Design, Synthesis, In Vitro and In Vivo Characterization of CDC42 GTPase Interaction Inhibitors for the Treatment of Cancer

Abstract

CDC42 GTPases (RHOJ, CDC42, and RHOQ) are overexpressed in multiple tumor types and activate pathways critical for tumor growth, angiogenesis, and metastasis. Recently, we reported the discovery of a novel lead compound, ARN22089, which blocks the interaction of CDC42 GTPases with specific downstream effectors. ARN22089 blocks tumor growth in BRAF mutant mouse melanoma models and patient-derived xenografts (PDXs) in vivo. ARN22089 also inhibits tumor angiogenesis in three-dimensional vascularized microtumor models in vitro. Notably, ARN22089 belongs to a novel class of trisubstituted pyrimidines. Based on these results, we describe an extensive structure–activity relationship of ∼30 compounds centered on ARN22089. We discovered and optimized two novel inhibitors (27, ARN25062, and 28, ARN24928), which are optimal back-up/follow-up leads with favorable drug-like properties and in vivo efficacy in PDX tumors. These findings further demonstrate the potential of this class of CDC42/RHOJ inhibitors for cancer treatment, with lead candidates ready for advanced preclinical studies.

Introduction

CDC42 GTPases (CDC42, RHOJ, and RHOQ) are small guanosine triphosphate (GTP)-binding proteins that are known to regulate tumor growth, angiogenesis, metastasis, and cell resistance to targeted therapies.¹⁻⁵ CDC42 GTPases are essential molecular switches within the cell. Their active/inactive state depends on whether they are bound to GTP or guanosine diphosphate (GDP).⁶ When CDC42 GTPases are bound to GTP, they change their structural conformation, allowing protein surface interactions that are complementary to their downstream effectors. These include, but are not limited to, p21-activated protein kinases (PAKs).⁷ Notably, PAKs are known to be involved in invasion, migration, and oncogenic transformation.^8,9 Many groups have sought to design small molecules that inhibit PAK kinases by targeting the large and flexible ATP binding pocket in the kinase domain or by targeting a large autoinhibitory region that is observed in group I PAKs (PAK1, PAK2, and PAK3).¹⁰ However, developed agents have failed to reach the clinic secondary to their poor selectivity. For example, existing PAK inhibitors act on multiple isoforms of PAKs, including PAK2, which is thought to induce cardiotoxicity with a narrow therapeutic window.¹¹

CDC42 GTPases themselves have been previously considered “undruggable” due to their globular structure with the lack of alternative binding pockets to the active site, where there is high affinity for GTP or GDP.^1,12 Despite these restrictions, one of the main approaches to targeting CDC42 involves designing agents that block CDC42 activation by targeting the CDC42-GEF protein–protein interaction interfaces.¹³ Several small molecules have been identified that inhibit CDC42 activation.¹³⁻¹⁶ Among them, ZCL278 and AZA197 (Figure 1) represent two notable examples that have been shown to inhibit CDC42 activation and reduce prostate and colon cancer migration and invasion.^17,18 Unfortunately, many agents that target CDC42 activation also block activation of the GTPase RAC1,¹³ which can inhibit platelet aggregation, resulting in hematologic side effects.¹⁹

Figure 1.

Examples of selective CDC42 inhibitors, previously reported: ZCL278, AZA197, and our lead ARN22089.¹²

We recently discovered a new compound, ARN22089 (Figure 1), designed to specifically inhibit the interaction between CDC42 and its downstream kinase PAK. This molecule differs from the other small molecules that interfere with CDC42-GEF interaction.¹ Notably, this compound does not block RAC1 effector interactions.¹² In our initial disclosure of the compound ARN22089 (Figure 1), we reported its in vitro and in vivo efficacy and overall drug-like profile, with a brief description of the medicinal chemistry effort for the discovery and characterization of this novel lead compound. We discovered that this lead has broad activity against a panel of cancer cell lines and could inhibit tumor growth and tumor angiogenesis.¹² Administration of this drug either biweekly intravenously (I.V.) or twice daily subcutaneous (S.Q.) did not induce any notable side effects in mice, in contrast to PAK inhibitors that induce cardiotoxicity even at low doses. Here, we report the extensive structure–activity relationship (SAR) exploration and the complete rational drug design study centered on the ARN22089 discovery, with seventeen additional new derivatives. The overall drug discovery effort includes 29 inhibitors (Figure 2). In addition to the resulting SAR, we highlight the identification and characterization in vitro and in vivo of two optimal back-up/follow-up leads, 27 and 28 (ARN25062 and ARN24928).

Figure 2.

Representation of the points subject to chemical manipulation of scaffold A.

Results and Discussion

Hit to Lead Process toward the Discovery of Compound ARN22089

To briefly recapitulate our drug discovery program, we applied a virtual screening campaign on a collection of ∼20,000 molecules to identify 68 promising compounds that were experimentally tested for their antiproliferative activity in SKM28 melanoma cells—a cell line that we previously determined was sensitive to RHOJ depletion.¹² This effort allowed us to identify ARN12405 as a starting hit that inhibited the interaction between RHOJ or CDC42 and its downstream effector PAK (IC₅₀ = 16.4 μM, Table 1, entry 1).¹² The hit features a pyrimidine core bearing a 3-piperidine, a 4-chloroaniline, and a 4-pyridine in 2, 4, and 6 position of the central heterocycle, respectively (Table 1, entry 1).¹² Looking at the hit structure, we envisioned four points of modification. Therefore, we systematically changed the functional groups of our scaffold A (Figure 2) to improve the potency and the drug-like properties of the hit compound. We mainly explored (i) pyridines with different positions of nitrogen and some simple aromatic rings on C6 of the pyrimidine core, (ii) anilines on position C4 with different substituents embedding electron-donating or electron-withdrawing groups in ortho, meta, and para positions, especially focusing on the dimethylamino and methoxy groups as the main substituents, (iii) several aliphatic six-member (hetero)cycles on C2 between two pyridine nitrogen, and (iv) the core of the scaffold retaining the best substituents on a triazine core.

Table 1. Modification at R¹.

^aAll IC₅₀ values have an R² > 0.90.

^bCompounds initially reported by Jahid et al.¹²

^cND = not determined.

The antiproliferative activity of all compounds was mainly assessed in SKM28 cells, and the antiproliferative activity of the largest part of analogues was further evaluated in four additional cancer cell lines: three melanoma cell lines (SKMel3, WM3248, and A375) and a colon cancer line SW480 (Tables 1–5). All three melanoma lines have a BRAF (Val600Glu) mutation, while SW480 cells have a KRAS (Gly12Val) mutation. These transformed cells have been shown to have RHOJ/CDC42 activity that synergizes with known Ras/Raf effector pathways (Raf-MEK-ERK and PI3K-Akt).^20,21

Table 5. Triazine Analogues and Related Antiproliferative Activity.

^aAll IC₅₀ values have an R² > 0.90.

