Discovery of 5-trifluoromethyl-2-aminopyrimidine derivatives as potent dual inhibitors of FLT3 and CHK1

Abstract

Here, we discover an FLT3/CHK1 dual inhibitor (30) that exhibits excellent kinase potency and antiproliferative activity against MV4-11 cells. Simultaneously, 30 possesses high selectivity over c-Kit enzyme and low hERG inhibitory ability. Compound 30, meanwhile, overcomes varied resistance in BaF3 cell lines carrying FLT3-TKD and FLT3-ITD mutations. Moreover, 30 demonstrates favorable oral PK properties and kinase selectivity. These conclusions support that compound 30 may be a promising potential FLT3/CHK1 dual agent for further development.

Novel FLT3/CHK1 dual agents, the representative compound 30, with favorable oral PK properties, can overcome multiple FLT3-TKD and FLT3-ITD mutations.

Introduction

Acute myeloid leukemia (AML) is a malignant cancer with an abnormal hematopoietic system, characterized by impaired differentiation and indefinite proliferation of immature cells.^1,2 At present, the “7 + 3” regimen is mainly recognized as the standard treatment method for AML patients, which contains seven continuous days of cytarabine infusion and daily boluses of anthracycline on the first three days. Unfortunately, although this treatment regimen has significant early efficacy, the prognosis of AML patients in the later stage is not satisfactory, with primary refractory diseases and recurrence.^3,4 The five-year survival period for AML patients receiving treatment is roughly 35% in middle aged and young people (age < 60) and less than 15% in elderly people (age > 60).^4–6 Hence, it is still necessary to develop AML treatment strategies and drugs with higher efficacy through molecular and genetic research.

Fms-like tyrosine kinase 3 (FLT3) belongs to the type III receptor tyrosine kinase family (RTK), which plays a notable role in regulating the propagation and hemopoiesis of progenitor cells in medulla ossium.^7,8 FLT3 gene mutation is one of the most common genetic abnormalities in AML, including internal tandem duplication (ITD) as well as missense mutations of the tyrosine kinase domain (TKD), which is closely related to poor prognosis and high relapse risk in patients.^9–11 Subsequently, a large number of FLT3 inhibitors (FLT3i) have been exploited for the treatment of AML, in response to the high correlation between FLT3 and AML.¹² The first-generation FLT3i were multitarget and nonselective inhibitors, including midostaurin,⁹ which possibly caused off-target side effects. The second-generation FLT3i had better specificity and less toxicity, such as gilteritinib¹³ and quizartinib¹⁴ (Fig. 1). Regretfully, acquired target mutations (acquired resistance) and abnormal activation of off-target pathways (adaptive resistance) pose substantial limitations to the use of FLT3 inhibitors in AML control.¹⁵ As a result, targeting FLT3 and related drug resistance mechanisms simultaneously may be a more effective treatment measure for AML.

Fig. 1. The structures of clinical FLT3i.

Checkpoint kinase 1 (CHK1) is a principal regulator of the DNA damage response pathway (DDR).¹⁶ The inhibition of CHK1 is going to force tumor cells into mitosis, carrying the wrong genetic material, and thereby induce cell apoptosis.¹⁷ Our earlier discovery identified that inhibiting FLT3 and CHK1 together is able to restore the p53 pathway activity, which is inactivated by FLT3i, and then overcome FLT3i resistance.¹⁸ Therefore, the concomitant targeting of FLT3 and CHK1 proteins achieves a better AML curative effect probably.

On the basis of our previous investigation, compound A with a 5-trifluoromethyl-2-aminopyrimidine core was identified as a potent dual FLT3/CHK1 inhibitor (Fig. 2A). We firstly analyzed the possible interactions between compound A and the two proteins through molecular docking. As shown in Fig. 2B and C, compound A built two hydrogen bonds with Glu85 and Cys87 of CHK1 and Cys694 of the FLT3 protein. The primary amine located at the 4-position of pyrimidine had hydrogen bonds with polar residues Glu91 of CHK1 and Arg745 and Asp759 of FLT3, respectively. The carbonyl, located at the 4-position of the phenyl ring, extended into the internal pocket of CHK1 and formed a hydrogen bond with the key residue Lys38. Meanwhile, the carbonyl group remained in the large solvent exposure region in the FLT3 protein, which was composed of residues Lys614, Try696, Gly697 and Asp698.

Fig. 2. The structure and binding model of compound A. (A) The chemical structure of compound A. (B) Docking of A with the CHK1 protein (PDB: 2YM8). (C) Docking of A with the homology model of the FLT3 protein.

Conformational restriction is a common strategy for molecule optimization in medicinal chemistry, which is able to enhance target affinity and kinase selectivity, as well as druggability of the lead compound.^19,20 Based on the results of molecular docking, we focused our attention on the 4-position of the phenyl ring and carried out a series of SAR exploration to obtain a highly efficient FLT3/CHK1 dual inhibitor via the conformational restriction strategy.

Results and discussion

Chemistry

The synthesis of compounds 7–19 and 22–40 is illustrated in Scheme 1. The substitution between starting material 1–5 and corresponding sulfonyl chloride obtained the intermediates 7–18. Compound 6 was oxidized by m-CPBA to generate intermediate 19. The raw material 20 was replaced with tert-butyl (3-aminopropyl)carbamate to gain the intermediate 21. And then 21 was subjected to Buchwald coupling with R²Br followed by deprotection of the Boc group to afford target compounds 22–40.

Scheme 1. Reactions and conditions: a. corresponding sulfonyl chloride, pyridine, N₂, 0 °C-rt, 1–6 h, 44–63%; b. corresponding sulfonyl chloride, TEA, DCM, N₂, 0 °C-rt, 1–3 h, 42–75%; c. m-CPBA, DCM, 0 °C-rt, 3 h, 92%; d. tert-butyl (3-aminopropyl)carbamate, TEA, ACN, reflux, overnight, 75%; e. R²Br, Pd₂(dba)₃, Xantphos, Cs₂CO₃, dioxane, N₂, reflux, overnight; f. TFA, DCM, 0 °C-rt, 1–3 h, 28–45%.

The inhibitory activity of FLT3 and CHK1 kinases

Primarily, we adopted a conformational restriction method to cyclize the 4-position and 3-position on the phenyl ring of lead compound A. The SAR analysis for the R² substituent is shown in Table 1. When the R² was a benzocyclic ketone derivative (22–28), the compound with a six membered ring on the right side of the phenyl ring had better inhibitory activities against FLT3-D835Y and CHK1 than the five membered ring derivatives (22vs.23). In addition, the introduction of heteroatoms within the aliphatic ring is unconducive to the improvement of FLT3-D835Y affinity (24vs.22, 25vs.23). If the oxygen atom of carbonyl was placed inside the aliphatic ring as ether, it would cause a slight decrease in the potency of FLT3-D835Y and CHK1 (24vs.26). As ketone carbonyl was replaced with a sulfonyl-substituted N atom (29–40), there was no obvious difference in the inhibitory abilities against FLT3-D835Y and CHK1, whether the R² group was the (iso)tetrahydroquinoline ring or the (iso)indoline ring (29vs.30vs.31vs.32). Considering that the six membered ring had better ring stability and drug-likeness, we chose the (iso)tetrahydroquinoline scaffold for further optimization. Introducing heteroatoms into the tetrahydroquinoline ring triggered a dramatic reduction in FLT3-D835Y and CHK1 kinase activities (33vs.29). Moreover, the substituent size of sulfonyl groups on the N atom of (iso)tetrahydroquinoline significantly affected the inhibitory potency of CHK1. Increasing the substituent volume was able to form a severe decrease in CHK1 kinase activity (39, 36, 34vs.29, 40, 38, 37, 35vs.30). However, replacing the methylsulfonyl group with an ethylsulfonyl or a cyclopropyl group would result in a 2–3 fold improvement of the FLT3-D835Y inhibitory effect (34vs.29, 36vs.29, 35vs.30, 37vs.30). In contrast, introducing a larger phenylsulfonyl group could reduce the inhibitory activity of FLT3-D835Y slightly (39vs.29, 40vs.30).

The SAR analysis of target compounds 22–40.

