Structure-Based Optimization of Pyrazinamide-Containing Macrocyclic Derivatives as Fms-like Tyrosine Kinase 3 (FLT3) Inhibitors to Overcome Clinical Mutations

Abstract

Secondary mutations in Fms-like tyrosine kinase 3-tyrosine kinase domain (FLT3-TKD) (e.g., D835Y and F691L) have become a major on-target resistance mechanism of FLT3 inhibitors, which present a significant clinical challenge. To date, no effective drugs have been approved to simultaneously overcome clinical resistance caused by these two mutants. Thus, a series of pyrazinamide macrocyclic compounds were first designed and evaluated to overcome the secondary mutations of FLT3. The representative 8v exhibited potent inhibitory activities against FLT3^D835Y and FLT3^D835Y/F691L with IC₅₀ values of 1.5 and 9.7 nM, respectively. 8v also strongly suppressed the proliferation against Ba/F3 cells transfected with FLT3-ITD, FLT3-ITD-D835Y, FLT3-ITD-F691L, FLT3-ITD-D835Y-F691L, and MV4–11 acute myeloid leukemia (AML) cell lines with IC₅₀ values of 12.2, 10.5, 24.6, 16.9, and 6.8 nM, respectively. Furthermore, 8v demonstrated ideal anticancer efficacy in a Ba/F3-FLT3-ITD-D835Y xenograft model. The results suggested that 8v can serve as a promising macrocycle-based FLT3 inhibitor for the treatment of AML.

Acute myeloid leukemia (AML) is the most common leukemia in adults and remains a highly fatal disease with a five-year survival rate of 30.5% in the United States.¹ Fms-like tyrosine kinase 3 (FLT3), a type III receptor tyrosine kinase, is the most frequently mutated gene in AML.² Approximately 30% of patients with AML harbor activating mutations in FLT3, either internal tandem duplication (ITD) mutations in the juxtamembrane domain or point mutations in the tyrosine kinase domain (TKD).³ FLT3-ITD is the most frequent mutation accounting for approximately 23% of AML patients, which is associated with increased risk of relapse and confers a poor prognosis in patients with AML.⁴ FLT3-TKD mutations, typically at the activation loop (AL) residue and gatekeeper (GK) residue, generally occupy close to 7–10% of AML patients with more complex prognostic significance.^5,6 The most common mutation in AL is the substitution of aspartic acid by tyrosine at position 835 (D835Y), while that in GK is the substitution of phenylalanine by leucine at position 691 (F691L).⁷ These mutations lead to ligand-independent dimerization and autophosphorylation and compositionally activate downstream FLT3 signaling pathways, such as STAT5, PI3K/AKT, and Ras/MAPK, ultimately promoting AML cell survival and proliferation.⁸ Therefore, targeting FLT3 mutations has become an effective strategy for the treatment of AML.

Several small-molecule FLT3 inhibitors have been developed for the treatment of AML. The first-generation FLT3 inhibitors, such as midostaurin (1),⁹ sunitinib (2),¹⁰ and sorafenib (3),¹¹ have been used as first-line targeted therapy (Figure 1). However, the toxicity issue limited the wide clinical application owing to lacking kinome selectivity. Hence, two selective second-generation FLT3 inhibitors quizartinib (4)¹² and gilteritinib (5)¹³ have also been developed with lower toxicity and significant responses in AML patients (Figure 1).^14,15 However, clinically acquired resistance occurred by the emergence of secondary mutations in TKD of FLT3 (ITD-D835Y or ITD-F691L) decreased the clinical efficacy of existing drugs.¹⁶ The D835Y mutation promoted the DFG-out conformation of FLT3 to DFG-in, which was detected in AML patients with FLT3-ITD mutations who relapsed after treatment with the type II inhibitor quizartinib.¹⁷ The type I inhibitor gilteritinib consistently showed potency against various AL mutations in FLT3-TKD including D835Y as well as ITD.¹³ However, the GK mutation F691L showed universal resistance to the available second-generation FLT3 inhibitors 4 and 5 due to the loss of interaction or inducing steric hindrance with them.^18,19 The macrocyclic inhibitor pacritinib (6) is currently in clinical phase II for the treatment of AML, but its inhibitory activity against FLT3^F691L is also unsatisfactory (Ba/F3-FLT3-ITD-F691L = 291 nM).²⁰ Several FLT3 inhibitors that can effectively overcome D835Y and F691L mutations simultaneously have been reported, such as FF-10101 and TLX83 (Figure S1),²¹⁻²⁴ which are still in biological studies or early stages of clinical testing. The secondary mutations, especially F691L, have emerged as the primary resistance mechanism of action of FLT3 inhibitors and served as an unmet clinic need. Therefore, the development of novel FLT3 inhibitors targeting resistance-related secondary mutations is urgently needed. Herein, we reported the first discovery of a pyrazinamide macrocycle-based FLT3 inhibitor 8v to overcome the D835Y and F691L mutations, which demonstrated ideal anticancer efficacy in a Ba/F3-FLT3-ITD-D835Y xenograft model.

Figure 1.

Chemical structures of representative FLT3 inhibitors.

Molecular Design

G-749 (7) is a type I FLT3 inhibitor, which displayed digital nanomolar activities against ITD and various AL mutants and moderate activity against FLT3^F691L (Ba/F3-FLT3-ITD-F691L = 38.1 nM).²⁵ Molecular docking studies showed that the lactam in the pyrimidinone core of 7 forms two hydrogen bonds with the residues Glu692 and Cys694 in the hinge region (Figure 2). The diphenyl ether moiety points to the solvent front region of FLT3 kinase. The bromine group extends to the gatekeeper residue Phe691. Based on the binding mode of 7 with FLT3, it is revealed that the pyrimidine core and the aminopiperidine moiety adopted a dihedral angle of 168° in an “unfolded” conformation. Further torsional angle scan analysis of compound 7 suggested that it has two local minimum energy conformations, including “folded” and unfolded, under water solution conditions, with the dihedral angles around 0 and 180°, respectively, revealing a conformational switch from folded to unfolded low-energy conformation when it binds with FLT3 (Figure 3). Therefore, we hypothesized that a macrocyclization strategy might be suitable to “lock” the molecule in the FLT3 dominant binding unfolded conformation to avoid the entropy-driven binding affinity loss, which can lead to high affinity and specificity with the target protein. Moreover, the macrocyclic conformational constrained strategy has been reported in the discovery of new generation inhibitors of several kinases, such as ALK, TRK, EGFR, and c-Met.²⁶⁻³⁰ Hence, we introduced an appropriate linker to compound 7 and obtained a series of macrocyclic derivatives as novel FLT3 inhibitors to enhance inhibitory activities against the D835Y and F691L mutants (Figure 4).

Figure 2.

Predicted binding mode of compound 7 with FLT3 (PDB: 6JQR). The hydrogen bonds are displayed as yellow dashed lines and the key residues are labeled as gray sticks.

Figure 3.

Torsion angle scan for compound 7 by metadynamic studies.

Figure 4.

Optimization design of macrocycle-based FLT3 inhibitors based on the lead 7.

Results and Discussion

Taken compound 7 as a lead, the 5-carbon atom length-linked compound 8a was first synthesized by considering the distance between the phenyoxyl and nitrogen methyl with 7.5 Å (Figure 2) and deleting the bromine atom to reduce the steric clash with the GK residue. Unexpectedly, the cyclized compound 8a totally lost the biochemical activity against FLT3^D835Y and cellular activities against the D835Y and F691L mutants (Table 1). Molecular docking studies suggested that it cannot bind well to the adenosine triphosphate (ATP) pocket of FLT3 due to the strained macrocycle linker. To reduce the tension of 8a, lengthening the linker of 8a as 6-carbon atom length or switching the oxygen atom in the linker from 4 to 3 position of phenyl led to compounds 8b and 8c, respectively. Both compounds displayed improved biochemical and cellular activities against FLT3 mutants. Especially, compound 8c exhibited inhibitory activities against FLT3^D835Y (IC₅₀ = 0.16 μM) and Ba/F3 cells transfected with FLT3-ITD (IC₅₀ = 0.17 μM), FLT3-ITD-D835Y (IC₅₀ = 0.13 μM), and FLT3-ITD-F691L (IC₅₀ = 0.16 μM), which provided a promising lead for further optimization. Subsequently, lengthening or shortening the linker of 8c led to two compounds 8d and 8e, which displayed slightly decreased cellular activities than that of 8c, though 8d exhibited the improved kinase inhibition against the FLT3^D835Y mutant (Table 1).

Table 1. Biochemical and Cellular Activities of Compounds 8a–e against the Multiple Mutants of FLT3^a.

^aThe IC₅₀ values (biochemical and cellular activities) are the means of duplicate experiments.

The molecular docking studies indicated that compound 8c bound to the ATP-binding pocket of FLT3 in a type I mode similar to that of 7. It is suggested that the pyrimidinone of 8c may make potential steric hindrance with the rotated Leu691 (Figure 5A). Thus, replacing the pyrimidinone core of 8c with a smaller volume pyrimidine-5-carboxamide scaffold gave 8f (Table 2). However, 8f showed 10-fold decreased activities against FLT3 mutants in both biochemical and cellular assays. It is speculated that the pyrimidine-5-carboxamide core of 8f is not suitable for the binding with FLT3 due to a back hydrophobic pocket formed by residues Cys828, Asp829, Lys644, and Val675 (Figure 5B). To explore this pocket, several hydrophobic substituents, such as methyl, ethyl, and isopropyl, were introduced at the R₁ position in the pyrazine-2-carboxamide scaffold (Table 2). The resulting compounds 8g–i exhibited improved biochemical and cellular activities than 8f. For instance, compounds 8g–i exhibited the potency against FLT3^D835Y with IC₅₀ values of 79, 30, and 29 nM, respectively. Compound 8h displayed the most cellular activities against Ba/F3 cells transfected with FLT3-ITD (IC₅₀ = 0.18 μM) and FLT3-ITD-D835Y (IC₅₀ = 0.32 μM) while poor against FLT3-ITD-F691L mutants (IC₅₀ = 1.18 μM). The predicted binding mode indicated that 8h bound into FLT3 DFG-in conformation with excellent shape complementarity. The amide in the pyrazine core formed key hydrogen bonds with hinge residues Glu692 and Cys694, and the ethyl group pointed into the aforementioned back hydrophobic pocket. Additionally, the aniline group extended to the solvent-accessible space of kinase, providing a suitable optimization site (Figure 5C). Subsequently, compounds 8j–n harboring various hydrophilic groups were designed and synthesized (Table 2). Encouragingly, 8j–n demonstrated a significantly improved inhibitory activity against FLT3^D835Y with IC₅₀ values of 14–78 nM. Among them, compound 8k exhibited obviously improved cellular activities (Ba/F3-FLT3-ITD: IC₅₀ = 140 nM, Ba/F3-FLT3-ITD-D835Y: IC₅₀ = 260 nM, Ba/F3-FLT3-ITD-F691L: IC₅₀ = 560 nM). This may be related to the fact that these compounds display poor membrane permeability and also with a higher ATP concentration in cell line than the biochemical level. The molecular docking model indicated that the morpholine group extended to the solvent region, which validated our design strategy and explained the improved potency (Figure 5D).