Optimization of R¹

All the evaluated modification of R¹ did not drastically alter the potency in the SKM28 cell line. In detail, moving pyridine nitrogen from 4 (1, ARN12405) to 3 or 2 positions as in derivatives 2 and 3 maintained the activity (IC₅₀ = 14.9 μM and 13.1 mM in SKM28 cell lines, respectively, entries 1–3, Table 1). However, removal of pyridine nitrogen in analogue 4 moderately increased the activity with an IC₅₀ of 10.2 μM in SKM28 (entry 4, Table 1). The introduction of the electron-donating methoxy group as in 5 in para position of the phenyl ring maintained the activity (entry 5, Table 1). Generally, all derivatives 1–5 exhibited a good potency in the other cell lines, showing an IC₅₀ in the range of 6.3–15.3 μM.

Optimization of R²

We then explored the variation of R² in order to alter the electronics of the aniline ring (Table 2). Moving the chlorine from para position to meta position at the aromatic ring in the scaffold with a pyridine and a 3-piperidine gave a 2-fold decrease in activity for 6 (IC₅₀ 38.3 μM in SKM28) compared to 1 (entry 1, Table 2). Instead, the activity was maintained in SKMel3 and A375. The substitution of chlorine with a fluorine in para position in derivative 7 and the naked phenyl ring in 8 dropped the activity in SKM28 (IC₅₀ > 50 μM, entries 2, 3, Table 2). On the other hand, compound 8 showed a slightly decreased potency in the other four cell lines, with IC₅₀ in the range of 20–30 μM. Since these modifications led to a marked decrease in potency in the cell SKM28, we decided to continue the exploration of R², using as the reference compound analogue 4, which displayed a comparable potency to the hit 1.

Table 2. Modification at R².

^aAll IC₅₀ values have an R² > 0.90.

^bCompounds initially reported by Jahid et al.¹²

^cND = not determined.

Maintaining the phenyl ring on C6 and a 3-piperidine on C2, we mainly investigated the effects of methoxy and dimethylamino groups in para and meta positions, which generated the derivatives 9–12 (entries 4–7, Table 2). The introduction of these substituents in para position resulted in a total or partial loss of activity in SKM28 (entries 4 and 5, Table 2). The same substituents in meta position moderately diminished the activity of the compounds. Specifically, we observed a 2- and 3-fold reduction for 11 (IC₅₀ = 24.2 μM) and 12 (IC₅₀ alone 31.4 μM), respectively. In the other four cell lines, derivative 9 led to only a slightly decreased activity, while compounds 10–12 exhibited a good activity with IC₅₀ in the 8.1–17.8 μM range.

Optimization of R³

Examining R³ substituents, the initial goal was to move the piperidine nitrogen from position 3 to 4, therefore removing the stereocenter and increasing the chemical tractability (synthesis and chirality). At the beginning, we synthesized the 4-piperidine counterpart of 4, which generated compound 13 with 4-chloroaniline on C4 and the phenyl substituent on C6 of pyrimidine, which exhibited one-digit micromolar potency in all cell lines (entry 1, Table 3). We then investigated in depth the role of methoxy and dimethylamino substituents on the aniline moiety in combination with 4-piperidine in position 2 (entries 2–5, Table 3), which exhibited a different effect compared to related analogues with a 3-piperidine moiety. This trend would suggest that there is a different way of interaction of 4-piperidine analogues compared to related 3-piperidine equivalents. Generally, the methoxy group in the para and meta position of compounds 14 and 16 has a detrimental effect on the activity in SKM28, showing approximately a 3-fold decrease in potency (entries 2 and 4, Table 3) compared to 4, while the related dimethylamino counterpart in 15 and 17 remained moderately active. On the other hand, compounds 14–17 showed a good antiproliferative effect toward the other four cell lines.

Table 3. Modification at R³.

^aAll IC₅₀ values have an R² > 0.90.

^bCompounds initially reported by Jahid et al.¹²

^cND = not determined.

The oxo counterparts 18 and 19 of compounds 14 and 15 resulted in complete loss of activity (IC₅₀ > 50 μM), indicating that a pyrane in C2 is not tolerated (entries 6 and 7, Table 3). Removing the heteroatom through the insertion of the cyclohexyl substituent as in compound 20 resulted in a 2- or 4-fold decrease in activity compared to 4. Finally, we masked the nitrogen of the piperidine with an acyl and a methoxy alkyl group as in 21 and 22 to understand the role of a free basic center and the potential interaction mediated by its substituents. While derivative 21 did not display a good potency, the insertion of a methoxyethyl group at the piperidine nitrogen in 22 restored the activity mainly in SKM28, SKMel3, and SW480, likely due to supplementary interactions established by the terminal methoxy group.

Further Expansion of Aniline as R²

Starting from the evidence that the type and conformation of the aliphatic cycle on C2 strongly affect the inhibitory capability of this chemical class, we further expanded the exploration of the aniline moiety maintaining a 4-piperidine on C2 and a phenyl on C6. This was well tolerated as shown for derivate 17 (ARN22089, Table 3). Thus, we completed the evaluation of the main investigated methoxy and dimethyl amino groups synthesizing the ortho-substituted analogues 23 and 24 (Table 4). The shifting of these groups to the ortho position increased the potency in SKM28 compared to the meta and para counterpart, restoring the activity with an IC₅₀ of ∼14 μM. On the other hand, the activity in the other cell lines decreased by ∼2–3-fold. A similar effect in the SKM28 cell line was observed with analogues 25 and 26, which feature a 3,4-dimethoxyphenyl- and a 3-methoxyethyloxyphenyl substituent, respectively. Despite 25 and 26 maintaining the activity in SKM28 and SKMel3, they displayed the weakest effect in other cell lines. Finally, the insertion of an electron-withdrawing H bond acceptor like the trifluoromethyl group in meta position as in compound 27 boosted the potency in all five cancer cell lines, with single-digit micromolar IC₅₀ (entry 5, Table 4).

Table 4. Expansion of R² Exploration.

^aAll IC₅₀ values have an R² > 0.90.

^bND = not determined.

In summary, meta-trifluorophenyl and meta-N,N-dimethylaminophenyl R² aniline substituents on C4 stand out as the best functionalities in combination with a phenyl on C6 and a 4-piperidine on C2. This returned a robust antiproliferative activity in all cancer cell lines. In view of these data, we shifted the dimethylaniline and trifluoromethylaniline groups into a triazine core, generating compounds 28 and 29 and symmetrizing the system. In general, this strategy confirmed the positive contribution of the selected substituents that exhibited a similar or slightly improved potency in all cell lines, with the exception of compound 29 toward SKM28 (IC₅₀ = 20.7 μM).