Cpd	R2	IC50a (nM)
		CHK1	FLT3-WT	FLT-D835Y
22		38.57 ± 1.03	27.66 ± 6.60	23.22 ± 0.81
23		143.60 ± 1.41	57.53 ± 6.78	69.13 ± 9.38
24		62.29 ± 6.14	83.91 ± 2.65	80.80 ± 0.30
25		93.65 ± 9.27	81.79 ± 3.39	129.05 ± 30.48
26		179.10 ± 13.29	98.49 ± 25.47	169.65 ± 44.05
27		236.60 ± 14.99	179.60 ± 22.34	197.60 ± 24.32
28		104.11 ± 22.33	19.29 ± 1.22	50.25 ± 1.91
29		28.93 ± 3.44	8.86 ± 2.29	31.30 ± 2.84
30		25.63 ± 2.30	16.39 ± 2.02	22.80 ± 5.00
31		32.53 ± 0.40	19.10 ± 1.61	29.83 ± 1.84
32		35.19 ± 0.41	16.86 ± 0.28	30.12 ± 0.04
33		183.35 ± 31.89	84.70 ± 0.51	105.92 ± 21.33
34		71.07 ± 2.92	8.10 ± 0.63	11.05 ± 0.30
35		88.44 ± 11.17	13.94 ± 2.67	13.82 ± 0.77
36		94.83 ± 6.50	11.23 ± 0.80	11.57 ± 1.77
37		66.91 ± 6.60	11.58 ± 1.98	10.71 ± 0.64
38		100.27 ± 10.22	28.98 ± 1.17	37.03 ± 2.77
39		166.50 ± 14.00	37.04 ± 5.23	51.62 ± 10.52
40		132.65 ± 10.82	31.32 ± 0.05	55.47 ± 5.64
CCT-245737	—	25.71 ± 0.33	909.75 ± 208.24	1968.50 ± 71.42
Quizartinib	—	>10 μM	19.50 ± 1.36	6937.00 ± 125.87

^aIC₅₀ values are shown as an average of three independent determinations.

In vitro antiproliferation activity of MV4-11 cells

On the basis of the potent CHK1 and FLT3-D835Y inhibitory activities, we evaluated the antiproliferation activities of ten representative compounds against MV4-11 cells. As shown in Table 2, all compounds displayed moderate to excellent antiproliferative activities against MV4-11 cells. Importantly, five compounds (22, 29, 30, 31, 32) showed a powerful inhibitory effect against MV4-11 cells with IC₅₀ values <4 nM.

Antiproliferative activity of selected compounds against MV4-11 cells.

Cpd	MV4-11 IC50a (nM)	Cpd	MV4-11 IC50a (nM)
22	2.88 ± 0.51	34	13.12 ± 1.22
24	12.89 ± 2.51	35	4.14 ± 0.35
29	3.26 ± 0.57	36	12.06 ± 0.44
30	2.96 ± 0.35	37	4.10 ± 0.29
31	3.15 ± 0.27	CCT-245737	211.13 ± 27.93
32	2.38 ± 0.37	Quizartinib	0.31 ± 0.04

^aIC₅₀ values are shown as an average of three independent determinations.

c-Kit inhibitory activity and hERG inhibitory potency

In order to develop efficient and safe FLT3/CHK1 inhibitors, five representative compounds were chosen for further evaluation of their kinase selectivity over c-Kit and inhibitory potency of the hERG ion channel (Table 3). All of the tested compounds exhibited strong selectivity over the c-Kit kinase (ratio of c-Kit/FLT3-D835Y ≥ 313-fold). In particular, compounds 30 and 32 showed excellent selective potency over c-Kit (c-Kit/FLT3-D835Y = 460-fold and 708-fold, respectively) and the lowest hERG inhibition rate at a concentration of 10 μM (36% and 23%, respectively).

c-Kit selectivity and hERG affinity of tested compounds.

Cpd	c-Kit IC50a (nM)	Fold (c-Kit/FLT3-D835Y)	hERGb (inhibition%@10 μM)
22	10887.50 ± 13.44	468	76.11%
29	9810.00 ± 299.81	313	64.21%
30	10497.50 ± 149.20	460	36.07%
31	12269.00 ± 1463.71	411	39.51%
32	21348.50 ± 4090.61	708	23.19%
Quizartinib	82.63 ± 5.20	—	—

^aIC₅₀ values are shown as an average of three independent determinations.

^bInhibition of hERG ion current in HEK cells overexpressing the hERG ion channel, single determination at a test concentration of 10 μM.

Inhibitory potency of BaF3 cells carrying multiple FLT3 mutations

Furthermore, we assessed the inhibitory potency against BaF3 cells, with a variety of FLT3 mutations, of compounds 30 and 32 (Table 4). Both of them represented superior antiproliferative activity against five tested BaF3 engineering cell lines. In particular, compounds 30 and 32 possessed apparently more potential antiproliferative activities against BaF3-FLT3-F691L, BaF3-FLT3-D835F and BaF3-FLT3-D835V cell lines, compared to Quizartinib.

Inhibitory potency of BaF3 cells with multiple FLT3 mutations of 30 and 32.

Cpd	IC50a (nM)
	BaF3-FLT3-F691L	BaF3-FLT3-D835F	BaF3-FLT3-D835V	BaF3-FLT3-ITD	BaF3-FLT3-ITD/D835Y
30	28.96 ± 1.54	29.30 ± 4.34	26.37 ± 0.84	30.24 ± 5.67	75.81 ± 3.35
32	29.26 ± 9.48	30.77 ± 5.14	27.22 ± 5.79	30.17 ± 5.47	121.60 ± 11.03
Quizartinib	128.80 ± 22.15	152.40 ± 1.93	146.60 ± 6.91	0.38 ± 0.07	25.53 ± 2.04

^aIC₅₀ values are shown as an average of three independent determinations.

In vivo PK parameters of 30 and 32

We then further evaluated the PK parameters of compounds 30 and 32 in ICR mice (Table 5 and Fig. 3). After oral administration of 20 mg kg⁻¹ in mice respectively, the C_max of 30 was determined to be 2213.07 ng mL⁻¹ and the AUC_(0−t) was 2736.58 h ng mL⁻¹, which was better than compound 32. The result above indicated that compound 30 had favorable PK properties for oral administration and was worth further development.

PK properties of 30 and 32.

Cpd	30	32
	Po (20 mg kg−1)	Po (20 mg kg−1)
T max (h)	0.50	0.50
T 1/2 (h)	1.10	1.20
C max (ng mL−1)	2213.07	949.05
AUC(0−t) (h ng mL−1)	2736.58	1657.82
AUC(0−∞) (h ng mL−1)	2753.85	1677.76
V d (mL kg−1)	11516.64	20616.34
CL (mL h−1 kg−1)	7262.57	11920.63

Fig. 3. PK study of compounds 30 and 32 in ICR mice.

Kinase selectivity of compound 30

As the most promising compound, 30 was determined against twelve kinases from six enzyme subfamilies (Fig. 4). Compound 30 exhibited efficient kinase selective abilities of FLT3-WT, FLT3-D835Y and CHK1 (IC₅₀ ≤ 25 nM). Meanwhile, compound 30 behaved only a little inhibitory potency of IRAK4, P70S6K, CDK2 and Aurora A, with IC₅₀ values of 864.60 nM, 834.85 nM, 833.50 nM and 1863.50 nM, respectively. In addition, compound 30 had no affinity with other kinases. In conclusion, compound 30 was a potential dual FLT3/CHK1 agent, with a favorable kinase selectivity.

Fig. 4. Profile of kinase selectivity of compound 30.

FLT3/CHK1-related signal pathway analysis of compound 30

To further identify the targeted effect of compound 30, we assessed the phosphorylation of FLT3/CHK1 and its downstream effectors (Fig. 5). Compound 30 could potently inhibit the phosphorylation of FLT3 and its main downstream effectors, STAT5 (Tyr694), AKT (Ser473) and ERK (Tyr204) in a dose-dependent manner. In particular, p-FLT3, p-STAT5 and p-ERK were inhibited almost completely at the concentration of 100 nM. In addition, compound 30 could decrease the p-S296-CHK1 level, which were the biomarkers of inhibiting the ATR-CHK1 pathway. Meanwhile, compound 30 could also downregulate the c-Myc protein expression level, which was used as an indicator for the efficacy of CHK1i and FLT3i against AML. Our previous research has evidenced that the inhibition of CHK1 could restore the p53 expression and overcome adaptive resistance in FLT3-ITD AML cells. In accordance with this, 30 could dose-dependently upregulate the p53 and downstream p21 protein levels.

