchak, Keli Song, Teresa Soucy, Yana Wang, He Xu, Xiaofeng Yang, Agatha Zawadzka, Ji Zhang, and Sai M. Pulukuri

ABSTRACT: SUMOylation is a reversible post-translational modiﬁcation that regulates protein function through covalent attachment of small ubiquitin-like modiﬁer (SUMO) proteins. The process of SUMOylating proteins involves an enzymatic cascade, the ﬁrst step of which entails the activation of a SUMO protein through an ATP-dependent process catalyzed by SUMO-activating enzyme (SAE). Here, we describe the identiﬁcation of TAK-981, a mechanism-based inhibitor of SAE which forms a SUMO−TAK-981 adduct as the inhibitory species within the enzyme catalytic site. Optimization of selectivity against related enzymes as well as enhancement of mean residence time of the adduct were critical to the identiﬁcation of compounds with potent cellular pathway inhibition and ultimately a prolonged pharmacody- namic eﬀect and eﬃcacy in preclinical tumor models, culminating in the identiﬁcation of the clinical molecule TAK-981.
Downloaded via 113.178.205.241 on May 14, 2021 at 09:30:38 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

⦁ INTRODUCTION
Small ubiquitin-like modiﬁer (SUMO) proteins are reversible
post-translational modiﬁers of intracellular proteins. The SUMOylation of proteins occurs through a three-step enzymatic cascade involving activation, transfer, and ligation of SUMO protein to lysine residues of target proteins. The ﬁrst step of the cascade is catalyzed by SUMO activating enzyme (SAE), which belongs to a small family of activating enzymes collectively known as E1s.1,2 SAE is a heterodimeric protein (SAE1, SAE2, or UBA2) and activates SUMO proteins through the carboxy terminus by reaction with ATP to form an acyl adenylate intermediate that is primed for acylation of a cysteine residue on the enzyme to form a thioester. The SUMO protein is subsequently transferred, through thioester exchange, to the SUMO speciﬁc conjugating enzyme E2, UBC9.3,4 Aided by SUMO speciﬁc E3 protein ligases, the SUMO proteins are transferred to lysine residues of target proteins to form oligomers through formation of an isopeptide bond.5

The enzymatic cascade is shown in Scheme 1, steps A−C. This three- step sequence is analogous to the process of ubiquitination and neddylation of The Authors. Published by American Chemical Society catalyzed by the homologue E1 ubiquitin activating enzyme (UAE) or Nedd8 activating enzyme (NAE), respectively.6,7 There are three SUMO family members: SUMO1, which conjugates to target proteins as a monomer, and SUMO2 and
SUMO3, which modify proteins through formation of oligomeric chains.5 The modiﬁcation of proteins by SUMO alters their molecular surface and can have eﬀects on the protein’s catalytic activity, cellular localization, and interaction with other proteins. In recent years, protein targets of SUMOylation have been identiﬁed that play key roles in multiple cellular pathways, including cell division and DNA repair, nuclear transport, gene transcription, chromosome structure and segregation, and immune response modulation.1,8 An early adenosine sulfamate and clinical compounds for NAE and UAE.

Elevated levels of the components of the SUMO pathway, e.g. UBC9, are associated with poor patient outcome in certain malignancies.9,10 Knockdown of SAE has been reported to confer synthetic lethality in cell lines and tumor models with high myc expression, and shRNAs targeting components of the SUMO pathway have demonstrated tumor growth inhibition in xenograft models.11−14 SUMOylation has also been shown to modulate innate immune responses.15 A net inhibitory eﬀect of SUMOylation on type 1 interferon (IFN) expression has been demonstrated such that inhibiting SUMOylation by genetic
means resulted in enhanced basal expression levels and sensitization of induction of type 1 IFNs by pathogenic stimuli.16,17
In addition, SUMOylation has been found to suppress a type1 interferon response to innate immune sensors, and depletion of UBC9 in hematopoietic cells enhanced an inﬂammatory response and protection to viral infection in mouse models.16,17 Collectively, these observations warrant the identiﬁcation of potent and selective small molecule inhibitors of SAE to explore the cellular eﬀects of SUMO pathway inhibition and potential therapeutic opportunity.

⦁ RESULTS AND DISCUSSION
Our group has previously identiﬁed a class of adenosine
sulfamate (AdoS) derived inhibitors of E1 enzymes that act through a unique type of mechanism-based inhibition termed substrate assisted inhibition.18,19 These inhibitors were shown to act through reaction of the sulfamate moiety with a cysteine thioester of the E1 enzyme and UBL to form a UBL−inhibitor adduct within the enzyme catalytic site, Scheme 1 step D, using SAE as an illustrative example. The conformation of the enzyme that binds the substrates ATP and SUMO is thought to be open, while a more closed structure resembling that captured in the crystal structure of a SUMO-adenosine vinylsulfone covalently trapped intermediate may also contribute to binding the SUMO-AdoS adduct.20,21

An early example of the series is the N-6 substituted adenosine sulfamate compound 1, a nonselective E1 inhibitor, was shown Transthiolation assay measuring inhibition of the transfer of SUMO1 to UBC9 (the E2 for SAE) at an [ATP] approximating Km (20 μM). bPyrophosphate exchange assay measuring inhibition of enzyme activity for either SUMO1 or SUMO2 substrates for SAE, Ub for UAE, and Nedd- 8 for NAE; all were conducted at 1 mM [ATP]. cCellular assay conducted in HCT116 cells assessing inhibition of conjugation (transthiolation) of SUMO proteins to UBC9. Similarly, for conjugation of ubiquitin to UBC12 and conjugation of Nedd-8 to UBC-10, the UBL and E2 substrates for UAE and NAE, respectively to be capable of forming an inhibitor adduct with E1 UBL substrates, including UAE, NAE, and SAE, Figure 1.18 Compound 1 was optimized through an iterative medicinal chemistry program to the selective inhibitor of NAE, Pevonedistat (TAK-924, MLN-4924), which is currently in clinical trial for solid and heme malignancies, including a phase III trial in combination with azacytidine for patients with high risk myelodysplastic syndrome (HR MDS), chronic myelomo- nocytic leukemia (CMML), or low-blast acute myeloid leukemia (AML) (NCT03268954).22−24

Pevonedistat was shown to selectively inhibit NAE by reacting with the NAE-Nedd-8 thioester to form the Nedd-8-Pevonedistat adduct, a tight binding bisubstrate analogue, within the enzyme active site.18,25 The chemical class was subsequently diversiﬁed, and the UAE inhibitor TAK-243 was identiﬁed, which acts via a similar mechanism forming a ubiquitin−TAK-243 adduct that inhibits UAE and also entered clinical evaluation within oncology (NCT03816319), Figure 1.26−29 Despite demonstrating that compound 1 was able to form SUMO−compound 1 adduct, it proved challenging to identify a selective compound capable of demonstrating robust inhibition of the SUMO pathway in cells. Compounds reported to be inhibitors of SAE were not useful as tool compounds due to weak activity and unclear selectivity.30−32 While conducting an extensive scaﬀold hopping exercise within the adenosine sulfamate series, compound 2 was prepared and demonstrated potent SAE activity in the HTRF or transthiolation assay.19 Although retaining signiﬁcant activity for UAE and NAE, it was an example of a distinct chemotype, and removal the 2′-hydroxyl (adenosine numbering) gave a somewhat more selective molecule 3, Table 1. The homogeneous time resolved ﬂuorescence (HTRF) assay was conducted at Km for ATP and utilized SUMO1 as the UBL substrate.