Evaluation of Drug-like Properties

After this initial evaluation, we assessed the drug-like profile of these novel RHOJ/CDC42 inhibitors by studying the kinetic solubility and the plasma and phase I microsomal stability of 19 selected compounds (Table 6). Previously, we selected 17 (ARN22089) as the most promising compound, since it displayed the best compromise between potency, solubility, metabolic stability, and chemical tractability.¹²17 was selected for further in vivo studies.¹² Notably, the starting hit 1 exhibited a good kinetic solubility of 222 μM, an excellent plasma stability over 120 min, and an acceptable microsomal stability of 48 min (Table 6, entry 1). All compounds exhibited an excellent kinetic solubility (>150 μM), with the exception of analogues 2–5, 13, and 29 (Table 6). From these data emerged that only the 4-pyridine on C6 in combination with a chloroaniline on C4 gave a good kinetic solubility, while a 3-pyridine, a 2-pyridine, a naked phenyl, or a p-methoxyphenyl drastically dropped the solubility. Indeed, phenyl and p-methoxy derivatives (4 and 5) exhibited the lowest kinetic solubility (<1 μM), followed by compounds 2 and 3 (38 ± 4 and 45 ± 3 μM, respectively); so, they were excluded for further investigations. Surprisingly, the introduction of the trifluoromethyl group on the aniline moiety moderately diminished the solubility in the pyrimidine derivative 27 compared to previously selected compound 17, while when −CF₃ was inserted in the triazine core as in compound 29, the solubility dropped to 1 μM. All compounds showed a good plasma stability (>70 min), with most of them showing acceptable microsomal stability (t_1/2 > 40 min). Furthermore, the solubility of compounds with Sk > 30 μM was confirmed by NMR measurement. This preliminary analysis is crucial in order to avoid false-positive results in the binding assays, which we performed via microscale thermophoresis (MST) and NMR. The compounds were tested by NMR in the binding assay buffer at different concentrations (from 10 to 100 μM) in the presence of an internal reference (4-trifluoromethyl benzoic acid, 400 μM) to assess their solubility and their aggregation state under these experimental conditions (SPAM filter,²²Table 6). The solubility data obtained by NMR were consistent with the kinetic solubility values (data not shown). Notably, derivatives 27 (ARN25062) and 28 (ARN24928) resulted to be the best compounds in terms of potency in all cell lines, kinetic solubility, also showing a strong improvement in plasma and microsomal stability.

Table 6. Kinetic Solubility and Plasma and Microsomal Stability in Mice of Selected Compounds^a.

entry	compound	S kinetic (μM)	aggregation by NMR (μM)	T1/2 plasma (min)	T1/2 microsomal (min)
1	1	222	ND	>120	48
2	2	38	no aggreg. up to 50	104	17
3	3	45	no aggreg. up to 50	>120	52
4	4b	<1	ND	>120	46
5	5	<1	ND	>120	ND
6	10b	246	100	>120	43
7	11b	237	no aggreg. up to 100	>120	21
8	12b	>250	no aggreg. up to 50	119	20
9	13	3	ND	>120	52
10	14	>250	no aggreg. up to 100	95	>60
11	15	180	30	>120	49
12	16b	>250	100	>120	27
13	17 (ARN22089)b	250	100	71	27
14	22	107	100	>120	12
14	23	219	50	>120	46
15	24	210	50	>60	>60
16	25	>250	no aggreg. up to 100	97	54
17	27 (ARN25062)	168	100	>120	45
18	28 (ARN24928)	209	100	>120	>60
19	29	1	ND	>120	>60

^aND = not determined.

^bCompound initially reported by Jahid et al.¹²

Microscale Thermophoresis and NMR for CDC42-Binding Validation

In order to evaluate target engagement of the most interesting compounds, we exploited two different techniques: microscale thermophoresis (MST) and NMR analysis. Among the selected 19 compounds, 2, 4, 5, 13, 15, and 29 were excluded from the analysis due to their low kinetic solubility or aggregation at a concentration of < 50 μM (Table 6).

MST is a biophysical technique that saw a rapid increase in popularity in the past few years, especially as a method for fragment screening. This was mainly due to the low amount of sample required, short run time, absence of sample-immobilization requirements, and the ability to detect binders in a wide range of affinities (both strong and weak binders). Target proteins are Red-NHS labeled to monitor their change in thermophoresis in a temperature gradient that is produced by an IR-laser on a small area of the sample solution. The interaction with a small molecule leads to a shift in the protein thermophoresis, which is measured by an UV detector and given as the output.

In this MST analysis, wild-type His-CDC42 (Ile4-Pro182) was used as target macromolecule. The fusion protein was expressed, purified, and loaded with GDP (90% efficiency of loading) or GppNHp (>98% efficiency of loading), as previously described.¹² The difference in normalized fluorescence (ΔF_norm [‰] = F_hot/F_cold) between a protein–compound sample and a protein-only sample was registered. Binding events were tested at a compound concentration of 50 μM throughout all the experiments (Figure 3 and Table S1, Supporting Information). Assays were set up in Tris–HCl buffer (for the binding to His-CDC42 loaded with GppNHp or GDP) and phosphate-buffered saline (PBS) buffer (for the binding to His-CDC42 loaded with GppNHp only).

Figure 3.

(A,B) MST analysis of selected compounds. An example of binding check by MST [compounds 24-(A) and 28-(B)]. The MST traces were recorded as follows: 3 s MST power off, 20 s MST power on, and 1 s MST power off. The amplitude of the response (A) and the signal-to-noise ratio (S/N) at 1.5−2.5 s were taken into account to evaluate binding. Compound 24 thermophoresis experiment was set up in Tris buffer pH 7.5 for the evaluation of the binding to inactive His-CDC42 (loaded with GDP). Compound 28 thermophoresis experiment was set up in PBS buffer for the evaluation of the binding to active His-CDC42 (loaded with GppNHp). (C) graph displays the response amplitude of the experiment setup in Tris buffer. The amplitude of the response is the difference in normalized fluorescence (ΔF_norm [‰] = F_hot/F_cold) between the protein/compound sample and a protein-only sample. Compounds (C) were tested at 50 μM toward the activated (loaded with GppNHp) or inactivated (loaded with GDP) His-CDC42. A single blue asterisk indicates a signal-to-noise ratio above 5, while double red asterisks indicate a signal-to-noise ratio above 12. (D−F) examples of binding check by NMR of compound 22 [¹H NMR T2 filter-(D)], compound 27 [¹⁹F T2 filter-(E)], and compound 12 [WaterLOGSY-(F)]. Compounds were tested: in the absence of protein, while in the presence of GppNHp (black) or GDP (gray), and in the presence of activated (loaded with GppNHp) His-CDC42 (red) or inactivated (loaded with GDP) His-CDC42 (green). The arrows indicate where a difference in the compound NMR signal is observed after the addition of the protein, highlighting a binding event.

Three compounds were highlighted for their confidence in binding and signal-to-noise ratio higher than 12 (Table S1, Supporting Information): 17,¹² the previously identified lead compound, 10, and 25. All three compounds carry a phenyl ring on C6 of the pyrimidine and a 4-piperidine cycle on C2. The aniline moiety is instead substituted at position 3 and/or 4 with either a dimethylamine group or a methoxy group. Additional binders were identified with a lower signal-to-noise ratio (between 5 and 12). They are 3, 12, 14, 22, 27, and 28 (Figure 3 and Table S1, Supporting Information).

