Designing Dual Inhibitors of Anaplastic Lymphoma Kinase (ALK) and Bromodomain-4 (BRD4) by Tuning Kinase Selectivity.

Concomitant inhibition of anaplastic lymphoma kinase (ALK) and bromodomain-4 (BRD4) is a potential therapeutic strategy for targeting two key oncogenic drivers that co-segregate in a significant fraction of high-risk neuroblastoma patients, mutation of ALK and amplification of MYCN. Starting from known dual polo-like kinase (PLK)-1-BRD4 inhibitor BI-2536, we employed structure-based design to redesign this series toward compounds with a dual ALK-BRD4 profile. These efforts led to compound ( R)-2-((2-ethoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-7-ethyl-5-methyl-8-((4-methylthiophen-2-yl)methyl)-7,8-dihydropteridin-6(5 H)-one (16k) demonstrating improved ALK activity and significantly reduced PLK-1 activity, while maintaining BRD4 activity and overall kinome selectivity. We demonstrate the compounds' on-target engagement with ALK and BRD4 in cells as well as favorable broad kinase and bromodomain selectivity.

Introduction

Neuroblastoma is a pediatric cancer of neural crest origin and is the most common extracranial solid tumor in childhood.[1] In high-risk patients, mutations within the kinase domain of the anaplastic lymphoma kinase (ALK), such as ALKF1174L, co-segregate with amplification of the MYCN gene. Since the ALK mutation increases transcription and expression of MYCN, both oncogenic gene changes cooperate to upregulate this well-established driver of neuroblastoma proliferation, resulting in poor prognosis and clinical outcome.[2]

Several selective inhibitors of the ALK kinase have been disclosed and entered clinical trials for different indications. However, crizotinib is clinically ineffective in neuroblastoma and many second-generation ALK inhibitors are predicted to be ineffective for neuroblastoma patients harboring the F1174L mutation due to insufficient inhibition of the mutant kinase.[3] Recently, the third-generation ALK inhibitor lorlatinib was shown to potently inhibit ALKF1174L and has now entered phase I clinical trials in relapsed or refractory neuroblastoma patients.[4]

Inhibition of bromodomain-4 (BRD4) has recently emerged as an essential transcriptional co-regulator of MYCN, and inhibition of the bromodomain has been shown to be an effective therapeutic approach to target dysregulated MYCN in neuroblastoma.[5−7] Several compounds have progressed to clinical trials for adult malignancies but have yet to reach pediatric trials.[8,9]

It is increasingly recognized that targeting multiple pathways that support cancer growth and survival is necessary to treat aggressive cancers, provide a more durable response, and overcome resistance.[10] Given the clinical challenge that high-risk neuroblastoma cases pose, combining ALK and BRD4 inhibition may represent an effective therapeutic approach for this high medical need. Combining both ALK and BRD4 inhibition would serve two purposes. First, it would target the two most common and co-segregating events that drive high-risk neuroblastoma and curb MYCN expression, potentially resulting in strong antiproliferative or proapoptopic effects. Moreover, blocking two targets at once reduces the risk of resistance to the therapy since the probability of clonal adaptation to targeted therapy is lower for combination therapies.[11]

A key barrier in clinical implementation of new agents or treatment strategies in children is that combination trials of multiple drugs are challenging in pediatric patients. This is in part due to the increased chance of off-target toxicity when two agents are tested and length of trials because tolerable dose must be established for each new agent separately in very small patient populations.

An alternative approach to using two drugs in combination is to explore dual inhibitors that block both targets of a therapeutic combination, in the case of high-risk neuroblastoma, BRD4 and ALKF1174L. A dual inhibitor is likely to reduce the liabilities associated with combination treatments, particularly, off-target toxicities, drug–drug interactions, and additive effects. Furthermore, combinatorial treatment in the form of a dual inhibitor reduces the length and complexity of trials as well as costs.[10,12,13]

Dual inhibitors are thus an attractive therapeutic approach, but the design and development of drugs that specifically inhibit two targets, particularly, where these are structurally distinct and not members of the same protein family, are challenging. In particular, combining two pharmacophores into a single druglike compound while also achieving selectivity and physicochemical and pharmacokinetics properties consistent with clinical development is regarded as very difficult.[10]

However, precedent for dual kinase–bromodomain inhibitors has recently emerged. Through systematic screening efforts, Ember et al. and Ciceri et al. identified a total of 24 kinase inhibitors that interact with BRD4.[14,15] Cocrystal structures of these dual inhibitors revealed insights into how the BRD4 and kinase pharmacophores can be combined into a single druglike molecule. Although these reports provide important precedence for dual kinase–bromodomain inhibition and structural insights, the combination of bromodomain and kinase inhibited by these dual inhibitors was discovered serendipitously by screening selective kinase inhibitors against the bromo- and extra-terminal domain (BET) bromodomains. To date, there are a few published reports of discovery efforts that aim to combine inhibition of a particular kinase with bromodomain inhibition into a single dual inhibitor to explore a specific disease hypothesis.[16−18]

Herein, we describe our efforts to discover dual ALK–BRD4 inhibitors to target both oncogenic drivers of high-risk neuroblastoma. We chose the dual polo-like kinase (PLK)-1–BRD4 inhibitor BI-2536 as our starting point and investigated if this inhibitor series can be reoptimized to show potent inhibition of mutant (F1174L) ALK kinase, reduced PLK-1 activity while maintaining BRD4 activity, and acceptable kinome selectivity.

Results and Discussion

Our goal at the start of the project was to discover starting points that showed significant activity against BRD4 and the ALK kinase. We were particularly intrigued by the dual kinase–bromodomain inhibitor BI-2536 (Figure ). The compound was discovered and developed as a PLK-1 kinase inhibitor but was found to potently inhibit BRD4 by Knapp and Schönbrunn’s labs.[14,19,20] BI-2536 has been reported to show high specificity within the kinase family, partially due to the methoxy substituent. Some kinases are not able to accommodate this substituent due to a steric clash with a larger tyrosine or tryptophan residue in the hinge region. Among the exceptions are PLK-1 and importantly ALK due to the presence of a smaller leucine at this position.[21,22] We thus hypothesized that although BI-2536 showed excellent overall kinase selectivity it may show sufficient activity against ALK and ALKF1174L to serve as a starting point. We analyzed the publicly available data, and indeed a Kd of 160 nM had been disclosed for ALK.[23] In our assays, BI-2536 showed comparable, albeit modest, activity (IC50s of 390 and 190 nM against ALK and ALKF1174L, respectively).

Figure 1

Biochemical potencies of the dual PLK-1–BRD4 inhibitor BI-2536 and moieties of the molecule that form key interactions with BRD4 (orange) and the PLK-1 hinge region (violet).

In addition, we confirmed inhibition of BRD4 and PLK-1. In fact, the potency of BI-2536 against PLK-1 was beyond the dynamic range of the assay (IC50 < 2.6 nM) consistent with a published Kd of 0.19 nM.[23]

BI-2536 thus was a suitable starting point for the discovery of a dual ALK–BRD4 inhibitor, not least because it inhibited a few other kinases apart from PLK-1 and ALK.[23] We thus decided to investigate if BI-2536 can be optimized toward a dual ALK–BRD inhibitor. In particular, we sought to reduce activity against PLK-1, improve ALK activity, and maintain BRD4 activity. To analyze the scope for modifications as well as derive design hypotheses, we first analyzed the published structures of BI-2536 bound to BRD4 and PLK-1.[14,21] In addition, we docked BI-2536 into ALK (2XB7) using Glide. We chose this structure for our docking experiment due to structural similarity between the ALK inhibitor NVP-TAE684[22] and BI-2536, in particular, the aminopyrimidine hinge binding motif and the ortho-methoxy phenyl group.

In BRD4, the methyl amide moiety acts as the acetylated lysine (KAc) mimetic, interacting with Asn140 (Figure a). As described in more detail previously,[14] additional interactions arise between the aminopyrimidine group and conserved waters in the ZA channel. The PLK-1-bound structure showed that the aminopyrimidine forms the key interactions with the hinge region of the kinase including Cys133 (Figure b). Docking into ALK not surprisingly predicted that aminopyrimidine of BI-2536 interacted with the ALK hinge region including residue Met1199 in a manner similar to that for PLK-1 (Figure c). The amide carbonyl was also predicted to interact with conserved water in the back of the ALK pocket.

Figure 2

X-ray structures of BI-2536 in complex with (a) BRD4 (4OGI) and (b) PLK-1 (2RKU) highlighting key interactions. (c) Docking pose generated using Glide showing BI-2536 in complex with ALK (2XB7), highlighting key interactions. (d) Four chosen areas for modification on BI-2536.

Based on the ALK docking and BRD4 X-ray structure, we wanted to maintain the key kinase and bromodomain binding motifs on the dihydropteridinone core and thus decided to leave these regions of the compound untouched.

Our analysis also suggested that several positions around the core structure offered significant scope for modification (Figure d). The first area chosen for modification was the (R)-ethyl group (R1), which fits into a hydrophobic pocket in both PLK-1 and BRD4. The docking of BI-2536 in ALK also suggests that the (R)-ethyl is pointing toward a pocket formed by residues Val1130, Ala1148, Lys1150, and Leu1196. It has been reported that changing the chirality of the ethyl group has little effect on the potency at PLK-1 and BRD4, whereas removing the group afforded a 26-fold reduction in potency at BRD4.[24,25]

We were also interested in the cyclopentyl substitution off the dihydropteridinone core (R2), which has been shown to be a region where BRD4 and PLK-1 selectivity can be tuned.[24] For example, changing the cyclopentyl to a 3-bromobenzyl group increases BRD4 affinity due to improved hydrophobic interactions with the WPF shelf but decreases PLK-1 activity, suggesting that modifications in this region could be used to improve selectivity against PLK-1.

Next, we considered the methoxy group at the R3 position. Docking of BI-2536 suggests that the alkoxy group is pointing toward a known region for achieving ALK selectivity (Figure c).[21] The majority of kinases have a bulkier residue at this region, allowing ALK selectivity to be achieved as seen with ceritinib and alectinib.[26,27] In addition, larger alkoxy groups were known to improve interactions at BRD4.[24] We thus hypothesized that larger alkoxy groups will lead to increased selectivity against PLK-1 while maintaining BRD4 activity.

The final area chosen for modification is the solvent channel methylpiperidine group (R4). Published structure–activity relationship (SAR) data for ALK inhibitors suggested that a wide variety of groups are tolerated in this region.[28−30] We particularly decided to prepare analogues without the amide group to reduce PLK-1 activity by removing a hydrogen-bonding interaction with Leu59 and a water molecule in the PLK-1 pocket.

Following this structure-based analysis and docking, we decided to focus modifications on four areas (R1, R2, R3, and R4) and to keep the dihydropteridinone core unchanged to maintain the key kinase and bromodomain binding interactions (Figure d).

Chemistry

We prepared a series of analogues of BI-2536 by adapting and optimizing previously described syntheses.[31] For aniline intermediates containing the R3 and R4 modifications, two routes were used as shown in Scheme . For the amide-containing intermediates, the isobutyl group was first introduced by alkylation to 3-hydroxy-4-nitrobenzoic acid 1 via protection and subsequent deprotection of the benzoic acid. The 4-nitrobenzoic acid 3a was coupled to 4-amino-1-methylpiperidine using O-(benzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium hexafluorophosphate (HBTU) to give amide 4a, followed by reduction using SnCl2 to give aniline 5a. The ethoxy analogue 5b was made via the same route from commercial compound 3b. The synthesis of anilines 10a and 10b began with a Suzuki coupling between an alkoxy-nitrobenzene (6a and 6b) and 4-pyridineboronic acid. Pyridines 7a and 7b were methylated to give 8a and 8b as the iodonium salts and then partially reduced using sodium borohydride (9a and 9b). Hydrogenation of the dihydropyridine and nitro group gave aniline intermediates 10a and 10b.

Scheme 1

Meanwhile, intermediates 15a–g were prepared according to Scheme . The appropriate amino acids 11a–c underwent reductive amination with various aldehydes to give intermediates 12a–g followed by regioselective SNAr with 2,5-dichloro-4-nitropyrimidine to give 13a–g. Reductive heterocyclization using iron and acetic acid formed the dihydropteridinone scaffold (14a–g), which was methylated to give intermediates 15a–g. The final step was acid-promoted SNAr between intermediates 15a–g and anilines 5a, 5b, 10a, and 10b to yield final compounds 16a–l. Hydrogenation of compounds 16f and 16h to remove the bromine yielded compounds 16g and 16i.

Scheme 2

A key obstacle to preparing a significant number of analogues at the R2 position was that this substituent was introduced early and preparation of each example required five steps (Scheme ). To our surprise, there was no precedent for synthetic routes for late stage variation at this position. To facilitate preparation of these derivatives, we developed an alternative approach. This approach maintained many elements of the initial route (Scheme ) but started with a dimethoxybenzyl (DMB) protecting group at the R2 position. This strategy allowed us to remove the DMB group by refluxing the advanced intermediate (20) in trifluoroacetyl (TFA) and to subsequently introduce a wide range of R2 groups. A key challenge was that the reductive cyclization step on intermediate 18 employing iron and acetic acid partially removed the DMB group. We tested a range of hydrogenation conditions and, gratifyingly, PtO2 and VO(acac)2 gave an excellent yield of 98%.[32] This route enabled us to prepare the key intermediate 21 on a multigram scale and convert it into a range of final products (23a–f) in only two steps. Chiral shift NMR and chiral high-performance liquid chromatography (HPLC) confirmed that no racemization took place under either the strongly basic or strongly acidic conditions. However, on a later repeat of this synthesis, we observed that epimerization could occur during the alkylation of 19 to yield 20. Reanalysis of previous batches confirmed that no epimerization was observed in the synthesis of the batch used to prepare the compounds described in this article (Supp. Figure 1). Furthermore, we reanalyzed products of reactions obtained under similar alkylating conditions (Scheme ), confirming that racemization had not occurred here either. All final products described here were thus unaffected by the epimerization. However, due to the potential for racemization, we sought to reoptimize this step. By reducing the number of equivalents of sodium hydride or replacing this with a weaker base such as sodium hydroxide, enantiomeric ratios of 9:1 or greater could be obtained.

Scheme 3

Biological Testing

Table shows the derivatives of BI-2536, modifying the R1, R2, R3, and R4 positions, and their activities at ALKF1174L, BRD4, and PLK-1. The primary assays for biochemical testing were LanthaScreen for ALKF1174L and thermal shift against BRD4 (10 μM compound concentration). For selected compounds, this was followed by isothermal calorimetry against BRD4 and/or a PLK-1 Z-Lyte assay.

Table 1

Structure–Activity Relationships of the R1, R2, R3, and R4 Groups

Data represents the geometric mean; see Supp. Table 1 for full statistics. n.d. = not determined.

Tm shift determined at compound concentration of 10 μM, n = 2.