The compounds were also shown to be active at millimolar ATP concentration for both SUMO1 and SUMO2 as substrates by employing a pyrophosphate exchange assay.19 Importantly, quantitative Western blot assays demonstrated robust inhibition of the SUMO pathway in cells, giving a selective reduction in the concentration of the E2 intermediate UBC9-SUMO compared to the E2-UBL conjugates for UAE and NAE, Table 1 and Figure Western blot demonstrating cellular pathway inhibition by compounds 2 and 3. Inhibition of SAE and NAE in cells was assessed by immunoblot analysis of SUMO2/3-conjugated proteins, SAE enzyme, and neddylated cullin levels in HCT-116 cell lysates. Multiple SUMO2/3 conjugate bands reﬂect the myriad proteins serving as substrates for SUMOylation.

SAE to explore the biology and potential therapeutic utility of SAE and consequently SUMO pathway inhibition both in cells and in vivo.
The Western blot assay served to demonstrate potent, selective pathway inhibition in cells. However, to support a medicinal chemistry program, we required a higher throughput assay that was amenable to automation. In cells, there exists a highly dynamic process of SUMOylation and deSUMOylation.1 SUMO conjugated proteins, which normally have high molecular weight, are mainly localized within the nucleus. The cellular inhibition of SAE results in an apparent loss of SUMOylated proteins in the nucleus due to rapid deSUMOy- lation, catalyzed by SUMO speciﬁc proteases, coupled with the inability to further SUMOylate existing or newly synthesized proteins. Additionally, free SUMO levels increase and redistribute throughout the cell. An immunoﬂuorescence assay (SUMO IF) was developed to monitor this redistribution of SUMO protein from nucleus to cytosol and quantify SAE inhibition in the HCT116 tumor cell line. Representative images are shown in Figure 3.
Initial SAR investigation focused on two areas of the molecule: exploration of the substitution pattern of the cyclopentane core and N1-substituents of the 3-substituted

2. We were thus encouraged to utilize this scaﬀold as the basis of
pyrazole. The 2′-deoxy compound 3 improved selectivity The SUMO2/3 immunoﬂuorescence assay. This assay measures the level of inhibition of the SAE by measuring the ratio of SUMO2/3 in the nuclear versus cytoplasmic cellular compartments. HCT-116 cells were seeded in 96-well plates and treated with compound 3 at the indicated concentrations for 4 h followed by staining with antihuman SUMO2/3 monoclonal antibody (MBL, clone 1E7). Imaging was performed on an IX- Ultra imager and analyzed for SUMO2/3 inner/outer ratio using MetaXpress Translocation-Enhanced module. Dose-dependent reduction of cytoplasmic SUMO2/3 positive staining reﬂects loss of SUMOylated proteins sequestered in this compartment under physiological conditions (DMSO). Right panel shows the dose response for compound 3 in HCT-116 cells as determined by the SUMO2/3 immunoﬂuorescence assay. while epimerization of the alcohol at this position as in 4 reduced activity for all three enzymes, Table 2. Fluorination of the 2′- position was tolerated as either epimer, see 6 and 7, and as the diﬂuorinated compound 8, with modest improvements in selectivity, although at the expense of cell potency and added synthetic complexity. Epimerization of the C4′-methylene sulfamate, as in 5, to give the antirelationship relative to the nucleobase mimetic, reminiscent of the conﬁguration for the NAE inhibitor TAK-924, reduced activity, Table 2.

An array of N1-pyrazole substituents was investigated, see Table 3. The benzyl group was found to be preferred over alkyl groups, including linear and branched alkyl substituents. The benzyl group, as in 3, was also preferred over phenyl, as in 9, and the corresponding methylene cyclohexane 10. Investigating the preferred substitution pattern around the benzyl group showed that the meta-position, as in 12, for chloro-substitution was preferred over ortho 11 and para 13, and typically, larger more hydrophobic groups were preferred at this position (ML-792, 15−19). Methylene heterocycles with halo substitution were also found to be potent, for example the 5-chloro-2-furanyl 19, corresponding thiophene 20, and the 6-halo-2-pyridines (21,

Alternative heterocycles to the 3-pyrazole were also explored. The N-substituted 3-pyrrole was found to be potent although
less selective toward NAE; see 23 and 25 compared to 3 and ML-792. The N-benzyl 4-pyrazole regioisomer 24 was less potent than the 3-pyrazole, Table 4. Heterocycles with C- benzylation were also explored, and the 4-benzylated 2- thiophene 26 showed promising potency, especially in the cellular SUMO IF assay, compared to the thiophene and furan isomers 27 and 28, Table 4.

ML-792 was chosen as a suitable representative compound to investigate the eﬀects of SAE inhibition on cancer cell biology because of its potency and selectivity. The eﬀects of ML-792 on several cancer lines have been previously reported.33 For the HCT116 cancer cell line, both an apoptotic and senescent phenotype were observed, resulting in a reduced cell proliferation in a 72 h viability assay. The next step was to explore the eﬀects of SAE and SUMO pathway inhibition in in vivo tumor xenograft models. Selective pathway inhibition was ﬁrst determined by Western blot assays showing loss of SUMO- Ubc9 thioester and SUMO-protein conjugates beyond 8 h following a single subcutaneous dose. The SUMO-ML-792 adduct from xenograft homogenates was quantiﬁed initially by mass spectrometry.18 Subsequently, an antibody to the SUMO- ML-792 adduct was developed to quantify levels of adduct by immunohistochemistry (IHC). Additionally, an IHC assay was developed to assess the levels of SUMO−protein conjugates in washed to remove residual compound, and the nuclear and cytoplasmic SUMO2/3 ﬂuorescence signal was measured immediately or 8 h following washout. As a result of pathway recovery, the SUMO2/3 ﬂuorescence signal was shown to relocalize to the nucleus over time, and an IC50 was calculated and reported at 0 and 8 h, Table

H-bond interaction with Ser95, potentially helping to stabilize the adduct within the enzyme and increase residence time, Figure 6.
The 3-chloro benzyl thiophene 29 was 5−10-fold less potent than the pyrazole ML-792 in the cellular SUMO IF assay both at 0 and 8 h following washout, Table 5. Substitution of the benzylic methylene with a hydroxyl or primary amine and cellular nuclei from xenograft homogenates. This facilitated observation of target engagement (presence of adduct) and SUMO pathway inhibition (loss of SUMO-protein conjugates) in the same tissue. The ratio of nuclei staining positive for SUMO-ML-792 adduct and SUMO−protein conjugates to total nuclei in HCT116 (colorectal carcinoma) xenograft homoge- nates is plotted in Figure 4. Importantly, an inverse relationship between the persistence of SUMO−inhibitor adduct with the rate of SUMOylated protein recovery was observed. The pathway recovers between 8 and 24 h despite signiﬁcant adduct remaining. This may be due to partial restoration of enzyme activity through adduct release. The eﬀect on tumor growth was investigated, and we were gratiﬁed to observe robust, dose- dependent tumor growth inhibition in the HCT116 model on both continuous and intermittent dosing schedulesWith establishment of PD markers for both target engage- ment and pathway inhibition as well as robust tumor growth inhibition, we were encouraged to continue to optimize the chemical series.