To further strengthen the binding evidence obtained through MST, we evaluated the binding of our derivatives by NMR.²³ Here, it was employed as a secondary assay to confirm the binders identified by MST: derivatives 3, 10, 12, 14, 17, 22, 25, 27, and 28. First, the stability and the aggregation state of both (loaded with GppNHp) His-CDC42 and (loaded with GDP) His-CDC42 proteins were assessed by the ¹⁹F transverse relaxation (T₂) filter²⁴ testing a mixture of 35 fluorinated fragments (soluble and not aggregating) in the absence and in the presence of the proteins just after purification (t₀) and 24 h later (t₁). No differences in the NMR spectra of the mixture in the absence and in the presence of proteins are visible at both t₀ and t₁, suggesting that there are no diffused bindings, hence no aggregation, and that the protein is stable at least for 24 h under our experimental conditions (Table 6 and Figure S1, Supporting Information). The MST binders were then tested in the absence and in the presence of both His-CDC42 by ¹H or ¹⁹F (fluorinated compounds) T₂ filters²⁴ and WaterLOGSY (water-ligand observation with gradient spectroscopy) experiments.²⁵ In T₂ filter experiments, the NMR signals of a compound that binds the protein show line broadening in the presence of the protein, resulting in a reduction in their intensities, as reported in the examples of Figures 3D and 3E as ¹H and ¹⁹F T₂ filter spectra, respectively. The WaterLOGSY experiment is based on magnetization transfer from bulk water to the compounds that interact with a macromolecule. In the presence of binding, the WaterLOGSY signal of the compound changes from negative to less negative or to positive, as shown in the example of Figure 3F. As reported in Table S1 in the Supporting Information, MST hits were confirmed by NMR, but according to NMR data, compounds 3, 14, 22, and 28 bind also to the inactivated form of His-CDC42. This result is not surprising as it is well known that NMR is highly sensitive even to very weak binders.²⁶

Molecular Modeling for Binding Validation

The previously identified drug binding pocket localized at the CDC42-PAK protein–protein interface was used for our computational studies.¹² Accordingly, we performed molecular docking calculations of the follow-up lead molecules 27 (ARN25062) and 28 (ARN24928) on the GTP-bound active form of CDC42.

Modeling predicts that both 27 (ARN25062) and 28 (ARN24928) fit within the effector pocket of CDC42 (Figure 4A,B), which is consistent with the binding mode of 17 (ARN22089) and the hit 1 binding mode in CDC42, as previously reported (Figure 4).¹²

Figure 4.

Model structure of compounds 27 [ARN25062, panel (A)] and 28 [ARN24928, panel (B)] bound to the allosteric drug-binding pocket of CDC42 is reported. The structure of CDC42 is represented as the white surface, while the identified drug-binding pocket is shown in both sticks and transparent blue surface. ARN25062 and ARN24928 are reported as green sticks and yellow, respectively.

The hydrophobic nature of the pocket easily accommodates the phenyl on C6, which represents the anchor point for binding of both the lead 17 (ARN22089) and for our back-up compounds, as shown in Figure S2. Noteworthily, a different orientation of the piperidine portion in the binding mode with respect to the other active compounds has been predicted for compound 27 (ARN25062), where the accommodation of the phenyl ring is maintained, while the positioning of the pyridine core is inverted (Figure 4). This is probably due to the close positively charged Lys5, which attracts the electronegative group −CF₃ (Figure S3). However, further structural analyses are needed in order to determine the accurate compound binding mode. In this regard, to further assess the predicted binding poses, we performed equilibrium molecular dynamics (MD) simulations of both 27 (ARN25062) and 28 (ARN24928) in complex with CDC42 (Figure 5).

Figure 5.

MD simulations of the protein–ligand complexes. The structural representation of CDC42 in complex with compound 27 [ARN25062, panel (A)] or compound 28 [ARN24928, panel (B)] is reported on the left. CDC42 is represented as the schematic, while the binding pocket is highlighted as both the sticks and transparent blue surface. Multiple MD snapshots of the 27 (ARN25062, green) and 28 (ARN24928, yellow) binding poses are shown as sticks. On the right, the rmsd over time for the CDC42-ligand complexes is reported with the rmsd running averages highlighted in bold.

As for the previously reported hit and lead compounds,¹² the binding pose of both 27 (ARN25062) and 28 (ARN24928) in the CDC42-ligand complexes is preserved during 500 ns-long MD simulations (rmsd = 3.90 ± 1.30 Å and 3.80 ± 1.64 for 27 (ARN25062) and 28 (ARN24928) CDC42 complexes, respectively (Figure 5A,B, right panels). Indeed, as shown in Figure 5 (left panels), 27 (ARN25062) and 28 (ARN24928) steadily bind the target pocket throughout the MD simulations.

In Vivo Pharmacokinetics of the Selected Follow-Up/Back-Up Leads 27 and 28

On the basis of the overall results and drug-like profile of our compounds, 27 (ARN25062) and 28 (ARN24928) were selected as candidates for in vivo pharmacokinetics (PK) studies, in view of further experiments to assess their in vivo efficacy in animal models (mouse species) of cancer and also compare the PK profile to the lead’s one (17, ARN22089, Figure 6).

Figure 6.

Mouse PK profiles of ARN25062, ARN24928, and ARN22089¹² following intravenous (I.V.) and oral (P.O) administration at 3 and 10 mg/kg, respectively, and the corresponding observed and calculated PK parameters.

We tested two different routes of administration: (i) I.V. injection at a concentration of 3 mg/kg (n = 3 animals for each time point), and ii) oral (P.O.) treatment at a dose of 10 mg/kg (n = 3 animals for each time point, Figure 6). The two back-ups showed PK profiles comparable to ARN22089, as previously evaluated.¹² During the PK studies, via either I.V. or P.O. administration, ARN25062 and ARN24928 were well tolerated by all animals, and no treatment-related toxicological signs were observed. Indeed, I.V. profiles of ARN25062 and ARN24928 reached a C_max of 1298 and 889 ng/mL in 5 min after administration, respectively, followed by a protracted elimination phase. Both compounds were still detectable after 2 h at a concentration of 87 and 17 ng/mL, respectively. After oral administration (10 mg/kg), ARN25062 achieved the maximum concentration (C_max = 216 ng/mL) in 1 h, significantly faster than compound ARN24928 (C_max = 86 ng/mL and T_max = 2 h), and both compounds were still detectable after 8 h at concentrations of 13 and 25 ng/mL, respectively. In particular, compound ARN24928 showed good exposure with a long half-life of 37 min (I.V.) and 184 min (P.O.). In summary, these data indicate that both compounds are well tolerated after a single injection and possess a favorable PK profile, comparable to that of the lead ARN22089.¹²