We started by analyzing the effect of the amide functionality in the solvent channel group (R4, Figure d). Removal of the amide group (compound 16a) had little effect against ALKF1174L and BRD4 potency. However, consistent with our hypothesis, it significantly decreased PLK-1-1 activity. Based on the measured IC50 value of 2.6 nM, 16a showed a four-fold reduction in potency against PLK-1. The real effect on PLK-1 potency of this structural change may be significantly greater, however, as the potency of BI-2536 is below the dynamic range of our assay. Next, we analyzed the effect of the R3 substituent (Figure d) and prepared analogues with an ethoxy and an isobutoxy group at this position. The ethoxy group on compound 16c gave a small reduction in potency against ALKF1174L but a greater decrease in PLK-1 potency. An increase in the size of the group further to a butoxy group (compound 16b) resulted in a much larger drop in potency against ALKF1174L. For BRD4, increasing the group to ethoxy and isobutoxy gave only a small decrease in activity, confirming the tolerance of larger groups at this position. Compound 16c provided the best balance of potency and selectivity thus far, so the ethoxy group was maintained.

We started testing the effect of the R1 group by removing the (R)-ethyl group (16e). This removal reduced potency at BRD4 and PLK-1 very likely due to the loss of favorable interactions with the aforementioned pockets in BRD4 and PLK-1. Similarly, removal of the (R)-Et group also reduced ALKF1174L potency 3-fold. We also synthesized the opposite enantiomer 16d, which also significantly reduced potency at all three targets. The reduction in PLK-1 and BRD4 activity was not in agreement with a published study but consistent with loss of productive interactions of the (R)-Et group with both PLK-1 and BRD4 and supported by modeling, which suggests that the ethyl group is too big to be accommodated on the lower face of the PLK-1 pocket. Finally, we analyzed substitution at the R2 position. Compounds 16f and 16g with a 3-bromobenzyl and benzyl group, respectively, maintained BRD4 activity and led to a 30–40-fold decrease of the PLK-1 activity, as expected from previous reports.[24] Compounds 16f and 16g were equipotent in ALKF1174L activity, compared with 16c. Importantly, the divergent SAR between ALK and PLK-1 with 16f and 16g supported our hypothesis that it is possible to optimize the ALK/PLK-1 selectivity window at the R2 position.

To conclude this initial SAR investigation, we combined modifications that showed a decrease in PLK-1 potency. Compounds 16h and 16i with the combined modifications of the benzyl or 3-bromobenzyl group at R2, ethoxy group at R3, and removal of the amide at R4 gave a greater decrease in PLK-1 activity (>100–200-fold). With benzyl compound 16i, we also saw a 2-fold increase in ALKF1174L potency and maintained potent BRD4 activity. This was the first analogue demonstrating greater potency at ALKF1174L than at PLK-1.

R2 SAR

So far, the greatest change in PLK-1 activity resulted from modifying the cyclopentyl to a benzyl or 3-bromobenzyl group (Table ). Further inspection of this region in ALK and PLK-1 structures shows that PLK-1 has a larger phenylalanine residue (Phe183) at the bottom of this pocket compared with a smaller leucine residue (Leu1286) in ALK (Figure a,b). We hypothesized that PLK-1 inhibition would be less tolerant to substitution at the R2 position due to this geometric constraint compared to that in the more open ALK structure. To investigate this hypothesis further, we performed docking studies of 3-bromobenzyl analogue 16h in PLK-1 (Figure c). The aminopyrimidine moiety was still predicted to bind to the hinge region including Cys133, but a shift in the core was observed. Additionally, the core of the molecule adopts a more strained, puckered conformation. Both differences from the observed binding mode were likely caused by clashes of the bromobenzyl group with the side chain of Phe183 and were consistent with the decrease in activity observed with benzyl analogues 16f–i. We thus decided to explore if this key amino acid difference can be exploited to discover compounds with improved ALK potency and PLK-1 selectivity.

Figure 3

Surfaces of (a) ALK (2XB7) and (b) PLK-1 (2RKU) highlighting residues Leu1256 and Phe183. (c) Docking pose generated using Glide showing 16h (orange) in complex with PLK-1 (2RKU), overlaid with X-ray structure of BI-2536 (yellow) in PLK-1 (2RKU).

To enable structure-based design, we attempted to solve cocrystal structures of our compounds bound to ALK. However, these attempts failed, likely due to the still modest activity. Due to the lack of structural information, we decided to probe the ALK pocket through systematic synthesis and testing.

Using the modified route in Scheme to facilitate preparation of analogues at the R2 position, we prepared compounds 23a–c to analyze the effect of different substitutions at the 3-benzyl position. While these compounds maintained the BRD4 activity, we did not observe improvements in ALKF1174L potency or PLK-1 selectivity, aside from compound 23b that contains a 3-cyanobenzyl group. The PLK-1 IC50 was 1.8 μM, a >700-fold decrease from starting compound BI-2536, leading to an improved (5-fold) ALK selectivity over PLK-1. Compound 23b demonstrated that PLK-1 activity can be substantially reduced by varying this position, but we were still yet to find a group that achieved this while maintaining or improving ALKF1174L potency. Next, we considered changing the substitution position on the benzyl ring. However, the ortho- and para-bromobenzyls 23d and 23e showed no improvement in ALKF1174L potency compared with meta-analogue 16h and were still not as potent as unsubstituted benzyl analogue 16i.

We then considered heterocycle substituents. Thiazole 23f showed a decrease in ALKF1174L activity, but exploring thiophenes proved to be more productive. Compound 16j gave encouraging IC50s against ALKF1174L and BRD4, yet the ALK/PLK ratio dropped to 1:1. Following from this result, we moved the methyl group around the thiophene ring. Derivative 16k with the methyl group at the 4-position showed potent inhibition of ALKF1174L (IC50 = 17 nM) and through this gain in potency also a 7-fold selectivity window over PLK-1. Gratifyingly, the compound retained BRD4 activity (Table ).

Table 2

Structure–Activity Relationships of the R2 Groupc

Data represents the geometric mean; see Supp. Table 2 for full statistics. n.d. = not determined.

Final compounds synthesized using Scheme .

Tm shift determined at compound concentration of 10 μM, n = 2.

We next determined the cocrystal structures of 16i and 16k with BRD4 (Figure and Supp. Table 3). The compounds bound into the acetyllysine-binding site with the methyl amide moiety retaining the key hydrogen bond interaction with asparagine N140. The methyl amide and pyrimidine moieties in both compounds retain interactions with conserved waters in the BRD4 pocket, whereas the thiophene and benzyl R2 substituents are situated in the hydrophobic WPF shelf region.

Figure 4

(a) Cocrystal structure of 16i with BRD4 (6Q3Y) and associated |2FO| – |FC| refined electron density map contoured at 1σ. (b) Cocrystal structure of 16k with BRD4 (6Q3Z) and associated |2FO| – |FC| refined electron density map contoured at 1σ.

With compound 16k, we had a compound in hand that showed satisfactory IC50s against ALKF1174L and BRD4 and a selectivity window over PLK-1. We decided to profile 16k further by examining the physicochemical properties and broader kinase and bromodomain selectivity. Lipophilicity was measured using the SiriusT3 potentiometric method, giving a high log P of 6.1 (Table ). It is notable that the measured log D7.4 of 4.2 is significantly lower due to protonation of the piperidine basic center, which can also account for the good solubility. In addition, the compound shows good permeability as measured in a PAMPA permeability assay.

Table 3

Physicochemical Properties of 16k

parameter	result
solubility (HPLC)a	90 μM
PAMPA pH7.4	49 × 10–6 cm/s
log P	6.1
log D7.4	4.2

Solubility measured via an in-house HPLC method from phosphate buffer at pH 7.4.

Next, we considered the broader kinase and bromodomain selectivity of 16k to assess if the compound maintained the selectivity across both protein families. Compound 16k was first tested against a panel of BET and non-BET bromodomains. In this panel, 16k exhibited excellent BET family selectivity, in common with BI-2536 and other BET inhibitors (Figure a and Supp. Table 4).[14,33,34] We then tested 16k against the DiscoverX scanEDGE panel of 97 kinases at a single point concentration of 1 μM.[35] Gratifyingly, only four kinases were identified as hits, confirming that we had maintained broad kinome selectivity; the greatest inhibition was observed for ALK wild type (WT) and PLK-1 followed by insulin receptor (INSR) and PLK-3 (Figure b and Table ). We attributed the observed broad kinase selectivity of 16k at least partially to be due to the ethoxy group.[21,27] This was based on the observation that the kinases that showed significant inhibition (ALK, PLK-1, PLK-3, and INSR) indeed feature a smaller leucine residue in the hinge region that can accommodate the ethoxy group (Supp. Table 5).

Figure 5

(a) Screening of 16k against a panel of 25 bromodomains, showing ΔTm as circles. Larger circles indicate greater ΔTm. (b) Screening of 16k against the DiscoverX scanEDGE panel of 97 kinases at a single point concentration of 1 μM. Larger circles indicate a stronger hit.

Table 4

Kds of 16k against the Kinase Screening Hits

kinase	hits at 1 μM (% control)	Kd (nM)
ALK WT	7.1	89
INSR	25	190
PLK-1	1	11
PLK-3	25	160
ALKF1174L		23

Demonstrating Target Engagement in Cells

We next tested our lead compounds in the cellular context to confirm that we had on-target engagement of ALK and BRD4 in cells. We first tested the effect of compound 16k on autophosphorylation of the ALKF1174L mutant in Kelly neuroblastoma cells using an meso-scale discovery (MSD) assay (Figure ). For comparison, we also tested our starting point BI-2536 and the Food and Drug Administration-approved ALK inhibitor ceritinib[26] as a positive control. Compound 16k showed inhibition of ALKF1174L phosphorylation levels, in line with what was expected from the biochemical IC50 and representing a significant improvement from starting compound BI-2536.

Figure 6

ALK MSD immunoassay measuring ALK phosphorylation in the Kelly cell line. The plot shows the ratio of phosphorylated ALK to total ALK.

We also measured the cellular potency of our lead compounds against BRD4 and ALK WT using NanoBRET assays (Figure and Supp. Table 6). Compound 16k showed a cellular potency of 470 nM against ALK WT and 260 nM against BRD4. The BRD4 cellular potency was comparable to that of starting compound BI-2536, consistent with the compound’s minimal disruption to the key interactions with the BRD4 pocket. The results of the ALK autophosphorylation assay and NanoBRET experiments demonstrated that 16k inhibited both ALKF1174L and BRD4 in cells.

Figure 7

NanoBRET data for lead compounds 16i–k against ALK and BRD4 in HEK293T cells.

Cellular Kinase Selectivity

The observation that the biochemical potency of 16k translated efficiently into cellular assays with a relatively minor drop-off consistent with other ALK inhibitors prompted us to investigate the selectivity against PLK-1 in cells. Despite very good overall selectivity in the DiscoverX scan, 16k showed similar Kds for the ALKF1174L mutant and PLK-1 (23 vs 11 nM, Table ). However, given the small cellular drop-off for ALK mentioned above, we speculated that a larger drop-off for PLK-1 may lead to a better selectivity window. To investigate this hypothesis, we investigated markers of PLK-1 inhibition in cells, namely, mitotic arrest in the G2/M phase and increased concentrations of PLK-1, cyclin B1, and phosphohistone H3. Indeed, BI-2536 has been published to show effects on these markers at concentration around 50 nM, which is a 250-fold drop-off from its Kd of 0.19 nM.[36] As compound 16k is structurally similar to BI-2536, we would expect to see a similar potency drop-off in cells, in this case to >2.5 μM. To assess the cellular selectivity, we assessed the effect of compounds 16k and BI-2536 on HeLa cells. Consistent with data mentioned above, BI-2536 induced mitotic arrest and increased concentrations of PLK-1, cyclin B1, and phosphohistone H3 at a concentration as low as 50 nM (Figure ).[36] In addition, we observed an increase in PLK-1 phosphorylation at sites T210 and S137. In strong contrast, compound 16k did not cause an increase in levels of these mitotic markers and phosphorylation sites up to concentrations of 10 μM and thus well beyond the concentration where ALKF1174L and BRD4 are inhibited in the cellular context. Quantitative analysis of the cell cycle distribution by flow cytometry supported the data obtained by Western blot analysis (Supp. Figure 2).

Figure 8

Induction of mitotic arrest by BI-2536 and 16k in HeLa cells. Changes in the fraction of cells arrested in mitosis were analyzed by Western blots against phosphorylation sites pT210 and pS137 and mitotic markers PLK-1, cyclin B1, and phosphohistone H3.

Based on the cellular data, the combination of the reduction in PLK-1 activity and the improvement in ALKF1174L activity thus resulted in a >10–20-fold selectivity window for ALKF1174L over PLK-1 in cells. Compound 16k thus represented a significant step toward discovering drugs that concomitantly target mutant ALK and BRD4.

Conclusions

To our knowledge, we report the first combined ALK–BRD4 inhibitor (16k). Moreover, our work is one of the few examples where medicinal chemistry design was applied to combine inhibition of a specific kinase (in this case ALK) with BRD4 inhibition to discover a dual inhibitor for a clearly defined therapeutic hypothesis.

We started from the known dual inhibitor BI-2536 that inhibits PLK-1 and BRD4. Interestingly, although tuning the PLK-1 and BRD4 activity of BI-2536 has been disclosed, changing the kinase activity to another target and hence to a new dual kinase–bromodomain combination has not been reported.[24,25] Exploring a structure-based design, we discovered compounds with significantly improved ALK activity and greatly reduced PLK-1 activity while maintaining BRD4 potency and overall kinome selectivity. Our lead compound demonstrates on-target engagement of ALK and BRD4 in cellular assays and favorable broad kinase and bromodomain selectivity. In particular, the compound showed selectivity over PLK-1 in cells underlining the design effort to change the kinase selectivity from PLK-1 to ALK.

Our work also highlighted the well-known challenges of designing and discovering dual inhibitors that inhibit two, structurally distinct proteins. A particular challenge is to incorporate two distinct pharmacophores that ensure not only potent inhibition of both targets but also selectivity within these protein families (in our case, kinases and bromodomains) into one druglike compound.

Despite these challenges, we achieved significant steps toward a dual ALK and BRD4 inhibitor. Key to this progress was choosing a starting point that contained an alkoxy group that is tolerated only by a few kinases including ALK and PLK-1. By maintaining this throughout, we achieved a compound with satisfactory overall kinome selectivity.

Our main optimization goal starting from BI-2536 was to significantly improve ALK activity and PLK-1 selectivity. The challenge here was that key regions of BI-2536 had to remain untouched to maintain the BRD4 activity. In particular, the methyl substituent that points toward the gatekeeper and the ethyl group would have been promising places to improve activity for ALK and selectivity over PLK-1. However, both moieties are essential for potent binding to BRD4 and could hence not be changed. Crucially, our structure-based analysis also suggested that the R2 substituent can be modified without losing BRD4 activity and that optimizing this part of the molecule was a promising approach to achieve our goal of shifting kinase selectivity, as exemplified by 16k.