The dose of ML-792 required for a robust PD response and eﬃcacy was high (150 mg/kg s.c.) and involved twice daily administration; thus, we aimed to prolong PD response to dosing frequency. The SAE enzyme is relatively long-lived, and we determined that the resynthesis rate, even in the presence of the SAE inhibitor ML-792, is comparable to cellular doubling time.33 It was thus anticipated that optimizing the binding aﬃnity of the adduct to the enzyme would increase the residence time of the inhibitory adduct species in the enzyme active site and result in prolonged pathway inhibition. To support this approach, the SUMO IF assay was adapted to assess duration of pathway inhibition by including a compound washout step. In this format, the cells were incubated with a range of compound concentrations for 2 h.

The cells were then separation of the epimers gave one stereoisomer with signiﬁcantly more activity over the other, see 31 compared to 30 and 33 compared to 32, Table 5. The more active isomer of the primary amine, compound 33, exhibited activity approach- ing the cellular potency of the pyrazole ML-792 and showed good selectivity over both NAE and UAE. Methylation of the amine to the secondary or tertiary amine signiﬁcantly reduced potency. Substitution of the thiophene was also investigated. Chlorination or methylation at the 5-position were tolerated with the methylated compound 36 being more potent than the unsubstituted thiophene 29. The polar and electron with- drawing cyano group at this position 38 was signiﬁcantly less active, Table 6. Combining methylation of the thiophene and hydroxylation or amination of the benzylic center yielded highly potent compounds. The more active stereoisomers exhibiting only a 2- to 3-fold drop oﬀ in potency following washout and maintaining pathway inhibition below 50 nM at 8 h. Incorporating methylation at the benzylic position to give the tertiary alcohol or corresponding primary amine was well- tolerated and again demonstrated only a small drop oﬀ in potency between 0 and 8 h for the more active isomers in the washout assay, e.g. see 45, Table 6.

The crystal structure of the adduct of the more active isomer of a tertiary alcohol analogue, compound 46, in which the meta- chlorophenyl is replaced with 6-bromopyridyl, demonstrates the anticipated H-bond between the benzylic methylene hydroxyl group and serine 95, Figure 7. It also speciﬁes the desired conﬁguration of the benzylic center as S. Assuming a similar preferred orientation for the benzylic amino analogues, a rationale can be put forward to explain why methylation to the secondary or tertiary amine (compounds 34, 35 in Table 5) is Pharmacodynamic eﬀects of ML-792 in HCT-116 tumor bearing female NCr nude mice. Each data point represents the averaged value across n = 3 mice. Error bars = SEM. At t = 0, mice received a single SC injection of ML-792 at 150 mg/kg. Red line = percent nuclei staining positive for SUMO2/3 conjugation normalized to vehicle- treated animals. Blue line = percent nuclei staining positive for SUMO- ML-792 adduct.

less active, as the methyl group would either preclude access to this orientation or clash with the enzyme. The internal H-bond between the pyridine substituent and tertiary alcohol suggests that cyclization of the heteroatom to the pendant aryl group may be favored.
To investigate the eﬀect of cyclization, we prepared both the isochromane and tetrahydroisoquinoline ring systems. The isochromane was indeed extremely potent and demonstrated prolonged cellular pathway inhibition; for example, ML-93 exhibited an IC50 of 15 nM at the 8 h time point in the SUMO IF assay. Similar to the open chain system, one isomer is highly preferred over the other, as can be seen by comparing ML-93 with 47, Table 7. Meta-substitution had been shown to be preferred in the open benzylic analogues. The unmethylated thiophene analogue 50 and replacement of the methyl group with chloro, as in 48, gave compounds of similar potency. Cyclization to the isochromane eﬀectively ﬁxes the orientation of the meta-substituents in two distinct areas of space. The 7- chloro-isochromane 50 was found to be signiﬁcantly more active than the corresponding substitution at the 5-position as in 51, Table 7. The matched pair of 48, the 5-membered Antitumor eﬃcacy of ML-792 at multiple dose/schedule regimens in Ncr/nude mice bearing HCT-116 tumor-bearing mice. (a) ML-792 was dosed subcutaneously at the dose and schedule indicated for each arm (n = 10 female NCr nude mice/arm) until the ﬁnal treatment day on day 21. Schedules: BID x21 = twice per day for 21 days (days 1−21); BID BIW = twice per day/twice per week (days 1, 4, 8, 11, 15, 18, 21); BID 2on-5oﬀ = twice per day/twice per week on 2 consecutive days followed by 5 nondosing days (days 1, 2, 8, 9, 15, 16, 21). dihydroisobenzofuran ring system 55, was not well-tolerated with a signiﬁcant loss of cell potency. The tetrahydroisoquino- line (THiQ) exempliﬁed by TAK-981 also exhibited potent and prolonged cellular pathway inhibition with improved selectivity over NAE. Removal of the methyl or chloro-substituents (58, 59) reduced cellular potency more so in the THiQ series than the isochromane series. Methylation of the THiQ nitrogen, as in 60, also signiﬁcantly reduced potency, and again, the 5- membered isoindoline ring system 61 suﬀered from reduced Figure 6.

Structure of SUMO1-compound 2 adduct within SAE ATP binding domain. Carbon is shown in cyan (2), green (SUMO), or gray (SAE); nitrogen, blue; oxygen, red; and sulfur, yellow. Secondary structure of SAE is shown in gray. Hydrogen bonds are indicated by dashed lines. PDB accession code 6XOI.isochromane oxygen and ser-95 are within contact distance to form an H-bond, Figure 8. The corresponding residues to Ser-95 are Lys-147 and Arg-551 in NAE and UAE, respectively, which may explain the particularly high selectivity for the basic amino-containing compounds such as TAK-981.18,34 The H-bonding to ser-95 may favor orientation of the isochromane and THiQ groups to form hydrophobic interactions with the enzyme and contribute to prolong binding of the adduct. The residence time of the adduct in the enzyme is thought to be particularly important for cellular potency as released adduct may be susceptible to cleavage by cellular endoproteases such as SUMO speciﬁc endoproteases (SENPs).19,35 Structure of SUMO1-compound 46 adduct bound to SAE. Carbon is shown in cyan for compound 46, green (SUMO), or gray (SAE); nitrogen, blue; oxygen, red; bromide, dark red; and sulfur, yellow. Secondary structure of SAE is shown in gray. Hydrogen bonds are indicated by dashed lines.

PDB accession code 6XOG. even at 16 h for TAK-981, Figure 10a. All three compounds exhibit similar relatively rapid decrease in plasma concentration to below detection by 8 h in the HCT-116 tumor bearing mice, Figure 11a and in the OCI-Ly10 tumor bearing mice, Figure 11b. Compound ML-93 and TAK-981 showed similar potency and viability eﬀects in the HCT116 cells cultured in vitro, as previously observed for compound ML-792, Table 8.33 An eﬃcacy study in HCT116 model demonstrated equivalent tumor growth inhibition (tumor stasis, Figure 12a) at much lower weekly dosage for ML-93 (50 mg/kg, QDX21, total weekly dose = 350 mg/kg) and TAK-981 (50 mg/kg QDX3/ week, total weekly dose = 150 mg/kg) compared to ML-792 (150 mg/kg, BID, total weekly dose = 2100 mg/kg), Figure 12b. Similar results and tumor regression were observed in the OCI- Ly10 model, Figure 12c, d. This is consistent with the Ly10 cells being intrinsically more sensitive than HCT-116 for example a cellular viability EC50 for TAK-981 of 4 and 63 nM in LY10 and HCT-116, respectively, Table 8.