In Vivo Efficacy of the Selected Follow-Up/Back-Up Leads 27 and 28

We previously showed that ARN22089 significantly inhibited the growth of patient-derived xenografts (PDXs) in NOD scid gamma (NSG) mice at 25 mg/kg I.V. twice a week.¹² Similarly, to test the efficacy of 27 (ARN25062) and 28 (ARN24928) in vivo, we inoculated PDX tumor chuck underneath the skin in NSG mice on both flanks and treated the mice with 10 mg/kg ARN25062 or ARN24928 or vehicle via I.V. daily for a week. We included the lead compound, ARN22089, to directly compare the efficacy of the two new derivatives with the initial lead’s one. ARN25062 significantly inhibited tumor growth, with similar efficacy as the lead compound (Figure 7A), compared to ARN24928, which had less efficacy (Figure 7A). To determine the efficacy of longer term administration on tumor growth, we extended the treatment for ARN25062 for a total of 2 weeks of daily I.V. administration at a dose of 10 mg/kg. After 2 weeks, the tumors were harvested, and the volume (mm³) was calculated for each tumor. 2 week daily treatment reduced tumor volume significantly for ARN25062 compared to vehicle (Figure 7B). Both ARN25062 and ARN22089 had a similar effect after 2 weeks, and there was no body weight change or overt toxicity observed in the drug-treated animals (data not shown).

Figure 7.

ARN25062 but not ARN24928 significantly inhibits tumor growth in the PDX model. (A) When initial tumors reached 150–250 mm³, mice were injected I.V. with 10 mg/kg ARN25062, ARN24928, or ARN22089 (4 tumors per group) or with vehicle (3 tumors per group), every day for a week. Values are given as individual tumor volume, error bars as SEM, two-tail T-test, *p-value ≤ 0.05, ^nsp-value ≤ 0.125. (B) Mice were treated with 10 mg/kg ARN25062 or with vehicle daily for 2 weeks. Tumors were extracted, and volumes were calculated. Values are tumor volumes (mm³) from ARN25062 (n = 6) or vehicle (n = 6), error bars as SD, two-tail T-test, p-value ≤ 0.0025.

Chemistry

We obtained the final targets 13–15, 20, and 23–27 using the five-step synthetic route outlined in Scheme 1. The first Suzuki cross-coupling reaction between commercial 2,4,6-trichloropyrimidine 30 and phenylboronic acid in the presence of PdCl₂(dppf)CH₂Cl₂ and K₂CO₃ afforded intermediate 31 with the phenyl group on C6 with 72% yield. The regioselectivity of the reaction was driven using a mild temperature of 60 °C in 1,4-dioxane under microwave irradiation for 1 h. The yield of the transformation was partially affected by the formation with the time of the undesired 2,4-diphenylpyrimidine byproduct. The regioselective amination on C4 of dichloropyridine core 31 with suitable functionalized anilines 32a–i was achieved through two different types of transformation: the C4-regioselective nucleophilic aromatic substitution employing LiHDMS as a strong base in tetrahydrofuran (THF) at −60 °C or Buchwald–Hartwig reaction using Pd(OAc)₂, rac-(BINAP), and Cs₂CO₃ in 1,4-dioxane at 60 °C under microwave irradiation.²⁷ The first procedure accessed di-substituted pyrimidine 33a and 33e–i with an excellent 80–92% range yield, while the second type of reaction generated compounds 33b−d with a 63–67% range yield, as previously reported.¹² The second Suzuki cross-coupling between suitable boronic esters 34a and 34b and the intermediate of type 33 in the presence of PdCl₂(dppf)CH₂Cl₂ and K₂CO₃ in 1,4-dioxane at 120 °C in microwave yielded trisubstituted pyrimidine of type 35 with a very good 60–89% range yield. Finally, the reduction of the double C–C bond of cyclohexenyl or dihydropyridine moieties on C2 and final Boc deprotection get easy access to final desired compounds. Except for compound 35aa, the double-bond reduction was reduced in the presence of ammonium formate and Pd(OH)₂/C in MeOH at 80 °C. Since a dechlorination side reaction occurred under this reductive condition, a milder procedure with Et₃SiH, Pd/C at −10 °C in a 3:3:1 mixture toluene/ethyl acetate/EtOH was used to obtain protected precursor 36aa from 35aa with 23% yield after chromatography purification. The low yield was due to the incomplete conversion of the starting material and the low formation of the undesired dechlorinated byproduct (5%). The final Boc deprotection with HCl (4 M in dioxane) afforded final derivatives 13–15 and 23–27 with 16–96% range yield.

Scheme 1. Synthesis of Additional Trisubstituted Pyrimidine Derivatives 13–15, 20, and 23–27.

Reagents and conditions: (a) PdCl₂(dppf)CH₂Cl₂, K₂CO₃ (2 M)_aq, 1,4-dioxane dry, 60 °C MW; (b) (i) LiHMDS, aniline 32a–i, THF dry, −60 to 0 °C or (ii) Pd(OAc)₂, rac-(BINAP), Cs₂CO₃, 1,4-dioxane dry, 60 °C, MW; (c) 34a–b, PdCl₂(dppf)CH₂Cl₂, K₂CO₃ (2 M)_aq, 1,4-dioxane dry, 120 °C MW; (d) (i) NH₄COOH, Pd(OH)₂/C, MeOH, 80–90 °C or (ii) Et₃SiH, Pd/C, from −10 to 0 °C; and (e) HCl (4 M in dioxane), 1,4-dioxane.

As shown in Scheme 2, derivatives 21 and 22 were obtained through direct functionalization of free piperidine of compound 17 via amide coupling with isobutyric acid and nucleophilic substitution on 2-bromoethyl methyl ether, respectively.

Scheme 2. Synthesis of Final Compounds 21 and 22.

Reagents and conditions: (a) IprCOOH, HATU, DIPEA, DMF and (b) BrCH₂CH₂OMe, DIPEA, CH₃CN, 60 °C.

The preparation of the two triazine analogues required a different specific strategy due to the peculiar reactivity of cyanuric chloride (2,4,6-trichlorotriazine) 37 (Scheme 3). It is well known that the C–Cl bond reactivity of the triazine substrate subsequently decreases from cyanuric chloride 37 to monochlorotriazine toward both metal cross-coupling and aromatic nucleophilic substitution transformations.^28,29 Accordingly, the orthogonal chemoselective substitution of thrichlorotriazine with the three specific functionalities of final compounds 28 and 29 was carefully evaluated, changing the reaction conditions and the order of synthetic sequence.

Scheme 3. Synthesis of Triazine Analogue 28.

Reagents and conditions: (a) Pd(PPh₃)₂Cl₂, K₂CO₃, toluene dry, 60 °C; (b) Pd(PPh₃)₂Cl₂, K₂CO₃, toluene dry, 80 °C; (c) aniline 32d, DIPEA, THF dry, 80 °C; (d) i) NH₄COOH, Pd(OH)₂/C, MeOH, 80–90 °C; and (e) HCl (4 M in dioxane), 1,4-dioxane.