Figure 5.

Binding modes of compounds 8c (A), 8f (B), 8h (C), and 8k (D) with FLT3 (PDB: 6JQR). The hydrogen bonds are displayed as yellow dashed lines and the key residues are labeled as gray sticks. The structural model showed the steric clashes between gatekeeper mutation residue (F691L) and 8b in panel (A). The back hydrophobic pocket is formed by residues Cys828, Asp829, Lys644, and Val675, which were shown as the surface in panels (B) and (C).

Table 2. Biochemical and Cellular Activities of Compounds 8f–n against the Multiple Mutants of FLT3^a.

^aThe IC₅₀ values (biochemical and cellular activities) are the means of duplicate experiments.

Taken 8k as a lead, further optimization was focused on improving the potency against FLT3 mutants. By analyzing the binding mode of 8k with FLT3, it was indicated that the acyl piperidine is adjacent to Asp698 (Figure 5D). We hypothesized that introducing potential ion–ion interaction between Asp698 and inhibitors might improve the inhibitory activity. Thus, the acyl piperidine of 8k was replaced by a highly alkaline ethylpiperidine. The resulting macrocycle compound 8o was indeed more potent than 8k for the FLT3^D835Y mutant in the biochemical assay (IC₅₀ = 2.6 nM) (Table 3). 8o also achieved nanomolar inhibitory activities against Ba/F3 cells transfected with FLT3-ITD (IC₅₀ = 10.7 nM) and FLT3-ITD-D835Y (IC₅₀ = 1.1 nM) but poor against FLT3-ITD-F691L mutants (IC₅₀ = 60.8 nM), which was comparable to those of G-749 and gilteritinib. The results also demonstrated our hypothesis that making additional interactions with Asp698 is an effective approach capable of improving the inhibitor’s potency against FLT3 mutants. Molecular docking analysis revealed that 4-piperidine in 8o was adjacent to the ribose region in FLT3 and formed potential ion–ion interactions with Asp698 (Figure 6). Previous studies have proved that the ribose region of FLT3 could be utilized to improve the binding affinity between inhibitors and FLT3.^31,32 However, replacement of the 4-piperidine moiety of 8o with R-/S- 3-piperidine and R-/S- 2-pyrrole resulted in compounds 8p–s, which significantly decreased both biochemical and cellular activities against FLT3 mutants. It is explained that the alkaline tertiary amine was away from Asp698, which weakened the direct interactions with FLT3. Further lengthening the carbon atom of the link of 8o led to compounds 8t and 8u, which maintained similar potency to 8o against FLT3^D835Y (Table 4) while displaying decreased cellular activities against three engineered Ba/F3 cell lines. Based on these results, it is suggested that the length of the macrocycle linker indeed has an effect on the potency of compounds. To explore the types of linkers, the alkyl ether and amide group were introduced to get compounds 8v and 8w. Encouragingly, compounds 8v and 8w displayed digital nanomolar activities against FLT3^D835Y with IC₅₀ values of 1.5 and 8.8 nM, respectively. More importantly, compound 8v achieved nanomolar inhibitory activities against Ba/F3-FLT3-ITD (IC₅₀ = 12.2 nM), Ba/F3-FLT3-ITD-D835Y (IC₅₀ = 10.5 nM), and Ba/F3-FLT3-ITD-F691L (IC₅₀ = 24.6 nM), which is more potent than gilteritinib (Table 4).

Table 3. Biochemical and Cellular Activities of Compounds 8o–s against Various Mutants of FLT3^a.

^aThe IC₅₀ values (biochemical and cellular activities) are the means of duplicate experiments.

Figure 6.

Predicted binding mode of compound 8o with FLT3 (PDB: 6JQR). The hydrogen bonds are displayed as yellow dashed lines and the ion–ion interactions are displayed as purple dashed lines, respectively. The key residues are labeled as gray sticks.

Table 4. Biochemical and Cellular Activities of Compounds 8t–w against Various Mutants of FLT3^a.

^aThe IC₅₀ values (biochemical and cellular activities) are the means of duplicate experiments.

To elucidate the binding mode of compounds with target, we obtained the crystal structure of wild-type FLT3 in complex with 8v (Figures 7A, S2, and Table S1). As we expected, the pyrazinamide core of 8v makes two hydrogen bonds with the residues Glu692 and Cys694 in the hinge region. Moreover, 4-piperidine of 8v was adjacent to the ribose region in FLT3 and formed potential ion–ion interactions with Asp698. Molecular modeling studies of 8v with FLT3^F691L indicated that it did not have any spatial collision with the mutated leucine 691 (Figure 7B), which verified our design strategy and explained the improved potency. The results also demonstrated our hypothesis that avoiding steric hindrance with F691L is an effective approach capable of overcoming the GK mutation of FLT3.

Figure 7.

Cocrystal structure of wild-type FLT3 in complex with 8v (A, PDB: 8XB1) and the predicted binding mode of compound 8v with FLT3^F691L (B). The hydrogen bonds are displayed as yellow dashed lines and the ion–ion interactions are displayed as purple dashed lines, respectively. The key residues are labeled as gray sticks. The GK residue Phe691 is mutated to Leu691 in panel (B).

Further kinase profiling screening of 8v against a panel of 378 wild-type kinases was conducted at Eurofins Cerep SA at a concentration of 1 μM. The results demonstrated that 8v displayed a moderate kinome selectivity profile with S (1) and S (10) scores of 0.029 and 0.172 (Figure 8, Tables S2 and S3). Compared to the KINOMEscan assay profile for gilteritinib at 100 nM,¹³8v exhibited improved selectivity. We next evaluated the inhibitory activity of 8v on kinases associated with significant immune and cardiovascular toxicity, such as VEGFR1/2/3, C-KIT, PDGFRα/β, and RET (Table 5). The study found that 8v exhibited potent inhibitory activity against VEGFR3 with an IC₅₀ of 21.3 nM while being less effective against VEGFR1/2, C-KIT, PDGFRα/β, and RET, with IC₅₀ values ranging from 78.8 to 2103.0 nM, which revealed that 8v is a relatively selective FLT3 inhibitor.

Figure 8.

KINOMEscan profiling results for compound 8v at a concentration of 1 μM against 378 kinases.

Table 5. Inhibitory Activity (IC₅₀) of 8v against Some Kinases.

kinase	%ctrl@1000 nM	IC50 (nM)
VEGFR1	1	224.2
VEGFR2	6	110.2
VEGFR3	1	21.3
C-KIT	5	78.8
PDGFRα	24	178.8
PDGFRβ	71	2103.0
RET	1	211.7

Given excellent inhibitory activities of compounds 8o and 8v against multiple clinical mutations of FLT3, we next investigated their antiproliferation activities against FLT3-ITD-positive AML cell lines MV4–11 (Table 6). The results showed that 8o and 8v selectively inhibit the proliferation of MV4–11 cells with IC₅₀ values of 11.0 and 6.8 nM, respectively. And they did not exhibit inhibitory activity in human normal cell lines, such as HEK-293. Moreover, considering that clinical patients with gilteritinib resistance carrying the FLT3-ITD-D835Y-F691L mutation,³³ we tested the inhibitory activities of 8o, 8v, and gilteritinib against FLT3^D835Y/F691L and Ba/F3 cells transfected with FLT3-ITD-D835Y-F691L. As shown in Table 6, 8o and 8v displayed digital nanomolar activities against FLT3^D835Y/F691L with IC₅₀ values of 3.5 and 9.7 nM, respectively, and exhibited potent antiproliferative activities against Ba/F3-FLT3-ITD-D835Y-F691L with IC₅₀ values of 33.8 and 16.9 nM, respectively, which is more potent than gilteritinib (Table 6).

Table 6. Biochemical and Cellular Activities of 8o, 8v, and Gilteritinib^a.

	biochemical kinase inhibition (IC50: nM)	cellular inhibition (IC50: nM)
cpds	FLT3D835Y/F691L	Ba/F3-FLT3-ITD-D835Y-F691L	MV4–11	HEK-293
8o	3.5	33.8	11.0	2792.3
8v	9.7	16.9	6.8	2735.9
gilteritinib	46.0	49.1	7.5	1387.6

^aThe IC₅₀ values (biochemical and cellular activities) are the means of duplicate experiments.

We next investigated the effects of 8v on the phosphorylation of FLT3 and its downstream mediators STAT5, ERK, and AKT by Western blot analysis in MV4–11, Ba/F3-FLT3-ITD-D835Y, and Ba/F3-FLT3-ITD-F691L cell lines. As shown in Figure 9, compound 8v dose-dependently inhibited the phosphorylation of FLT3 and the downstream signaling mediators STAT5, ERK, and AKT in MV4–11 and Ba/F3-FLT3-ITD-D835Y cell lines with comparable potency to gilteritinib. Notably, a similar trend was observed for compound 8v in Ba/F3-FLT3-ITD-F691L cell lines, while gilteritinib was significantly less effective against Ba/F3-FLT3-ITD-F691L and its downstream signaling. Taken together, these results demonstrated that compound 8v indeed interfered the transduction of FLT3 signaling pathways in MV4–11, Ba/F3-FLT3-ITD-D835Y, and Ba/F3-FLT3-ITD-F691L cell lines.

Figure 9.

Effects of compound 8v and gilteritinib on the FLT3-mediated signaling pathway in MV4–11 (A), Ba/F3-FLT3-ITD-D835Y (B), and Ba/F3-FLT3-ITD-F691L (C) cell lines.