We thus conclude that, not surprisingly, the quality and properties of the selected starting point matter even more for programs aiming at dual inhibitors since the options for chemical optimization will be significantly more limited. Directly related to that, the challenge of discovering dual inhibitors is somewhat less daunting when both targets tolerated a range of different chemotypes and inhibitors, thus increasing the chance that a starting point with already favorable properties can be explored.

Experimental Section

General Synthetic Information

All anhydrous solvents and reagents were obtained from commercial suppliers and used without further purification. All reactions were carried out under a positive pressure of N2, and moisture-sensitive reagents were transferred via a syringe. Evaporation of solvent was carried out using a rotary evaporator under reduced pressure at a bath temperature of up to 60 °C. Column chromatography was carried out using a Biotage SP4 purification system using Biotage SNAP Kp-Si cartridges. Semipreparative separations were carried out using a 1200 series preparative HPLC over a 15 min gradient elution (Grad15min20mls.m) from 90:10 to 0:100 water/methanol (both modified with 0.1% formic acid) at a flow rate of 20 mL/min. Microwave-assisted reactions were carried out using a Biotage Initiator microwave system. In general, the course of reactions was followed by thin layer chromatography or mass spectroscopy. All final compounds were purified to ≥95% purity.

NMR spectra were recorded on a Bruker AMX 500 (500 MHz) spectrometer using a deutrated solvent. NMR data is presented in the form of chemical shift δ (multiplicity, coupling constants, and integration) for major diagnostic protons, given in parts per million (ppm) relative to tetramethylsilane as an internal standard. High-resolution mass spectrometry (HRMS) was assessed using an Agilent 1200 series HPL and diode array detector coupled to a 6120 time-of-flight mass spectrometer with a dual multimode atmospheric-pressure chemical ionization/electrospray ionization (ESI) source. Analytical separation was carried out at 30 °C on a Merck Purospher STAR column (RP-18e, 30 × 4 mm2) using a flow rate of 1.5 mL/min in a 4 min gradient elution; solvents: aqueous (0.1% formic aid) and methanol, monitored at a wavelength of 254 nm. Optical rotation was determined by an ADP 440 polarimeter. Specific rotations [α]D are given in deg cm3/(g dm).

Methyl 3-Isobutoxy-4-nitrobenzoate (2)

Step 1: 3-Hydroxy-4-nitrobenzoic acid 1 (500 mg, 2.73 mmol) was suspended in MeOH (0.1 M), and SOCl2 (0.40 mL, 5.46 mmol) was added slowly at 0 °C. The reaction was heated at reflux for 4 h. Upon completion, the volatiles were removed in vacuo and the residue was triturated with diethyl ether. The resulting solid was filtered and dried under vacuum to afford methyl 3-hydroxy-4-nitrobenzoate (HCl salt) as a yellow solid (576 mg, 90%). δH (500 MHz, CDCl3): 10.51 (s, 1H), 8.18 (d, J = 8.5 Hz, 1H), 7.84 (d, J = 1.6 Hz, 1H), 7.62 (dd, J = 8.5, 1.6 Hz, 1H), 3.97 (s, 3H). Step 2: To the crude product in DMF (0.1 M) was added K2CO3 (592 mg, 4.28 mmol) and 1-iodo-2-methylpropane (142 μL, 1.20 mmol), and the reaction was stirred at 50 °C for 16 h. Upon completion, EtOAc was added and the mixture was washed with water (×2) and brine (×1), dried over MgSO4, filtered, and concentrated in vacuo. The residue was purified by Biotage column chromatography (cHex/EtOAc, 0–20%) to afford 2 as a colorless oil (158 mg, 0.62 mmol 73%). δH (500 MHz, CDCl3): 7.81 (d, J = 8.2 Hz, 1H), 7.71 (d, J = 1.6 Hz, 1H), 7.67–7.64 (m, 1H), 3.96 (s, 3H), 3.92 (d, J = 6.3 Hz, 1H), 2.20–2.11 (m, 1H), 1.05 (d, J = 6.9 Hz, 6H).

3-Isobutoxy-4-nitrobenzoic Acid (3a)

To 2 (150 mg, 0.59 mmol) in a solvent mixture of THF/H2O (1:1, 0.1 M) was added LiOH (142 mg, 5.92 mmol), and the reaction was stirred for 18 h. Upon completion, the aqueous layer was acidified with 1 M HCl until pH 1 was reached and extracted with EtOAc (×3). The combined organic phases were dried over MgSO4, filtered, and concentrated in vacuo to afford 3a as a yellow solid (100 mg, 0.42 mmol, 71%). δH (500 MHz, CDCl3): 7.85 (d, J = 8.2 Hz, 1H), 7.79 (d, J = 1.3 Hz, 1H), 7.76 (dd, J = 8.2, 1.3 Hz, 1H), 3.95 (d, J = 6.3 Hz, 1H), 2.23–2.14 (m, 1H), 1.08 (d, J = 6.9 Hz, 6H).

3-Isobutoxy-N-(1-methylpiperidin-4-yl)-4-nitrobenzamide (4a)

To a solution of 3a (50 mg, 0.21 mmol) in DMF (0.1 M) was added HBTU (127 mg, 0.33 mmol), Et3N (58 μL, 0.41 mmol), and 1-methylpiperidin-4-amine (23.9 mg, 0.21 mmol), and the reaction was stirred for 18 h. Upon completion, the reaction mixture was partitioned between EtOAc and water and the aqueous layer was extracted with EtOAc (×2). The organic phases were combined, dried over MgSO4, filtered, and concentrated in vacuo. The residue was purified by Biotage column chromatography (dichloromethane (DCM)/MeOH, 0–10%) to afford 4a as a yellow solid (60 mg, 0.18 mmol, 86%). HRMS (ESI +ve): found [M]+ 336.1927, [C17H24N3O4]+ requires 336.1923; δH (500 MHz, CDCl3) 7.81 (d, J = 8.2 Hz, 1H), 7.55 (d, J = 1.6 Hz, 1H), 7.28 (dd, J = 8.2, 1.6 Hz, 1H), 6.44 (d, J = 7.6 Hz, 1H), 4.06–3.97 (m, 1H), 3.92 (d, J = 6.3 Hz, 2H), 2.94 (d, J = 11.7 Hz, 2H), 2.37 (s, 3H), 2.30–2.25 (m, 2H), 2.18–2.10 (m, 1H), 2.09–2.03 (m, 2H), 1.78–1.69 (m, 2H), 1.04 (d, J = 6.6 Hz, 6H).

3-Ethoxy-N-(1-methylpiperidin-4-yl)-4-nitrobenzamide (4b)

4b was synthesized in 77% yield (orange solid, 111 mg, 0.36 mmol) according to the same procedure as 4a, from 3b (100 mg, 0.47 mmol) and 1-methylpiperidin-4-amine (54.1 mg, 0.47 mmol). HRMS (ESI +ve): found [M]+ 308.1624 [C15H21N3O4]+ requires 308.1610; δH (500 MHz, CDCl3): 7.82 (d, J = 8.2 Hz, 1H), 7.58 (d, J = 1.9 Hz, 1H), 7.23 (dd, J = 8.2, 1.9 Hz, 1H), 5.99 (d, J = 7.3 Hz, 1H), 4.26 (q, J = 7.2 Hz, 2H), 4.05–3.94 (m, 1H), 2.87 (d, J = 11.2 Hz, 2H), 2.33 (s, 3H), 2.19 (t, J = 11.2 Hz, 2H), 2.10–2.03 (m, 2H), 1.65–1.56 (m, 2H), 1.49 (t, J = 6.9 Hz, 3H).

4-Amino-3-isobutoxy-N-(1-methylpiperidin-4-yl)benzamide (5a)

4a (60 mg, 0.18 mmol) was dissolved in a solvent mixture of EtOAc/EtOH (10:3, 0.1 M) to which tin(II) chloride dihydrate (202 mg, 0.89 mmol) was added. The reaction was stirred for 16 h at 50 °C. Upon completion, NaHCO3 was added and the aqueous layer was extracted with EtOAc (×3). The combined organic phases were dried over MgSO4, filtered, and concentrated in vacuo to afford 5a as a yellow solid (30 mg, 99 μmol, 55%). HRMS (ESI +ve): found [M]+ 306.22161, [C17H28N3O2]+ requires 306.2176; δH (500 MHz, CDCl3) 7.32 (d, J = 1.9 Hz, 1H), 7.10 (d, J = 8.2, 1.9 Hz, 1H), 6.66 (d, J = 8.2 Hz, 1H), 5.86 (d, J = 6.3 Hz, 1H), 4.12 (s, 2H), 4.02–3.92 (m, 1H), 3.83 (d, J = 6.3 Hz, 2H), 2.83 (d, J = 11.0 Hz, 2H), 2.31 (s, 3H), 2.22–2.09 (m, 3H), 2.07–2.00 (m, 2H), 1.57 (qd, J = 11.7, 3.6 Hz, 2H), 1.04 (d, J = 6.9 Hz, 6H).

4-Amino-3-ethoxy-N-(1-methylpiperidin-4-yl)benzamide (5b)

5b was synthesized in 53% yield (yellow solid, 46 mg, 0.16 mmol) according to the same procedure as 5a, from 4b (95 mg, 0.31 mmol). HRMS (ESI +ve): found [M]+ 278.1874 [C15H23N3O2]+ requires 278.1863; δH (500 MHz, CDCl3) 7.33 (d, J = 1.9 Hz, 1H), 7.11 (dd, J = 8.1, 1.9 Hz, 1H), 6.66 (d, J = 8.1 Hz, 1H), 5.86 (d, J = 7.3 Hz, 1H), 4.18–4.09 (m, 4H), 4.05–3.94 (m, 1H), 2.86 (d, J = 11.0 Hz, 2H), 2.33 (s, 3H), 2.23–2.18 (m, 2H), 2.04 (d, J = 10.4 Hz, 2H), 1.61–1.57 (m, 2H), 1.45 (t, J = 6.9 Hz, 3H).

4-(3-Ethoxy-4-nitrophenyl)pyridine (7a)

4-Bromo-2-ethoxy-1-nitrobenzene 6a (1.5 g, 6.10 mmol), pyridine-4-ylboronic acid (749 mg, 6.10 mmol), Na2CO3 (1.1 g, 10.0 mmol), and Pd(PPh3)2Cl2 (160 mg, 0.23 mmol) were dissolved in a solvent mixture of dioxane/water (6:1, 0.38 M) and heated at 110 °C for 45 min under microwave irradiation. Upon completion, the mixture was partitioned between EtOAc and water and the aqueous phase was extracted with EtOAc (×2) and DCM (×1). The organic phases were combined, dried over MgSO4, filtered, and concentrated in vacuo. The residue was purified by Biotage column chromatography (DCM/MeOH, 9:1) to afford 7a as a yellow solid (81 mg, 3.32 mmol, 54%). HRMS (ESI +ve): found [M]+ 245.0917 [C13H12N2O3]+ requires 245.0921; δH (500 MHz, CDCl3): 8.74 (dd, J = 4.4, 1.19 Hz, 2H), 7.96 (d, J = 8.5 Hz, 1H), 7.49 (d, J = 4.4, 1.6 Hz, 2H), 7.27–7.24 (m, 2H), 4.28 (q, J = 7.0 Hz, 2H), 1.53 (dt, J = 7.0 Hz, 3H)

4-(3-Methoxy-4-nitrophenyl)pyridine (7b)

7b was synthesized in 65% yield (yellow solid, 200 mg, 0.87 mmol) according to the same procedure as 7a, from 4-chloro-2-methoxy-1-nitrobenzene 6b (250 g, 1.33 mmol) and pyridine-4-ylboronic acid (164 mg, 1.33 mmol). HRMS (ESI +ve): found [M]+ 231.0803 [C12H10N2O3]+ requires 231.0770; δH (500 MHz, CDCl3): 8.74–8.72 (dd, J = 4.4, 1.6 Hz, 2H), 7.99 (dd, J = 8.8 Hz, 1H), 7.51 (dd, J = 4.4, 1.6 Hz, 2H), 7.28–7.25 (m, 2H), 4.05 (s, 3H).

4-(3-Ethoxy-4-nitrophenyl)-1-methylpyridin1-ium Iodide (8a)

7a (600 mg, 2.47 mmol) and methyl iodide (0.54 mL, 8.65 mmol) were dissolved in acetonitrile (0.1 M), and the reaction was heated for 4 h at 50 °C. Upon completion, the mixture was concentrated in vacuo to afford 8a as a yellow solid (945 mg, 2.45 mmol, 99%). HRMS (ESI +ve): found [M]+ 259.1076 [C14H15N2O3]+ requires 259.1082; δH (500 MHz, MeOD4) 8.99 (d, J = 6.7 Hz, 2H), 8.48 (d, J = 6.7 Hz, 2H), 7.99 (d, J = 8.5 Hz, 1H), 7.79 (d, J = 1.8 Hz, 1H), 7.65 (dd, J = 8.5, 1.8 Hz, 1H), 4.46 (s, 3H), 4.38 (q, J = 6.9 Hz, 2H), 1.49 (t, J = 6.9 Hz, 3H).

4-(3-Methoxy-4-nitrophenyl)-1-methylpyridin1-ium Iodide (8b)

8b was synthesized in 91% yield (yellow solid, 295 mg, 0.79 mmol) according to the same procedure as 8a, from 7b (200 mg, 0.87 mmol) and methyl iodide (0.19 mL, 3.04 mmol). HRMS (ESI +ve): found [M]+ 245.0920 [C13H13N2O3]+ requires 245.0926; δH (500 MHz, MeOD4) 8.99 (d, J = 6.7 Hz, 2H), 8.49 (d, J = 6.7 Hz, 2H), 8.01 (d, J = 8.2 Hz, 1H), 7.81 (d, J = 1.9 Hz, 1H), 7.66 (dd, J = 8.2, 1.9 Hz, 1H), 4.45 (s, 3H), 4.10 (s, 3H).

4-(3-Ethoxy-4-nitrophenyl)-1-methyl-1,2,3,6-tetrahydropyridine (9a)

8a (945 mg, 2.45 mmol) was dissolved in MeOH (0.05 M) and cooled to 0 °C. NaBH4 (943 mg, 24.5 mmol) was slowly added in batches, and the reaction was stirred at rt for 2 h. The reaction was quenched with 1 M HCl, and MeOH was partially removed in vacuo. The residue was partitioned between EtOAc and 1 M NaOH until pH 12 was reached. The EtOAc layer was washed with 1 M NaOH, dried over MgSO4, filtered, and concentrated in vacuo to give 9a as a yellow oil (609 mg, 2.32 mmol, 95%). HRMS (ESI +ve): found [M]+ 263.1387 [C14H18N2O3]+ requires 263.1390; δH (500 MHz, CDCl3): 7.85 (d, J = 8.5 Hz, 1H), 7.04–7.00 (m, 2H), 6.21–6.18 (m, 1H), 4.20 (q, J = 6.9 Hz, 2H), 3.17–3.14 (m, 2H), 2.69 (t, J = 5.7 Hz, 2H), 2.60–2.56 (m, 2H), 2.43 (s, 3H), 1.49 (t, J = 6.9 Hz, 3H).