As depicted in Scheme 3, the synthesis toward triazine derivative 28 began from inexpensive cyanuric chloride 37. The first Suzuki reaction with the simple boronic acid occurred in the presence of Pd(PPh₃)₂Cl₂ and K₂CO₃ in toluene dry at 60 °C, yielding monophenyl-substituted 38 with 51% yield. At first, the second Suzuki between intermediate 38 and boronic ester 34a was attempted, but the reaction did not give any conversion into desired product 40 regardless of the type of catalyst (Pd(dppf)Cl₂·DCM, Pd(PPh)₃Cl₂, and Pd(PPh₃)₄), the base (K₂CO₃ and K₃PO₄), or the temperature (80, 100, and 120 °C) employed. Thus, we changed the intrinsic reactivity of the boronic partner, transforming the boronic ester 34a into corresponding boronic acid 39 with NaIO₄ and NH₄OAc in the acetone/water mixture (see the Experimental Section). Notably, boronic acid 39 selectively and efficiently reacted with phenyl triazine 38 to give di-substituted product 40 by using PdCl₂(PPh₃)₂ and K₂CO₃ in toluene at 80 °C with 71% yield. Finally, the aniline moiety was introduced by exploiting a nucleophilic aromatic substitution between aniline 32d and monochloro triazine 40 with DIPEA in THF at 80 °C, affording trisubstituted triazine 41 as a mixture of isomers with 94% combined yield. Indeed, under these conditions, the double C–C bond underwent partial isomerization from C1′–C2′ positions to C2′–C3′ position, but this side reaction was irrelevant since both isomeric forms of 41 were reduced to related saturated derivatives with NH₄COOH and Pd(OH)₂/C. The final Boc deprotection gave desired analogue 28 with 51% yield after two steps.

Synthesis of analogue 29 required a different synthetic sequence due to the weaker nucleophilic nature of aniline 32i compared to 32d. Every attempt of aromatic nucleophilic substitution with 32i on the advanced intermediate 40 failed in the presence of different bases (NaH, LiHMDS, and DIPEA) and solvents (acetone, THF, and DCM). The alternative Buchwald reaction with NaOtBu and Pd₂(dba)₃ gave impure conversion, too. To overcome these issues, we introduce aniline 32i at the beginning of the synthetic route on trichlorotriazine 37, which is more prone to undergo nucleophilic substitution than mono- and di-substituted triazine (Scheme 4). In confirmation of this, reaction between aniline 32i and trichlorotrizine 37 under mild conditions (DIPEA and DCM at 0 °C) afforded desired substrate 42 with 24% yield. Then, two sequential Suzuki couplings with boronic acid 39 and phenyl boronic acid were successfully performed through a proper stoichiometry of the reagent and temperature in the presence of PdCl₂(PPh₃)₂ and K₂CO₃, giving advanced intermediate 44 with 21% yield after two steps (Scheme 3). Given the high selectivity in stepwise functionalization of dichloro triazine, we developed a one-pot procedure for obtaining intermediate 44 starting by pure compound 42, improving the two-step yield from 21 to 54% (Scheme 5). Finally, after the classic double C–C reduction and Boc removal, we obtained final compound 29.

Scheme 4. Synthesis of Intermediate 44 toward Final Compound 29.

Reagents and conditions: (a) aniline 32i, DIPEA, dichloromethane (DCM), 0 °C; (b) Pd(PPh₃)₂Cl₂, K₂CO₃ (2 M)_aq, toluene dry, 90 °C; and (c) phenyl boronic acid, Pd(PPh₃)₂Cl₂, K₂CO₃ (2 M)_aq, toluene dry, 120 °C.

Scheme 5. One-Pot Synthesis Optimization toward Compound 29.

Reagents and conditions: (a) Pd(PPh₃)₂Cl₂, K₂CO₃ (2 M)_aq, toluene dry, 90 °C; then (b) phenyl boronic acid, Pd(PPh₃)₂Cl₂, K₂CO₃ (2 M)_aq, toluene dry, 120 °C; (c) (i) NH₄COOH, Pd(OH)₂/C, MeOH, 80–90 °C; and (d) HCl (4 M in dioxane), 1,4-dioxane.

In summary, through the optimization of the key Suzuki and aromatic nucleophilic substitution reaction, we built up a simple and efficient five-step synthetic route to get access to a widely diversified class of new pyrimidine and triazine RHOJ/CDC42 inhibitors.

Conclusions

Here, we have presented additional efforts of our drug discovery program targeting the CDC42 GTPase (RHOJ, CDC42, and RHOQ) proteins, which are overexpressed in multiple tumor types. Based on the recent discovery of a lead compound 17, ARN22089,¹² which has shown anticancer activity in vivo, we have expanded this new chemical class of RHOJ/CDC42 inhibitors. Importantly, we have identified and characterized two follow-up/back-up compounds, namely, 27 (ARN25062) and 28 (ARN24928), out of an SAR generated with ∼30 close analogues bearing different substituents at the pyrimidine and triazine core of this new chemical scaffold (Figure 2).

The resulting SAR elucidates the key structural features that enhance the potency and ameliorate the drug-like profile of our new trisubstituted pyrimidine/triazine derivatives. Indeed, we explored three points of modification of the pyrimidine class. The first subset includes compounds 1–5 with several substituents on C6; then, we explored the aniline moiety on C4, synthesizing 6–13 and 23–27. We evaluated the effect of different aliphatic (hetero)cycles on C2 with compounds 14–22. This study identified the best functionalities on a pyrimidine core that were also retained on the symmetric triazine core, like in 28–29. The nineteen most potent compounds were further investigated for their kinetic solubility and in vitro plasma and metabolic stability.

The measured antiproliferative activity and in vitro binding data at the target, together with our extensive analyses of the drug-likeness profile, indicate the novel derivatives 27 (ARN25062) and 28 (ARN24928) as the most drug-like CDC42/RHOJ inhibitors of this novel chemical series, which are comparable to the previously identified drug candidate 17 (ARN22089). These back-up/follow-up compounds have also displayed a stable binding at the target pocket through MD simulations and a favorable PK profile comparable to 17 (ARN22089).

Notably, we also measured the in vivo efficacy of the two lead compounds 27 and 28, with analogue 27 (ARN25062) that has exhibited a significant ability to inhibit tumor in PDXs in vivo, with an efficacy comparable to the lead 17.¹² These overall results and extended drug discovery effort further demonstrate the potential of our new chemical class of CDC42/RHOJ inhibitors, with lead compounds ready for advanced preclinical studies to develop new cancer treatments.