Biparametric flow cytometric analysis was performed to explore the apoptosis induction of compound 8v in MV4–11 cells. As shown in Figure 10, notable apoptotic cells (32.52% in 20 nM and 34.84% in 100 nM) were observed in MV4–11 cells under the treatment of 8v. The results are consistent with its antiproliferative activity against MV4–11 cells. Besides, compound 8v potently arrested cell cycle progression in the G0/G1 phase in MV4–11 cells. Western blot analysis also showed that 8v dose-dependently induced the cleaved activation of caspase-9 and decreased the protein levels of CDK4, CDK6, and cyclin D3 in MV4–11 cells (Figure 10).

Figure 10.

8v induced apoptosis and G0/G1 phase arrest in MV4–11 cells. (A) Effects of 8v on apoptosis in MV4–11 cells (48 h). Representative flow cytometry profiles are shown here. (B) Effects of 8v on cell cycle in MV4–11 cells (24 h). Representative flow cytometry profiles of cell cycle distribution are shown here. (C) Effects of 8v on apoptosis-related proteins. (D) Effects of 8v on cell cycle-related proteins.

With the encouraging biochemical and cellular profiles of 8v, we next evaluated the pharmacokinetic (PK) properties of compound 8v in Sprague–Dawley rats by intravenous and oral administration (Table 7). Compound 8v displayed an ideal PK profile with a reasonable half-life (T_1/2) of 7.75 h, an acceptable area under the curve (AUC) of 3538.61 ng·h/mL, and an ideal oral bioavailability (F) of 32.82%, indicating that 8v was suitable for oral application in the following animal efficacy study.

Table 7. Pharmacokinetic Studies of 8v by Intravenous and Oral Administration in Sprague–Dawley Rats.

administration route	dose level (mg/kg)	T1/2 (h)	Tmax (h)	Cmax (ng/mL)	AUC (0–t) (h*ng/mL)	AUC (0–∞) (h*ng/mL)	Cl (mL/h/kg)	F (%)
IV	2	8.58	0.083	230.03	1951.23	2279.81	894.89
PO	10	7.75	6.00	204.64	3201.72	3538.61		32.82

Further, in vivo anticancer efficacy of compound 8v was investigated in the established Ba/F3-FLT3-ITD-D835Y inoculated xenograft tumor model in nude mice. As shown in Figure 11, 8v dose-dependently inhibited the tumor growth with TGI (tumor growth inhibition) of 88.2 and 51.6% at the dosage of 30 and 10 mg/kg on the sixth day in the Ba/F3-FLT3-ITD-D835Y xenograft model, respectively. And there were no nude mice fatalities observed in the 8v 30 mg/kg group until the end point of the study. Meanwhile, no obvious weight loss was observed during the whole process. Taken together, these results demonstrated that compound 8v possessed significant anticancer potency in the Ba/F3-FLT3-ITD-D835Y xenograft tumor model without potential toxicity (Figure 11).

Figure 11.

In vivo antitumor effect of compound 8v in the Ba/F3-FLT3-ITD-D835Y xenograft model. Anticancer efficacy (A, B), body weight effect (C), and percent survival (D) in the Ba/F3-FLT3-ITD-D835Y xenograft model.

Chemistry

The synthesis of macrocycles 8a–e is described in Scheme 1. Briefly, nucleophilic coupling of 4′ or 3′ hydroxyl-tert-butyl phenylcarbamates 9a and 9b with various brominated ethyl esters 10a–c gave 11a–e. Further, deprotection of 11a–e with trifluoroacetic acid (TFA), followed by the nucleophilic reaction with ethyl 4-chloro-6-methyl-2-(methylthio)pyrimidine-5-carboxylate (12), obtained 13a–e. Then, 13a–e were treated with 1,1-dimethoxy-N,N-dimethylmethanamine (DMF-DMA), followed by ammonia under 30% NH₄OH to yield the compounds 14a–e. Oxidization of 14a–e by meta-Chloroperoxybenzoic acid (m-CPBA), followed by the nucleophilic reaction with tert-butyl 4-aminopiperidine-1-carboxylate, gave the key intermediates 15a–e, which were followed by hydrolysis, deprotection, and amidation to yield the desired macrocycles 8a–e.

Scheme 1. Synthetic Route of Macrocycles 8a–e.

Reagents and conditions: (a) NaH, DMF, room temperature (rt), 6 h, yield: 85–90%; (b) (i) TFA, dichloromethane (DCM), rt, 3 h; (ii) acetic acid, 80 °C, 6 h, yield of two steps: 60–68%; (c) (i) 1,1-dimethoxy-N,N-dimethylmethanamine, DMF, 100 °C, 8 h; (ii) 30% NH₄OH, EtOH, 85 °C, 10 h, yield of two steps: 28–36%; (d) (i) m-CPBA, DMF, rt, 4 h; (ii) tert-butyl 4-aminopiperidine-1-carboxylate, triethylamine, DMF, 50 °C, 6 h, yield of two steps: 11–18%; (e) (i) LiOH·H₂O, tetrahydrofuran (THF)/H₂O, 50 °C, overnight; (ii) TFA, DCM, rt, 2 h; (iii) pentafluorophenyl diphenylphosphinate (FDPP), N,N-diisopropylethylamine (DIPEA), DMF, rt, 10 h, yield of three steps: 6–13%.

The synthesis of compounds 8g–n is depicted in Scheme 2. First, 18 and 18a underwent a direct nucleophilic substitution reaction with ethyl 5-bromopentanoate to yield 19 and 19a. Then, 19a reacted with various hydrophilic groups to obtain 20a–e. Reduction of 19 and 20a–e with iron dust/ammonium chloride produced the arylamines 20 and 21a–e, which underwent Buchwald–Hartwig coupling with various substituted pyrazine derivatives 17a–c to give the key intermediates 21 and 22a–g, respectively. Hydrolyzation of 21 and 22a–g, followed by deprotection and amidation, gave the macrocycles 22 and 23a–g. Subsequently, hydrolyzation of 22 and 23a–g under NaOH/H₂O₂ obtained the desired compounds 8g–n.

Scheme 2. Synthetic Route of Macrocycles 8g–n.

Reagents and conditions: (a) tert-butyl-4-aminopiperidine-1-carboxylate, DIPEA, 1,4-dioxane, 50 °C, 8 h, yield: 73–76%; (b) ethyl 5-bromopentanoate, K₂CO₃, DMF, 60 °C, 6 h, yield: 80%; (c) K₂CO₃, DMF, 60 °C, overnight, yield: 76–83%; (d) Fe, NH₄Cl, EtOH/H₂O, 70 °C, 6 h, yield: 65–75%; (e) Pd(OAc)₂, Xantphos, Cs₂CO₃, 60 °C, 3 h, yield: 39–50%; (f) (i) LiOH·H₂O, THF/H₂O, 50 °C, overnight; (ii) TFA, DCM, rt, 2 h; (iii) FDPP, DIPEA, DMF, rt, 10 h, yield of three steps: 13–20%; (g) 2 mol/L NaOH(aq), 30% H₂O₂(aq), dimethyl sulfoxide (DMSO), rt, 10 min, yield: 14–31%.

The synthesis of compounds 8o–v is depicted in Scheme 3. As shown in Scheme 3, nucleophilic coupling of 18a with benzyl bromide and morpholine gave intermediate 26, which was reduced by iron dust/ammonium giving 27. Nucleophilic coupling of 16b with various Boc-protected hydrophilic amines yielded 17b and 24a–e. Further, Buchwald–Hartwig coupling of 27 with various substituted pyrazine derivatives (17b, 24a–e) provided the key intermediates 28a–e, which were hydrolyzed to obtain 29a–e. Subsequently, 29a–e were reduced by H₂ and underwent the nucleophilic reaction with 1,5-dibromopentane to obtain derivatives 30a–e. Finally, deprotection of 30a–e with trifluoroacetic acid, followed by the nucleophilic reaction, gave the desired compounds 8o–s. Similarly, 29a was reduced by hydrogen (H₂) and underwent the nucleophilic reaction with 31a–c to obtain 32a–c, which were deprotected with trifluoroacetic acid, followed by the nucleophilic reaction, to obtain the desired compounds 8t–v. Additionally, the synthesis routes of macrocycles 8f and 8w are shown in Schemes S1 and S2.

Scheme 3. Synthetic Route of Macrocycles 8o–v.

Reagents and conditions: (a) DIPEA, 1,4-dioxane, 50 °C, 8 h, yield: 75–82%; (b) (bromomethyl)benzene, K₂CO₃, DMF, 70 °C, 3 h, yield: 85%; (c) morpholine, K₂CO₃, DMF, 70 °C, 8 h, yield: 75%; (d) Fe, NH₄Cl, EtOH/H₂O, 70 °C, 6 h, yield: 70%; (e) Pd(OAc)₂, Xantphos, Cs₂CO₃, 60 °C, 3 h, yield: 60–68%; (f) 2 mol/L NaOH(aq), 30% H₂O₂(aq), DMSO, rt, 30 min, yield: 64–70%; (g) (i) H₂, Pd/C, EtOH, 40 °C, overnight; (ii) K₂CO₃, DMF, 70 °C, 4 h, yield of two steps: 16–28%; (h) (i) TFA, DCM, rt, 2 h; (ii) K₂CO₃, DMF, 65 °C, 4 h, yield of two steps: 20–35%.

Conclusions

In summary, a series of pyrazinamide macrocycle FLT3 inhibitors were first designed and synthesized to overcome multiple clinical mutations of FLT3-TKD (ITD-D835Y or ITD-F691L). One representative compound 8v exhibited strong kinase inhibition against FLT3^D835Y and FLT3^D835Y/F691L with IC₅₀ values of 1.5 and 9.7 nM, respectively. It also strongly suppressed the proliferation of Ba/F3 cells driven by FLT3-ITD, FLT3-ITD-D835Y, FLT3-ITD-F691L, and FLT3-ITD-D835Y-F691L with IC₅₀ values of 12.2, 10.5, 24.6, and 16.9 nM, respectively, and AML MV4–11 cell lines with the IC₅₀ value of 6.8 nM. Moreover, 8v demonstrated ideal antitumor efficacy in a Ba/F3-FLT3-ITD-D835Y xenograft model with TGI values of 88.2% at the dosage of 30 mg/kg. The study provides a promising new macrocycle compound for future drug discovery combating multiple clinical resistance in AML patients.