4-(3-Methoxy-4-nitrophenyl)-1-methyl-1,2,3,6-tetrahydropyridine (9b)

9b was synthesized in 56% yield (yellow oil, 80 mg, 0.32 mmol) according to the same procedure as 9a, from 8b (212 mg, 0.57 mmol). HRMS (ESI +ve): found [M]+ 245.1239 [C13H13N2O3]+ requires 249.1243; δH (500 MHz, CDCl3): 7.85 (d, J = 8.5 Hz, 1H), 7.04 (d, J = 1.9 Hz, 1H), 7.02 (dd, J = 8.5, 1.9 Hz, 1H), 6.20 (tt, J = 3.5, 1.6 Hz, 1H), 3.96 (s, 3H), 3.16 (d, J = 2.85 Hz, 2H), 2.71–2.67 (m, 2H), 2.60–2.56 (m, 2H), 2.42 (s, 3H).

2-Ethoxy-4-(1-methylpiperidin-4-yl)aniline (10a)

9a (600 mg, 2.32 mmol) and PtO2 (158 mg, 0.70 mmol) were dissolved in acetic acid (0.03 M), and the mixture was purged with N2. The mixture was placed under 50 psi of H2 gas at rt for 16 h. The mixture was filtered through celite, washed with MeOH, and concentrated in vacuo. The residue was purified by Biotage column chromatography (DCM/MeOH, 9:1) to afford 10a as an orange oil (410 mg, 1.75 mmol, 75%). HRMS (ESI +ve): found [M]+ 235.1819 [C14H22N2O]+ requires 235.1809; δH (500 MHz, CDCl3) 6.68–6.60 (m, 3H), 4.05 (q, J = 7.0 Hz, 2H), 3.20 (d, J = 11.7 Hz, 2H), 2.46–2.38 (4H, m) 2.29 (td, J = 11.7, 2.5 Hz, 2H), 1.99–1.89 (m, 2H), 1.88–1.82 (m, 2H), 1.42 (t, J = 7.0 Hz, 3H).

2-Methoxy-4-(1-methylpiperidin-4-yl)aniline (10b)

10b was synthesized in 85% yield (yellow oil, 30 mg, 0.14 mmol) according to the same procedure as 10a, from 9b (600 mg, 2.32 mmol). HRMS (ESI +ve): found [M]+ 221.1649 [C13H20N2O]+ requires 221.1648; δH (500 MHz, CDCl3) 6.65–6.61 (m, 3H), 3.83 (s, 3H), 3.08 (dt, J = 12.1, 2.5 Hz, 2H), 2.46–2.37 (m, 1H), 2.38 (s, 3H), 2.16 (td, J = 12.1, 4.1 Hz, 2H), 1.91–1.81 (m, 4H).

(R)-Methyl 2-(Cyclopentylamino)butanoate (12a)

(R)-Methyl 2-aminobutanoate 11a (HCl salt, 1.50 g, 9.77 mmol) and cyclopentanone (0.87 mL, 9.77 mmol) were dissolved in DCE (0.1 M). The reaction was cooled to 0 °C, and NaOAc (801 mg, 9.77 mmol) and NaBH(OAc)3 (4.14 g, 19.5 mmol) were added. The reaction was stirred at rt for 16 h. Upon completion, sat. NaHCO3 was added and the aqueous phase was extracted with DCM (×3). The combined organic phases were washed with water, dried over MgSO4, filtered, and concentrated in vacuo to afford compound 12a as a yellow oil (1.45 g, 7.83 mmol, 80%). HRMS (ESI +ve): found [M]+ 186.1498, [C10H20NO2]+ requires 186.1494; [α]D21.8: −12.0 (c 1.0, MeOH); δH (500 MHz, CDCl3): 3.73 (s, 3H), 3.22 (t, J = 6.6 Hz, 1H), 2.97 (quin, J = 6.7 Hz, 1H), 1.83–1.60 (m, 6H), 1.55–1.47 (m, 2H), 1.35–1.27 (m, 2H), 0.92 (t, J = 7.6 Hz, 3H).

(S)-Methyl 2-(Cyclopentylamino)butanoate (12b)

12b was synthesized in 88% yield (brown oil, 1.42 g, 7.68 mmol) according to the same procedure as 12d, from (S)-methyl 2-aminobutanoate 11b (HCl salt, 1.34 g, 8.72 mmol) and cyclopentanone (0.78 mL, 8.72 mmol). HRMS (ESI +ve): found [M]+ 186.1498, [C10H20NO2]+ requires 186.1494; [α]D21.9: +11.8 (c 1.0, MeOH); δH (500 MHz, CDCl3): 3.71 (s, 3H), 3.20 (t, J = 6.6 Hz, 1H), 2.96 (quin, J = 6.7 Hz, 1H), 1.83–1.58 (m, 6H), 1.55–1.47 (m, 2H), 1.35–1.27 (m, 2H), 0.91 (t, J = 7.6 Hz, 3H).

Methyl Cyclopentylglycinate (12c)

12c was synthesized in 39% yield (yellow oil, 535 mg, 3.40 mmol) according to the same procedure as 12d, from glycine methyl ester 11c (HCl salt, 1.10 g, 8.77 mmol) and cyclopentanone (0.62 mL, 7.01 mmol). HRMS (ESI +ve): found [M]+ 158.1176, [C8H15NO2]+ requires 158.1176; δH (500 MHz, CDCl3): 3.73 (s, 3H), 3.42 (s, 2H), 3.07 (quin, J = 6.5 Hz, 1H), 1.85–1.77 (m, 2H), 1.75–1.66 (m, 2H), 1.60–1.50 (m, 2H), 1.40–1.32 (m, 2H).

(R)-Methyl 2-((3-Bromobenzyl)amino)butanoate (12d)

12d was synthesized in 92% yield (yellow oil, 1.37 g, 4.79 mmol) according to the same procedure as 12a, from (R)-methyl 2-aminobutanoate 11a (HCl salt, 800 mg, 5.21 mmol) and 3-bromobenzaldehyde (0.61 mL, 5.21 mmol). HRMS (ESI +ve): found [M]+ 286.0431, [C12H17BrNO2]+ requires 286.0437; [α]D21.9: +22.9 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.52 (t, J = 1.6 Hz, 1H), 7.38 (dt, J = 7.9, 1.6 Hz, 1H), 7.28–7.25 (m, 1H), 7.20–7.17 (m, 1H), 3.80 (d, J = 13.2 Hz, 1H), 3.74 (s, 3H), 3.60 (d, J = 13.2 Hz, 1H), 3.20 (t, J = 6.6 Hz, 1H), 1.76–1.62 (m, 2H), 0.96 (t, J = 7.6 Hz, 3H).

(R)-Methyl 2-(((3-Methylthiophen-2-yl)methyl)amino)butanoate (12e)

12e was synthesized in 74% yield (yellow oil, 720 mg, 3.17 mmol) according to the same procedure as 12d, from (R)-methyl 2-aminobutanote 11a (HCl salt, 500 mg, 4.27 mmol) and 3-methylthiophene-2-carbaldehyde (0.46 mL, 4.27 mmol). HRMS (ESI +ve): found [M]+ 228.1061, [C11H18NO2S]+ requires 228.1053; [α]D22.6: +23.6 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.10 (d, J = 5.0 Hz, 1H), 6.78 (d, J = 5.4 Hz, 1H), 3.93 (d, J = 13.9 Hz, 1H), 3.76–3.73 (m, 4H), 3.27 (t, J = 6.6 Hz, 1H), 2.17 (s, 3H), 1.74–1.62 (m, 2H), 0.95 (t, J = 6.9 Hz, 3H).

(R)-Methyl 2-(((4-Methylthiophen-2-yl)methyl)amino)butanoate (12f)

12f was synthesized in 95% yield (yellow oil, 1.70 g, 7.48 mmol) according to the same procedure as 12d, from (R)-methyl 2-aminobutanote 11a (HCl salt, 1.20 mg, 7.81 mmol) and 4-methylthiophene-2-carbaldehyde (0.83 mL, 6.72 mmol). HRMS (ESI +ve): found [M]+ 228.1062, [C11H18NO2S]+ requires 228.1053; [α]D22.5: +28.4 (c 1.0, MeOH); δH (500 MHz, CDCl3): 6.78 (t, J = 1.2 Hz, 1H), 6.73 (s, 1H), 3.98 (dd, J = 14.0, 0.8 Hz, 1H), 3.78 (dd, J = 14.0, 0.8 Hz, 1H), 3.74 (s, 3H), 3.29 (t, J = 6.5 Hz, 1H), 2.22 (d, J = 1.2 Hz, 3H), 1.74–1.62 (m, 2H), 0.95 (t, J = 7.6 Hz, 3H).

(R)-Methyl 2-(((5-Methylthiophen-2-yl)methyl)amino)butanoate (12g)

12g was synthesized in 77% yield (yellow oil, 892 mg, 3.92 mmol) according to the same procedure as 12d, from (R)-methyl 2-aminobutanote 11a (HCl salt, 600 mg, 5.12 mmol) and 5-methylthiophene-2-carbaldehyde (0.43 mL, 4.10 mmol). HRMS (ESI +ve): found [M]+ 228.1049, [C11H18NO2S]+ requires 228.1053; [α]D22.2: +23.6 (c 1.0, MeOH); δH (500 MHz, CDCl3): 6.68 (d, J = 3.2 Hz, 1H), 6.58–6.55 (m, 1H), 3.95 (d, J = 13.9 Hz, 1H), 3.75 (d, J = 13.9 Hz, 1H), 3.73 (s, 3H), 3.28 (t, J = 6.6 Hz, 1H), 2.45 (s, 3H), 1.89 (s, 1H), 1.74–1.61 (m, 2H), 0.94 (t, J = 6.9 Hz, 3H).

(R)-Methyl 2-((2-Chloro-5-nitropyrimidin-4-yl)(cyclopentyl)amino)butanoate (13a)

12a (700 mg, 3.78 mmol) and NaHCO3 (635 mg, 7.6 mmol) were dissolved in cyclohexane (0.1 M) and stirred for 30 min. 2,4-Dichloro-5-nitropyrimidine (806 mg, 4.16 mmol) was added, and the reaction was stirred at 60 °C for 16 h. Upon completion, the reaction mixture was filtered, washed with CH2Cl2, and concentrated in vacuo. The residue was purified by Biotage column chromatography (cHex/EtOAc, 0–20%) to afford the title compound 13a as a yellow solid (1.03 g, 2.99 mmol, 79%). HRMS (ESI +ve): found [M]+ 343.1165, [C14H20ClN4O4]+ requires 343.1173; [α]D21.8: +228.5 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.67 (s, 1H), 3.78–3.72 (m, 1H), 3.76 (s, 3H), 3.60–3.52 (m, 1H), 2.47–2.36 (m, 1H), 2.26–2.17 (m, 1H), 2.09–1.98 (m, 1H), 1.98–1.91 (m, 1H), 1.84–1.56 (m, 6H), 1.05 (t, J = 7.6 Hz, 3H).

(S)-Methyl 2-((2-Chloro-5-nitropyrimidin-4-yl)(cyclopentyl)amino)butanoate (13b)

13b was synthesized in 32% yield (yellow solid, 830 mg, 2.42 mmol) according to the same procedure as 13d, from 12b (1.40 g, 7.56 mmol) and 2,4-dichloro-5-nitropyrimidine (1.61 g, 8.31 mmol). HRMS (ESI +ve): found [M]+ 343.1171, [C14H20ClN4O4]+ requires 343.1173; [α]D22.0: −229.9 (c 1.0, MeOH); δH (CDCl3, 500 MHz): 8.67 (s, 1H), 3.78–3.72 (m, 1H), 3.76 (s, 3H), 3.60–3.51 (m, 1H), 2.46–2.36 (m, 1H), 2.25–2.17 (m, 1H), 2.09–2.01 (m, 1H), 1.99–1.90 (m, 1H), 1.84–1.47 (m, 6H), 1.05 (t, J = 7.6 Hz, 3H).

Methyl N-(2-Chloro-5-nitropyrimidin-4-yl)-N-cyclopentylglycinate (13c)

13c was synthesized in 78% yield (yellow solid, 416 mg, 1.32 mmol) according to the same procedure as 13d, from 12c (265 mg, 1.69 mmol) and 2,4-dichloro-5-nitropyrimidine (360 mg, 1.85 mmol). HRMS (ESI +ve): found [M]+ 315.0843, [C12H15ClN4O4]+ requires 315.0860; δH (CDCl3, 500 MHz): 8.65 (s, 1H), 4.17 (s, 2H), 3.98–3.88 (m, 1H), 3.79 (s, 3H), 2.16–2.07 (m, 2H), 1.78–1.70 (m, 2H), 1.64–1.53 (m, 4H).

(R)-Methyl 2-((3-Bromobenzyl)(2-chloro-5-nitropyrimidin-4-yl)amino)butanoate (13d)

13d was synthesized in 90% yield (yellow oil, 1.81 g, 4.08 mmol) according to the same procedure as 13a, from 12d (1.30 g, 4.54 mmol) and 2,4-dichloro-5-nitropyrimidine (970 mg, 5.00 mmol). HRMS (ESI +ve): found [M]+ 443.0118, [C16H17BrClN4O4]+ requires 443.0122; [α]D22.0: +12.0 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.64 (s, 1H), 7.45 (s, 1H), 7.41 (d, J = 7.8 Hz, 1H), 7.23 (d, J = 7.8 Hz, 1H), 7.17–7.13 (m, 1H), 4.75 (d, J = 15.8 Hz, 1H), 4.73–4.68 (m, 1H), 4.59 (d, J = 15.8 Hz, 1H), 3.82 (s, 3H), 2.32–2.23 (m, 1H), 2.10–2.00 (m, 1H), 1.07 (t, J = 7.4 Hz, 3H).

(R)-Methyl 2-((2-Chloro-5-nitropyrimidin-4-yl)((3-methylthiophen-2-yl)methyl)amino)butanoate (13e)

13e was synthesized in 90% yield (yellow solid, 1.18 g, 3.07 mmol) according to the same procedure as 13d, from 12e (770 mg, 3.39 mmol) and 2,4-dichloro-5-nitropyrimidine (723 mg, 3.72 mmol). HRMS (ESI +ve): found [M]+ 385.0732, [C15H17ClN4O4S]+ requires 385.0732; [α]D22.1: +4.85 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.65 (s, 1H), 7.15 (d, J = 5.2 Hz, 1H), 6.70 (d, J = 5.2 Hz, H2), 4.92 (dd, J = 5.4, 4.1 Hz, 1H), 4.75 (s, 2H), 3.80 (s, 3H), 2.30–2.20 (m, 1H), 2.16 (s, 3H), 2.07–2.00 (m, 1H), 1.09 (t, J = 7.4 Hz, 3H).