The ADME properties of ML-93 and TAK-981 are similar to those of ML-792, exhibiting high extraction ratios in liver S9 subcellular fractions, indicating these compounds are subject to a relatively high degree of metabolism in the liver, Table 8. The results from the permeability assay using Caco-2 cells showed a solid tumor xenograft (HCT-116; colorectal carcinoma) and a hematological tumor xenograft (OCI-Ly10; DLBCL) model. A PD time course for the isochromane ML-93 and the corresponding THiQ TAK-981 compared to the initial lead ML-792 in HCT-116 is shown in Figure 9a and in OCI-Ly10 is shown in Figure 9b. In HCT-116, the depletion of SUMO- conjugated proteins is much more sustained for ML-93 and TAK-981 than for ML-792. Even with lower dose of ML-93 (50 mg/kg) and TAK-981 (50 mg/kg in HCT116 and 25 mg/kg in LY10), we observed less than 25% recovery through 16 h.

Consistently, similar improvement in the durability of the SUMO−inhibitor adduct was observed for ML-93 and TAK-
981 compared to ML-792 in the HCT-116 model, Figure 10a and OCI-Ly10 model, Figure 10b. The adduct to SAE enzyme ratio as determined by mass spectrometry, remains close to unity high eﬄux ratios of >16, and the resulting low apparent permeability for all three compounds indicate these compounds are substrates of eﬄux transporters, Table 8. The PK of the compounds in rat are similar, exhibiting high blood to plasma ratios, relatively short half-lives, and low exposures following an oral administration due to relatively poor oral bioavailability, Table 9. The high blood to plasma partitioning may be attributed to reversible binding to carbonic anhydrase (CA), Table 8, which present in high levels in red blood cells, through the sulfamate group.36,37 The PK proﬁles in blood and plasma exhibited parallel terminal phases with comparable half-lives in both matrixes, indicting the high blood to plasma partitioning of these compounds is considered to be in rapid equilibrium.

Even though these compounds showed a relatively short half- life, both ML-93 and Tak-981 demonstrated a durable PD Crystal structure of SAE with SUMO1-compound 49 adduct. Carbon is shown in cyan, 49, green (SUMO), or gray (SAE); nitrogen, blue; oxygen, red; and sulfur, yellow. Secondary structure of SAE is shown in gray. Hydrogen bonds are indicated by dashed lines. PDB accession code 6XOH. response and robust eﬃcacy, including on intermittent dosing schedules, at greatly reduced dose compared to the initial lead ML-792. In light of this disconnect between PK and PD, driven by the tight binding nature of the inhibitor adduct to the enzyme, the team saw no requirement for improved exposure kinetics as a criterion for development as a parentally administered agent. With this in mind, the equilibrium solubility of TAK-981 compared to ML-93 was substantially higher at mildly acidic pH due to the presence of the basic THiQ system (pKa 7.5, Table 8). In the presence of hydroxypropylcyclodex- trin (hpβcd) excipient, the equilibrium solubility at approx- imately 20 mg/mL was acceptable for development as an i.v. formulation.

ML-93 has recently been investigated in pancreatic ductal carcinoma (PDAC), providing evidence of sensitivity to SAE inhibition and activation of MYC.14 Exploration of the activity of TAK-981 in mouse models and primary cell explants revealed upregulation of Type I IFN signaling in immune cells and Type I IFN dependent activation of innate immune cells, including macrophages, NK cells and dendritic cells, as well as T cells.38 In mouse syngeneic tumor models, TAK-981 was shown to promote an innate immune response capable of bridging to an adaptive antitumor immune response, representing a novel and mechanistically diﬀerentiated strategy for stimulating Type I Reduction of SUMO2/3 conjugation in xenograft tumors following IV administration of SUMOylation inhibitors. Each data point represents the averaged value across n = 3 female Ncr nude mice. Error bars = SEM. Percent SUMO2/3 inhibition following compound administration represented as percent nuclei staining positive by IHC for SUMO2/3 conjugates/total nuclei normalized to vehicle-treated animals (time = 0 h). SUMOylation inhibitors were administered at time = 0. ML-792 was dosed at 150 mg/kg SC (dashed line/open square). ML-93 was dosed at 50 mg/ kg IV (solid line/open circle). TAK-981 was dosed at 50 mg/kg IV (solid line/open triangle). (a) Reduction of SUMO2/3 conjugation at 2, 4, 8, and 16 h postinjection in HCT-116 tumors. (b) Reduction of SUMO2/3 conjugation at 2, 4, 8, 16, and 24 h postinjection in OCI-Ly10 tumors.

Gain of SUMO−compound adduct in xenograft tumors following IV administration of SUMOylation inhibitors. Adduct formation in xenograft tumors following compound administration. Each data point represents the averaged value across n = 3 female Ncr nude mice. Error bars = SEM. (a) Single-dose ML-792 at 150 mg/kg IV (dashed line/open square), ML-93 at 50 mg/kg IV (solid line/open circle) or TAK-981 (solid line/ open triangle) was administered to Ncr/nude mice bearing HCT116 tumors. Tumors were harvested 2, 4, 8, and 16 h post IV injection. Adduct/SAE ratio = quantity SUMO−compound adduct/quantity total SAE enzyme as measured by mass spectrometry. (b) Single-dose ML-792 at 150 mg/kg IV (dashed line/open square) or TAK-981 at 25 mg/kg IV (solid line/open triangle) was administered to CB17 SCID mice bearing OCI-Ly10 tumors. Tumors were harvested 2, 4, 8, 24, and 48 h post IV injection. Percent nuclei staining positive for adduct was measured by IHC.

Plasma exposure following single-dose IV administration. Plasma collected at indicated time-point postinjection by cardiac blood collection followed by centrifugation. Each data point represents the averaged value across n = 3 female Ncr nude mice. Error bars = SEM. (a) Single-dose injection at t = 0 delivered to Ncr/nude mice bearing HCT-116 tumors with the indicated compounds and dosages or (b) single-dose injection at t =0 delivered to CB17 SCID mice bearing OCI-Ly10 tumors with the indicated compounds and dosages. Plasma samples were collected at 0.5, 2, 4, 8, and 16 h postinjection via cardiac blood collection followed by centrifugation.

IFN signaling and antitumor immune responses.39 The activity of TAK-981 is currently being assessed in Ph1 clinical trials in adult patients with solid tumors and lymphomas (NCT03648372, NCT04074330, and NCT04381650). Chemistry. The synthetic route used to synthesize the analogues described in Tables 1 and 2 is detailed in Scheme 2. Benzylation of pyrazol-3-carbaldehyde provided the desired N1- benzyl pyrazole in a highly regiospeciﬁc manner. Lithium halogen metal exchange of 4-chloro-5-iodopyrimidine followed Antitumor activity versus total weekly dosage in human xenograft cancer models. Line graphs = average tumor volumes over time. Error bars = SEM. Mice were randomized into groups on Day 0. Treatment started on Day 1. Tumor volume measurements were taken twice per week. Bar graphs = total weekly dose for each treatment arm. (a) Female Ncr/nude mice bearing HCT-116 tumors (∼200 mm3 volume) were randomized into 5 groups (n = 8/group). Treatment started on Day 1. Tumor volumes were measured twice per week. Error bars represent SEM. (b) Total weekly dose administered to mice in each arm of the HCT-116 eﬃcacy study (panel a) is calculated, demonstrating reduced compound required to achieve eﬃcacy with TAK-981. (c) Female CB17 SCID mice bearing OCI-Ly10 tumors (∼200 mm3 volume) were randomized into 5 groups (n = 8/group). Treatment started on Day 1. Tumor volumes were measured twice per week. Error bars represent SEM. (d) Total weekly dose administered to mice in each arm of the OCI-Ly10 eﬃcacy study (panel c) is calculated, demonstrating reduced compound required to achieve eﬃcacy with TAK-981. Dosing schedules: BID x21 = twice-per-day (Days 1−21); QD x2/week = twice per week (Days 1, 4, 8, 11, 15, 18); QD x21 = once-per-day (Days 1−21); QD
x3/week = thrice-per-week (Days 1−21); QD x3/week = three consecutive dosing days followed by four nondosing days (Days 1, 2, 3, 8, 9, 10, 15, 16,
17) by addition of the aldehyde aﬀorded the benzylic alcohol, which was oxidized with MnO2 to provide an intermediate chloro- pyrimidine ketone (I-3). Displacement of the chloride with the appropriately substituted amino cyclopentane in the presence of base yielded intermediates which were further elaborated to ﬁnal products by protecting group manipulation and sulfamation.