Experimental Section

General Synthetic Methods

Starting materials, reagents, and solvents were obtained via commercial sources and used without further purification. Flash column chromatography was performed using 200–300 mesh or 300–400 mesh silica gel. All ¹H and ¹³C NMR spectra were obtained on a Bruker AV-400 spectrometer at 400 MHz or a Bruker AV-600 spectrometer at 151 MHz in DMSO-d₆ or CDCl₃. Coupling constants (J) are expressed in hertz (Hz). Chemical shifts (δ) of NMR were reported in parts per million (ppm) units relative to an internal standard (TMS). Low-resolution electrospray ionization mass spectrometry (ESI-MS) was performed on an Agilent 1200 HPLC-MSD mass spectrometer and high-resolution ESI-MS was performed on an Applied Biosystems Q-STAR Elite ESI-LC-MS/MS mass spectrometer at Jinan University. Target compounds 8a–w were found to be >95% pure based on analytical high-performance liquid chromatography [HPLC, Dionex Summit HPLC (column: Diamonsil C18, 5.0 μm, 4.6 × 250 mm (Dikma Technologies); detector: PDA-100 photodiode array; injector: ASI-100 autoinjector; pump: p-680A)] analysis. A flow rate of 0.6–1.0 mL/min was used with a mobile phase of MeOH in H₂O (containing 0.1% triethylamine).

3⁵,3⁶-Dihydro-6-oxa-2,4-diaza-3(2,4)-pyrido[4,3-d]pyrimidina-1(4,1)-piperidina-5(1,4)-benzenacycloundecaphane-3⁵,11-dione (8a)

Lithium hydroxide monohydrate (48 mg, 2.0 mmol) was added to a solution of 15a (230 mg, 0.40 mmol) in a mixture of THF and H₂O (12 mL, 5:1, v/v). The resulting mixture was stirred at 50 °C overnight. Then the reaction mixture was concentrated and purified to afford an intermediate of acid (150 mg) as a white solid. The crude product was dissolved in DCM (10 mL) and added 3 mL of trifluoroacetic acid (TFA). The reaction solution was stirred for 2 h at room temperature, and then the solvent was evaporated. A solution of the residue, FDPP (230 mg, 0.6 mmol), and DIPEA (0.35 mL, 2.0 mmol) in DMF (100 mL) was stirred for 10 h at room temperature. The mixture was extracted with DCM (3 × 40 mL). The combined organic extracts were dried with anhydrous Na₂SO₄ and evaporated. The crude product was purified by column chromatography on silica gel to obtain 8a (14 mg, yield of three steps: 8%) as a white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 12.33 (s, 1H), 11.53 (s, 1H), 8.77 (s, 1H), 7.75 (t, J = 5.7 Hz, 1H), 7.28 (s, 2H), 7.04 (d, J = 7.7 Hz, 2H), 6.32 (d, J = 6.3 Hz, 1H), 4.32–4.22 (m, 3H), 3.75–3.62 (m, 2H), 2.84–2.78 (m, 2H), 2.43–2.37 (m, 2H), 2.18–2.08 (m, 2H), 1.81–1.60 (m, 4H), 1.50–1.42 (m, 2H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.20, 161.94, 160.86, 156.95, 129.05, 127.71, 116.38, 97.91, 96.60, 67.86, 52.31, 44.36, 40.80, 31.08, 29.89, 29.61, 26.46, 21.14. HRMS (ESI) for C₂₃H₂₇O₃N₆ [M + H]⁺: calcd, 435.2139, found, 435.2131. HPLC analysis: MeOH–H₂O (70:30), 12.48 min, 100% purity.

3⁵,3⁶-Dihydro-6-oxa-2,4-diaza-3(2,4)-pyrido[4,3-d]pyrimidina-1(4,1)-piperidina-5(1,4)-benzenacyclododecaphane-3⁵,12-dione (8b)

In a manner similar to that described for 8a, using 15b (310 mg, 0.52 mmol), 8b (19 mg, yield of three steps: 11%) was obtained as a white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.65 (s, 1H), 11.43 (s, 1H), 7.49 (t, J = 6.7 Hz, 1H), 7.34 (d, J = 8.4 Hz, 2H), 6.89 (d, J = 8.4 Hz, 2H), 6.13 (d, J = 7.1 Hz, 1H), 4.37 (d, J = 13.2 Hz, 1H), 4.24–4.07 (m, 2H), 3.72–3.69 (m, 1H), 3.25–3.17 (m, 2H), 2.74–2.56 (m, 2H), 2.27–2.12 (m, 2H), 1.98–1.92 (m, 2H), 1.83–1.75 (m, 2H), 1.68–1.55 (m, 4H), 1.47–1.34 (m, 2H). ¹³C NMR (151 MHz, DMSO-d₆) δ 171.45, 162.07, 160.24, 159.16, 158.94, 156.50, 129.19, 126.72, 115.93, 96.53, 68.08, 52.03, 44.48, 41.07, 31.19, 29.73, 29.65, 26.91, 24.74, 23.25. HRMS (ESI) for C₂₄H₂₉O₃N₆ [M + H]⁺: calcd, 449.2296, found, 449.2294. HPLC analysis: MeOH–H₂O (70:30), 14.26 min, 97.46% purity.

3⁵,3⁶-Dihydro-6-oxa-2,4-diaza-3(2,4)-pyrido[4,3-d]pyrimidina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵,11-dione (8c)

In a manner similar to that described for 8a, using 15c (240 mg, 0.41 mmol), 8c (20 mg, yield of three steps: 11%) was obtained as a white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 12.11 (s, 1H), 11.28 (d, J = 4.5 Hz, 1H), 7.92 (s, 1H), 7.74 (d, J = 6.8 Hz, 1H), 7.34 (t, J = 6.7 Hz, 1H), 7.24 (t, J = 8.1 Hz, 1H), 6.64 (d, J = 8.3 Hz, 1H), 6.55 (d, J = 8.1 Hz, 1H), 6.03 (d, J = 7.2 Hz, 1H), 3.99–3.90 (m, 3H), 3.76–3.72 (m, 1H), 3.46 (s, 1H), 3.23–3.09 (m, 2H), 2.44 (s, 2H), 2.24–2.08 (m, 2H), 1.90–1.71 (m, 6H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.97, 163.80, 162.40, 162.29, 160.46, 159.55, 140.06, 136.56, 130.16, 113.08, 109.31, 106.77, 105.42, 97.05, 68.25, 49.45, 43.64, 31.51, 29.98, 27.01, 22.93. HRMS (ESI) for C₂₃H₂₇O₃N₆ [M + H]⁺: calcd, 435.2139, found, 435.2130. HPLC analysis: MeOH–H₂O (70:30), 14.00 min, 97.51% purity.

3⁵,3⁶-Dihydro-6-oxa-2,4-diaza-3(2,4)-pyrido[4,3-d]pyrimidina-1(4,1)-piperidina-5(1,3)-benzenacyclododecaphane-3⁵,12-dione (8d)

In a manner similar to that described for 8a, using 15d (0.32 mg, 0.54 mmol), 8d (15 mg, yield of three steps: 6%) was obtained as a white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 12.03 (s, 1H), 11.32 (d, J = 6.0 Hz, 1H), 8.29 (s, 1H), 7.77 (d, J = 6.8 Hz, 1H), 7.36 (t, J = 6.9 Hz, 1H), 7.24 (t, J = 8.1 Hz, 1H), 6.59 (dd, J = 13.9, 8.5 Hz, 2H), 6.07 (d, J = 7.3 Hz, 1H), 4.29–4.26 (m, 1H), 4.07–4.04 (m, 1H), 3.97–3.88 (m, 2H), 3.82–3.72 (m, 1H), 2.94–2.77 (m, 2H), 2.31–2.28 (m, 1H), 1.95–1.79 (m, 5H), 1.64–1.42 (m, 6H). ¹³C NMR (151 MHz, DMSO-d₆) δ 175.39, 163.73, 162.57, 161.87, 160.58, 159.88, 140.53, 136.63, 130.73, 112.60, 108.68, 106.00, 105.58, 97.39, 67.91, 49.65, 45.59, 41.78, 32.25, 30.07, 28.84, 26.97, 24.73. HRMS (ESI) for C₂₄H₂₉O₃N₆ [M + H]⁺: calcd, 449.2296, found, 449.2291. HPLC analysis: MeOH–H₂O (85:15), 6.56 min, 95.82% purity.

3⁵,3⁶-Dihydro-6-oxa-2,4-diaza-3(2,4)-pyrido[4,3-d]pyrimidina-1(4,1)-piperidina-5(1,3)-benzenacyclodecaphane-3⁵,10-dione (8e)

In a manner similar to that described for 8a, using 15e (250 mg, 0.44 mmol), 8e (25 mg, yield of three steps: 13%) was obtained as a white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 12.19 (s, 1H), 11.26 (s, 1H), 7.85–7.82 (m, 2H), 7.33 (d, J = 7.3 Hz, 1H), 7.24 (t, J = 8.0 Hz, 1H), 6.68 (dd, J = 8.2, 2.3 Hz, 1H), 6.56 (dd, J = 8.0, 2.0 Hz, 1H), 6.02 (d, J = 7.2 Hz, 1H), 4.20–4.09 (m, 2H), 3.90–3.84 (m, 1H), 3.01–2.74 (m, 4H), 2.42–2.34 (m, 2H), 2.08–2.01 (m, 2H), 1.89–1.85 (m, 2H), 1.65–1.55 (m, 2H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.18, 163.81, 162.86, 162.25, 160.43, 159.96, 139.77, 136.49, 130.13, 113.44, 110.18, 107.62, 105.44, 97.01, 67.35, 52.20, 49.07, 45.45, 30.63, 30.22, 27.46, 23.59. HRMS (ESI) for C₂₂H₂₅O₃N₆ [M + H]⁺: calcd, 421.1983, found, 421.1975. HPLC analysis: MeOH–H₂O (85:15), 7.50 min, 97.48% purity.

11-Oxo-6-oxa-2,4-diaza-3(2,4)-pyrimidina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8f)

To a solution of 36 (130 mg, 0.33 mmol) in DMSO (5 mL) was added 3 mL of 2 mol/L NaOH(aq) and 1.2 mL of 30% H₂O₂(aq). The mixture was stirred at room temperature for 10 min. The mixture was extracted with EA (3 × 10 mL). The combined organic extracts were dried with anhydrous Na₂SO₄ and evaporated. The crude product was purified by column chromatography on silica gel to obtain 8f (30 mg, yield: 22%) as a white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.76 (s, 1H), 8.53 (s, 1H), 7.88–7.74 (m, 3H), 7.20 (t, J = 8.2 Hz, 2H), 6.59 (d, J = 8.1 Hz, 1H), 6.48 (d, J = 8.0 Hz, 1H), 4.02–3.94 (m, 3H), 3.76–3.73 (m, 1H), 3.43 (s, 1H), 3.24–3.18 (m, 1H), 3.05 (s, 1H), 2.46–2.38 (m, 2H), 2.23–2.09 (m, 2H), 1.88–1.66 (m, 6H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.90, 170.10, 162.37, 159.85, 159.53, 158.71, 140.38, 130.05, 112.93, 109.08, 106.24, 97.85, 68.17, 49.07, 43.65, 31.49, 29.76, 29.13, 27.05, 22.93. HRMS (ESI) for C₂₁H₂₇O₃N₆ [M + H]⁺: calcd, 411.2139, found, 411.2133. HPLC analysis: MeOH–H₂O (80:20), 4.56 min, 98.74% purity.