Experimental Section

Chemistry

Chemistry General Considerations

All the commercially available reagents and solvents were used as purchased from vendors without further purification. Dry solvents were purchased from Sigma-Aldrich. Automated column chromatography purifications were done using a Teledyne ISCO apparatus (CombiFlash Rf) with pre-packed silica gel columns of different sizes (from 4 g up to 24 g) and mixtures of increasing polarity of cyclohexane and ethyl acetate (EtOAc) or DCM and methanol (MeOH). NMR data were collected on 400 or 600 MHz (¹H) and 100 or 150 MHz (¹³C). Spectra were acquired at 300 K, using deuterated dimethyl sulfoxide (DMSO-d₆) or deuterated chloroform (CDCl₃) as solvents. For ¹H NMR, data are reported as follows: chemical shift, multiplicity (s = singlet, d = doublet, dd = double of doublets, t = triplet, q = quartet, and m = multiplet), coupling constants (Hz), and integration. Ultra-performance liquid chromatography-mass spectrometry (UPLC-MS) analyses were run on a Waters ACQUITY UPLC-MS instrument consisting of a single quadrupole detector (SQD) mass spectrometer. The analyses were performed on an ACQUITY UPLC BEH C₁₈ column (50 × 2.1 mm ID and particle size 1.7 μm) with a VanGuard BEH C₁₈ pre-column (5 × 2.1 mm ID and particle size 1.7 μm) (Log D > 1). The mobile phase was 10 mM NH₄OAc in H₂O at pH 5 adjusted with AcOH (A) and 10 mM NH₄OAc in CH₃CN–H₂O (95:5) at pH 5 (B). Electrospray ionization in positive and negative modes was applied in the mass scan range 100–500 Da. Depending on the analysis method used, a different gradient increasing the proportion of mobile phase B was applied. For analysis method 1, the mobile phase B proportion increased from 5 to 95% in 2.5 min. For analysis method 2, the mobile phase B proportion increased from 50 to 100% in 2.5 min. For analysis method 3, the mobile phase B proportion increased from 70 to 100% in 2.5 min. High-resolution mass spectrometry (HRMS) was carried out on a Waters Synapt G2 Quadrupole-TOF instrument equipped with an ESI ion source. The analyses were run on an ACQUITY UPLC BEH C₁₈ column (50 × 2.1 mm ID and particle size 1.7 μm), using H₂O + 0.1% HCOOH (A) and CH₃CN + 0.1% HCOOH as the mobile phase. The QC analysis was performed starting from a 10 mM stock solution of the test compound in DMSO-d₆ and further diluted 20-fold with CH₃CN–H₂O (1:1) for analysis. The QC analyses were run on a Waters ACQUITY UPLC-MS system consisting of an SQD mass spectrometer equipped with an electrospray ionization interface and a photodiode array detector. Electrospray ionization in positive and negative modes was applied in the mass scan range 100–500 Da. The PDA range was 210–400 nm. The analyses were run on an ACQUITY UPLC BEH C₁₈ column (100 × 2.1 mm ID and particle size 1.7 μm) with a VanGuard BEH C₁₈ pre-column (5 × 2.1 mm ID and particle size 1.7 μm). The mobile phase was 10 mM NH₄OAc in H₂O at pH 5 adjusted with AcOH (A) and 10 mM NH₄OAc in CH₃CN–H₂O (95:5) at pH 5 (B) with 0.5 mL/min as flow rate. A linear gradient was applied: 0–0.2 min: 10% B, 0.2–6.2 min: 10–90% B, 6.2–6.3 min: 90–100%, 6.3–7.0 min: 100% B (QC method). All final compounds displayed ≥95% purity as determined by NMR and UPLC-MS analysis (unless otherwise indicated). Compounds 1–8 were purchased from Sigma-Aldrich and used as such without further purification. Compounds 9–12 and 16–19 and intermediates 33b–d and 35–36da were synthesized as previously reported.¹²

General Procedure 1: Preparation of Compounds of Type 33 via Regioselective Nucleophilic Aromatic Sostitution

A solution (0.2 M) of 6-phenyl-2,4-dichloropyrimidine 30 (1.00 equiv) with a suitable aniline of type 32 (1.00 mmol) in THF was cooled to −60 °C. To this solution was added dropwise LiHMDS (1.0 M in THF, 2.5 equiv). After 1 h, water was added, and the mixture was extracted with EtOAc. The combined organic layers were washed with brine, dried (Na₂SO₄), and concentrated in vacuo. Final normal-phase chromatographic purification (cyclohexane/EtOAc) provided the desired product of type 33.

General Procedure 2: Preparation of Compounds of Type 33 Via Buchwald–Hartwig Amination

A mixture of Pd(OAc)₂ (0.05 equiv) and rac-BINAP (12.5 mg, 0.02 mmol, 0.05 equiv) in 1,4-dioxane (0.02 M solution) was stirred under Ar flushing for 10 min. Then, a solution (1 M) of intermediate 31 (1 equiv) and a solution (1 M) of suitable aniline of type 32 (1 equiv) in 1,4-dioxane (1 M mL) and Cs₂CO₃ (1.2 equiv) were stepwise added. The reaction mixture was stirred in a CEM microwave apparatus at 60 °C (200 W) for 4 h, filtrated through a Celite coarse patch, rinsed with DCM, and concentrated to dryness at low pressure. Final normal-phase chromatographic purification afforded the pure desired compound of type 33.

General Procedure 3: Preparation of Compounds of Type 35 through Suzuki Cross-Coupling Reaction

A suspension of intermediate 33 (1 equiv), 4,4,5,5-tetramethylboronate of type 34 (1.2 equiv), dichloro[1,1′-bis(diphenylphosphino)ferrocene] palladium DCM complex (0.2 equiv), and K₂CO₃ 2 M solution (3 equiv) in 1,4-dioxane (0.04 M solution) was stirred in a CEM microwave apparatus at 120 °C (200 W) for 2 h. The resulting crude was portioned between DCM and NaHCO₃ saturated solution, and the organic layer was dried over Na₂SO₄ and concentrated to dryness at low pressure. Final normal-phase chromatographic purification (cyclohexane/EtOAc) afforded the pure title compound of type 35.

General Procedure 4. Preparation of Compounds of Type 36 or 20 after C=C Double-Bond Reduction

Under a N₂ atmosphere, a suspension of the intermediate of type 35 (1 equiv), ammonium formate (4 equiv), and Pd(OH)₂/C (20% of starting material weight) in MeOH dry (0.04 M solution) was stirred at reflux temperature until reaction completion. The catalyst was filtered off through a Celite coarse patch, and the resulting filtrate was concentrated to dryness at low pressure. Final chromatographic normal-phase purification (cyclohexane/EtOAc) afforded the pure desired compound of type 36.

General Procedure 5: Final Boc Deprotection

To a 0 °C solution of the protected intermediate of type 36 (1 equiv) in 1,4-dioxane (0.02 M solution) was dropwise added a HCl (4 M) solution in 1,4-dioxane (10 equiv), and the reaction mixture was stirred at room temperature until completion of the conversion (from 4 to 24 h); then, the reaction crude was concentrated to dryness at low pressure, the resulting crude was portioned between EtOAc and NaOH 0.1 M, and the organic layer was dried over Na₂SO₄ and concentrated to dryness at low pressure. Final chromatographic purification by silica or alumina (DCM/MeOH NH₃ 1 M) afforded pure final compounds.