Scheme 3 details the synthetic route used to synthesize the analogues described in Table 3. The appropriately substituted
N-Boc pyrazole (I-25) intermediate was prepared as described in Scheme 2. Substitution of the pyrazole was performed in an array format by one of two methods, Scheme 3. Either the primary alcohol was initially sulfamated, followed by N-Boc deprotection to provide an intermediate that was selectively N- alkylated through parallel array, or alternatively, the Boc group was removed ﬁrst to provide an intermediate that was subsequently N-alkylated followed by sulfamation. Schemes 4 and 5 detail the synthetic route used to prepare the benzylic alcohol or benzylic amine analogues described in Tables 5 and 6. Metal halogen exchange of the bromothiophene and reaction either with the appropriate benzaldehyde or benzylic sulfonyl imine gave the secondary alcohol or sulfonyl amine. Protecting group manipulation gave the thiophene aldehyde intermediate suitable for reaction with 4-chloro-5- iodopyrimidine, and subsequent elaboration analogous to that shown in Scheme 2 gave the target compound, Scheme 4. The tertiary alcohols were prepared in a similar manner starting from the appropriate acetophenone. The alpha-methyl amines were prepared through oxidation of the secondary alcohol to the ketone, formation of the sulfonylimine, followed by reaction with methyllithium, Scheme 5. Separation of the epimers was

Synthetic Routes Employed for Pyrazolo-Substitution by Parallel Array achieved through SFC or prep HPLC on the ﬁnal sulfamated compounds.
Scheme 6 details the synthetic route used to prepare the isochromane and tetrahydroisoquinoline analogues described in Table 7. Metal halogen exchange of the bromothiophene and reaction with a benzaldehyde derivatized at the ortho position with an appropriately protected alcohol. Treatment with acid deprotected the primary alcohol and facilitated cyclization to give the isochromane (I-107). The thiophene aldehyde was then progressed to the ﬁnal compounds by an analogous route to Scheme 2. Synthesis of Key Intermediates for Benzylic Alcohol and Benzylic Amines Described in Tables 5 and 6
Synthesis of Key Intermediates for Alpha-Methyl Benzyl Amines The tetrahydroisoquinoline analogues (I-117) were prepared in a similar fashion by reaction with the appropriate dihydroisoquinoline and prosecution to the target compound. Separation of the epimers was achieved through SFC or prep HPLC on the ﬁnal sulfamated compounds.

The conﬁguration of the more active isomer at the THiQ center was designated as R from a small molecule crystal structure of a late stage, presulfamation, synthetic intermediate on route to TAK-981. This was subsequently conﬁrmed through a micro-ED crystal structure of TAK-981, Figure 13.

⦁ CONCLUSION
In conclusion, a series of small molecule mechanism-based
inhibitors of SAE have been identiﬁed and described in this report. These compounds act through a substrate assisted mechanism and demonstrate robust and selective inhibition of SAE and the SUMO pathway in cells and in vivo models. A distinct allosteric covalent inhibitor approach demonstrating in vivo activity has also recently been described.40,41 An iterative medicinal chemistry program optimized a nonselective com- pound, identiﬁed from a scaﬀold hopping exercise, into inhibitors that form tight binding SUMO−inhibitor adducts that exhibit extended duration of target and pathway inhibition. This work has culminated in the identiﬁcation of TAK-981 which has entered clinical trials in metastatic solid tumors in combination with pembrolizumab, and in Non-Hodgkin’s lymphoma in combination with rituximab (NCT03648372,
NCT04074330, and NCT04381650).

⦁ EXPERIMENTAL SECTION
All animal studies were conducted under the approval of the Takeda
Oncology Institutional Animal Care and Use Committee and complied with all relevant ethical regulations. Mice were maintained in a speciﬁc pathogenfree facility in accordance with American Association for Crystal structure of TAK-981 solved by microelectron diﬀraction (microED) technique. Carbon is shown in gray; hydrogen, white; nitrogen, blue; oxygen, red; sulfur, yellow; chlorine, green. CCD accession code CCDC 2013738. Laboratory Animal Science guidelines. See Supporting Information for in vivo protocols.

Chemistry. NMR spectra were recorded in the solvent reported on a 400 MHz Bruker spectrometer using residual solvent peaks as the reference. Compound purity was determined by analysis of the diode array UV trace of an LC-MS spectrum using the following procedure: compounds were dissolved in DMSO, methanol, or acetonitrile, and the solutions were analyzed using a Hewlett-Packard HP1100 or Agilent 1100 Series LC system connected to a Micromass mass spectrometer using reverse phase C18 columns. One of two gradients was used to elute the compounds: either a formic acid (FA) gradient (acetonitrile containing 0−100% 0.1% formic acid in water) or an ammonium acetate (AA) gradient (acetonitrile containing 0−100% 10 mM
ammonium acetate in water). All compounds were determined to be >95% pure unless otherwise noted.

Synthesis of 2. Step 1:1-Benzyl-1H-pyrazole-3-carbaldehyde (I- 1). A 100 mL round-bottom ﬂask was charged with 1-pyrazole-3- carbaldehyde (250 mg, 2.60 mmol), Cs2CO3 (2.12 g, 6.50 mmol), and DMF (10 mL). To the suspension was added benzyl bromide (0.33 mL,
2.70 mmol), and the reaction was stirred for 1 h. The reaction mixture was poured into water (50 mL), and the mixture was extracted with EtOAc (50 mL) three times. The combined organic layers were washed with brine, dried over Na2SO4, ﬁltered, and concentrated in vacuo. The residue was puriﬁed on silica gel to provide I-1 (372 mg, 77%). 1H

(1-Benzyl-1H-pyrazol-3-yl)(4-chloropyrimidin-5-yl)- methanol (I-2). 4-Chloro-5-iodopyrimidine (300 mg, 1.25 mmol) was weighed into a 100 mL 2-necked round-bottom ﬂask, and the ﬂask was purged with argon. This starting material was dissolved in THF (10 mL), and the solution was cooled to −78 °C. To the solution was added n-B uLi (2.50 M in hexane; 1.0 mL, 2.5 mmol) at −78 °C, and then the mixture was stirred for 30 min. To this mixture was added dropwise a solution of 1-benzyl-1H-pyrazole-3-carbaldehyde (211 mg, 1.1 mmol) in THF (4 mL), and the resulting mixture was stirred for 30 min. The reaction was quenched by addition of saturated NH4Cl (50 mL) and extracted with EtOAc (50 mL) four times. The combined organic layers were washed with brine, dried over Na2SO4, ﬁltered, and concentrated in vacuo. The residue was puriﬁed on silica gel to provide I-2 (304 mg, 85%) as a light-yellow oil. LCMS (FA): m/z = 301.4 (M+H).