3³-Methyl-11-oxo-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8g)

In a manner similar to that described for 8f, using 23a (82 mg, 0.20 mmol), 8g (20 mg, yield: 23%) was obtained as a white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.42 (s, 1H), 7.77 (s, 1H), 7.65 (s, 1H), 7.24 (s, 1H), 7.17–7.11 (m, 2H), 6.48 (d, J = 8.2 Hz, 1H), 6.42 (d, J = 8.0 Hz, 1H), 3.96–3.88 (m, 3H), 3.72–3.66 (m, 2H), 3.31–3.17 (m, 2H), 2.42 (s, 2H), 2.25 (s, 3H), 2.20–2.16 (m, 2H), 1.91–1.71 (m, 6H). ¹³C NMR (151 MHz, DMSO-d₆) δ 171.12, 170.20, 159.68, 153.34, 150.55, 141.52, 129.94, 128.66, 112.37, 112.20, 107.33, 104.82, 68.08, 55.36, 40.85, 31.57, 29.49, 28.78, 27.02, 22.92, 20.24. HRMS (ESI) for C₂₂H₂₉O₃N₆ [M + H]⁺: calcd, 425.2296, found, 425.2286. HPLC analysis: MeOH–H₂O (80:20), 4.68 min, 97.15% purity.

3³-Ethyl-11-oxo-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8h)

In a manner similar to that described for 8f, using 23b (83 mg, 0.20 mmol), 8h (25 mg, yield: 29%) was obtained as a yellow–white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.40 (s, 1H), 7.78 (s, 1H), 7.58 (d, J = 3.2 Hz, 1H), 7.27 (d, J = 3.2 Hz, 1H), 7.18–7.14 (m, 2H), 6.50 (dd, J = 8.2, 2.4 Hz, 1H), 6.43 (dd, J = 8.0, 2.1 Hz, 1H), 3.97 (d, J = 4.3 Hz, 2H), 3.88 (s, 1H), 3.75–3.72 (m, 2H), 3.30–3.16 (m, 2H), 2.60–2.54 (m, 2H), 2.45–2.38 (m, 2H), 2.26–2.15 (m, 2H), 1.96–1.70 (m, 6H), 1.20 (t, J = 7.3 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 171.23, 170.26, 159.68, 152.80, 150.31, 141.51, 132.75, 129.99, 112.39, 112.11, 107.31, 104.94, 68.07, 43.46, 40.81, 31.65, 29.47, 28.73, 27.04, 24.94, 22.91, 11.31. HRMS (ESI) for C₂₃H₃₁O₃N₆ [M + H]⁺: calcd, 439.2452, found, 439.2446. HPLC analysis: MeOH–H₂O (80:20), 5.62 min, 97.12% purity.

3³-Isopropyl-11-oxo-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8i)

In a manner similar to that described for 8f, using 23c (85 mg, 0.20 mmol), 8i (28 mg, yield: 31%) was obtained as a white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.36 (s, 1H), 7.78 (s, 1H), 7.54 (s, 1H), 7.27 (s, 1H), 7.20 (d, J = 7.1 Hz, 1H), 7.16 (t, J = 8.1 Hz, 1H), 6.50 (dd, J = 8.2, 2.4 Hz, 1H), 6.43 (d, J = 8.0 Hz, 1H), 3.99–3.95 (m, 2H), 3.80–3.69 (m, 3H), 3.24 (s, 2H), 3.14–3.08 (m, 1H), 2.47–2.39 (m, 2H), 2.24–2.14 (m, 2H), 1.99–1.72 (m, 6H), 1.16 (dd, J = 6.7, 3.3 Hz, 6H). ¹³C NMR (151 MHz, DMSO-d₆) δ 171.30, 170.24, 159.67, 152.07, 150.08, 141.50, 136.33, 129.99, 112.39, 112.03, 107.27, 104.98, 68.05, 49.43, 43.38, 31.67, 29.45, 28.66, 28.14, 27.03, 22.89, 21.37, 21.26. HRMS (ESI) for C₂₄H₃₂O₃N₆²³Na [M + Na]⁺: calcd, 475.2428, found, 475.2424. HPLC analysis: MeOH–H₂O (65:35), 15.39 min, 99.58% purity.

3³-Ethyl-5⁴-(4-methylpiperazin-1-yl)-11-oxo-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8j)

In a manner similar to that described for 8f, using 23d (85 mg, 0.16 mmol), 8j (15 mg, yield: 18%) was obtained as a yellow solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.10 (s, 1H), 7.68 (s, 1H), 7.53 (s, 1H), 7.19 (s, 1H), 6.94 (d, J = 7.2 Hz, 1H), 6.79 (d, J = 8.4 Hz, 1H), 6.42 (dd, J = 8.4, 2.1 Hz, 1H), 3.94–3.89 (m, 2H), 3.79–3.74 (m, 1H), 3.68–3.49 (m, 6H), 3.27 (s, 2H), 2.97 (s, 4H), 2.60–2.54 (m, 2H), 2.27 (s, 3H), 2.03–1.85 (m, 10H), 1.19 (t, J = 7.4 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 175.79, 172.03, 170.23, 152.64, 151.71, 150.57, 136.39, 135.62, 132.20, 118.56, 111.88, 66.37, 55.33, 50.37, 45.95, 40.42, 29.45, 27.42, 24.84, 22.55, 20.79, 11.43. HRMS (ESI) for C₂₈H₄₁O₃N₈ [M + H]⁺: calcd, 537.3296, found, 537.3289. HPLC analysis: MeOH–H₂O (90:10), 5.62 min, 97.12% purity.

3³-Ethyl-5⁴-morpholino-11-oxo-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8k)

In a manner similar to that described for 8f, using 23e (98 mg, 0.19 mmol), 8k (20 mg, yield: 20%) was obtained as a yellow–white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.11 (s, 1H), 7.69 (s, 1H), 7.53 (s, 1H), 7.20 (s, 1H), 6.95 (d, J = 7.1 Hz, 1H), 6.79 (d, J = 8.4 Hz, 1H), 6.44 (dd, J = 8.4, 2.3 Hz, 1H), 3.95–3.89 (m, 2H), 3.81–3.75 (m, 1H), 3.72 (t, J = 4.7 Hz, 4H), 3.61–3.46 (m, 4H), 3.00–2.93 (m, 4H), 2.60–2.53 (m, 2H), 2.20–2.03 (m, 2H), 1.90–1.79 (m, 8H), 1.19 (d, J = 7.4 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 171.99, 170.23, 152.63, 151.76, 150.57, 136.33, 135.77, 132.21, 118.44, 112.36, 111.90, 66.96, 66.41, 51.41, 40.81, 40.43, 31.42, 28.53, 28.31, 24.84, 20.78, 11.42. HRMS (ESI) for C₂₇H₃₈O₄N₇ [M + H]⁺: calcd, 524.2980, found, 524.2973. HPLC analysis: MeOH–H₂O (85:15), 10.38 min, 97.46% purity.

5⁴-((2-(Dimethylamino)ethyl)(methyl)amino)-3³-ethyl-11-oxo-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8l)

In a manner similar to that described for 8f, using 23f (89 mg, 0.17 mmol), 8l (16 mg, yield: 17%) was obtained as a yellow–white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.17 (s, 1H), 7.71 (s, 1H), 7.54 (s, 1H), 7.22 (s, 1H), 6.92 (d, J = 7.2 Hz, 1H), 6.80 (d, J = 8.4 Hz, 1H), 6.40 (dd, J = 8.4, 2.3 Hz, 1H), 4.00–3.88 (m, 2H), 3.76–3.73 (m, 1H), 3.60 (s, 2H), 3.49 (s, 2H), 3.04 (d, J = 6.2 Hz, 2H), 2.69 (s, 3H), 2.61–2.55 (m, 2H), 2.37 (t, J = 7.2 Hz, 2H), 2.13 (s, 6H), 2.09–2.01 (m, 2H), 1.96–1.87 (m, 8H), 1.19 (t, J = 7.4 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 172.17, 170.26, 152.63, 151.96, 150.57, 136.54, 135.23, 132.16, 119.68, 111.84, 66.72, 57.26, 53.12, 45.84, 40.76, 40.30, 31.72, 31.59, 30.27, 24.83, 20.83, 11.40. HRMS (ESI) for C₂₈H₄₃O₃N₈ [M + H]⁺: calcd, 539.3453, found, 539.3436. HPLC analysis: MeOH–H₂O (90:10), 6.80 min, 99.09% purity.

3³-Ethyl-5⁴-(4-(oxetan-3-yl)piperazin-1-yl)-11-oxo-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8m)

In a manner similar to that described for 8f, using 23g (103 mg, 0.18 mmol), 8m (16 mg, yield: 15%) was obtained as a yellow–white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.11 (s, 1H), 7.67 (s, 1H), 7.53 (d, J = 2.5 Hz, 1H), 7.21 (d, J = 2.4 Hz, 1H), 6.94 (d, J = 6.9 Hz, 1H), 6.79 (d, J = 8.4 Hz, 1H), 6.43 (dd, J = 8.4, 2.1 Hz, 1H), 4.56 (t, J = 6.5 Hz, 2H), 4.47 (t, J = 6.0 Hz, 2H), 3.92–3.88 (m, 2H), 3.78–3.75 (m, 1H), 3.61 (s, 2H), 3.49–3.43 (m, 1H), 3.00–2.96 (m, 4H), 2.60–2.56 (m, 2H), 2.55 (s, 4H), 2.40 (s, 4H), 2.03–1.81 (m, 8H), 1.19 (t, J = 7.4 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 172.02, 170.23, 152.64, 151.65, 150.57, 136.47, 135.54, 132.20, 130.12, 118.43, 112.33, 111.85, 74.94, 66.25, 59.05, 50.43, 49.90, 40.78, 31.74, 31.25, 27.38, 24.84, 20.79, 11.43. HRMS (ESI) for C₃₀H₄₃O₄N₈ [M + H]⁺: calcd, 579.3402, found, 579.3392. HPLC analysis: MeOH–H₂O (85:15), 10.85 min, 97.00% purity.