Fig. 5. Compound 30 blocked FLT3 and CHK1 signaling in MV4-11 cells.

Molecular docking of compound 30 with FLT3 and CHK1

To further verify the interaction model of compound 30, we performed molecular docking of compound 30 with FLT3 and CHK1 proteins. The molecular docking result is shown in Fig. 6. The 1-N and 2-NH on the pyrimidine core had two important hydrogen bonds with the hinge region residue Cys87 of CHK1 or Cys694 of FLT3, respectively. The primary amine on the 4-position of the pyrimidine ring formed two hydrogen bond interactions with the polar residues Arg745 as well as Asp759 of FLT3. Besides, there was a vital hydrogen bond between the Lys614 of FLT3 and the methylsulfonyl group on the quinoline ring. In addition, the primary amine on the 4-position of the pyrimidine ring engaged in two hydrogen bonds with two acidic residues Glu91 and Glu134 in the ribose region of the CHK1 enzyme.

Fig. 6. Docking model of compound 30. (A) Docking of 30 with the CHK1 protein (PDB: 2YM8). (B) Docking of 30 with the homology model of the FLT3 protein.

Conclusions

Originating from compound A, we design a series of FLT3/CHK1 dual inhibitors, with a 5-trifluoromethyl-2-aminopyrimidine scaffold, using a conformational restriction strategy. Among them, compound 30 exhibits excellent FLT3-D835Y and CHK1 enzyme inhibitory activities. Furthermore, it is demonstrated that compound 30 has high selectivity over c-Kit as well as inferior hERG affinity. Simultaneously, 30 is discovered with good antiproliferative abilities on MV4-11 cells and various BaF3 cells carrying diversified FLT3 mutations. Moreover, compound 30 is equipped with good oral PK properties and favorable kinase selectivity. In summary, compound 30 is an efficient and low toxicity FLT3/CHK1 dual target inhibitor, which deserves further pharmacodynamic research.

Chemistry

All chemical reagents and solvents were purchased from commercial suppliers and used without further purification. ¹H and ¹³C NMR spectra were obtained from a Brüker 500 MHz verispectrometer (Brüker Bioscience, Billerica, USA). The NMR solvent was CDCl₃, CD₃OD or DMSO-d₆ with tetramethylsilane (TMS) as an internal standard. LC-MS analysis was performed on a Shimadzu LCMS-2020 mass spectrometer (mobile phase A: methanol and mobile phase B: water with 0.1% formic acid). HRMS spectra were recorded on an Agilent Technologies 6224 time-of-flight LC/MS. HPLC analysis was performed on an Agilent 1260 Series system with a COSMOSIL 5C18-AR-II (4.6 mm × 250 mm) column and detected at 254 nm UV wavelength, with a flow rate of 1.0 mL min⁻¹. The detailed HPLC methods could be referred to the ESI.†

General procedure A for the synthesis of intermediates (7–12)

Compounds 1–3 (1.00 mmol) were dissolved in pyridine (2.0 mL). And then the corresponding sulfonyl chloride (1.50–2.00 mmol) was added to the solution with an atmosphere of 0 °C and nitrogen. The mixture was stirred at room temperature for 1–3 h. The solvent was evaporated, and the residue was purified by column chromatography to afford the products 7–12.

6-Bromo-1-(methylsulfonyl)-1,2,3,4-tetrahydroquinoline (7)

White solid; yield: 63%; ¹H NMR (500 MHz, CDCl₃) δ 7.60–7.58 (m, Ar–H, 1H), 7.30–7.26 (m, Ar–H × 2, 2H), 3.81–3.79 (m, CH₂, 2H), 2.90 (s, CH₃, 3H), 2.83 (t, J = 6.5 Hz, CH₂, 2H), 2.02–1.97 (m, CH₂, 2H); LC-MS (ESI): m/z calcd. for C₁₀H₁₂BrNO₂S: 288.98 [M + H]⁺; found: 289.90, 291.91.

5-Bromo-1-(methylsulfonyl)indoline (8)

White solid; yield: 58%; ¹H NMR (500 MHz, DMSO-d₆) δ 7.48–7.47 (m, Ar–H, 1H), 7.38–7.36 (m, Ar–H, 1H), 7.18 (d, J = 8.5 Hz, Ar–H, 1H), 3.94 (t, J = 8.5 Hz, CH₂, 2H), 3.12 (t, J = 8.5 Hz, CH₂, 2H), 3.01 (s, CH₃, 3H); LC-MS (ESI): m/z calcd. for C₉H₁₀BrNO₂S: 274.96 [M + H]⁺; found: 276.11, 278.17.

7-Bromo-4-(methylsulfonyl)-3,4-dihydro-2H-benzo[b][1,4]oxazine (9)

White solid; yield: 46%; ¹H NMR (500 MHz, DMSO-d₆) δ 7.37 (d, J = 8.0 Hz, Ar–H, 1H), 7.33 (dd, J = 8.0, 2.5 Hz, Ar–H, 1H), 7.30 (d, J = 2.5 Hz, Ar–H, 1H), 4.11–4.06 (m, CH₂, 2H), 3.66–3.64 (m, CH₂, 2H), 3.03 (s, CH₃, 3H); LC-MS (ESI†): m/z calcd. for C₉H₁₀BrNO₃S: 290.96 [M + H]⁺; found: 292.11, 294.15.

6-Bromo-1-(ethylsulfonyl)-1,2,3,4-tetrahydroquinoline (10)

Yellowish solid; yield: 60%; ¹H NMR (500 MHz, DMSO-d₆) δ 7.26–7.22 (m, Ar–H × 2, 2H), 7.03–7.01 (m, Ar–H, 1H), 3.77–3.75 (m, CH₂, 2H), 3.52 (q, J = 6.0 Hz, CH₂, 2H), 2.71 (t, J = 8.5 Hz, CH₂, 2H), 2.01–1.96 (m, CH₂, 2H), 1.66 (t, J = 6.5 Hz, CH₃, 3H); LC-MS (ESI†): m/z calcd. for C₁₁H₁₄BrNO₂S: 302.99 [M + H]⁺; found: 304.37, 306.33.

6-Bromo-1-(cyclopropylsulfonyl)-1,2,3,4-tetrahydroquinoline (11)

Yellowish oil; yield: 44%; ¹H NMR (500 MHz, DMSO-d₆) δ 7.49 (d, J = 8.5 Hz, Ar–H, 1H), 7.39 (d, J = 2.5 Hz, Ar–H, 1H), 7.32 (dd, J = 8.5, 2.0 Hz, Ar–H, 1H), 3.70–3.68 (m, CH₂, 2H), 2.83 (t, J = 7.0 Hz, CH₂, 2H), 2.77–2.72 (m, CH, 1H), 1.98–1.93 (m, CH₂, 2H), 0.97–0.92 (m, CH₂ × 2, 4H); LC-MS (ESI†): m/z calcd. for C₁₂H₁₄BrNO₂S: 314.99 [M + H]⁺; found: 317.13, 319.13.

6-Bromo-1-(phenylsulfonyl)-1,2,3,4-tetrahydroquinoline (12)

White solid; yield: 51%; ¹H NMR (500 MHz, DMSO-d₆) δ 7.70–7.64 (m, Ar–H, 1H), 7.63–7.52 (m, Ar–H × 5, 5H), 7.38 (dd, J = 9.0, 2.5 Hz, Ar–H, 1H), 7.31–7.30 (m, Ar–H, 1H), 3.78–3.74 (m, CH₂, 2H), 2.44–2.39 (m, CH₂, 2H), 1.60–1.51 (m, CH₂, 2H); LC-MS (ESI): m/z calcd. for C₁₅H₁₄BrNO₂S: 350.99 [M + H]⁺; found: 352.03, 354.03.

General procedure B for the synthesis of intermediates (13–18)

Compounds 4 and 5 (1.00 mmol) and TEA (5.00 mmol) were dissolved in DCM (3.0 mL). The corresponding sulfonyl chloride (1.50–2.00 mmol) was added to the solution with an atmosphere of 0 °C and nitrogen. The mixture was stirred at room temperature for 1–3 h. The solvent was evaporated, and the residue was purified by column chromatography to afford the products 13–18.