Step 3: (1-Benzyl-1H-pyrazol-3-yl)(4-chloropyrimidin-5-yl)- methanone (I-3). To a solution of (1-benzyl-1H-pyrazol-3-yl)(4- chloropyrimidin-5-yl)methanol (285 mg, 0.95 mmol) in DCM (10 mL) was added MnO2(0.82 g, 9.5 mmol), and the mixture was stirred for 15 h at rt. The reaction was ﬁltered through a pad of Celite, and the residual solid was washed with DCM several times. The ﬁltrate was concentrated in vacuo, and the residue was puriﬁed on silica gel to give I-3 (257 mg, 91%) as a colorless solid. 1H NMR (CDCl3) δ 9.09 (s, 1H) 8.90 (s, 1H) 7.45 (d, 1H, J = 2.5 Hz) 7.32−7.41 (m, 3H) 7.22 (m, 2H) 7.02 (d, 1H, J = 2.5 Hz) 5.33 (s, 2H); LCMS (FA): m/z = 299.4 (M +H). Step 4: (1-Benzyl-1H-pyrazol-3-yl)[4-({(1R,3R,4S)-3-(hydroxy- methyl)-4-[(triisopropylsilyl)oxy]cyclopentyl}amino)pyrimidin-5-yl]- methanone (I-9). To a solution of {(1R,2S,4R)-4-amino-2 [(triisopropylsilyl)oxy]cyclopentyl}methanol-TFA (I-8, 1.04 g, 2.58 mmol) in DMF (20 mL) was added K2CO3 (1.1 g, 7.7 mmol) followed by (1-benzyl-1H-pyrazol-3-yl)(4-chloropyrimidin-5-yl)methanone (I- 3, 0.85 g, 2.8 mmol) at rt, and the mixture was stirred for 13 h. The reaction was then concentrated in vacuo.

⦁ ASSOCIATED CONTENT
*sı Supporting Information
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jmedchem.0c01491.
Molecular formula strings ﬁle and some data (CSV)
Protocols for the in vitro and cell assays and in vivo experiments, synthetic methods for the remaining examples, HPLC traces of ﬁnal compound, protein synthesis and puriﬁcation methodology, and X-ray crystallographic and microelectron diﬀraction protocols (PDF)

Accession Codes
PDB depositions are available for SAE SUMO adducts of compounds 2, 46, 49, and TAK-981. The authors will release the atomic coordinates and experimental data upon article publication.

⦁ AUTHOR INFORMATION
Corresponding Authors
Steven P. Langston − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States; Imageorcid.org/0000-0001-5840-0005; Email: [email protected]
Stephen Grossman − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States; Email: [email protected]

Authors
Dylan England − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Roushan Afroze − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Neil Bence − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Douglas Bowman − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Nancy Bump − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Ryan Chau − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Bei-Ching Chuang − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Christopher Claiborne − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Larry Cohen − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States; Takeda Pharmaceuticals, Fujisawa, Kanagawa 251-0012, Japan
Kelly Connolly − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Matthew Duﬀey − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States; Takeda
Pharmaceuticals, Fujisawa, Kanagawa 251-0012, Japan
Nitya Durvasula − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States; Takeda Pharmaceuticals, Fujisawa, Kanagawa 251-0012, Japan
Scott Freeze − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States; Takeda
Pharmaceuticals, Fujisawa, Kanagawa 251-0012, Japan
Melissa Gallery − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States; Takeda Pharmaceuticals, Fujisawa, Kanagawa 251-0012, Japan
Katherine Galvin − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Jeﬀrey Gaulin − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Rachel Gershman − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Paul Greenspan − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Jessica Grieves − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Jianping Guo − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Nanda Gulavita − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Shumet Hailu − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Xingyue He − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Kara Hoar − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Yongbo Hu − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Zhigen Hu − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Mitsuhiro Ito − Takeda Pharmaceuticals, Fujisawa, Kanagawa 251-0012, Japan
Mi-Sook Kim − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Scott Weston Lane − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
David Lok − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Anya Lublinsky − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
William Mallender − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Charles McIntyre − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
James Minissale − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Hirotake Mizutani − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Miho Mizutani − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Nina Molchinova − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Koji Ono − Takeda Pharmaceuticals, Fujisawa, Kanagawa 251-0012, Japan
Ashok Patil − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Mark Qian − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States

Jessica Riceberg − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Vaishali Shindi − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Michael D. Sintchak − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Keli Song − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Teresa Soucy − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Yana Wang − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
He Xu − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Xiaofeng Yang − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Agatha Zawadzka − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Ji Zhang − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Sai M. Pulukuri − Millennium Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Ltd., Cambridge, Massachusetts 02139, United States
Complete contact information is available at: https://pubs.acs.org/10.1021/acs.jmedchem.0c01491

Notes
The authors declare no competing ﬁnancial interest.

⦁ ACKNOWLEDGMENTS
We would like to thank Dr. Dennis Huszar for reading the
manuscript. We would like to thank Drs. John Ringeling, John Donovan, and Saurabh Menon for advice on assay development. We would also like to thank NanoImaging Inc. for performing the microED data collection.

⦁ ABBREVIATIONS USED
BID x21, twice per day for 21 days; BID BIW, twice per day/
twice per week; BID 2on-5oﬀ, twice per day on 2 consecutive days, followed by 5 nondosing days; CA, carbonic anhydrase; QD x21, once per day for 21 consecutive days; QD x3/week, Three consecutive dosing days followed by 4 nondosing days; SC, subcutaneously; DLBCL, diﬀuse large B-cell lymphoma; IHC, immunohistochemistry; IF, immunoﬂuorescence; IFN, type 1 interferon; IV, intravenously; PD, pharmacodynamic; SCID, severe combined immune deﬁciency; SEM, standard error of the mean