(R)-Methyl 2-((2-Chloro-5-nitropyrimidin-4-yl)((4-methylthiophen-2-yl)methyl)amino)butanoate (13f)

13f was synthesized in 81% yield (yellow solid, 2.32 g, 6.03 mmol) according to the same procedure as 13d, from 12f (1.70 g, 7.48 mmol) and 2,4-dichloro-5-nitropyrimidine (1.60 g, 8.23 mmol). HRMS (ESI +ve): found [M]+ 385.0738, [C15H17ClN4O4S]+ requires 385.0732; [α]D22.5: +5.54 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.70 (s, 1H), 6.82–6.81 (m, 1H), 6.71 (s, 1H), 4.95–4.80 (m, 3H), 3.82 (s, 3H), 2.32–2.22 (m, 1H), 2.18 (s, 3H), 2.09–2.00 (m, 1H), 1.07 (t, J = 7.4 Hz, 3H).

(R)-Methyl 2-((2-Chloro-5-nitropyrimidin-4-yl)((5-methylthiophen-2-yl)methyl)amino)butanoate (13g)

13g was synthesized in 86% yield (yellow gum, 1.30 g, 3.38 mmol) according to the same procedure as 13d, from 12g (892 mg, 3.92 mmol) and 2,4-dichloro-5-nitropyrimidine (837 mg, 4.31 mmol). HRMS (ESI +ve): found [M]+ 385.0732, [C15H17ClN4O4S]+ requires 385.0732; [α]D22.4: −2.78 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.68 (s, 1H), 6.67 (d, J = 2.5 Hz, 1H), 6.52 (d, J = 2.5 Hz, H2), 4.98–4.93 (m, 1H), 4.84 (d, J = 15.9 Hz, 1H), 4.81 (d, J = 15.9 Hz, 1H), 3.81 (s, 3H), 2.38 (s, 3H), 2.31–2.21 (m, 1H), 2.08–2.00 (m, 1H), 1.06 (t, J = 7.4 Hz, 3H).

(R)-2-Chloro-8-cyclopentyl-7-ethyl-7,8-dihydropteridin-6(5H)-one (14a)

A mixture of 13a (1.00 g, 2.92 mmol) in AcOH (0.4 M) was heated to 100 °C. Iron powder (196 mg, 3.5 mmol) was added batchwise, and the reaction was stirred for 6 h. Upon completion, the mixture was filtered through celite, washed with MeOH, and concentrated in vacuo. The residue was purified by Biotage column chromatography (cHex/EtOAc, 0–40%) to afford the title compound 14a as a yellow solid (345 mg, 1.23 mmol, 25%). HRMS (ESI +ve): found [M]+ 281.1163, [C13H18ClN4O]+ requires 281.1169; [α]D22.1: −96.2 (c 1.0, MeOH); δH (500 MHz, CDCl3): 9.39 (s, 1H), 7.68 (s, 1H), 4.37–4.29 (m, 1H), 4.21 (dd, J = 7.4, 3.6 Hz, 1H), 2.11–2.05 (m, 1H), 2.00–1.74 (m, 7H), 1.70–1.61 (m, 2H), 0.96 (t, J = 7.6 Hz, 3H).

(S)-2-Chloro-8-cyclopentyl-7-ethyl-7,8-dihydropteridin-6(5H)-one (14b)

14b was synthesized in 28% yield (yellow solid, 186 mg, 0.66 mmol) according to the same procedure as 14d, from 13b (815 mg, 2.38 mmol). HRMS (ESI +ve): found [M]+ 281.1163, [C13H18ClN4O]+ requires 281.1169; [α]D22.1: +95.3 (c 1.0, MeOH); δH (500 MHz, CDCl3): 9.67 (s, 1H), 7.70 (s, 1H), 4.32 (quin, J = 8.4 Hz, 1H), 4.21 (dd, J = 7.3, 3.5 Hz, 1H), 2.12–2.04 (m, 1H), 2.00–1.73 (m, 7H), 1.70–1.60 (m, 2H), 0.94 (t, J = 7.6 Hz, 3H).

2-Chloro-8-cyclopentyl-7,8-dihydropteridin-6(5H)-one (14c)

14c was synthesized in 31% yield (orange solid, 101 mg, 0.40 mmol) according to the same procedure as 14d, from 13c (410 mg, 1.30 mmol). HRMS (ESI +ve): found [M]+ 281.1163, [C13H18ClN4O]+ requires 281.1169; [α]D22.1: +95.3 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.98 (s, 1H), 7.60 (s, 1H), 6.15 (quin, J = 8.4 Hz, 1H), 4.11 (s, 2H), 1.98–1.89 (m, 2H), 1.80–1.57 (m, 6H).

(R)-2-Chloro-8-(3-bromobenzyl)-7-ethyl-7,8-dihydropteridin-6(5H)-one (14d)

14d was synthesized in 42% yield (orange solid, 387 mg, 1.01 mmol) according to the same procedure as 14a, from 13d (1.80 g, 4.06 mmol). HRMS (ESI +ve): found [M]+ 383.0078, [C15H15BrClN4O]+ requires 383.0091; [α]D22.0: −18.0 (c 1.0, MeOH); δH (500 MHz, CDCl3): 9.33 (s, 1H), 7.71 (s, 1H), 7.48–7.45 (m, 1H), 7.45 (s, 1H), 7.26–7.23 (m, 2H), 5.62 (d, J = 15.1, 1H), 4.16–4.12 (m, 1H), 4.08 (d, J = 15.1, 1H), 2.05–1.97 (m, 1H), 1.96–1.86 (m, 1H), 0.91 (t, J = 7.4 Hz, 3H).

(R)-2-Chloro-7-ethyl-8-((3-methylthiophen-2-yl)methyl)-7,8-dihydropteridin-6(5H)-one (14e)

14e was synthesized in 47% yield (orange solid, 380 mg, 1.17 mmol) according to the same procedure as 14d, from 13e (960 mg, 2.49 mmol). HRMS (ESI +ve): found [M]+ 323.0724, [C14H15ClN4OS]+ requires 323.0728; [α]D22.2: −11.1 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.70 (s, 1H), 7.65 (s, 1H), 6.86 (d, J = 3.3 Hz, 1H), 6.63–6.61 (m, 1H), 5.55 (d, J = 15.5, 1H), 4.33–4.29 (m, 2H), 2.46 (s, 3H), 2.07–1.88 (m, 2H), 0.90 (t, J = 7.4 Hz, 3H).

(R)-2-Chloro-7-ethyl-8-((4-methylthiophen-2-yl)methyl)-7,8-dihydropteridin-6(5H)-one (14f)

14f was synthesized in 42% yield (orange solid, 300 mg, 0.93 mmol) according to the same procedure as 14d, from 13f (860 mg, 2.23 mmol). HRMS (ESI +ve): found [M]+ 323.0728, [C14H15ClN4OS]+ requires 323.0728; [α]D22.8: −18.0 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.73 (s, 1H), 6.86 (s, 1H), 6.83 (d, J = 1.0 Hz, 1H), 5.58 (d, J = 15.5 Hz, 1H), 4.32–4.28 (m, 2H), 2.23 (s, 3H), 2.07–1.99 (m, 1H), 1.97–1.86 (m, 1H), 0.89 (t, J = 7.4 Hz, 3H).

(R)-2-Chloro-7-ethyl-8-((5-methylthiophen-2-yl)methyl)-7,8-dihydropteridin-6(5H)-one (14g)

14g was synthesized in 18% yield (orange solid, 200 mg, 0.62 mmol) according to the same procedure as 14d, from 13g (1.30 g, 3.38 mmol). HRMS (ESI +ve): found [M]+ 323.0727, [C14H15ClN4OS]+ requires 323.0728; [α]D22.5: +18.7 (c 1.0, MeOH); δH (500 MHz, CDCl3): 9.20 (s, 1H), 7.69 (s, 1H), 7.18 (d, J = 5.2 Hz, 1H), 6.83 (d, J = 5.2 Hz, 1H), 5.64 (d, J = 15.5 Hz, 1H), 4.29 (d, J = 15.5 Hz, 1H), 4.24 (dd, J = 6.3, 3.5 Hz, 1H), 2.26 (s, 3H), 2.05–1.99 (m, 1H), 1.98–1.87 (m, 1H), 0.92 (t, J = 7.6 Hz, 3H).

(R)-2-Chloro-8-cyclopentyl-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (15a)

14a (200 mg, 0.72 mmol) was dissolved in DMF (0.1 M), methyl iodide (58 μL, 0.93 mmol) was added, and the mixture was cooled to −10 °C. NaH (22 mg, 0.93 mmol) was added, and the reaction was stirred for 16 h. Upon completion, ice was added and the aqueous phase was extracted with EtOAc (×3). The combined organic phases were washed with water, dried over MgSO4, filtered, and concentrated in vacuo. The residue was purified by Biotage column chromatography (cHex/EtOAc, 0–20%) to afford the title compound 15a as a white solid (200 mg, 0.68 mmol, 71%). HRMS (ESI +ve): found [M]+ 295.1321, [C14H20ClN4O]+ requires 295.1326; [α]D21.4: −88.3 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.62 (s, 1H), 4.41–4.29 (m, 1H), 4.25 (dd, J = 7.6, 3.5 Hz, 1H), 3.33 (s, 3H), 2.12–2.04 (m, 1H), 2.02–1.58 (m, 9H), 0.86 (t, J = 7.4 Hz, 3H).

(S)-2-Chloro-8-cyclopentyl-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (15b)

15b was synthesized in 95% yield (white solid, 180 mg, 0.61 mmol) according to the same procedure as 15d, from 14b (180 mg, 0.64 mmol) and methyl iodide (52 μL, 0.83 mmol). HRMS (ESI +ve): found [M]+ 295.1321, [C14H20ClN4O]+ requires 295.1326; [α]D21.5: +93.5 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.65 (s, 1H), 4.37–4.28 (m, 1H), 4.23 (dd, J = 7.6, 3.8 Hz, 1H), 3.31 (s, 3H), 2.10–2.02 (m, 1H), 2.00–1.58 (m, 9H), 0.85 (t, J = 7.6 Hz, 3H).

2-Chloro-8-cyclopentyl-5-methyl-7,8-dihydropteridin-6(5H)-one (15c)

15c was synthesized in 59% yield (white solid, 53 mg, 0.20 mmol) according to the same procedure as 15d, from 14c (85 mg, 0.37 mmol) and methyl iodide (27 μL, 0.43 mmol). HRMS (ESI +ve): found [M]+ 267.0995, [C12H15ClN4O]+ requires 267.1007; δH (500 MHz, CDCl3): 7.67 (s, 1H), 5.14 (quin, J = 8.4 Hz, 1H), 4.13 (s, 2H), 4.37–4.28 (m, 1H), 3.33 (s, 3H), 1.97–1.91 (m, 2H), 1.81–1.65 (m, 4H), 1.64–1.56 (m, 2H).

(R)-8-(3-Bromobenzyl)-2-chloro-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (15d)

15d was synthesized in 71% yield (yellow solid, 200 mg, 0.51 mmol) according to the same procedure as 15a, from 14d (270 mg, 0.71 mmol) and methyl iodide (58 μL, 0.92 mmol). HRMS (ESI +ve): found [M]+ 395.0269, [C16H17BrClN4O]+ requires 395.0274; [α]D22.1: +15.2 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.73 (s, 1H), 7.48–7.44 (m, 2H), 7.25–7.22 (m, 2H), 5.62 (d, J = 15.1 Hz, 1H), 4.17 (dd, J = 6.3, 3.5 Hz, 1H), 4.09 (d, J = 15.1 Hz, 1H), 3.35 (s, 3H), 2.02–1.93 (m, 1H), 1.90–1.80 (m, 1H), 0.84 (t, J = 7.4 Hz, 3H).

(R)-2-Chloro-7-ethyl-5-methyl-8-(3-methylthiophen-2-yl)-7,8-dihydropteridin-6(5H)-one (15e)

15e was synthesized in 48% yield (yellow solid, 186 mg, 0.55 mmol) according to the same procedure as 15d, from 14e (370 mg, 1.15 mmol) and methyl iodide (72 μL, 1.15 mmol). HRMS (ESI +ve): found [M]+ 337.0901, [C15H17ClN4OS]+ requires 337.0884; [α]D23.3: +6.9 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.69 (s, 1H), 7.17 (d, J = 5.4 Hz, 1H), 6.83 (d, J = 5.4 Hz, 1H), 5.61 (d, J = 15.3 Hz, 1H), 4.29 (d, J = 15.3 Hz, 1H), 4.27 (dd, J = 6.6, 3.5 Hz, 1H), 3.32 (s, 3H), 2.25 (s, 3H), 2.03–1.93 (m, 1H), 1.91–1.81 (m, 1H), 0.84 (t, J = 7.6 Hz, 3H).

(R)-2-Chloro-7-ethyl-5-methyl-8-(4-methylthiophen-2-yl)-7,8-dihydropteridin-6(5H)-one (15f)

15f was synthesized in 53% yield (yellow oil, 260 mg, 0.77 mmol) according to the same procedure as 15d, from 14f (470 mg, 1.46 mmol) and methyl iodide (19 μL, 1.90 mmol). HRMS (ESI +ve): found [M]+ 337.0881, [C15H17ClN4OS]+ requires 337.0884; [α]D23.3: +4.9 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.69 (s, 1H), 6.86 (s, 1H), 6.82 (s, 1H), 5.56 (d, J = 15.5 Hz, 1H), 4.33–4.29 (2H, m), 3.31 (s, 3H), 2.22 (s, 3H), 2.03–1.95 (m, 1H), 1.91–1.84 (m, 1H), 0.81 (t, J = 7.4 Hz, 3H).

(R)-2-Chloro-7-ethyl-5-methyl-8-(5-methylthiophen-2-yl)-7,8-dihydropteridin-6(5H)-one (15g)

15g was synthesized in 53% yield (yellow solid, 110 mg, 0.33 mmol) according to the same procedure as 15d, from 14g (200 mg, 0.62 mmol) and methyl iodide (39 μL, 0.62 mmol). HRMS (ESI +ve): found [M]+ 337.0866, [C15H17ClN4OS]+ requires 337.0884; [α]D23.5: +1.7 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.69 (s, 1H), 6.85 (d, J = 3.5 Hz, 1H), 6.61–6.59 (m, 1H), 5.53 (d, J = 15.5 Hz, 1H), 4.33–4.29 (2H, m), 3.32 (s, 3H), 2.45 (s, 3H), 2.04–1.95 (m, 1H), 1.91–1.83 (m, 1H), 0.82 (t, J = 7.4 Hz, 3H).