6-Bromo-2-(methylsulfonyl)-1,2,3,4-tetrahydroisoquinoline (13)

White solid; yield: 75%; ¹H NMR (500 MHz, CDCl₃) δ 7.34–7.32 (m, Ar–H × 2, 2H), 6.97 (d, J = 8.5 Hz, Ar–H, 1H), 4.40 (s, CH₂, 2H), 3.54 (t, J = 6.0 Hz, CH₂, 2H), 2.96 (t, J = 6.0 Hz, CH₂, 2H), 2.85 (s, CH₃, 3H); LC-MS (ESI†): m/z calcd. for C₁₀H₁₂BrNO₂S: 288.98 [M + H]⁺; found: 290.17, 292.15.

5-Bromo-2-(methylsulfonyl)isoindoline (14)

Yellowish solid; yield: 70%; ¹H NMR (500 MHz, DMSO-d₆) δ 7.58 (d, J = 2.0 Hz, Ar–H, 1H), 7.51 (dd, J = 8.0, 2.0 Hz, Ar–H, 1H), 7.31 (d, J = 8.0 Hz, Ar–H, 1H), 4.63 (s, CH₂, 2H), 4.59 (s, CH₂, 2H), 2.98 (s, CH₃, 3H); LC-MS (ESI†): m/z calcd. for C₉H₁₀BrNO₂S: 274.96 [M + H]⁺; found: 276.11, 278.14.

6-Bromo-2-(ethylsulfonyl)-1,2,3,4-tetrahydroisoquinoline (15)

White solid; yield: 51%; ¹H NMR (500 MHz, DMSO-d₆) δ 7.41 (d, J = 2.5 Hz, Ar–H, 1H), 7.37 (dd, J = 8.5, 2.5 Hz, Ar–H, 1H), 7.15 (d, J = 8.0 Hz, Ar–H, 1H), 4.38 (s, CH₂, 2H), 3.48 (t, J = 6.0 Hz, CH₂, 2H), 3.12 (q, J = 7.5 Hz, CH₂, 2H), 2.88 (t, J = 6.0 Hz, CH₂, 2H), 1.21 (t, J = 7.0 Hz, CH₃, 3H); LC-MS (ESI†): m/z calcd. for C₁₁H₁₄BrNO₂S: 302.99 [M + H]⁺; found: 304.01, 306.01.

6-Bromo-2-(cyclopropylsulfonyl)-1,2,3,4-tetrahydroisoquinoline (16)

Colorless oil; yield: 42%; ¹H NMR (500 MHz, DMSO-d₆) δ 7.40 (d, J = 2.5 Hz, Ar–H, 1H), 7.37 (dd, J = 8.5, 2.0 Hz, Ar–H, 1H), 7.17 (d, J = 8.0 Hz, Ar–H, 1H), 4.39 (s, CH₂, 2H), 3.49 (t, J = 6.0 Hz, CH₂, 2H), 2.91 (t, J = 6.0 Hz, CH₂, 2H), 2.64–2.59 (m, CH, 1H), 1.00–0.92 (m, CH₂ × 2, 4H); LC-MS (ESI†): m/z calcd. for C₁₂H₁₄BrNO₂S: 314.99 [M + H]⁺; found: 316.03, 318.03.

6-Bromo-2-(cyclohexylsulfonyl)-1,2,3,4-tetrahydroisoquinoline (17)

White solid; yield: 46%; ¹H NMR (500 MHz, DMSO-d₆) δ 7.40 (d, J = 2.0 Hz, Ar–H, 1H), 7.37 (dd, J = 8.0, 2.0 Hz, Ar–H, 1H), 7.13 (d, J = 8.5 Hz, Ar–H, 1H), 4.41 (s, CH₂, 2H), 3.52 (t, J = 5.5 Hz, CH₂, 2H), 3.22–3.16 (m, CH, 1H), 2.85 (t, J = 6.0 Hz, CH₂, 2H), 2.00–1.97 (m, CH₂, 2H), 1.79–1.74 (m, CH₂, 2H), 1.63–1.59 (m, CH, 1H), 1.42–1.34 (m, CH₂, 2H), 1.30–1.21 (m, CH₂, 2H), 1.16–1.07 (m, CH, 1H); LC-MS (ESI†): m/z calcd. for C₁₅H₂₀BrNO₂S: 357.04 [M + H]⁺; found: 358.11, 360.13.

6-Bromo-2-(phenylsulfonyl)-1,2,3,4-tetrahydroisoquinoline (18)

White solid; yield: 44%; ¹H NMR (500 MHz, DMSO-d₆) δ 7.84–7.83 (m, Ar–H × 2, 2H), 7.73–7.69 (m, Ar–H, 1H), 7.66–7.60 (m, Ar–H × 2, 2H), 7.35–7.32 (m, Ar–H × 2, 2H), 7.13 (d, J = 9.0 Hz, Ar–H, 1H), 4.17 (s, CH₂, 2H), 3.29 (t, J = 6.5 Hz, CH₂, 2H), 2.84 (t, J = 5.5 Hz, CH₂, 2H); LC–MS (ESI): m/z calcd. for C₁₅H₁₄BrNO₂S: 350.99 [M + H]⁺; found: 352.03, 354.01.

Synthesis procedure of 5-bromobenzo[b]thiophene 1,1-dioxide (19)

5-Bromobenzo[b]thiophene 6 (1.00 mmol) was dissolved in DCM (3.0 mL), with an atmosphere of 0 °C, the oxidant m-CPBA was added to the solution. The reaction mixture was stirred at room temperature for 3 h. The reaction solution was treated with saturated NaOH and was extracted with DCM and water. The organic layer was collected, washed with brine, dried over Na₂SO₄, and concentrated under reduced pressure. The residue was purified by column chromatography to afford intermediate 19.

White solid; yield: 92%; ¹H NMR (500 MHz, DMSO-d₆) δ 7.87–7.86 (m, Ar–H, 1H), 7.83–7.82 (m, Ar–H × 2, 2H), 7.59 (d, J = 7.0 Hz, CH, 1H), 7.45 (d, J = 7.0 Hz, CH, 1H); LC-MS (ESI†): m/z calcd. for C₈H₅BrO₂S: 243.92 [M + H]⁺; found: 245.16, 247.16.

Synthesis procedure of tert-butyl (3-((2-amino-5-(trifluoromethyl)pyrimidin-4-yl)amino)propyl)carbamate (21)

To a solution of 4-chloro-5-(trifluoromethyl)pyrimidin-2-amine 20 (1.0 mmol) and tert-butyl (3-aminopropyl)carbamate (1.2 mmol) in 10 mL ACN was added TEA (1.5 mmol) at room temperature. Then the reaction mixture was heated to 80 °C and stirred overnight. The solution was evaporated, and the residue was purified by column chromatography to afford the product 21.

White solid; yield: 75%; ¹H NMR (500 MHz, DMSO-d₆) δ 7.96 (s, Ar–H, 1H), 6.83 (t, J = 6.0 Hz, NH, 1H), 6.70 (m, NH₂, 2H), 3.36 (q, J = 6.5 Hz, CH₂, 2H), 2.94 (q, J = 6.5 Hz, CH₂, 2H), 1.60 (m, CH₂, 2H), 1.37 (s, CH₃ × 3, 9H); LC-MS (ESI†): m/z calcd. for C₁₃H₂₀F₃N₅O₂: 335.16 [M + H]⁺; found: 336.20.

General procedure C for the synthesis of compounds (22–40)

A mixture of tert-butyl(3-((2-amino-5-(trifluoromethyl)pyrimidin-4-yl)amino)propyl)carbamate 21 (1.00 mmol), corresponding R²Br (1.20–1.50 mmol), Pd₂(dba)₃ (0.10 mmol), Xantphos (0.20 mmol), and Cs₂CO₃ (2.00 mmol) in dry dioxane (2.0 mL) was heated to 110 °C overnight under nitrogen. The cooled reaction mixture was purified by column chromatography to give the intermediates. To the solution of the intermediates in 2.0 mL DCM at 0 °C was added 1.0 mL TFA slowly. The mixture was stirred at room temperature for 1–3 h and the reaction was evaporated under reduce pressure. The residues were purified by column chromatography to afford products 22–40.