⦁ REFERENCES
(1) Geiss-Friedlander, R. (1) Concepts in sumoylation: a decade on. Nat.
Rev. Mol. Cell Biol. 2007, 8 (12), 947−956.
(2) Schulman, B. A.; Harper, J. W. Ubiquitin-like protein activation by E1 enzymes: the apex for downstream signalling pathways. Nat. Rev. Mol. Cell Biol. 2009, 10 (5), 319−331.
(3) Gareau, J. R.; Lima, C. D. The SUMO pathway: emerging
mechanisms that shape specificity, conjugation and recognition. Nat. Rev. Mol. Cell Biol. 2010, 11, 861−871.
(4) Bernier-Villamor, V.; Sampson, D. A.; Matunis, M. J.; Lima, C. D.
Structural basis for E2-mediated SUMO conjugation revealed by a complex between ubiquitin-conjugating enzyme Ubc9 and RanGAP1. Cell 2002, 108 (3), 345−356.
(5) Johnson, E. S. Protein modification by SUMO. Annu. Rev.
Biochem. 2004, 73, 355−382.
(6) Hershko, A.; Ciechanover, A. The ubiquitin system. Annu. Rev. Biochem. 1998, 67, 425−479.
(7) Gong, L.; Yeh, E. T. Identification of the activating and
conjugating enzymes of the NEDD8 conjugation pathway. J. Biol. Chem. 1999, 274 (17), 12036−12042.
(8) Bergink, S.; Jentsch, S. Principles of ubiquitin and SUMO modifications in DNA repair. Nature 2009, 458 (7237), 461−467.
(9) Chen, S. F.; Gong, C.; Luo, M.; Yao, H. R.; Zeng, Y. J.; Su, F. X.
Ubc9 expression predicts chemoresistance in breast cancer. Aizheng
2011, 30 (9), 638−644.
(10) Seeler, J. S.; Dejean, A. SUMO and the robustness of cancer. Nat. Rev. Cancer 2017, 17 (3), 184−197.
(11) Kessler, J. D.; Kahle, K. T.; Sun, T.; Meerbrey, K. L.; Schlabach,
M. R.; Schmitt, E. M.; Skinner, S. O.; Xu, Q.; Li, M. Z.; Hartman, Z. C.; Rao, M.; Yu, P.; Dominguez-Vidana, R.; Liang, A. C.; Solimini, N. L.; Bernardi, R. J.; Yu, B.; Hsu, T.; Golding, I.; Luo, J.; Osborne, C. K.; Creighton, C. J.; Hilsenbeck, S. G.; Schiff, R.; Shaw, C. A.; Elledge, S. J.; Westbrook, T. F. A SUMOylation-dependent transcriptional subpro- gram is required for Myc-driven tumorigenesis. Science 2012, 335, 348− 353.
(12) He, X.; Riceberg, J.; Pulukuri, S. M.; Grossman, S.; Shinde, V.; Shah, P.; Brownell, J. E.; Dick, L.; Newcomb, J.; Bence, N. Characterization of the loss of SUMO pathway function on cancer cells and tumor proliferation. PLoS One 2015, 10 (4), No. e0123882.
(13) Liu, X.; Xu, Y.; Pang, Z.; Guo, F.; Qin, Q.; Yin, T.; Sang, Y.; Feng, C.; Li, X.; Jiang, L.; Shu, P.; Wang, Y. Knockdown of SUMO-activating enzyme subunit 2 (SAE2) suppresses cancer malignancy and enhances chemotherapy sensitivity in small cell lung cancer. J. Hematol. Oncol. 2015, 8, 1−10.
(14) Biederstadt, A.; Hassan, Z.; Schneeweis, C.; Schick, M.;
Schneider, L.; Muckenhuber, A.; Hong, Y.; Siegers, G.; Nilsson, L.; Wirth, M.; Dantes, Z.; Steiger, K.; Schunck, K.; Langston, S.; Lenhof, H. P.; Coluccio, A.; Orben, F.; Slawska, J.; Scherger, A.; Saur, D.; Muller, S.; Rad, R.; Weichert, W.; Nilsson, J.; Reichert, M.; Schneider, G.; Keller, U. SUMO pathway inhibition targets an aggressive pancreatic cancer subtype. Gut 2020, 8, 1472−1482.
(15) Adorisio, S.; Fierabracci, A.; Muscari, I.; Liberati, A. M.; Ayroldi,
E.; Migliorati, G.; Thuy, T. T.; Riccardi, C.; Delfino, D. V. SUMO proteins: Guardians of immune system. J. Autoimmun. 2017, 84, 21−28.
(16) Decque, A.; Joffre, O.; Magalhaes, J. G.; Cossec, J. C.; Blecher-
Gonen, R.; Lapaquette, P.; Silvin, A.; Manel, N.; Joubert, P. E.; Seeler, J. S.; Albert, M. L.; Amit, I.; Amigorena, S.; Dejean, A. Sumoylation coordinates the repression of inflammatory and anti-viral gene- expression programs during innate sensing. Nat. Immunol. 2016, 17 (2), 140−149.
(17) Crowl, J. T.; Stetson, D. B. SUMO2 and SUMO3 redundantly
prevent a noncanonical type I interferon response. Proc. Natl. Acad. Sci. U. S. A. 2018, 115 (26), 6798−6803.
(18) Brownell, J. E.; Sintchak, M. D.; Gavin, J. M.; Liao, H.; Bruzzese,
F. J.; Bump, N. J.; Soucy, T. A.; Milhollen, M. A.; Yang, X.; Burkhardt, A. L.; Ma, J.; Loke, H. K.; Lingaraj, T.; Wu, D.; Hamman, K. B.; Spelman, J. J.; Cullis, C. A.; Langston, S. P.; Vyskocil, S.; Sells, T. B.; Mallender, W. D.; Visiers, I.; Li, P.; Claiborne, C. F.; Rolfe, M.; Bolen, J. B.; Dick, L. R. Substrate-assisted inhibition of ubiquitin-like protein-activating enzymes: the NEDD8 E1 inhibitor MLN4924 forms a NEDD8-AMP mimetic in situ. Mol. Cell 2010, 37 (1), 102−111.
(19) Chen, J. J.; Tsu, C. A.; Gavin, J. M.; Milhollen, M. A.; Bruzzese, F. J.; Mallender, W. D.; Sintchak, M. D.; Bump, N. J.; Yang, X.; Ma, J.; Loke, H. K.; Xu, Q.; Li, P.; Bence, N. F.; Brownell, J. E.; Dick, L. R. Mechanistic studies of substrate-assisted inhibition of ubiquitin- activating enzyme by adenosine sulfamate analogues. J. Biol. Chem. 2011, 286 (47), 40867−40877.
(20) Lu, X.; Olsen, S. K.; Capili, A. D.; Cisar, J. S.; Lima, C. D.; Tan, D.
S. Designed semisynthetic protein inhibitors of Ub/Ubl E1 activating enzymes. J. Am. Chem. Soc. 2010, 132 (6), 1748−1749.
(21) Olsen, S. K.; Capili, A. D.; Lu, X.; Tan, D. S.; Lima, C. D. Active
site remodelling accompanies thioester bond formation in the SUMO E1. Nature 2010, 463 (7283), 906−912.
(22) Swords, R. T.; Erba, H. P.; DeAngelo, D. J.; Bixby, D. L.; Altman,
J. K.; Maris, M.; Hua, Z.; Blakemore, S. J.; Faessel, H.; Sedarati, F.; Dezube, B. J.; Giles, F. J.; Medeiros, B. C. Pevonedistat (MLN4924), a First-in-Class NEDD8-activating enzyme inhibitor, in patients with acute myeloid leukaemia and myelodysplastic syndromes: a phase 1 study. Br. J. Haematol. 2015, 169 (4), 534−543.
(23) Mizutani, H.; Langston, S. P.; Vyskocil, S. Synthesis of a Potent
NAE Inhibitor: Pevonedistat. In Comprehensive Accounts of Pharma- ceutical Research and Development: From Discovery to Late-Stage Process Development Vol. 2; Ahmed, F. A.-M., Pesti, J. A.., Vaidyanathan, R., Eds.; ACS Symposium Series: 2016; Vol. 1240, pp 1−11.