(R)-8-Cyclopentyl-7-ethyl-2-((2-methoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-5-methyl-7,8-dihydropteridin-6(5H)-one (16a)

To 15a (25 mg, 80 μmol) and 10b (18 mg, 80 μmol) in a solvent mixture of dioxane/EtOH/H2O (1:1:1, 0.1 M) was added conc. HCl (24 μL, 110 μmol). The reaction was stirred at 120 °C for 48 h. Upon completion, the reaction mixture was partitioned between EtOAc and NaOH (1 M) and the aqueous layer was further extracted with EtOAc (×2). The organic phase was combined, dried over MgSO4, and concentrated in vacuo. The residue was purified by Biotage column chromatography (DCM/20% amm. MeOH, 0–40%) and reverse phase column chromatography (water/MeOH, 0–100%) to afford 16a as a yellow solid (9.0 mg, 19 μmol, 22%). HRMS (ESI +ve): found [M]+ 479.3101, [C27H38N6O2]+ requires 479.3129; [α]D21.8: −9.1 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.20 (d, J = 8.2 Hz, 1H), 7.88 (s, 1H), 7.61 (s, 1H), 6.80 (dd, J = 8.2, 1.9 Hz, 1H), 6.75 (d, J = 1.9 Hz, 1H), 4.42 (quin, J = 8.0 Hz, 1H), 4.21 (dd, J = 7.7. 3.6 Hz, 1H), 3.90 (s, 3H), 3.59 (d, J = 10.7 Hz, 2H), 3.31 (s, 3H), 2.75 (s, 3H), 2.77–2.67 (m, 3H), 2.27 (q, J = 12.5 Hz, 2H), 2.14–1.92 (m, 4H), 1.92–1.60 (m, 8H), 0.87 (t, J = 7.6 Hz, 3H).

(R)-4-((8-Cyclopentyl-7-ethyl-5-methyl-6-oxo-5,6,7,8-tetrahydropteridin-2-yl)amino)-3-isobutoxy-N-(1-methylpiperidin-4-yl)benzamide (16b)

16b was synthesized in 28% yield (white solid, 7.0 mg, 12 μmol) according to the same procedure as 16h, from 15a (13 mg, 44 μmol) and 5a (14 mg, 44 μmol). HRMS (ESI +ve): found [M]+ 564.3652, [C31H45N7O3]+ requires 564.3657; [α]D20.8: −31.2 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.55 (d, J = 8.4 Hz, 1H), 7.68 (s, 1H), 7.67 (s, 1H), 7.40 (d, J = 1.6 Hz, 1H), 7.22 (dd, J = 8.4, 1.6 Hz, 1H), 6.50 (d, J = 6.9 Hz, 1H), 4.40–4.32 (m, 1H), 4.23 (dd, J = 7.4, 3.3 Hz, 1H), 4.22–4.17 (m, 1H), 3.90 (d, J = 6.3 Hz, 2H), 3.41 (d, J = 10.4 Hz, 2H), 3.33 (s, 3H), 2.68 (t, J = 8.0 Hz, 2H), 2.33 (s, 3H), 2.26–1.65 (m, 15H), 1.09 (d, J = 6.6 Hz, 6H), 0.87 (t, J = 7.4 Hz, 3H).

(R)-4-((8-Cyclopentyl-7-ethyl-5-methyl-6-oxo-5,6,7,8-tetrahydropteridin-2-yl)amino)-3-ethoxy-N-(1-methylpiperidin-4-yl)benzamide (16c)

16c was synthesized in 24% yield (white solid, 13 mg, 24 μmol) according to the same procedure as 16h, from 15a (30 mg, 102 μmol) and 5b (28 mg, 102 μmol). HRMS (ESI +ve): found [M]+ 536.3345, [C29H41N7O3]+ requires 536.3349; [α]D21.8: −62.3 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.55 (d, J = 8.2 Hz, 1H), 7.73 (s, 1H), 7.68 (s, 1H), 7.41 (d, J = 1.9 Hz, 1H), 7.35 (dd, J = 8.3, 1.9 Hz, 1H), 6.56 (s, 1H), 4.52–4.44 (m, 1H), 4.28–4.18 (m, 4H), 3.50 (d, J = 10.4 Hz, 2H), 3.34 (s, 3H), 2.72–2.64 (m, 2H), 2.74 (s, 3H), 2.20–1.97 (m, 6H), 1.93–1.68 (m, 8H), 1.51 (t, J = 6.9 Hz, 3H), 0.88 (t, J = 7.4 Hz, 3H).

(S)-4-((8-Cyclopentyl-7-ethyl-5-methyl-6-oxo-5,6,7,8-tetrahydropteridin-2-yl)amino)-3-ethoxy-N-(1-methylpiperidin-4-yl)benzamide (16d)

16d was synthesized in 42% yield (white solid, 23 mg, 43 μmol) according to the same procedure as 16h, from 15b (30 mg, 102 μmol) and 5b (28 mg, 102 μmol). HRMS (ESI +ve): found [M]+ 536.3329, [C29H41N7O3]+ requires 536.3349; [α]D21.8: +51.9 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.55 (d, J = 8.2 Hz, 1H), 7.89 (s, 1H), 7.67 (s, 1H), 7.45 (d, J = 1.9 Hz, 1H), 7.35 (dd, J = 8.3, 1.9 Hz, 1H), 6.56 (s, 1H), 4.52–4.44 (m, 1H), 4.33–4.23 (m, 2H), 4.20 (q, J = 6.9 Hz, 2H), 3.50 (d, J = 10.4 Hz, 2H), 3.33 (s, 3H), 2.72–2.64 (m, 2H), 2.78 (s, 3H), 2.20–1.97 (m, 6H), 1.93–1.68 (m, 8H), 1.50 (t, J = 6.9 Hz, 3H), 0.88 (t, J = 7.6 Hz, 3H).

4-((8-Cyclopentyl-5-methyl-6-oxo-5,6,7,8-tetrahydropteridin-2-yl)amino)-3-ethoxy-N-(1-methylpiperidin-4-yl)benzamide (16e)

16e was synthesized in 6.6% yield (white solid, 2.0 mg, 3.9 μmol) according to the same procedure as 16h, from 15c (16 mg, 60 μmol) and 5b (16 mg, 57 μmol). HRMS (ESI +ve): found [M]+ 508.3028, [C29H41N7O3]+ requires 508.3031; δH (500 MHz, CDCl3): 8.53 (d, J = 8.4 Hz, 1H), 7.69 (s, 1H), 7.63 (s, 1H), 7.41 (d, J = 1.8 Hz, 1H), 7.23 (dd, J = 8.4, 1.8 Hz, 1H), 5.91 (s, 1H), 5.16 (quin, J = 8.1 Hz, H3), 4.22 (q, J = 7.6 Hz, 2H), 4.09 (s, 2H), 4.03–3.94 (m, 1H), 3.33 (s, 3H), 2.88–2.80 (m, 2H), 2.32 (s, 3H), 2.20–2.15 (m, 2H), 2.06 (m, 2H), 2.01–1.93 (m, 2H), 1.84–1.58 (m, 8H), 1.52 (t, J = 7.6 Hz, 3H).

(R)-4-((8-(3-Bromobenzyl)-7-ethyl-5-methyl-6-oxo-5,6,7,8-tetrahydropteridin-2-yl)amino)-3-ethoxy-N-(1-methylpiperidin-4-yl)benzamide (16f)

16f was synthesized in 53% yield (white solid, 55 mg, 86 μmol) according to the same procedure as 16h, from 15d (65 mg, 164 μmol) and 5b (46 mg, 164 μmol). HRMS (ESI +ve): found [M]+ 622.2302, [C31H38BrN7O3]+ requires 636.2298; [α]D22.1: +5.1 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.23 (d, J = 8.5 Hz, 1H), 7.74 (s, 1H), 7.63 (s, 1H), 7.50 (t, J = 1.6 Hz, 1H), 7.44 (dt, J = 7.6, 1.6 Hz, 1H), 7.42 (d, J = 1.9 Hz, 1H), 7.28–7.20 (m, 2H), 7.14 (dd, J = 8.5, 1.9 Hz, 1H), 5.96 (d, J = 7.9 Hz, 1H), 5.51 (d, J = 15.8 Hz, 1H), 4.23–4.16 (m, 4H), 4.04–3.94 (m, 1H), 3.38 (s, 3H), 2.76–2.67 (m, 2H), 2.31 (s, 3H), 2.17 (t, J = 10.9 Hz, 2H), 2.09–2.01 (m, 2H), 2.00–1.93 (m, 1H), 1.89–1.80 (m, 1H), 1.63–1.58 (m, 2H) 1.49 (t, J = 7.1 Hz, 3H), 0.87 (t, J = 7.4 Hz, 3H).

(R)-4-((8-Benzyl-7-ethyl-5-methyl-6-oxo-5,6,7,8-tetrahydropteridin-2-yl)amino)-3-ethoxy-N-(1-methylpiperidin-4-yl)benzamide (16g)

16g was synthesized in 46% yield (pale yellow solid, 4.0 mg, 7.2 μmol) according to the same procedure as 16i, from 16f (10 mg, 16 μmol). HRMS (ESI +ve): found [M]+ 558.3161, [C31H39N7O3]+ requires 558.3193; [α]D21.9: −9.7 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.45 (d, J = 8.5 Hz, 1H), 7.75 (s, 1H), 7.73 (s, 1H), 7.42 (d, J = 1.6 Hz, 1H), 7.40–7.29 (m, 6H), 6.61 (d, J = 8.2 Hz, 1H), 5.66 (d, J = 15.1 Hz, 1H), 4.28–4.19 (m, 4H), 4.18 (dd, J = 6.3, 3.5 Hz, 1H), 3.48 (d, J = 11.7 Hz, 2H), 3.37 (s, 3H), 2.80–2.72 (m, 2H), 2.73 (s, 3H), 2.20–2.12 (m, 4H), 2.01–1.92 (m, 1H), 1.91–1.82 (m, 1H), 1.51 (t, J = 6.9 Hz, 3H), 0.87 (t, J = 7.9 Hz, 3H).

(R)-8-(3-Bromobenzyl)-2-((2-ethoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (16h)

16h was synthesized in 29% yield (white solid, 30 mg, 51 μmol) according to the same procedure as 16a, from 15d (69 mg, 174 μmol) and 10a (39 mg, 166 μmol). HRMS (ESI +ve): found [M]+ 593.2209, [C30H38BrN6O2]+ requires 593.2234; [α]D23.7: +1.7 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.13 (s, 1H), 7.96 (d, J = 8.2 Hz, 1H), 7.68 (s, 1H), 7.46–7.42 (m, 2H), 7.23–7.20 (m, 2H), 6.74–6.70 (m, 2H), 5.51 (d, J = 15.5 Hz, 1H), 4.19–4.13 (m, 2H), 4.11 (q, J = 6.9 Hz, 2H), 3.67–3.63 (m, 2H), 3.35 (s, 3H), 2.85–2.75 (m, 2H), 2.82 (s, 3H), 2.70–2.61 (m, 1H), 2.28 (q, J = 12.3 Hz, 2H), 2.05–1.94 (m, 3H), 1.90–1.80 (m, 1H), 1.47 (t, J = 6.9 Hz, 3H), 0.85 (t, J = 7.6 Hz, 3H).

(R)-8-Benzyl-2-((2-ethoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (16i)

A solution of 16h (14 mg, 24 μmol) and 10% Pd·C (1.5 mg, 14 μmol) in MeOH (0.01 M) was stirred at rt under a hydrogen atmosphere (1 atm) for 1 h. The mixture was filtered through celite and washed with MeOH. The solvent was removed in vacuo, and the residue was purified by reverse phase column chromatography (water/MeOH, 0–100%) to give the title compound 16i as a white solid (5.0 mg, 9.7 μmol, 41%). HRMS (ESI +ve): found [M]+ 515.3114, [C30H40N6O2]+ requires 515.3129; [α]D23.7: +24.2 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.22 (d, J = 8.8 Hz, 1H), 7.68 (s, 1H), 7.58 (s, 1H), 7.37–7.29 (m, 5H), 6.72 (d, J = 1.6 Hz, 1H), 6.70 (dd, J = 8.2, 1.6 Hz, 1H), 5.62 (d, J = 15.1 Hz, 1H), 4.18–4.15 (m, 3H), 4.11 (q, J = 7.2 Hz, 2H), 3.64 (d, J = 10.4 Hz, 2H), 3.34 (s, 3H), 2.73 (s, 3H), 2.73–2.58 (m, 3H), 2.25–2.15 (m, 2H), 2.02–1.92 (m, 3H), 1.90–1.81 (m, 1H), 1.48 (t, J = 6.9 Hz, 3H), 0.85 (t, J = 7.4 Hz, 3H).

(R)-2-((2-Ethoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-7-ethyl-5-methyl-8-((3-methylthiophen-2-yl)methyl)-7,8-dihydropteridin-6(5H)-one (16j)

16j was synthesized in 9.4% yield (yellow solid, 6.0 mg, 11 μmol) according to the same procedure as 16h, from 15e (40 mg, 120 μmol) and 10a (26 mg, 110 μmol). HRMS (ESI +ve): found [M]+ 535.2828, [C29H39N6O2S]+ requires 535.2850; [α]D23.5: +7.6 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.31 (d, J = 8.2 Hz, 1H), 7.67 (s, 1H), 7.64 (s, 1H), 7.15 (d, J = 5.0 Hz, 1H), 6.83 (d, J = 5.0 Hz, 1H), 6.78 (dd, J = 8.2, 1.6 Hz, 1H), 6.73 (d, J = 1.6 Hz, 1H), 5.63 (d, J = 15.1 Hz, 1H), 4.33 (d, J = 15.1 Hz, 1H), 4.19 (dd, J = 6.5, 3.6 Hz, 1H), 4.13 (q, J = 6.9 Hz, 2H), 3.57 (d, J = 9.1 Hz, 2H), 3.32 (s, 3H), 2.76–2.60 (m, 6H), 2.25 (s, 3H), 2.24–2.15 (m, 2H), 2.02–1.91 (m, 3H), 1.88–1.79 (m, 1H), 1.49 (t, J = 6.9 Hz, 3H), 0.85 (t, J = 7.4 Hz, 3H).

(R)-2-((2-Ethoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-7-ethyl-5-methyl-8-((4-methylthiophen-2-yl)methyl)-7,8-dihydropteridin-6(5H)-one (16k)

16k was synthesized in 23% yield (yellow solid, 20 mg, 37 μmol) according to the same procedure as 16h, from 15f (55 mg, 163 μmol) and 10a (36 mg, 155 μmol). HRMS (ESI +ve): found [M]+ 535.2835, [C29H39N6O2S]+ requires 535.2850; [α]D22.6: +7.6 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.34 (d, J = 8.2 Hz, 1H), 7.67 (s, 1H), 7.45 (s, 1H), 7.14 (d, J = 1.3 Hz, 1H), 6.85 (s, 1H), 6.82–6.76 (m, 3H), 5.61 (d, J = 15.5 Hz, 1H), 4.34 (d, J = 15.5 Hz, 1H), 4.25 (dd, J = 6.3, 3.5 Hz 1H), 4.11 (q, J = 6.9 Hz, 2H), 3.35–3.29 (m, 5H), 2.60 (s, 3H), 2.58–2.46 (m, 3H), 2.22 (s, 3H), 2.20–2.12 (m, 2H), 2.00–1.92 (m, 3H), 1.89–1.80 (m, 1H), 1.48 (t, J = 6.9 Hz, 3H), 0.85 (t, J = 7.6 Hz, 3H).