6-((4-((3-Aminopropyl)amino)-5-(trifluoromethyl)pyrimidin-2-yl)amino)-3,4-dihydronaphthalen-1(2H)-one (22)

Yellowish solid; yield: 30%; HPLC retention time: 11.845 min, HPLC purity: 98.2% (method A); ¹H NMR (500 MHz, DMSO-d₆) δ 9.98 (br, NH, 1H), 8.24 (s, Ar–H, 1H), 7.88 (d, J = 2.0 Hz, Ar–H, 1H), 7.79 (d, J = 8.5 Hz, Ar–H, 1H), 7.68 (dd, J = 8.5, 2.0 Hz, Ar–H, 1H), 3.57 (t, J = 6.5 Hz, CH₂, 2H), 2.90 (t, J = 6.0 Hz, CH₂, 2H), 2.66 (t, J = 6.5 Hz, CH₂, 2H), 2.55–2.53 (m, CH₂, 2H), 2.05–2.00 (m, CH₂, 2H), 1.74–1.68 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 196.51, 161.14, 158.37, 154.89 (q, J_CF = 5.0 Hz), 146.12, 145.33, 127.84, 126.53, 125.30 (q, J_CF = 268.8 Hz), 117.79, 117.49, 98.52 (q, J_CF = 30.0 Hz), 49.05, 40.07, 38.89, 32.02, 29.97, 23.46; HRMS (ESI): m/z calcd. for C₁₈H₂₀F₃N₅O: 379.1620 [M + H]⁺; found: 380.1694.

5-((4-((3-Aminopropyl)amino)-5-(trifluoromethyl)pyrimidin-2-yl)amino)-2,3-dihydro-1H-inden-1-one (23)

Yellowish solid; yield: 28%; HPLC retention time: 10.075 min, HPLC purity: 98.9% (method A); ¹H NMR (500 MHz, CD₃OD) δ 8.23 (s, Ar–H, 1H), 8.02 (d, J = 1.0 Hz, Ar–H, 1H), 7.68 (dd, J = 8.5, 1.5 Hz, Ar–H, 1H), 7.64 (d, J = 8.5 Hz, Ar–H, 1H), 3.68 (t, J = 6.5 Hz, CH₂, 2H), 3.14 (t, J = 6.0 Hz, CH₂, 2H), 3.01–2.98 (m, CH₂, 2H), 2.69–2.66 (m, CH₂, 2H), 2.07–2.02 (m, CH₂, 2H); ¹³C NMR (125 MHz, CD₃OD) δ 208.44, 162.21, 160.26, 159.17, 155.78, 148.11, 131.97, 126.01 (q, J_CF = 286.3 Hz), 125.28, 120.17, 116.93, 101.27, 38.78, 38.66, 37.39, 28.52, 26.91; LC-MS (ESI†): m/z calcd for C₁₇H₁₈F₃N₅O: 365.15 [M + H]⁺; found: 366.14.

7-((4-((3-Aminopropyl)amino)-5-(trifluoromethyl)pyrimidin-2-yl)amino)chroman-4-one (24)

Yellowish solid; yield: 33%; HPLC retention time: 7.929 min, HPLC purity: 98.3% (method A); ¹H NMR (500 MHz, DMSO-d₆) δ 10.04 (br, NH, 1H), 8.25 (s, Ar–H, 1H), 7.71 (d, J = 2.0 Hz, Ar–H, 1H), 7.66 (d, J = 9.0 Hz, Ar–H, 1H), 7.35 (dd, J = 8.5, 2.0 Hz, Ar–H, 1H), 4.50 (t, J = 6.5 Hz, CH₂, 2H), 3.55 (t, J = 7.0 Hz, CH₂, 2H), 2.71 (t, J = 6.0 Hz, CH₂, 2H), 2.67 (t, J = 6.5 Hz, CH₂, 2H), 1.72–1.67 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 190.41, 162.89, 161.07, 158.33, 154.86 (q, J_CF = 6.3 Hz), 147.62, 127.47, 125.22 (q, J_CF = 268.8 Hz), 115.56, 113.00, 105.91, 98.81 (q, J_CF = 31.3 Hz), 67.41, 40.24, 40.07, 37.48, 32.08; LC-MS (ESI†): m/z calcd. for C₁₇H₁₈F₃N₅O₂: 381.14 [M + H]⁺; found: 382.16.

5-((4-((3-Aminopropyl)amino)-5-(trifluoromethyl)pyrimidin-2-yl)amino)isobenzofuran-1(3H)-one (25)

White solid; yield: 42%; HPLC retention time: 7.317 min, HPLC purity: 95.9% (method A); ¹H NMR (500 MHz, DMSO-d₆) δ 10.19 (br, NH, 1H), 8.25 (s, Ar–H, 1H), 8.21 (d, J = 2.0 Hz, Ar–H, 1H), 7.85 (dd, J = 8.5, 1.5 Hz, Ar–H, 1H), 7.74 (d, J = 8.5 Hz, Ar–H, 1H), 5.36 (s, CH₂, 2H), 3.55 (t, J = 7.0 Hz, CH₂, 2H), 2.66 (t, J = 6.5 Hz, CH₂, 2H), 1.73–1.67 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 170.89, 161.14, 158.35, 154.88 (q, J_CF = 3.8 Hz), 149.38, 146.36, 125.86, 125.22 (q, J_CF = 285.0 Hz), 120.21, 117.84, 111.44, 98.85 (q, J_CF = 31.3 Hz), 69.85, 39.91, 39.88, 31.86; LC-MS (ESI†): m/z calcd. for C₁₆H₁₆F₃N₅O₂: 367.13 [M + H]⁺; found: 368.19.

N ⁴-(3-Aminopropyl)-N²-(2,3-dihydrobenzo[b][1,4]dioxin-6-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (26)

White solid; yield: 30%; HPLC retention time: 7.424 min, HPLC purity: 95.0% (method A); ¹H NMR (500 MHz, DMSO-d₆) δ 9.41 (br, NH, 1H), 8.13 (s, Ar–H, 1H), 7.43 (d, J = 2.5 Hz, Ar–H, 1H), 7.13 (dd, J = 8.5, 2.0 Hz, Ar–H, 1H), 6.75 (d, J = 8.5 Hz, Ar–H, 1H), 4.23–4.18 (m, CH₂ × 2, 4H), 3.50 (t, J = 6.5 Hz, CH₂, 2H), 2.64 (t, J = 6.5 Hz, CH₂, 2H), 1.70–1.64 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 161.26, 158.37, 154.80 (q, J_CF = 5.0 Hz), 143.27, 138.85, 134.31, 125.62 (q, J_CF = 267.5 Hz), 116.92, 113.19, 108.90, 97.23, 64.68, 64.37, 40.24, 40.07, 32.01; LC-MS (ESI†): m/z calcd for C₁₆H₁₈F₃N₅O₂: 369.14 [M + H]⁺; found: 370.22.

N ⁴-(3-aminopropyl)-N²-(benzo[d][1,3]dioxol-5-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (27)

White solid; yield: 29%; HPLC Retention time: 6.355 min, HPLC purity: 97.4% (method A); ¹H NMR (500 MHz, DMSO-d₆) δ 9.48 (br, NH, 1H), 8.13 (s, Ar–H, 1H), 7.50 (d, J = 2.0 Hz, Ar–H, 1H), 7.10 (dd, J = 8.5, 2.0 Hz, Ar–H, 1H), 6.82 (d, J = 8.5 Hz, Ar–H, 1H), 5.96 (s, CH₂, 2H), 3.49 (t, J = 6.5 Hz, CH₂, 2H), 2.63 (t, J = 6.5 Hz, CH₂, 2H), 1.69–1.63 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 161.30, 158.36, 154.82 (q, J_CF = 6.3 Hz), 147.40, 142.41, 135.09, 125.59 (q, J_CF = 267.5 Hz), 112.60, 108.28, 102.18, 101.22, 97.30, 40.24, 40.08, 32.28; LC-MS (ESI†): m/z calcd. for C₁₅H₁₆F₃N₅O₂: 355.13 [M + H]⁺; found: 356.11.