3³-Ethyl-5⁴-(4-(4-methylpiperazin-1-yl)piperidin-1-yl)-11-oxo-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8n)

In a manner similar to that described for 8f, using 23h (95 mg, 0.16 mmol), 8n (14 mg, yield: 14%) was obtained as a yellow–white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.11 (s, 1H), 7.67 (s, 1H), 7.53 (d, J = 2.7 Hz, 1H), 7.21 (d, J = 2.0 Hz, 1H), 6.97 (d, J = 7.2 Hz, 1H), 6.78 (d, J = 8.5 Hz, 1H), 6.41 (dd, J = 8.4, 1.9 Hz, 1H), 4.03 (q, J = 7.1 Hz, 2H), 3.94–3.89 (m, 1H), 3.77–3.61 (m, 4H), 3.28 (s, 4H), 2.77 (s, 6H), 2.60–2.55 (m, 2H), 2.45 (s, 4H), 2.24–2.16 (m, 1H), 2.08–2.01 (m, 2H), 1.99 (s, 3H), 1.91–1.79 (m, 8H), 1.62–1.57 (m, 2H), 1.19 (t, J = 3.7 Hz, 3H). ¹³C NMR (151 MHz, Chloroform-d) δ 172.44, 170.25, 152.19, 151.40, 150.96, 137.12, 135.48, 130.48, 124.38, 118.48, 113.42, 112.85, 66.38, 62.02, 55.03, 51.28, 50.49, 48.76, 45.70, 43.19, 34.45, 31.93, 28.76, 28.47, 24.65, 22.70, 14.13. HRMS (ESI) for C₃₃H₅₀O₃N₉ [M + H]⁺: calcd, 620.4031, found, 620.4015. HPLC analysis: MeOH–H₂O (85:15), 15.01 min, 95.08% purity.

3³-Ethyl-5⁴-morpholino-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8o)

TFA (3 mL) was added to a suspension of 30a (200 mg, 0.29 mmol) in DCM (10 mL). The resulting solution was stirred for 2 h at room temperature, then the mixture was washed with saturated sodium bicarbonate and extracted with DCM (3 × 20 mL) and the organic layers were dried over Na₂SO₄ and concentrated in vacuo to give a yellow solid, which was used without any further purification. The crude product was dissolved in DMF (60 mL) and then K₂CO₃ (60 mg, 0.44 mmol) was added. The mixture was then heated at 65 °C for 4 h. Then the reaction solution was concentrated and purified to give product 8o as a yellow solid (30 mg, yield of two steps: 20%). ¹H NMR (400 MHz, DMSO-d₆) δ 11.35 (s, 1H), 7.99 (s, 1H), 7.56 (s, 1H), 7.25 (s, 1H), 7.17 (d, J = 7.6 Hz, 1H), 6.91 (d, J = 8.4 Hz, 1H), 6.58 (d, J = 7.9 Hz, 1H), 4.27 (s, 2H), 3.99 (s, 1H), 3.73 (s, 4H), 3.27–3.17 (m, 2H), 3.05 (s, 2H), 2.94–2.86 (m, 6H), 2.56 (q, J = 7.5 Hz, 2H), 2.02–1.87 (m, 4H), 1.70 (s, 6H), 1.18 (t, J = 7.3 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.14, 153.65, 152.22, 151.58, 139.02, 135.08, 131.48, 118.34, 113.18, 111.40, 69.83, 67.03, 57.32, 56.09, 52.75, 51.30, 31.76, 30.85, 30.29, 29.46, 28.61, 24.77, 11.54. HRMS (ESI) for C₂₇H₄₀O₃N₇ [M + H]⁺: calcd, 510.3187, found, 510.3179. HPLC analysis: MeOH–H₂O (80:20), 7.51 min, 97.06% purity.

(1³R)-3³-Ethyl-5⁴-morpholino-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(3,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8p)

In a manner similar to that described for 8o, using 30b (150 mg, 0.21 mmol), 8p (32 mg, yield of two steps: 29%) was obtained as a yellow–white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.19 (s, 1H), 7.89 (s, 1H), 7.52 (s, 1H), 7.20 (s, 1H), 6.79 (d, J = 8.4 Hz, 1H), 6.63 (d, J = 8.8 Hz, 1H), 6.42 (d, J = 6.5 Hz, 1H), 4.26 (s, 1H), 4.16–4.11 (m, 1H), 4.03–3.98 (m, 1H), 3.71 (t, J = 4.6 Hz, 4H), 3.06–3.03 (m, 1H), 2.99–2.83 (m, 5H), 2.59–2.53 (m, 2H), 2.41–2.25 (m, 2H), 2.03–1.84 (m, 2H), 1.70–1.67 (m, 4H), 1.54–1.43 (m, 6H), 1.18 (t, J = 7.3 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.18, 152.29, 150.71, 136.19, 136.00, 131.86, 118.49, 111.90, 111.84, 106.12, 68.86, 66.91, 59.69, 56.68, 52.53, 51.35, 47.54, 30.78, 28.61, 25.17, 24.80, 23.53, 11.44. HRMS (ESI) for C₂₇H₄₀O₃N₇ [M + H]⁺: calcd, 510.3187, found, 510.3180. HPLC analysis: MeOH–H₂O (80:20), 7.17 min, 100% purity.

(1³S)-3³-Ethyl-5⁴-morpholino-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(3,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8q)

In a manner similar to that described for 8o, using 30c (160 mg, 0.23 mmol), 8q (35 mg, yield of two steps: 30%) was obtained as a yellow–white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.19 (s, 1H), 7.89 (d, J = 2.4 Hz, 1H), 7.52 (d, J = 3.4 Hz, 1H), 7.20 (d, J = 3.3 Hz, 1H), 6.79 (d, J = 8.4 Hz, 1H), 6.62 (d, J = 8.8 Hz, 1H), 6.42 (dd, J = 8.4, 2.3 Hz, 1H), 4.27–4.24 (m, 1H), 4.16–4.11 (m, 1H), 4.03–3.97 (m, 1H), 3.71 (t, J = 4.6 Hz, 4H), 3.04 (dd, J = 10.3, 4.1 Hz, 1H), 2.98–2.83 (m, 5H), 2.59–2.53 (m, 2H), 2.40–2.36 (m, 1H), 2.28–2.24 (m, 1H), 1.93–1.86 (m, 1H), 1.81–1.67 (m, 6H), 1.55–1.40 (m, 5H), 1.17 (t, J = 7.4 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.18, 152.30, 152.28, 150.72, 136.17, 136.02, 131.82, 118.47, 111.87, 106.11, 68.86, 66.92, 59.72, 56.69, 52.54, 51.35, 47.55, 42.56, 30.79, 28.62, 25.18, 25.14, 24.81, 23.54, 11.45. HRMS (ESI) for C₂₇H₄₀O₃N₇ [M + H]⁺: calcd, 510.3187, found, 510.3179. HPLC analysis: MeOH–H₂O (80:20), 8.85 min, 98.21% purity.

(1³S)-3³-Ethyl-5⁴-morpholino-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(3,1)-pyrrolidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8r)

In a manner similar to that described for 8o, using 30d (130 mg, 0.19 mmol), 8r (29 mg, yield of two steps: 30%) was obtained as a yellow solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.34 (s, 1H), 8.07 (s, 1H), 7.56 (d, J = 3.4 Hz, 1H), 7.24 (d, J = 3.3 Hz, 1H), 6.92 (d, J = 7.2 Hz, 1H), 6.82 (d, J = 8.5 Hz, 1H), 6.48 (dd, J = 8.5, 2.4 Hz, 1H), 4.51–4.41 (m, 1H), 4.16–4.11 (m, 1H), 3.89–3.83 (m, 1H), 3.71–3.64 (m, 5H), 3.07–3.02 (m, 2H), 2.97 (d, J = 9.0 Hz, 1H), 2.83–2.79 (m, 2H), 2.61–2.56 (m, 2H), 2.53 (s, 1H), 2.36–2.29 (m, 2H), 2.15–2.00 (m, 2H), 1.94–1.81 (m, 2H), 1.70–1.60 (m, 2H), 1.48–1.44 (m, 3H), 1.18 (t, J = 7.3 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.24, 153.09, 152.10, 150.73, 137.23, 136.36, 131.98, 118.85, 113.07, 112.12, 108.57, 68.49, 66.99, 58.59, 54.04, 52.81, 51.26, 50.68, 28.81, 28.44, 24.72, 22.39, 11.44. HRMS (ESI) for C₂₆H₃₈O₃N₇ [M + H]⁺: calcd, 496.3031, found, 496.3021. HPLC analysis: MeOH–H₂O (80:20), 8.18 min, 100% purity.

(1³R)-3³-Ethyl-5⁴-morpholino-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(3,1)-pyrrolidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8s)

In a manner similar to that described for 8o, using 30e (150 mg, 0.22 mmol), 8s (38 mg, yield of two steps: 35%) was obtained as a yellow solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.34 (s, 1H), 8.07 (d, J = 2.4 Hz, 1H), 7.56 (d, J = 3.3 Hz, 1H), 7.23 (d, J = 3.4 Hz, 1H), 6.91 (d, J = 7.2 Hz, 1H), 6.82 (d, J = 8.5 Hz, 1H), 6.48 (dd, J = 8.5, 2.4 Hz, 1H), 4.50–4.41 (m, 1H), 4.16–4.11 (m, 1H), 3.88–3.83 (m, 1H), 3.71–3.63 (m, 5H), 3.07–3.02 (m, 2H), 2.94 (d, J = 5.8 Hz, 1H), 2.84–2.79 (m, 2H), 2.61–2.56 (m, 2H), 2.51 (s, 1H), 2.34–2.25 (m, 2H), 2.18–1.99 (m, 2H), 1.91–1.81 (m, 2H), 1.68–1.59 (m, 2H), 1.48–1.39 (m, 3H), 1.18 (t, J = 7.3 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.23, 153.09, 152.11, 150.73, 137.18, 136.40, 131.94, 118.83, 113.00, 112.13, 108.50, 68.49, 66.99, 58.62, 54.03, 52.79, 51.26, 50.71, 28.81, 28.46, 25.49, 24.72, 11.04. HRMS (ESI) for C₂₆H₃₈O₃N₇ [M + H]⁺: calcd, 496.3031, found, 496.3019. HPLC analysis: MeOH–H₂O (80:20), 8.86 min, 99.34% purity.