(R)-2-((2-Ethoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-7-ethyl-5-methyl-8-((5-methylthiophen-2-yl)methyl)-7,8-dihydropteridin-6(5H)-one (16l)

16l was synthesized in 25% yield (yellow oil, 18 mg, 34 μmol) according to the same procedure as 16h, from 15g (45 mg, 134 μmol) and 10a (30 mg, 127 μmol). HRMS (ESI +ve): found [M]+ 535.2833, [C29H39N6O2S]+ requires 535.2850; [α]D23.6: +11.1 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.34 (d, J = 8.2 Hz, 1H), 7.67 (s, 1H), 7.63 (s, 1H), 6.83 (d, J = 3.2 Hz, 1H), 6.79 (dd, J = 8.2, 1.9 Hz, 1H), 6.75 (d, J = 1.8 Hz, 1H), 6.60 (d, J = 3.2 Hz, 1H), 5.57 (d, J = 15.5 Hz, 1H), 4.43 (d, J = 15.5 Hz, 1H), 4.26 (dd, J = 6.2, 3.6 Hz, 1H), 4.14 (q, J = 6.9 Hz, 2H), 3.57 (d, J = 10.1 Hz, 2H), 3.32 (s, 3H), 2.73 (s, 3H), 2.73–2.61 (m, 3H), 2.44 (s, 3H), 2.25–2.15 (m, 2H), 2.02–1.92 (m, 3H), 1.89–1.80 (m, 1H), 1.60 (t, J = 6.9 Hz, 3H), 0.84 (t, J = 7.4 Hz, 3H).

(R)-Methyl 2-((2,4-Dimethoxybenzyl)amino)butanoate (17)

17 was synthesized in 97% yield (yellow oil, 27.0 g, 101 mmol) according to the same procedure as 12d, from (R)-methyl 2-aminobutanoate 11a (HCl salt, 16.0 g, 104 mmol) and 2,4-dimethoxybenzaldehye (17.3 g, 101 mmol). HRMS (ESI +ve): found [M]+ 268.1541, [C14H21NO4]+ requires 268.1543; [α]D23.9: +11.8 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.13 (d, J = 7.9 Hz, 1H), 6.45–6.41 (m, 2H), 3.81 (s, 3H), 3.80 (s, 3H), 3.71 (d, J = 13.2 Hz, 1H), 3.67–3.74 (m, 4H), 3.21 (t, J = 6.5 Hz, 1H), 1.73–1.63 (m, 2H), 0.91 (t, J = 7.4 Hz, 3H).

(R)-Methyl 2-((2-Chloro-5-nitropyrimidin-4-yl)(2,4-dimethoxybenzyl)amino)butanoate (18)

18 was synthesized in 78% yield (yellow gum, 33.4 g, 78.6 mmol) according to the same procedure as 13d, from 17 (27.0 g, 101 mmol) and 2,4-dichloro-5-nitropyrimidine 15 (21.6 g, 111 mmol). HRMS (ESI +ve): found [M]+ 425.1211, [C18H21ClN4O6]+ requires 425.1222; [α]D23.7: −6.9 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.65 (s, 1H), 7.04 (d, J = 8.2 Hz, 1H), 6.40–6.37 (m, 2H), 4.47–4.43 (m, 2H), 4.40 (d, J = 16.1 Hz, 1H), 3.79 (s, 3H), 3.71 (s, 3H), 3.62 (s, 3H), 2.28–2.18 (m, 1H), 2.07–1.99 (m, 1H), 1.05 (t, J = 7.4 Hz, 3H).

(R)-2-Chloro-8-(2,4-dimethoxybenzyl)-7-ethyl-7,8-dihydropteridin-6(5H)-one (19)

A solution of 18 (13.0 g, 30.6 mmol), 10% Pt·C (597 mg, 3.06 mmol), and VO(acac)2 (650 mg, 2.45 mmol) in THF (0.3 M) was stirred at rt under a hydrogen atmosphere (1 atm) for 48 h. The mixture was filtered through celite, and the solvent was removed in vacuo to give 19 as a brown solid (11.0 g, 30.3 mmol, 98%). HRMS (ESI +ve): found [M]+ 363.1198, [C17H19ClN4O3]+ requires 363.1218; [α]D23.4: +15.2 (c 1.0, MeOH); δH (500 MHz, CDCl3): 9.72 (1H), 7.65 (s, 1H), 7.30 (d, J = 9.1 Hz, 1H), 6.47–6.45 (m, 2H), 5.35 (d, J = 14.7 Hz, 2H), 4.28 (dd, J = 5.4, 4.4 Hz, 1H), 4.20 (d, J = 14.7 Hz, 1H), 3.80 (s, 6H), 2.03–1.94 (m, 2H), 0.94 (t, J = 7.3 Hz, 3H).

(R)-2-Chloro-8-(2,4-dimethoxybenzyl)-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (20)

20 was synthesized in 86% yield (yellow oil, 9.66 g, 25.6 mmol) according to the same procedure as 15d, from 19 (10.8 g, 29.8 mmol) and methyl iodide (2.42 mL, 38.7 mmol). HRMS (ESI +ve): found [M]+ 377.1366, [C18H21ClN4O3]+ requires 377.1375; [α]D23.7: +35.3 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.62 (s, 1H), 7.32 (d, J = 8.8 Hz, 1H), 6.47–6.45 (m, 2H), 5.34 (d, J = 14.5 Hz, 1H), 4.34 (dd, J = 6.3, 3.8 Hz, 1H), 4.20 (d, J = 14.7 Hz, 1H), 3.81 (s, 6H), 3.30 (s, 3H), 1.99–1.85 (m, 2H), 0.85 (t, J = 7.9 Hz, 3H).

(R)-2-Chloro-7-5-methyl-7,8-dihydropteridin-6(5H)-one (21)

A solution of 20 (5.09 g, 13.5 mmol) in TFA (1.3 M) was stirred at 80 °C for 4 h. Upon completion, the reaction was concentrated in vacuo to remove the majority of the TFA. The remaining TFA was quenched with sat. NaHCO3 until pH 7–8 was reached and extracted with DCM (×3). The organic layers were washed with brine, dried over MgSO4, filtered, and concentrated in vacuo. The residue was purified by Biotage column chromatography (DCM/MeOH, 0–15%) to give 21 as a brown solid (2.02 g, 8.91 mmol, 66%). HRMS (ESI +ve): found [M]+ 227.0695, [C9H11ClN4O]+ requires 227.0694; [α]D23.7: +35.3 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.73 (s, 1H), 6.39 (s, 1H), 4.34 (dd, J = 6.3, 4.7 Hz, 1H), 3.34 (s, 3H), 2.03–1.87 (m, 2H), 0.99 (t, J = 7.4 Hz, 3H).

(R)-2-Chloro-8-(3-chlorobenzyl)-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (22a)

To a solution of 21 (50 mg, 0.22 mmol) in DMF (0.2 M) was added 1-(bromomethyl)-3-chlorobenzene (36 μL, 0.28 mmol). The mixture was cooled to −10 °C, NaH (12 mg, 0.50 mmol) was added, and then the mixture was stirred at rt for 18 h. Upon completion, ice was added and the reaction was partitioned between EtOAc and water. The aqueous layer was further extracted with EtOAc (×3). The combined organic phases were washed with water (×2), dried over MgSO4, filtered, and concentrated in vacuo. The residue was purified by Biotage column chromatography (cHex/EtOAc, 0–20%) to give 22a as an orange solid (51 mg, 0.15 mmol, 66%). HRMS (ESI +ve): found [M]+ 351.0775, [C16H16Cl2N4O]+ requires 351.0774; [α]D23.6: −1.4 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.73 (s, 1H), 7.32–7.28 (m, 3H), 7.20–7.18 (m, 1H), 5.62 (d, J = 15.1 Hz, 1H), 4.17 (dd, J = 6.2, 3.2 Hz, 1H), 4.08 (d, J = 15.1 Hz, 1H), 3.35 (s, 3H), 2.00–1.81 (m, 3H), 0.84 (t, J = 7.6 Hz, 3H).

(R)-3-((2-Chloro-7-ethyl-5-methyl-6-oxo-6,7-dihydropteridin-8(5H)-yl)methyl)benzonitrile (22b)

22b was synthesized in 37% yield (white solid, 28 mg, 0.08 mmol) according to the same procedure as 22a, from 21 (50 mg, 0.22 mmol) and 3-(bromomethyl)benzonitrile (35 μL, 0.28 mmol). HRMS (ESI +ve): found [M]+ 341.1112, [C17H16ClN5O]+ requires 342.1116; [α]D23.7: −1.4 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.74 (s, 1H), 7.63–7.56 (m, 2H), 7.56 (d, J = 7.7 Hz, 1H), 7.47 (t, J = 7.7 Hz, 1H), 5.58 (d, J = 15.5 Hz, 1H), 4.18 (d, J = 15.5 Hz, 1H), 4.15 (dd, J = 6.5, 3.6 Hz, 1H), 3.35 (s, 3H), 2.01–1.94 (m, 1H), 1.87–1.78 (m, 1H), 0.83 (t, J = 7.6 Hz, 3H).

(R)-2-Chloro-7-ethyl-8-(3-methoxybenzyl)-5-methyl-7,8-dihydropteridin-6(5H)-one (22c)

22c was synthesized in 65% yield (colorless oil, 50 mg, 0.14 mmol) according to the same procedure as 22a, from 21 (50 mg, 0.22 mmol) and 3-methoxybenzyl bromide (39 μL, 0.28 mmol). HRMS (ESI +ve): found [M]+ 347.1269, [C17H19ClN4O2]+ requires 347.1269; [α]D23.6: −9.7 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.69 (s, 1H), 7.28–7.24 (m, 1H), 6.87–6.84 (m, 3H), 5.63 (d, J = 15.1 Hz, 1H), 4.19 (dd, J = 6.3, 3.5 Hz, 1H), 4.10 (d, J = 15.1 Hz, 1H), 3.79 (s, 3H), 3.35 (s, 3H), 2.00–1.93 (m, 1H), 1.93–1.81 (m, 1H), 0.82 (t, J = 7.6 Hz, 3H).

(R)-2-Chloro-8-(2-bromobenzyl)-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (22d)

22d was synthesized in 62% yield (orange solid, 87 mg, 0.22 mmol) according to the same procedure as 22a, from 21 (80 mg, 0.35 mmol) and 1-(bromomethyl)-2-chloro-benzene (111 mg, 0.45 mmol). HRMS (ESI +ve): found [M]+ 397.0250, [C16H16BrClN4O]+ requires 397.0274; [α]D21.5: −20.1 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.72 (s, 1H), 7.58 (dd, J = 7.8, 1.2 Hz, 1H), 7.36 (dd, J = 7.8, 1.7 Hz, 1H), 7.30 (m, 1H), 7.19 (m, 1H), 5.63 (d, J = 15.1 Hz, 1H), 4.33 (d, J = 15.1 Hz, 1H), 4.19 (dd, J = 6.3, 3.8 Hz, 1H), 3.35 (s, 3H), 1.98–1.86 (m, 2H), 0.87 (t, J = 7.6 Hz, 3H).

(R)-2-Chloro-8-(4-bromobenzyl)-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (22e)

22e was synthesized in 21% yield (white solid, 35 mg, 0.09 mmol) according to the same procedure as 22a, from 21 (95 mg, 0.42 mmol) and 1-(bromomethyl)-4-chloro-benzene (83 μL, 0.50 mmol). HRMS (ESI +ve): found [M]+ 397.0249, [C16H16Cl2N4O]+ requires 397.0247; [α]D22.3: −47.1 (c 1.0, MeOH); δH (500 MHz, CDCl3): 7.71 (s, 1H), 7.47 (d, J = 8.5 Hz, 2H), 7.18 (d, J = 8.5 Hz, 2H), 5.52 (d, J = 15.1 Hz, 1H), 4.14 (dd, J = 6.3, 3.5 Hz, 1H), 4.10 (d, J = 15.1 Hz, 1H), 3.33 (s, 3H), 1.98–1.90 (m, 1H), 1.86–1.77 (m, 1H), 0.81 (t, J = 7.6 Hz, 3H).

(R)-2-Chloro-7-ethyl-5-methyl-8-((4-methylthiazol-5-yl)methyl)-7,8-dihydropteridin-6(5H)-one (22f)

22f was synthesized in 39% yield (white solid, 41 mg, 0.12 mmol) according to the same procedure as 22a, from 21 (70 mg, 0.31 mmol) and 5-(chloromethyl)-4-methyl-1,3-thiazole (HCl salt, 68 mg, 0.37 mmol). HRMS (ESI +ve): found [M]+ 338.0838, [C14H16ClN5OS]+ requires 338.0837; [α]D22.7: −14.5 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.68 (s, 1H), 7.71 (s, 1H), 5.46 (d, J = 15.5 Hz, 1H), 4.46 (d, J = 15.5 Hz, 1H), 4.27 (dd, J = 6.3, 3.5 Hz, 1H), 3.33 (s, 3H), 2.53 (s, 3H), 2.04–1.98 (m, 1H), 1.88–1.70 (m, 1H), 0.83 (t, J = 7.4 Hz, 3H).

(R)-8-(3-Chlorobenzyl)-2-((2-ethoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (23a)

23a was synthesized in 13% yield (white solid, 8.0 mg, 15 μmol) according to the same procedure as 16h, from 22a (40 mg, 110 μmol) and 10a (25 mg, 110 μmol). HRMS (ESI +ve): found [M]+ 549.2727, [C30H37ClN6O2]+ requires 549.2739; [α]D23.1: +15.2 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.16 (d, J = 8.2 Hz, 1H), 7.71 (s, 1H), 7.46 (s, 1H), 7.30–7.27 (m, 2H), 7.23–7.20 (m, 1H), 6.76–6.70 (m, 2H), 5.58 (d, J = 15.5 Hz, 1H), 4.18–4.09 (m, 4H), 3.50 (d, J = 11.0 Hz, 2H), 3.36 (s, 3H), 2.69 (s, 3H), 2.69–2.59 (m, 3H), 2.17 (q, J = 12.3 Hz, 2H), 2.00–1.91 (m, 3H), 1.88–1.80 (m, 1H), 1.47 (t, J = 7.6 Hz, 3H), 0.87 (t, J = 7.0 Hz, 3H).