N ⁴-(3-Aminopropyl)-N²-(1-(methylsulfonyl)-1,2,3,4-tetrahydroquinolin-6-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (29)

White solid; yield: 44%; HPLC retention time: 18.392 min, HPLC purity: 99.7% (method B); ¹H NMR (500 MHz, DMSO-d₆) δ 9.55 (br, NH, 1H), 8.15 (s, Ar–H, 1H), 7.67 (d, J = 2.5 Hz, Ar–H, 1H), 7.48 (dd, J = 9.0, 2.5 Hz, Ar–H, 1H), 7.43 (d, J = 9.0 Hz, Ar–H, 1H), 3.68–3.66 (m, CH₂, 2H), 3.53 (t, J = 6.5 Hz, CH₂, 2H), 2.97 (s, CH₃, 3H), 2.77 (t, J = 6.5 Hz, CH₂, 2H), 2.63 (t, J = 6.5 Hz, CH₂, 2H), 1.94–1.89 (m, CH₂, 2H), 1.69–1.64 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 161.30, 158.39, 154.86 (q, J_CF = 6.3 Hz), 137.16, 131.34, 130.43, 125.56 (q, J_CF = 267.5 Hz), 123.58, 120.30, 118.08, 97.59, 49.06, 46.36, 40.04, 38.63, 32.29, 27.08, 22.12; HRMS (ESI†): m/z calcd. for C₁₈H₂₃F₃N₆O₂S: 444.1555 [M + H]⁺; found: 445.1631.

N ⁴-(3-Aminopropyl)-N²-(2-(methylsulfonyl)-1,2,3,4-tetrahydroisoquinolin-6-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (30)

White solid; yield: 42%; HPLC retention time: 17.684 min, HPLC purity: 98.6% (method B); ¹H NMR (500 MHz, DMSO-d₆) δ 9.56 (br, NH, 1H), 8.16 (s, Ar–H, 1H), 7.68 (d, J = 2.0 Hz, Ar–H, 1H), 7.53 (dd, J = 8.5, 2.0 Hz, Ar–H, 1H), 7.08 (d, J = 8.5 Hz, Ar–H, 1H), 4.30 (s, CH₂, 2H), 3.52 (t, J = 6.5 Hz, CH₂, 2H), 3.42 (t, J = 6.0 Hz, CH₂, 2H), 2.93 (s, CH₃, 3H), 2.88 (t, J = 6.0 Hz, CH₂, 2H), 2.63 (t, J = 6.5 Hz, CH₂, 2H), 1.70–1.65 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 160.81, 157.84, 154.33, 138.57, 133.14, 126.32, 125.31, 124.99 (q, J_CF = 267.5 Hz), 118.99, 117.62, 97.06, 46.52, 43.01, 39.53, 39.36, 34.53, 31.64, 28.45; HRMS (ESI†): m/z calcd. for C₁₈H₂₃F₃N₆O₂S: 444.1555 [M + H]⁺; found: 445.1631.

N ⁴-(3-Aminopropyl)-N²-(1-(methylsulfonyl)indolin-5-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (31)

White solid; yield: 38%; HPLC retention time: 13.611 min, HPLC purity: 97.8% (method B); ¹H NMR (500 MHz, DMSO-d₆) δ 9.55 (br, NH, 1H), 8.14 (s, Ar–H, 1H), 7.72 (d, J = 2.0 Hz, Ar–H, 1H), 7.53 (dd, J = 8.5, 2.0 Hz, Ar–H, 1H), 7.16 (d, J = 9.0 Hz, Ar–H, 1H), 3.92 (t, J = 8.5 Hz, CH₂, 2H), 3.51 (t, J = 6.5 Hz, CH₂, 2H), 3.10 (t, J = 8.5 Hz, CH₂, 2H), 2.94 (s, CH₃, 3H), 2.62 (t, J = 6.5 Hz, CH₂, 2H), 1.68–1.63 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 160.77, 157.82, 154.29 (q, J_CF = 5.0 Hz), 136.33, 136.31, 132.19, 125.02 (q, J_CF = 267.5 Hz), 118.58, 116.79, 113.39 (2C), 50.03, 48.50, 39.51, 33.64, 31.76, 27.64; HRMS (ESI†): m/z calcd. for C₁₇H₂₁F₃N₆O₂S: 430.1399 [M + H]⁺; found: 431.1472.

N ⁴-(3-Aminopropyl)-N²-(2-(methylsulfonyl)isoindolin-5-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (32)

White solid; yield: 40%; HPLC retention time: 17.280 min, HPLC purity: 97.6% (method B); ¹H NMR (500 MHz, DMSO-d₆) δ 9.67 (br, NH, 1H), 8.17 (s, Ar–H, 1H), 7.84 (d, J = 2.0 Hz, Ar–H, 1H), 7.61 (dd, J = 8.0, 2.0 Hz, Ar–H, 1H), 7.22 (d, J = 8.0 Hz, Ar–H, 1H), 4.61 (s, CH₂, 2H), 4.57 (s, CH₂, 2H), 3.52 (t, J = 7.0 Hz, CH₂, 2H), 2.98 (s, CH₃, 3H), 2.63 (t, J = 6.5 Hz, CH₂, 2H), 1.69–1.64 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 160.79, 157.82, 154.32, 139.76, 136.58, 129.31, 125.05, 124.95 (q, J_CF = 267.5 Hz), 122.62, 118.91, 113.15, 53.36, 52.96, 39.69, 33.33, 31.61, 20.66; HRMS (ESI†): m/z calcd. for C₁₇H₂₁F₃N₆O₂S: 430.1399 [M + H]⁺; found: 431.1472.

N ⁴-(3-Aminopropyl)-N²-(4-(methylsulfonyl)-3,4-dihydro-2H-benzo[b][1,4]oxazin-7-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (33)

White solid; yield: 45%; HPLC retention time: 16.030 min, HPLC purity: 98.0% (method B); ¹H NMR (500 MHz, DMSO-d₆) δ 9.59 (br, NH, 1H), 8.16 (s, Ar–H, 1H), 7.54 (d, J = 2.5 Hz, Ar–H, 1H), 7.44 (d, J = 9.0 Hz, Ar–H, 1H), 7.24 (dd, J = 9.0, 2.5 Hz, Ar–H, 1H), 4.27–4.25 (m, CH₂, 2H), 3.79–3.77 (m, CH₂, 2H), 3.52 (t, J = 6.5 Hz, CH₂, 2H), 3.07 (s, CH₃, 3H), 2.64 (t, J = 6.0 Hz, CH₂, 2H), 1.69–1.64 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 161.23, 158.37, 154.84, 146.73, 138.41, 125.50 (q, J_CF = 267.5 Hz), 123.39, 118.65, 112.46, 108.13, 97.76, 64.21, 44.18, 40.25, 40.09, 38.51, 32.25; LC-MS (ESI†): m/z calcd. for C₁₇H₂₁F₃N₆O₃S: 446.13 [M + H]⁺; found: 447.13.

N ⁴-(3-Aminopropyl)-N²-(1-(ethylsulfonyl)-1,2,3,4-tetrahydroquinolin-6-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (34)

White solid; yield: 33%; HPLC retention time: 21.586 min, HPLC purity: 97.9% (method B); ¹H NMR (500 MHz, DMSO-d₆) δ 9.53 (br, NH, 1H), 8.15 (s, Ar–H, 1H), 7.66 (d, J = 2.5 Hz, Ar–H, 1H), 7.46 (dd, J = 9.0, 3.0 Hz, Ar–H, 1H), 7.39 (d, J = 9.0 Hz, Ar–H, 1H), 3.68–3.65 (m, CH₂, 2H), 3.52 (t, J = 6.5 Hz, CH₂, 2H), 3.19 (q, J = 7.5 Hz, CH₂, 2H), 2.78 (t, J = 6.5 Hz, CH₂, 2H), 2.62 (t, J = 6.5 Hz, CH₂, 2H), 1.95–1.90 (m, CH₂, 2H), 1.69–1.64 (m, CH₂, 2H), 1.19 (t, J = 7.5 Hz, CH₃, 3H); ¹³C NMR (125 MHz, DMSO-d₆) δ 161.30, 158.39, 154.86, 136.83, 131.57, 129.96, 125.56 (q, J_CF = 267.5 Hz), 123.02, 120.33, 118.08, 110.49, 46.38, 46.20, 40.24, 40.08, 32.25, 27.07, 22.66, 8.29; LC-MS (ESI†): m/z calcd. for C₁₉H₂₅F₃N₆O₂S: 458.17 [M + H]⁺; found: 459.17.