3³-Ethyl-5⁴-morpholino-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacyclododecaphane-3⁵-carboxamide (8t)

In a manner similar to that described for 8o, using 32a (140 mg, 0.20 mmol), 8t (32 mg, yield of two steps: 30%) was obtained as a yellow solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.28 (s, 1H), 7.83 (s, 1H), 7.58 (s, 1H), 7.28 (s, 1H), 7.06 (d, J = 4.8 Hz, 1H), 6.91 (d, J = 8.4 Hz, 1H), 6.59 (d, J = 7.7 Hz, 1H), 4.16–4.13 (m, 2H), 4.01 (s, 1H), 3.75 (s, 4H), 3.42 (s, 2H), 3.14 (s, 2H), 2.94 (s, 4H), 2.59 (q, J = 7.4 Hz, 2H), 2.14 (s, 2H), 1.94–1.86 (m, 5H), 1.76 (s, 2H), 1.67–1.62 (m, 5H), 1.19 (t, J = 7.3 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.22, 152.48, 152.38, 150.60, 138.44, 136.43, 132.34, 119.47, 115.10, 112.32, 70.24, 67.06, 55.38, 53.12, 51.29, 50.43, 31.62, 30.29, 27.70, 25.40, 24.73, 24.21, 11.42. HRMS (ESI) for C₂₈H₄₂O₃N₇ [M + H]⁺: calcd, 524.3344, found, 524.3332. HPLC analysis: MeOH–H₂O (80:20), 8.25 min, 100% purity.

3³-Ethyl-5⁴-morpholino-6-oxa-2,4-diaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacyclotridecaphane-3⁵-carboxamide (8u)

In a manner similar to that described for 8o, using 32b (130 mg, 0.18 mmol), 8u (20 mg, yield of two steps: 21%) was obtained as a yellow solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.11 (s, 1H), 7.74 (d, J = 2.5 Hz, 1H), 7.57 (d, J = 3.3 Hz, 1H), 7.27 (d, J = 3.3 Hz, 1H), 7.05 (d, J = 7.0 Hz, 1H), 6.89 (d, J = 8.5 Hz, 1H), 6.59 (dd, J = 8.5, 2.5 Hz, 1H), 4.10 (d, J = 5.3 Hz, 2H), 4.01 (s, 1H), 3.75 (t, J = 4.4 Hz, 4H), 3.40 (s, 2H), 3.10 (s, 4H), 2.96 (t, J = 4.3 Hz, 4H), 2.60 (q, J = 7.4 Hz, 2H), 2.12 (d, J = 9.6 Hz, 2H), 1.98–1.89 (m, 2H), 1.77–1.72 (m, 2H), 1.68–1.63 (m, 4H), 1.49 (s, 4H), 1.19 (t, J = 7.3 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.15, 152.32, 150.59, 138.66, 136.03, 132.33, 119.16, 115.47, 112.88, 112.33, 70.70, 67.00, 54.35, 51.34, 50.27, 47.15, 28.96, 28.28, 26.01, 25.80, 24.72, 23.27, 20.10, 11.45. HRMS (ESI) for C₂₉H₄₄O₃N₇ [M + H]⁺: calcd, 538.3500, found, 538.3488. HPLC analysis: MeOH–H₂O (80:20), 8.24 min, 100% purity.

3³-Ethyl-5⁴-morpholino-6,9-dioxa-2,4-diaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8v)

In a manner similar to that described for 8o, using 32c (140 mg, 0.20 mmol), 8v (31 mg, yield of two steps: 31%) was obtained as a yellow solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.32 (s, 1H), 8.27 (d, J = 2.5 Hz, 1H), 7.58 (s, 1H), 7.30–7.25 (m, 2H), 6.94 (d, J = 8.6 Hz, 1H), 6.63 (dd, J = 8.6, 2.6 Hz, 1H), 4.34 (t, J = 3.4 Hz, 2H), 4.21–4.17 (m, 1H), 3.88 (d, J = 4.7 Hz, 2H), 3.79 (d, J = 3.4 Hz, 2H), 3.72 (t, J = 4.5 Hz, 4H), 3.61 (s, 2H), 3.26 (s, 4H), 2.96–2.93 (m, 4H), 2.61–2.56 (m, 2H), 2.06–1.99 (m, 2H), 1.84–1.81 (m, 2H), 1.19 (t, J = 7.4 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.09, 152.24, 151.08, 150.29, 138.32, 136.89, 132.12, 119.81, 115.01, 112.60, 112.41, 70.36, 70.12, 66.97, 64.88, 54.16, 51.16, 51.06, 48.26, 31.75, 31.61, 24.67, 11.45. HRMS (ESI) for C₂₆H₃₈O₄N₇ [M + H]⁺: calcd, 512.2980, found, 512.2968. HPLC analysis: MeOH–H₂O (80:20), 8.81 min, 99.38% purity.

3³-Ethyl-5⁴-morpholino-6-oxo-2,4,7-triaza-3(2,6)-pyrazina-1(4,1)-piperidina-5(1,3)-benzenacycloundecaphane-3⁵-carboxamide (8w)

In a manner similar to that described for 8a, using 42 (130 mg, 0.25 mmol), 8w (16 mg, yield of three steps: 12%) was obtained as a yellow–white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 11.09 (s, 1H), 8.83 (t, J = 5.7 Hz, 1H), 8.30 (d, J = 2.6 Hz, 1H), 7.57 (d, J = 3.3 Hz, 1H), 7.27 (d, J = 3.0 Hz, 1H), 7.19 (d, J = 7.9 Hz, 1H), 7.11 (d, J = 8.6 Hz, 1H), 7.01 (dd, J = 8.6, 2.7 Hz, 1H), 3.92 (s, 1H), 3.72 (t, J = 4.5 Hz, 4H), 3.14 (s, 2H), 3.01 (s, 2H), 2.89 (s, 2H), 2.85 (t, J = 4.5 Hz, 4H), 2.57 (q, J = 7.5 Hz, 2H), 2.03–1.97 (m, 2H), 1.92–1.89 (m, 2H), 1.83–1.76 (m, 6H), 1.17 (d, J = 7.3 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆) δ 170.08, 168.82, 152.13, 150.22, 145.13, 136.05, 132.65, 123.32, 121.43, 120.55, 112.32, 67.38, 53.83, 53.18, 49.91, 40.53, 30.30, 28.10, 24.67, 22.56, 14.43. HRMS (ESI) for C₂₇H₃₉O₃N₈ [M + H]⁺: calcd, 523.3140, found, 523.3130. HPLC analysis: MeOH–H₂O (70:30), 13.12 min, 98.32% purity.

Molecular Modeling Studies

Method A: Protein structures were obtained from the Protein Data Bank (http://www.rcsb.org/pdb). The X-ray crystal structure of FLT3 (PDB ID 6JQR) was chosen for the molecular docking analysis in Maestro and the FLT3 mutant form (F691L) was modeled from the determined X-ray crystallographic structure in Maestro. The proteins were prepared by the Protein Preparation Wizard (Schrödinger, LLC, New York, NY, 2019). All compounds were prepared by Ligprep, and molecular docking was performed by Glide with standard precision. Method B: Metadynamics is a popular enhanced sampling algorithm employed for the calculation of the free energy landscape of rare events through molecular dynamics simulations.^34,35 In this context, Desmond was used to conduct metadynamics simulations to explore the conformational space of compound 7. The dihedral angle of the CH₂NH moiety in 7 was employed as the collective variable. The missing dihedral angle parameters for the ligand were supplemented using the Force Field Builder. Both the protein and ligand were modeled using the OPLS4 force field, while the SPC water model was utilized as the solvent. The Gaussian height was set to 0.005 kcal/mol, with a biasing potential added every 1 ps during the simulation. Other parameters were maintained at their default settings. The metadynamics simulation was run for a duration of 200 ns.

Protein Preparation and Structure Determination

Protein expression and purification were performed as described previously with a few modifications.³⁶ The deoxyribonucleic acid (DNA) sequence encoding the kinase domain of FLT3 was cloned into pFastBacHT-A with codon optimization, which is from H564 to V958 with deletion from S711 to H761. It fuses a 6× His tag at the N-terminus with a TEV protease cleavage site. Recombinant FLT3 was expressed in the Sf9 cells. The proteins were first purified using nickel beads (500 mM NaCl, 20 mM Tris-HCl pH 8.0, 10% glycerol). The eluted product was then cleaved with TEV overnight at 4 °C to remove the His tag. After cleavage, anion exchange chromatography and size-exclusion chromatography were performed for further polishing. Purified protein was concentrated to no less than 5 mg/mL in 50 mM Tris (pH 8.0), 100 mM NaCl, 5% glycerol, and 1 mM dithiothreitol (DTT). The final protein was stored at −80 °C after flash freezing in liquid nitrogen. For crystallization, the crystals of Apo FLT3 were grown by the hanging-drop vapor diffusion method at 18 °C with the reservoir solution (0.2 M Li₂SO₄, 0.1 M CAPS (pH 10.0), 1.2 M NaH₂PO₄, 1.0 M K₂HPO₄). Typically, the crystals took about 1 week to appear and grow as hexagonal bipyramids. Crystal was soaked overnight with the reservoir solution containing 0.5 mM compound. All crystals were cryoprotected by supplementing the reservoir buffer with 25% (v/v) glycerol and 0.5 mM compound. All data sets were collected on beamline BL02Ul at the Shanghai Synchrotron Radiation Facility. The data were processed using the HKL3000 package.³⁷ The structure was determined by molecular replacement using the program Phaser using the FLT3 structure (PDB code: 6JQR) as the searching model.³⁸ The refinement and model building were performed using the program WinCoot and Phenix.^39,40 The statistics of data collection and structure refinement are listed in Table S1.