(24) Milhollen, M. A.; Traore, T.; Adams-Duffy, J.; Thomas, M. P.;
Berger, A. J.; Dang, L.; Dick, L. R.; Garnsey, J. J.; Koenig, E.; Langston,
S. P.; Manfredi, M.; Narayanan, U.; Rolfe, M.; Staudt, L. M.; Soucy, T. A.; Yu, J.; Zhang, J.; Bolen, J. B.; Smith, P. G. MLN4924, a NEDD8- activating enzyme inhibitor, is active in diffuse large B-cell lymphoma models: rationale for treatment of NF-{kappa}B-dependent lymphoma. Blood 2010, 116 (9), 1515−1523.
(25) Soucy, T. A.; Smith, P. G.; Milhollen, M. A.; Berger, A. J.; Gavin,
J. M.; Adhikari, S.; Brownell, J. E.; Burke, K. E.; Cardin, D. P.; Critchley, S.; Cullis, C. A.; Doucette, A.; Garnsey, J. J.; Gaulin, J. L.; Gershman, R. E.; Lublinsky, A. R.; McDonald, A.; Mizutani, H.; Narayanan, U.; Olhava, E. J.; Peluso, S.; Rezaei, M.; Sintchak, M. D.; Talreja, T.; Thomas, M. P.; Traore, T.; Vyskocil, S.; Weatherhead, G. S.; Yu, J.; Zhang, J.; Dick, L. R.; Claiborne, C. F.; Rolfe, M.; Bolen, J. B.; Langston,
S. P. An inhibitor of NEDD8-activating enzyme as a new approach to treat cancer. Nature 2009, 458 (7239), 732−736.
(26) Hyer, M. L.; Milhollen, M. A.; Ciavarri, J.; Fleming, P.; Traore,
T.; Sappal, D.; Huck, J.; Shi, J.; Gavin, J.; Brownell, J.; Yang, Y.; Stringer,
B.; Griffin, R.; Bruzzese, F.; Soucy, T.; Duffy, J.; Rabino, C.; Riceberg, J.; Hoar, K.; Lublinsky, A.; Menon, S.; Sintchak, M.; Bump, N.; Pulukuri,
S. M.; Langston, S.; Tirrell, S.; Kuranda, M.; Veiby, P.; Newcomb, J.; Li, P.; Wu, J. T.; Powe, J.; Dick, L. R.; Greenspan, P.; Galvin, K.; Manfredi, M.; Claiborne, C.; Amidon, B. S.; Bence, N. F. A small-molecule inhibitor of the ubiquitin activating enzyme for cancer treatment. Nat. Med. 2018, 24 (2), 186−193.
(27) Ciavarri, J.; Langston, S., The Discovery of First-in-Class
Inhibitors of the Nedd8-Activating Enzyme (NAE) and the Ubiquitin- Activating Enzyme (UAE). In Comprehensive Medicinal Chemistry III, Chackalamannil, S., Rotella, D., Ward, S., Eds.; Elsevier: Oxford, 2017; Vol. 8, pp 95−112.
(28) An, H.; Statsyuk, A. V. Development of activity-based probes for
ubiquitin and ubiquitin-like protein signaling pathways. J. Am. Chem. Soc. 2013, 135 (45), 16948−16962.
(29) Misra, M.; Kuhn, M.; Lobel, M.; An, H.; Statsyuk, A. V.; Sotriffer,
C.; Schindelin, H. Dissecting the Specificity of Adenosyl Sulfamate Inhibitors Targeting the Ubiquitin-Activating Enzyme. Structure 2017, 25 (7), 1120−1129.
(30) Fukuda, I.; Ito, A.; Hirai, G.; Nishimura, S.; Kawasaki, H.; Saitoh,
H.; Kimura, K.; Sodeoka, M.; Yoshida, M. Ginkgolic acid inhibits protein SUMOylation by blocking formation of the E1-SUMO intermediate. Chem. Biol. 2009, 16 (2), 133−140.
(31) Takemoto, M.; Kawamura, Y.; Hirohama, M.; Yamaguchi, Y.;
Handa, H.; Saitoh, H.; Nakao, Y.; Kawada, M.; Khalid, K.; Koshino, H.; Kimura, K.; Ito, A.; Yoshida, M. Inhibition of protein SUMOylation by davidiin, an ellagitannin from Davidia involucrata. J. Antibiot. 2014, 67 (4), 335−338.
(32) Fukuda, I.; Ito, A.; Uramoto, M.; Saitoh, H.; Kawasaki, H.; Osada, H.; Yoshida, M. Kerriamycin B inhibits protein SUMOylation. J. Antibiot. 2009, 62 (4), 221−224.
(33) He, X.; Riceberg, J.; Soucy, T.; Koenig, E.; Minissale, J.; Gallery,
M.; Bernard, H.; Yang, X.; Liao, H.; Rabino, C.; Shah, P.; Xega, K.; Yan,
Z. H.; Sintchak, M.; Bradley, J.; Xu, H.; Duffey, M.; England, D.; Mizutani, H.; Hu, Z.; Guo, J.; Chau, R.; Dick, L. R.; Brownell, J. E.; Newcomb, J.; Langston, S.; Lightcap, E. S.; Bence, N.; Pulukuri, S. M. Probing the roles of SUMOylation in cancer cell biology by using a selective SAE inhibitor. Nat. Chem. Biol. 2017, 13 (11), 1164−1171.
(34) Lv, Z.; Williams, K. M.; Yuan, L.; Atkison, J. H.; Olsen, S. K.
Crystal structure of a human ubiquitin E1-ubiquitin complex reveals conserved functional elements essential for activity. J. Biol. Chem. 2018, 293 (47), 18337−18352.
(35) Minissale, J.; Burkhardt, A. Unpublished results.
(36) Shank, R. P.; Doose, D. R.; Streeter, A. J.; Bialer, M. Plasma and whole blood pharmacokinetics of Subasumstat topiramate: the role of carbonic anhydrase. Epilepsy Res. 2005, 63 (2−3), 103−112.
(37) Winum, J. Y.; Scozzafava, A.; Montero, J. L.; Supuran, C. T.
Sulfamates and their therapeutic potential. Med. Res. Rev. 2005, 25 (2), 186−228.
(38) Lightcap, Y. P.; Grossman, S.; Song, K.; Khattar, M.; Xega, K.;
He, X.; Gavin, J. G.; Imaichi, H.; Garnsey, J. J.; Koenig, E.; Zhang, H.;
Lu, Z.; Shah, P.; Hatton, B. A.; Riceberg, J.; Shinde, V.; Li, C.; Minissale, J.; Yang, X.; England, D.; Klinghoﬀer, R. A.; Langston, S.; Galvin, K.; Shapiro, G.; Pulukuri, S. M.; Fuchs, S. Y.; Huszar, D. Unpublished results.
(39) Huszar, D. TAK-981: A ﬁrst-in-class SUMOylation inhibitor in phase 1 clinical trials promotes a Type I interferon response and antitumor immunity in preclinical models, Session DDT01; American Association for Cancer Research: Atlanta, GA, Mar 29−Apr 03, 2019.
(40) Li, Y. J.; Du, L.; Wang, J.; Vega, R.; Lee, T. D.; Miao, Y.; Aldana-
Masangkay, G.; Samuels, E. R.; Li, B.; Ouyang, S. X.; Colayco, S. A.; Bobkova, E. V.; Divlianska, D. B.; Sergienko, E.; Chung, T. D. Y.; Fakih, M.; Chen, Y. Allosteric Inhibition of Ubiquitin-like Modifications by a Class of Inhibitor of SUMO-Activating Enzyme. Cell Chem. Biol. 2019, 26 (2), 278−288.
(41) Lv, Z.; Yuan, L.; Atkison, J. H.; Williams, K. M.; Vega, R.;
Sessions, E. H.; Divlianska, D. B.; Davies, C.; Chen, Y.; Olsen, S. K. Molecular mechanism of a covalent allosteric inhibitor of SUMO E1 activating enzyme. Nat. Commun. 2018, 9, 1−12.