N ⁴-(3-Aminopropyl)-N²-(2-(ethylsulfonyl)-1,2,3,4-tetrahydroisoquinolin-6-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (35)

White solid; yield: 36%; HPLC retention time: 21.117 min, HPLC purity: 95.0% (method B); ¹H NMR (500 MHz, DMSO-d₆) δ 9.55 (br, NH, 1H), 8.16 (s, Ar–H, 1H), 7.67 (s, Ar–H, 1H), 7.51 (dd, J = 8.5, 2.5 Hz, Ar–H, 1H), 7.07 (d, J = 8.5 Hz, Ar–H, 1H), 4.35 (s, CH₂, 2H), 3.53–3.47 (m, CH₂ × 2, 4H), 3.11 (q, J = 7.5 Hz, CH₂, 2H), 2.85 (t, J = 6.0 Hz, CH₂, 2H), 2.62 (t, J = 6.0 Hz, CH₂, 2H), 1.69–1.64 (m, CH₂, 2H), 1.21 (t, J = 6.5 Hz, CH₃, 3H); ¹³C NMR (125 MHz, DMSO-d₆) δ 161.36, 158.40, 154.87, 139.07, 133.86, 126.74, 126.27, 125.55 (q, J_CF = 266.3 Hz), 119.65, 118.15, 97.59, 46.81, 43.71, 43.41, 40.23, 40.06, 32.09, 29.22, 8.06; LC-MS (ESI†): m/z calcd. for C₁₉H₂₅F₃N₆O₂S: 458.17 [M + H]⁺; found: 459.19.

N ⁴-(3-Aminopropyl)-N²-(1-(cyclopropylsulfonyl)-1,2,3,4-tetrahydroquinolin-6-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (36)

White solid; yield: 33%; HPLC retention time: 21.570 min, HPLC purity: 99.1% (method B); ¹H NMR (500 MHz, DMSO-d₆) δ 9.55 (br, NH, 1H), 8.15 (s, Ar–H, 1H), 7.67 (d, J = 2.5 Hz, Ar–H, 1H), 7.45 (dd, J = 9.0, 3.0 Hz, Ar–H, 1H), 7.40 (d, J = 9.0 Hz, Ar–H, 1H), 3.69–3.67 (m, CH₂, 2H), 3.52 (t, J = 6.5 Hz, CH₂, 2H), 2.80 (t, J = 6.5 Hz, CH₂, 2H), 2.67–2.65 (m, CH, 1H), 2.62 (t, J = 6.5 Hz, CH₂, 2H), 2.00–1.95 (m, CH₂, 2H), 1.69–1.64 (m, CH₂, 2H), 0.93–0.87 (m, CH₂ × 2, 4H); ¹³C NMR (125 MHz, DMSO-d₆) δ 161.30, 158.40, 154.88, 137.35, 131.52, 130.88, 125.55 (q, J_CF = 268.8 Hz), 124.36, 120.12, 117.91, 107.43, 46.57, 40.25, 40.08, 32.27, 29.54, 26.90, 22.38, 5.43 (2C); LC-MS (ESI†): m/z calcd. for C₂₀H₂₅F₃N₆O₂S: 470.17 [M + H]⁺; found: 471.18.

N ⁴-(3-Aminopropyl)-N²-(2-(cyclopropylsulfonyl)-1,2,3,4-tetrahydroisoquinolin-6-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (37)

White solid; yield: 37%; HPLC retention time: 21.226 min, HPLC purity: 98.3% (method B); ¹H NMR (500 MHz, DMSO-d₆) δ 9.57 (br, NH, 1H), 8.16 (s, Ar–H, 1H), 7.68 (d, J = 2.5 Hz, Ar–H, 1H), 7.53 (dd, J = 8.5, 2.5 Hz, Ar–H, 1H), 7.08 (d, J = 8.5 Hz, Ar–H, 1H), 4.37 (s, CH₂, 2H), 3.53 (t, J = 6.5 Hz, CH₂, 2H), 3.50 (t, J = 5.5 Hz, CH₂, 2H), 2.88 (t, J = 5.5 Hz, CH₂, 2H), 2.64 (t, J = 6.5 Hz, CH₂, 2H), 2.61–2.58 (m, CH, 1H), 1.71–1.65 (m, CH₂, 2H), 0.99–0.95 (m, CH₂ × 2, 4H); ¹³C NMR (125 MHz, DMSO-d₆) δ 161.36, 158.40, 154.84 (q, J_CF = 5.0 Hz), 139.11, 133.69, 126.79, 126.11, 125.55 (q, J_CF = 267.5 Hz), 119.57, 118.14, 97.44, 47.41, 43.98, 40.06, 32.26, 29.06, 26.04, 21.20, 4.53 (2C); LC-MS (ESI†): m/z calcd. for C₂₀H₂₅F₃N₆O₂S: 470.17 [M + H]⁺; found: 471.18.

N ⁴-(3-Aminopropyl)-N²-(2-(cyclohexylsulfonyl)-1,2,3,4-tetrahydroisoquinolin-6-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (38)

White solid; yield: 41%; HPLC retention time: 24.443 min, HPLC purity: 98.5% (method B); ¹H NMR (500 MHz, DMSO-d₆) δ 9.55 (br, NH, 1H), 8.16 (s, Ar–H, 1H), 7.67 (d, J = 2.0 Hz, Ar–H, 1H), 7.52 (dd, J = 8.5, 2.5 Hz, Ar–H, 1H), 7.05 (d, J = 8.5 Hz, Ar–H, 1H), 4.40 (s, CH₂, 2H), 3.54–3.51 (m, CH₂ × 2, 4H), 3.21–3.15 (m, CH₂, 2H), 2.82 (t, J = 6.0 Hz, CH₂, 2H), 2.63 (t, J = 6.5 Hz, CH₂, 2H), 2.01–1.98 (m, CH₂, 2H), 1.78–1.74 (m, CH₂, 2H), 1.70–1.65 (m, CH₂, 2H), 1.63–1.59 (m, CH, 1H), 1.43–1.35 (m, CH₂, 2H), 1.31–1.22 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 161.36, 158.40, 154.84 (q, J_CF = 5.0 Hz), 139.01, 133.99, 126.62, 126.56, 125.55 (q, J_CF = 267.5 Hz), 119.76, 118.13, 97.35, 60.22, 60.09, 46.96, 43.69, 32.27, 29.63, 26.62 (2C), 25.28, 24.96 (2C), 21.21; LC-MS (ESI†): m/z calcd. for C₂₃H₃₁F₃N₆O₂S: 512.22 [M + H]⁺; found: 513.25.

N ⁴-(3-Aminopropyl)-N²-(1-(phenylsulfonyl)-1,2,3,4-tetrahydroquinolin-6-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (39)

White solid; yield: 37%; HPLC retention time: 23.650 min, HPLC purity: 99.6% (method B); ¹H NMR (500 MHz, DMSO-d₆) δ 9.58 (br, NH, 1H), 8.16 (s, Ar–H, 1H), 7.67–7.64 (m, Ar–H, 1H), 7.57–7.52 (m, Ar–H × 7, 7H), 3.74–3.72 (m, CH₂, 2H), 3.50 (t, J = 6.5 Hz, CH₂, 2H), 2.61 (t, J = 6.5 Hz, CH₂, 2H), 2.32 (t, J = 6.5 Hz, CH₂, 2H), 1.67–1.62 (m, CH₂, 2H), 1.53–1.48 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 161.26, 158.37, 154.86 (q, J_CF = 5.0 Hz), 139.35, 137.77, 133.68, 131.43, 130.63, 129.85 (2C), 127.22 (2C), 125.53 (q, J_CF = 267.5 Hz), 124.98, 119.84, 117.96, 97.66, 46.60, 40.25, 40.08, 32.23, 26.59, 21.29; LC-MS (ESI†): m/z calcd. for C₂₃H₂₅F₃N₆O₂S: 506.17 [M + H]⁺; found: 507.21.