Enzymatic Activity Assay In Vitro

Method A: Inhibition activities of target compounds against FLT3^D835Y were determined using the LanthaScreen Eu Kinase Binding Assay systems according to the manufacturer’s instructions (Invitrogen). The reactions were carried out in 384-well plates in a 5 μL reaction volume with appropriate amounts of kinase/antibody mixture. Various concentrations of the test compounds were added, and then, after incubation, 5 μL tracer was added for additional 1 h at room temperature incubation, and the fluorescence signal was detected by a multilabel microplate detector. The excitation wavelength was 340 nm, and the emission wavelength was 665 and 615 nm. The data were analyzed using Graphpad Prism5 (Graphpad Software, Inc.). Method B: Inhibition activities of compounds 8o, 8v, and gilteritinib against FLT3^D835Y/F691L were determined using the fluorescence resonance energy transfer (FRET)-based Z′-Lyte assay systems according to the manufacturer’s instructions (Life Technologies, Carlsbad, CA). The reactions were carried out in 384-well plates in a 10 μL reaction volume with appropriate amounts of kinases in 50 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) (pH 7.5), 5 mM MgCl₂, 2 mM MnCl₂, 1 mM ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA), and 0.01% Brij-35. The reactions were incubated for 1 h at room temperature in the presence of 2 μM of Tyr 02 peptide substrate with 250 μM ATP for kinase FLT3^D835Y/F691L and in the presence of various concentrations of the compounds, and then, after incubation, 5 μL of development reagent was added for additional 1 h at room temperature, followed by the addition of 5 μL of stop solution. The fluorescence signal ratio of 445 nm (coumarin)/520 nm (fluorescein) was examined with an EnVision Multilabel Reader (PerkinElmer, Inc.). The data were analyzed using Graphpad Prism5 (Graphpad Software, Inc.). The inhibitory activities of 8v against VEGFR1/2/3, C-KIT, PDGFRα/β, and RET were determined using method B described above.

Reagents and Antibodies

Compounds were dissolved in dimethyl sulfoxide (DMSO, Sigma-Aldrich) at a concentration of 10 mM and stored at −20 °C. The primary antibodies against FLT3 (3462), Phospho-FLT3 (Tyr589/591) (3464), Stat5 (94205), Phospho-Stat5 (Tyr694) (4322), p44/42 MAPK (Erk1/2) (4695), Phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) (4370), Akt (pan) (4691), Phospho-Akt (Ser473) (4060), Caspase-9 (9508), β-Actin (4967), CDK4 (12790), CDK6 (13331), Cyclin D3 (2936), and anti-rabbit or anti-mouse IgG horseradish peroxidase (HRP)-linked secondary antibodies were purchased from Cell Signaling Technology (CST, Boston, MA). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH, AG019) was purchased from Beyotime Biotechnology (Beyotime, Shanghai, China).

Cell Culture

The cell lines used in this study were obtained from the Cell Resources Center of Shanghai Academy of Life Sciences or American Type Culture Collection (ATCC). MV4–11 cells were grown in IMDM (Gibco, Thermo, Inc.), while HEK-293 cells were grown in RPMI 1640 (Gibco, Thermo, Inc.) supplemented with 10% fetal bovine serum (FBS) (Excel Scientific) and 1% penicillin–streptomycin (Gibco, Thermo, Inc.). All cells were maintained at 37 °C in a humidified 5% CO₂ atmosphere. The cells were passaged for less than 3 months before renewal from frozen, early passage stocks obtained from the indicated sources.

Construction of Ba/F3 Model and Cell Culture

The interleukin-3 (IL-3)-dependent murine pro-B cell line (Ba/F3) was purchased from Deutsche Sammlung von Mikroorganismen and Zellkulturen (DSMZ). The Ba/F3 cell lines harboring various on-target point mutations in FLT3-ITD were established through electroporation, using an Amaxa Cell Line Nucleofector Kit V (Lonza, Cologne, Germany). The MIGR1 vector plasmid containing FLT3-ITD was provided by Dr. Wang Peihong of Guangzhou First People’s Hospital and confirmed by sequencing. Stable strains were selected by the withdrawal of interleukin-3 (IL-3, R&D) after transfection for 48 h. All Ba/F3 stable cell lines were verified by DNA sequencing, protein expression, and antiproliferation against positive drugs. Parental Ba/F3 cells were cultured in RPMI 1640 media supplemented with 10% FBS and 10 ng/mL IL-3, while all FLT3-ITD Ba/F3 stable cell lines were cultured in the same medium without IL-3.

Cell Proliferation Assay

For cell proliferation assay, the Ba/F3-FLT3-ITD model cells were plated at 2000 cells (the other cell lines were plated at 6000–18,000 cells) per well in 96-well plates in a complete medium. After overnight incubation, the cells were treated with gradient concentrations of the test compounds (0.0001–10 μM) and incubated for additional 3 days. Cell viability was assessed using the Cell Counting Kit-8 (CCK8, K1018, APExBIO Technology LLC, Houston) by adding 10 μL of the kit reagent to each well and incubating for 2–4 h. OD450 and OD650 were determined by a BioTeK Synergy H1 Hybrid Reader (BioTek). The IC₅₀ values were calculated by fitting concentration–response curves using GraphPad Prism 8.0.1 software. Each assay was repeated at least three times.

Western Blotting Analysis

Cells were treated with various concentrations of the compound for 4 h at 37 °C. Then, the cells were collected and washed once with ice-cold phosphate-buffered saline (PBS) and lysed in 1× sodium dodecyl sulfate (SDS) sample lysis buffer with protease and phosphatase inhibitors. Cell lysates were loaded and electrophoresed onto 10% SDS–polyacrylamide gel electrophoresis (PAGE) gel; then, the separated proteins were transferred to a poly(vinylidene fluoride) (PVDF) film. The film was blocked with 5% fat-free milk in Tris-buffered saline (TBS) solution containing 0.5% Tween-20 for 1 h at room temperature and then incubated with the corresponding primary antibody (1:1000–1:2000) overnight at 4 °C. After washing with TBST, horseradish peroxidase (HRP)-conjugated secondary antibody was incubated for 1 h. The protein signals were visualized by a StarSignal Plus Chemiluminescent Assay Kit (GenStar, Beijing, China) and detected with an Amersham Imager 600 system (GE, Boston, MA).

Cell Cycle

Cells were seeded in a six-well plate and treated with the test compound at various concentrations for 24 h. The cell cycle was measured using a Cell Cycle Analysis Kit (Sizhengbai Biotech, China) according to the manufacturer’s instructions. Briefly, cells were harvested after treatment, washed twice with precooled PBS, and then fixed in 95% ethanol overnight, and the cell number was adjusted to 1.0 × 10⁶ cells/mL before PI staining. The deoxyribonucleic acid (DNA) content of stained samples was analyzed using a Guava easyCyte flow cytometer (Merck).

Apoptosis Analysis

Cells were treated with indicated concentrations of the test compound or DMSO for 48 h. After incubation, the cells were collected and washed twice with precold PBS. About 1 × 10⁵ cells were resuspended in 100 μL of 1× BD binding buffer solution (#556454, BD) and then stained with 7-ADD (#559925, BD) and Annexin V (#556422, BD) in the dark for 15 min. Finally, 400 μL of 1 × BD binding buffer solution was added to stop dyeing. The cells were then measured on a Guava easyCyte flow cytometer (Merck).

Kinome Profiling

Kinase profiling screening of 8v at 1 μM against a panel of 378 wild-type kinases was conducted at Eurofins Cerep SA.

Pharmacokinetic In Vivo

Sprague–Dawley rats (male, three rats per group) weighing 200 ± 20 g were dosed with the test compound 8v intravenously at 2.0 mg/kg or by oral gavage at 10.0 mg/kg. Compound 8v was dissolved in mixed solvents [5% DMSO, 40% PEG400, and 55% cyclodextrin (20%)]. Blood was collected approximately 0.20 mL via jugular vein at different time points (i.v.: 0.083, 0.25, 0.5, 1, 2, 4, 6, 8, 24 h; p.o.: 0.25, 0.5, 1, 2, 4, 6, 8, 24 h) and then placed in tubes containing K2-EDTA and stored on ice until centrifugation. The blood samples were centrifuged at 6800g for 6 min at 2–8 °C within 1 h after they were collected and stored frozen at approximately −80 °C. An aliquot of 30 μL plasma sample was protein-precipitated with 300 μL of MeOH, which contains 10 ng/mL IS. The mixture was vortexed for 1 min and centrifuged at 18,000g for 7 min. 200 μL of supernatant was transferred to 96-well plates. An aliquot of 10 μL of supernatant was injected for LC-MS/MS analysis, and PK parameters were calculated by using Phoenix WinNonlin7.0 software.

Ba/F3-FLT3-ITD-D835Y Xenograft Model

The in vivo efficacy studies were approved by the Biomedical Model Animal Center (approval number, 2023098) of South China University of Technology, Guangzhou, China. Male nude mice (4–5 weeks old and weighed 17–20 g) were purchased from Gempharmatech Co., Ltd. (Guangdong, China). 5 × 10⁶ Ba/F3-FLT3-ITD-D835Y model cells combined with Matrigel (v/v, 3:1) were injected subcutaneously in the right flank of nude mice. Twenty-four hours after inoculation, the mice were randomized into different treatment groups with a total of 8 mice in the vehicle group and 7 mice in the 8v test group. Compound 8v (10 or 30 mg/kg, qd) or vehicle (0.5% methyl cellulose) was administrated orally from day 6 to day 22 postinoculation. Body weight and tumor size were measured every 2 days. Tumor size was measured with an electronic caliper, and tumor volume was calculated as (length × width²)/2. TGI was determined on day 6 as (1 – RTV_treated/RTV_control) × 100.

Glossary

Abbreviations

D835Y

Asp835 to Tyr835

F691L

Phe691 to Leu691

ALK

anaplastic lymphoma kinase

TRK

tropomyosin receptor kinase

EGFR

epidermal growth factor receptor

C-Met

cellular-mesenchymal epithelial transition factor

IC₅₀

half-maximum inhibitory concentration

PDB

Protein Data Bank

DMF

N,N-diisopropylethylamine

TFA

trifluoroacetic acid

DCM

dichloromethane

THF

tetrahydrofuran

FDPP

pentafluorophenyl diphenylphosphinate

DIPEA

N,N-diisopropylethylamine

DMSO

dimethyl sulfoxide

NMR

nuclear magnetic resonance

HPLC

high-performance liquid chromatography

HRMS

high-resolution mass spectrometry

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsptsci.4c00071.

• Molecular formula strings (XLSX)

• The synthesis methods of macrocycles 8f and 8w and other intermediates; kinase selectivity study of 8v; X-ray data collection of 8v, HPLC and high-resolution mass spectrometry (HRMS) analytics of representative macrocycles 8a–w (PDF)

• Crystallographic data (PDB)

Author Contributions

^⊥ X.Z., Z.C., and M.G. contributed equally to this work.

The authors declare no competing financial interest.