(R)-3-((2-((2-Ethoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-7-ethyl-5-methyl-6-oxo-6,7-dihydropteridin-8(5H)-yl)methyl)benzonitrile (23b)

23b was synthesized in 10% yield (white solid, 6.6 mg, 12 μmol) according to the same procedure as 16h, from 22b (40 mg, 120 μmol) and 10a (26 mg, 110 μmol). HRMS (ESI +ve): found [M]+ 540.3071, [C31H37N7O2]+ requires 540.3081; [α]D24.0: −2.1 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.08 (d, J = 8.2 Hz, 1H), 7.73 (s, 1H), 7.62–7.55 (m, 3H), 7.47 (t, J = 7.9 Hz, 1H), 7.41 (s, 1H), 6.73–6.69 (m, 2H), 5.55 (d, J = 15.5 Hz, 1H), 4.23 (d, J = 15.5 Hz, 1H), 4.14–4.07 (m, 3H), 3.44 (d, J = 11.7 Hz, 2H), 3.37 (s, 3H), 2.64 (s, 3H), 2.63–2.52 (m, 3H), 2.12 (q, J = 13.2 Hz, 2H), 2.00–1.91 (m, 3H), 1.86–1.77 (m, 1H), 1.46 (t, J = 6.9 Hz, 3H), 0.87 (t, J = 7.4 Hz, 3H).

(R)-8-(3-Methoxybenzyl)-2-((2-ethoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (23c)

23c was synthesized in 32% yield (yellow solid, 20 mg, 37 μmol) according to the same procedure as 16h, from 22c (40 mg, 120 μmol) and 10a (26 mg, 110 μmol). HRMS (ESI +ve): found [M]+ 545.3221, [C31H40N6O3]+ requires 545.3235; [α]D24.0: +13.2 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.19 (d, J = 8.8 Hz, 1H), 7.71 (s, 1H), 7.67 (1H), 7.29–7.26 (m, 1H), 6.91 (d, J = 7.6 Hz, 1H), 6.88–6.84 (m, 2H), 6.75–6.72 (m, 2H), 5.61 (d, J = 15.1 Hz, 1H), 4.18–4.09 (m, 4H), 3.77 (s, 3H), 3.61 (d, J = 8.8 Hz, 2H), 3.34 (s, 3H), 2.80–2.69 (m, 2H), 2.75 (s, 3H), 2.67–2.60 (m, 1H), 2.19 (q, J = 12.4 Hz, 2h), 2.02–1.90 (m, 3H), 1.89–1.80 (m, 1H), 1.48 (t, J = 6.9 Hz, 3H), 0.85 (t, J = 7.6 Hz, 3H).

(R)-8-(2-Bromobenzyl)-2-((2-ethoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (23d)

23d was synthesized in 13% yield (yellow solid, 7.1 mg, 12 μmol) according to the same procedure as 16h, from 22d (36 mg, 91 μmol) and 10a (20 mg, 86 μmol). HRMS (ESI +ve): found [M]+ 593.2210, [C30H37BrN6O2]+ requires 593.2231; [α]D22.0: −19.39 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.06 (d, J = 8.8 Hz, 1H), 7.70 (s, 1H), 7.57 (d, J = 7.6 Hz, 1H), 7.52 (s, 1H), 7.29–7.26 (m, 2H), 7.19–7.16 (m, 1H), 6.72–6.70 (m, 2H), 5.41 (d, J = 16.1 Hz, 1H), 4.30 (d, J = 16.1 Hz, 1H), 4.15 (dd, J = 6.6, 3.8 Hz, 1H), 4.10 (q, J = 6.9 Hz, 2H), 3.54 (d, J = 10.7 Hz, 2H), 3.37 (s, 3H), 2.74–2.60 (m, 6H), 2.17 (m, 2H), 2.02–1.86 (m, 4H), 1.47 (t, J = 6.9 Hz, 3H).

(R)-8-(4-Bromobenzyl)-2-((2-ethoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-one (23e)

23e was synthesized in 6% yield (yellow solid, 3.4 mg, 5.7 μmol) according to the same procedure as 16h, from 22e (35 mg, 89 μmol) and 10b (20 mg, 84 μmol). HRMS (ESI +ve): found [M]+ 593.2242, [C30H37BrN6O2]+ requires 593.2234; [α]D23.4: +6.2 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.16 (d, J = 7.9 Hz, 1H), 7.69 (s, 1H), 7.59 (s, 1H), 7.47 (d, J = 8.2 Hz, 2H), 7.20 (d, J = 8.2 Hz, 2H), 6.74–6.69 (m, 2H), 5.51 (d, J = 15.5 Hz, 1H), 4.19–4.09 (m, 4H), 3.58 (d, J = 9.1 Hz, 2H), 3.34 (s, 3H), 2.74 (s, 3H), 2.75–2.59 (m, 3H), 2.19 (m, 2H), 2.02–1.90 (m, 3H), 1.85–1.76 (m, 1H), 1.47 (t, J = 6.9 Hz, 3H).

(R)-2-((2-Ethoxy-4-(1-methylpiperidin-4-yl)phenyl)amino)-7-ethyl-5-methyl-8-((4-methylthiazol-5-yl)methyl)-7,8-dihydropteridin-6(5H)-one (23f)

23f was synthesized in 31% yield (yellow solid, 17 mg, 32 μmol) according to the same procedure as 16h, from 22f (35 mg, 100 μmol) and 10a (23 mg, 100 μmol). HRMS (ESI +ve): found [M]+ 536.2787, [C28H37N7O2S]+ requires 536.2802; [α]D22.8: −8.3 (c 1.0, MeOH); δH (500 MHz, CDCl3): 8.65 (s, 1H), 8.24 (d, J = 8.3 Hz, 1H), 7.69 (s, 1H), 7.64 (s, 1H), 6.77 (dd, J = 8.3, 1.8 Hz, 1H), 6.74 (d, J = 1.8 Hz, 1H), 5.47 (d, J = 15.8 Hz, 1H), 4.48 (d, J = 15.8 Hz, 1H), 4.17 (d, J = 6.5, 3.6 Hz, 1H), 4.13 (q, J = 7.0 Hz, 2H), 3.61 (d, J = 10.4 Hz, 2H), 3.32 (s, 3H), 2.75 (s, 3H), 2.75–2.70 (m, 2H), 2.66–2.63 (m, 1H), 2.52 (s, 3H), 2.26–2.15 (m, 2H), 2.03–1.94 (m, 3H), 1.86–1.76 (m, 1H), 1.49 (t, J = 7.0 Hz, 3H), 0.84 (t, J = 7.6 Hz, 3H).

ALKF1174L Kinase Assay

ALKF1174L activity was measured in a LanthaScreen Eu kinase binding assay. The assay was performed in 384-well plates containing ALKF1174L enzyme (5 nM, Carna Biosciences), Kinase Tracer 236 (30 nM, Thermo Fisher Scientific), LanthaScreen Eu-anti-GST antibody (2 nM, Thermo Fisher Scientific), either 1% (v/v) dimethyl sulfoxide (DMSO) or the test compound (in the range from 0.5 nM to 100 μM in 1% (v/v) DMSO), and assay buffer (50 mM N-(2-hydroxyethyl)piperazine-N′-ethanesulfonic acid (HEPES) pH 7.5; 10 mM MgCl2; 1 mM ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid and 0.01% (w/v) Brij-35; 1 mM dithiothreitol). The reaction was incubated for 60 min at room temperature. The plate was read on an EnVision multilabel plate reader (PerkinElmer, U.K.) calculating an emission ratio between the acceptor/tracer emission (665 nM) and the antibody/donor emission (615 nM). IC50 values were determined using a nonlinear regression fit of the log(inhibitor concentration) versus emission ratio with a variable slope equation.

Bromodomain Thermal Shift

BRD4 thermal shift was measured in a differential scanning fluorimetry (DSF) assay. The assay was performed on a Stratagene Mx3005P system quantitative polymerase chain reaction (qPCR). The assay was performed in 96-well plates with 20 μL of protein at 2 μM in DSF-assay buffer (10 mM HEPES pH 7.5, 500 mM NaCl) to which 4 μL of Sypro orange dye and 0.4 μL of 10 μM compound in DMSO was added. The plate was sealed and centrifuged (1 min, 1000 rpm), and the qPCR machine was used to detect SYPRO fluorescence while increasing the temperature from 25 to 95 °C in 71 cycles. The sigmoidal part of the fluorescence signal was fitted using the Boltzmann equation, the points of change were defined, and the thermal shift was calculated by subtracting the protein/DMSO Tm shift from the protein/compound Tm shift.

BRD4 Isothermal Calorimetry

BRD4 activity was measured by isothermal calorimetry. The sample cell, sample syringe, and injection syringe were all equilibrated with gel filtration buffer (10 mM HEPES, 150 mM NaCl, 0.5 mM tris(2-carboxyethyl)phosphine (TCEP), 5% glycerol). The sample syringe was filled with compound solution (30–40 μM in gel filtration buffer), and the injection syringe was filled with protein solution (300 μM in gel filtration buffer). After equilibration of the machine, protein was injected over 21–30 injections, 8 μL each time. For the analysis of the data, the baseline and integration points were defined to determine binding heats. Data was fitted to the Boltzmann equation to determine thermodynamic parameters.

PLK-1 Kinase Assay

All PLK-1 IC50s and percentage inhibitions at 10 and 100 nM were tested by Thermo Fisher Scientific in a Ź-LYTE activity assay using their SelectScreen biochemical kinase profiling service.

Kinase Selectivity Profiling

Compound X was assayed at 1 μM in the DiscoveRx KINOMEscan (scanEDGE) assay platform. The results for single-concentration binding interactions are reported as “%Ctrl” (DMSO = 100%Ctrl), where lower numbers indicate stronger hits.

Immunoassay and Immunoblotting

The human neuroblastoma Kelly cell line was maintained in RPMI with 10% fetal calf serum. For dose–response experiments, cells were seeded into 10 cm dishes and, following attachment, treated with the indicated compound concentration in fresh media. After 3 h, the plates were transferred onto ice, the media were taken off, and the cells were washed once with ice-cold phosphate-buffered saline (PBS), before adding 5% CHAPS lysis buffer. After 20 min in lysis buffer, cells were scraped, transferred into Eppendorfs, and spun at 4 °C for 12 min at 12 000 rpm. The supernatant was transferred to a fresh tube, and the protein content was quantified using the Direct Detect system (Millipore). For ALK immunoassay analysis, 15 μg of protein per well was added in duplicate for each sample to measure either total or pY1586 ALK, and for immunoblotting, 5 μg of denatured lysates were loaded into 4–12% Bis/Tris gels, both according to previously published methods.[37]

Crystallization, Data collection, and Processing

Purified protein in 25 mM HEPES pH 7.5, 300 mM NaCl, 0.5 mM TCEP, 5% glycerol was cocrystallized with compounds in 1:3 molar ratio at 293 K using sitting drop vapor diffusion. Reservoir solutions contained 25% poly(ethylene glycol)3350, 0.2 M ammonium sulfate, 0.1 M Bis–Tris pH 6.5 or 30% pentaerythritol ethoxylate 15/4, 0.05 ammonium sulfate, 0.1 M Bis–Tris pH 6.5 respectively.

Diffraction data were collected at the SLS X06S overlaid with X-ray A and BESSY II 14.2 and processed using either XDS (PMID 20124692) or iMOSFLM (PMID 28569763). The data was scaled using aimless from the CCP4 suite (PMID 29533233) and phased with PHASER (PMID 28573584) using PDB ID 4OGI (PMID 24584101) as initial model. Manual model rebuilding alternated with structure refinement was performed in COOT (PMID 28291755) and REFMAC (PMID 15572771), respectively. Geometry of the final models was verified using MolProbity (PMID 29067766). Data collection and refinement statistics are summarized in Supp. Table 3.

NanoBRET Target Engagement Assay

The NanoBRET target engagement assay in principle was performed as described previously.[38,39] The full-length ALK and BRDBD1 plasmid containing C-terminal placements of NanoLuc were obtained by the manufacturer (Promega). To lower intracellular expression levels of the reporter fusion, the NanoLuc/kinase fusion construct was diluted into carrier DNA (pGEM3ZF-, Promega) at a mass ratio of 1:10 (mass/mass), prior to forming FuGENE HD complexes according to the manufacturer’s instructions (Promega). DNA–FuGENE complexes were formed at a ratio of 1:3 (μg DNA/μL FuGENE). One part of the transfection complexes was then mixed with 20 parts (v/v) of HEK293T cells suspended at a density of 2 × 105/mL in Dulbecco’s modified Eagle’s medium (Gibco) + 10% fetal bovine serum (GE Healthcare), seeded into T75 flasks, and allowed to express for 20 h. For target engagement, both serially diluted test compound and NanoBRET Kinase Tracer K5 (ALK) and BRD-Tracer (BRD4BD1) (Promega) at a final concentration of 2 and 0.5 μM, respectively, were pipetted into white 96-well plates (Corning 3600). The corresponding ALK or BRD4BD1-transfected cells were added and reseeded at a density of 2 × 105/mL after trypsinization and resuspending in Opti-MEM without phenol red (Life Technologies). The system was allowed to equilibrate for 2 h at 37 °C/5% CO2 prior to BRET measurements. To measure BRET, NanoBRET NanoGlo substrate + extracellular NanoLuc inhibitor (Promega) was added as per the manufacturer’s protocol, and filtered luminescence was measured on a CLARIOstar plate reader (BMG Labtech) equipped with a 450 nm band-pass filter (donor) and a 610 nm low-pass filter (acceptor). Competitive displacement data were then graphed using GraphPad Prism 7 software using a four-parameter curve fit with the following equation: Y = bottom + (top – bottom)/(1 + 10((Log IC).

Western Blot Analysis

Protein extracts of cells were prepared by lysis in radioimmunoprecipitation assay buffer (Sigma) supplemented with protease inhibitors (complete protease inhibitor cocktail, Roche). Protein extracts (25 μg) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto poly(vinylidene difluoride) membranes using the Trans-Blot Turbo transfer system (BioRad). After blocking with 5% bovine serum albumin in PBS with 0.1% Tween-20 for 30 min, the membrane was incubated with primary antibodies for 1 h at room temperature. Horseradish peroxidase-linked secondary antibodies were incubated 30 min at room temperature followed by enhanced chemiluminescence (ECL) detection (ECL Western blot substrate, Pierce).

Antibodies

Primary antibodies were obtained from the following sources: PLK-1 (05-844) and phosphohistone H3 (Ser10) (05806) from Millipore; pPlk1-T210 (5472) from Cell Signaling; pPlk1-S137 (21738) from Abcam; and cyclin B1 (GNS1) and β-actin (A2228) from Sigma-Aldrich served as loading control.

Docking

Docking studies were performed using Glide, Schrödinger, LLC, New York, NY, 2018. Preparation of ALK using PDB codes 2XB7 was performed using Schrödinger Suite 2018-3 Protein Preparation Wizard; Epik, Schrödinger, LLC, New York, NY, 2018, and preparation of ligands was performed using LigPrep, Schrödinger, LLC, New York, NY, 2018.