N ⁴-(3-Aminopropyl)-N²-(2-(phenylsulfonyl)-1,2,3,4-tetrahydroisoquinolin-6-yl)-5-(trifluoromethyl)pyrimidine-2,4-diamine (40)

White solid; yield: 39%; HPLC retention time: 23.509 min, HPLC purity: 99.2% (method B); ¹H NMR (500 MHz, DMSO-d₆) δ 9.53 (br, NH, 1H), 8.15 (s, Ar–H, 1H), 7.84–7.82 (m, Ar–H × 2, 2H), 7.73–7.70 (m, Ar–H, 1H), 7.66–7.64 (m, Ar–H × 2, 2H), 7.61 (d, J = 2.5 Hz, Ar–H, 1H), 7.48 (dd, J = 8.0, 2.0 Hz, Ar–H, 1H), 7.05 (d, J = 8.5 Hz, Ar–H, 1H), 4.14 (s, CH₂, 2H), 3.50 (t, J = 6.5 Hz, CH₂, 2H), 3.29 (t, J = 5.5 Hz, CH₂, 2H), 2.82 (t, J = 6.0 Hz, CH₂, 2H), 2.62 (t, J = 6.5 Hz, CH₂, 2H), 1.68–1.62 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 161.32, 158.37, 154.85, 139.15, 136.34, 133.68, 133.40, 129.90 (2C), 127.86 (2C), 126.86, 125.53 (q, J_CF = 267.5 Hz), 125.28, 119.37, 118.12, 97.60, 47.45, 44.07, 40.25, 40.08, 32.22, 28.82; LC-MS (ESI†): m/z calcd for C₂₃H₂₅F₃N₆O₂S: 506.17 [M + H]⁺; found: 507.21.

Synthesis procedure of 5-((4-((3-aminopropyl)amino)-5-(trifluoromethyl)pyrimidin-2-yl)amino)-2,3-dihydrobenzo[b]thiophene 1,1-dioxide (28)

A mixture of 5-Bromobenzo[b]thiophene 1,1-dioxide 19 (1.50 mmol), tert-butyl (3-((2-amino-5-(trifluoromethyl)pyrimidin-4-yl)amino)propyl)carbamate 21 (1.00 mmol), Pd₂(dba)₃ (0.10 mmol), Xantphos (0.20 mmol), and Cs₂CO₃ (2.00 mmol) in dry dioxane (2.0 mL) was refluxed overnight under nitrogen. The reaction mixture was purified by column chromatography to give the intermediate. And then the intermediate was dissolved in 3.0 mL DCM/MeOH (1 : 1), and the Pd/C was added to the solution under hydrogen at room temperature for 3 h. The precipitate was filtered off, and the filtrate was evaporated under reduced pressure. To the residue in 2.0 mL DCM at 0 °C was added 1.0 mL TFA slowly. The mixture was stirred at room temperature for 1 h and the reaction was evaporated under reduced pressure. The residue was purified by column chromatography to afford product 28.

White solid; yield: 37%; HPLC retention time: 10.066 min, HPLC purity: 98.7% (method A); ¹H NMR (500 MHz, DMSO-d₆) δ 10.07 (br, NH, 1H), 8.25 (s, Ar–H, 1H), 8.05 (d, J = 2.0 Hz, Ar–H, 1H), 7.82 (dd, J = 9.0, 2.0 Hz, Ar–H, 1H), 7.64 (d, J = 8.5 Hz, Ar–H, 1H), 3.56 (t, J = 7.0 Hz, CH₂ × 2, 4H), 3.33 (t, J = 7.0 Hz, CH₂, 2H), 2.66 (t, J = 6.5 Hz, CH₂, 2H), 1.73–1.67 (m, CH₂, 2H); ¹³C NMR (125 MHz, DMSO-d₆) δ 161.14, 158.35, 154.90, 145.41, 139.68, 131.74, 125.26 (q, J_CF = 267.5 Hz), 121.71, 119.59, 116.42, 98.68 (q, J_CF = 32.5 Hz), 51.06, 49.05, 40.07, 32.13, 25.52; LC-MS (ESI†): m/z calcd. for C₁₆H₁₈F₃N₅O₂S: 401.11 [M + H]⁺; found: 402.17.

Biological evaluation

Kinase inhibition assay

The proteins FLT3-WT, FLT-D835Y, c-Kit, JAK3, and EGFR were purchased from Eurofins Scientific. CHK1, IRAK4 and BTK were purchased from Sino Biological. PIM1, CDK2, and P70S6K were purchased from Carna Biosciences, Inc. DRAK2 and Aurora A were expressed in E. coli.

A part of enzymic inhibitory activities was determined using an HTRF® KinEASE™ kit (PerkinElmer, Waltham, USA). Briefly, tested kinases, peptide substrates, ATP and different concentrations of compounds were incubated at room temperature for 1 h in 384-well plates. The final volume of the reactions is 10 μL in reaction buffer (1× kinase buffer, 5 mM MgCl₂, 1 mM DTT). After the kinase reactions were incubated at room temperature for 1 h, the antibody was added for detection. Fluorescence signals were collected on an EnVision® multimode microplate reader (PerkinElmer) with excitation at 340 nm and emission at 620 nm and 665 nm.

DRAK2 enzymic inhibitory activity was determined using an ADP-Glo assay kit (Promega). The enzyme buffer, substrate buffer and tested compound with different concentrations were added to a 384-well plate and incubated for 2 h at room temperature. And 5 μL ADP-Glo termination buffer was added to terminate the kinase reaction, and the remaining ATP was consumed. After one hour of reaction, the kinase detection reagent was added and incubated for half an hour, which converted ADP to ATP and the newly synthesized ATP was detected using the conjugated luciferase/luciferin reaction.

CDK2 enzymic inhibitory activity was determined using a LANCE Ultra detection kit (PerkinElmer, Waltham, USA). The tested compound was mixed with enzyme buffer. The substrate was added to a 384-well plate, incubated for 1 hour at room temperature. Then XL665 and the antibody were incubated for another hour, and finally the ratio of the fluorescence signal 665 nm to 620 nm was detected with an EnVision® multimode microplate reader.

The ratio of the concentration to the active percentage was plotted, and then the fitting and curve were calculated using nonlinear regression, and the IC₅₀ value of the compound was calculated using the software GraphPad Prism 9.

In vitro antiproliferative assay

MV4-11 cells and BaF3 cells with FLT3 ITD/TKD mutations (80 μL) were inoculated into 96-well plates according to the growth rate. And then cells were incubated with different concentrations of compounds (20 μL DMSO solution) for 72 h under the conditions of 37 °C and 5% CO₂. The CellTiter 96 Aqueous One Solution Reagent (G3581, Promega, Madison, WI) was added to culture wells, and the mixture was incubated for 3 h. The absorbance of the system at 490 nm and 690 nm was recorded via a microplate reader. The inhibition rate of the compounds on the tested cells was calculated by the following formula. Inhibition rate = (control group A value-blank control)/control group A value × 100%, and the IC₅₀ values of the tested compounds were calculated using the software GraphPad Prism 9.

hERG inhibition assay

The inhibitory activity of the representative compounds on hERG ion channels was evaluated on HEK-293 cells, and the test method was borrowed from the report of Lee et al. in 2019.²¹

PK study

The PK study was carried out in strict accordance with the Laboratory Animal Management Regulations (State Scientific and Technological Commission Publication No. 8-27 Rev. 2017) and was approved by Hang Zhou Leading Pharmatech Co., Ltd. (Hangzhou, China). ICR mice were purchased from JOINN Laboratories (Suzhou). The blood sample was collected at eight points (0.25, 0.50, 1.00, 2.00, 4.00, 6.00, 8.00, 24.00 h), and then the plasma sample was separated via centrifugation at 5000 rpm for 10 min at 4 °C and analyzed by LC-MS/MS (Qtrap 5500). The PK properties were analyzed via WinNonLin 8.0. (IACUC: A23-072).

Molecular docking

Discovery Studio 2.1 was used to perform molecular docking and the graphical image was drew through PyMOL v0.99. The FLT3 homology model was referenced from Li's lab.²² The X-ray crystal structure of the CHK1 protein (PDB: 2YM8) was obtained from https://www.rcsb.org/. For details of molecular docking, please refer to our previous study.²³

Conflicts of interest

There are no conflicts of interest to declare.