3-(4-((4-Chlorophenyl)amino)-6-(pyridin-4-yl)pyrimidin-2-yl)piperidin-1-ium Chloride (1)

Compound 1 was purchased from Aldrich. HRMS (ESI) m/z calcd for C₂₀H₂₁ClN₅ [M + H]⁺, 366.1485; found, 366.1483. ¹H NMR (400 MHz, DMSO-d₆): δ 10.44 (s, 1H), 9.01 (bs, 2H), 9.00–8.94 (m, 2H), 8.44–8.34 (m, 2H), 7.83–7.73 (m, 2H), 7.46 (s, 1H), 7.44–7.40 (m, 2H), 3.73–3.61 (m, 2H), 3.33–3.21 (m, 2H), 2.93 (m, 1H), 2.25 (m, 1H), 1.95–1.67 (m, 3H).

N-(4-Chlorophenyl)-2-(piperidin-3-yl)-6-(pyridin-3-yl)pyrimidin-4-amine (2)

Compound 2 was purchased from Aldrich. HRMS (ESI) m/z: calcd for C₂₀H₂₁ClN₅ [M + H]⁺, 366.1485; found, 366.1481. ¹H NMR (400 MHz, DMSO-d₆): δ 9.85 (s, 1H), 9.18 (dd, J = 2.3, 0.9 Hz, 1H), 8.68 (dd, J = 4.8, 1.6 Hz, 1H), 8.37 (dt, J = 8.1, 1.9 Hz, 1H), 7.83–7.74 (m, 2H), 7.55 (ddd, J = 8.0, 4.8, 0.9 Hz, 1H), 7.43–7.34 (m, 2H), 7.10 (s, 1H), 3.22 (d, J = 10.8 Hz, 1H), 2.91 (d, J = 12.4 Hz, 1H), 2.88–2.70 (m, 2H), 2.10 (d, J = 12.8 Hz, 1H), 1.76 (qd, J = 12.6, 3.8 Hz, 1H), 1.70–1.61 (m, 1H), 1.48 (qt, J = 12.5, 3.6 Hz, 1H). Some signals overlap with water.

N-(4-Chlorophenyl)-2-(piperidin-3-yl)-6-(pyridin-2-yl)pyrimidin-4-amine (3)

Compound 3 was purchased from Aldrich. HRMS (ESI) m/z: calcd for C₂₀H₂₁ClN₅ [M + H]⁺, 366.1485; found, 366.1483. [M + H]⁺ calcd. ¹H NMR (600 MHz, DMSO-d₆): δ 9.85 (s, 1H), 8.70 (ddd, J = 4.8, 1.8, 0.9 Hz, 1H), 8.39 (dt, J = 7.9, 1.1 Hz, 1H), 7.98 (td, J = 7.7, 1.8 Hz, 1H), 7.82–7.75 (m, 2H), 7.64 (s, 1H), 7.52 (ddd, J = 7.5, 4.7, 1.2 Hz, 1H), 7.42–7.37 (m, 2H), 3.27 (d, J = 11.2 Hz, 1H), 2.95 (dd, J = 12.4, 3.8 Hz, 1H), 2.86 (tt, J = 10.4, 3.4 Hz, 1H), 2.83–2.76 (m, 1H), 2.54 (dd, J = 12.0, 3.0 Hz, 1H), 2.15–2.11 (m, 1H), 1.82–1.76 (m, 1H), 1.69 (dp, J = 13.4, 3.4 Hz, 1H), 1.52 (qt, J = 12.7, 3.9 Hz, 1H). ¹³C NMR (151 MHz, DMSO-d₆): δ 171.8 (Cq), 161.6 (Cq), 161.0 (Cq), 154.4 (Cq), 150.0 (CH), 139.6 (Cq), 137.9 (CH), 129.1 (CH, 2C), 126.1 (Cq), 125.8 (CH), 121.5 (CH, 2C), 121.4 (CH), 100.8 (CH), 51.4 (CH₂), 46.5 (CH₂), 46.3 (CH), 30.3 (CH₂), 26.3 (CH₂).

N-(4-Chlorophenyl)-6-phenyl-2-(piperidin-3-yl)pyrimidin-4-amine (4)

Compound 4 was purchased from Aldrich. HRMS (ESI) m/z: calcd for C₂₁H₂₂ClN₄ [M + H]⁺, 365.1533; found, 365.1529. ¹H NMR (400 MHz, DMSO-d₆): δ 9.78 (s, 1H), 0.8.04–8.02 (m, 2H), 7.84–7.75 (m, 2H), 7.53–7.51 (m, 3H), 7.42–7.34 (m, 2H), 7.08 (s, 1H), 3.35–3.29 (m, 1H), 3.01 (d, J = 12.3 Hz, 1H), 2.94–2.84 (m, 2H), 2.60 (td, J = 12.0, 2.6 Hz, 1H), 2.15–2.12 (m, 1H), 1.83–1.67 (m, 2H), 1.56 (qt, J = 12.3, 3.4 Hz, 1H). ¹³C NMR (101 MHz, DMSO-d₆): δ 170.8 (Cq), 161.4 (Cq), 161.0 (Cq), 139.2 (Cq), 137.2 (Cq), 130.3 (CH), 128.9 (CH, 2C), 128.6 (CH, 2C), 126.5 (CH, 2C), 125.5 (Cq), 121.0 (CH, 2C), 99.9 (CH), 50.2 (CH₂), 45.6 (CH₂), 45.1 (CH), 29.4 (CH₂), 25.1 (CH₂).

N-(4-Chlorophenyl)-6-(4-methoxyphenyl)-2-(piperidin-3-yl)pyrimidin-4-amine (5)

Compound 5 was purchased from Aldrich. HRMS (ESI) m/z: calcd for C₂₂H₂₄N₄OCl [M + H]⁺, 395.1635; found, 395.1639. ¹H NMR (600 MHz, DMSO-d₆): δ 9.71 (s, 1H), 8.05–7.95 (m, 2H), 7.80–7.74 (m, 2H), 7.41–7.34 (m, 2H), 7.10–7.04 (m, 2H), 7.01 (s, 1H), 3.83 (s, 3H), 3.36–3.32 (m, 1H), 3.04 (td, J = 12.4, 3.5 Hz, 1H), 2.94–2.85 (m, 2H), 2.64 (td, J = 12.1, 3.0 Hz, 1H), 2.17–2.09 (m, 1H), 1.82–1.70 (m, 2H), 1.63–1.50 (qt, J = 12.4, 3.7 Hz, 1H). ¹³C NMR (151 MHz, DMSO-d₆): δ 170.4 (Cq), 161.1 (Cq), 161.0 (Cq), 160.9 (Cq), 139.2 (Cq), 129.4 (Cq), 128.6 (CH, 2C), 128.0 (CH, 2C), 125.4 (Cq), 120.9 (CH, 2C), 114.2 (CH, 2C), 98.7 (CH), 55.3 (CH₃), 49.8 (CH₂), 45.3 (CH₂), 44.7 (CH), 29.2 (CH₂), 24.7 (CH₂).

N-(3-Chlorophenyl)-2-(piperidin-3-yl)-6-(pyridin-4-yl)pyrimidin-4-amine (6)

Compound 6 was purchased from Aldrich. HRMS (ESI) m/z: calcd for C₂₀H₂₁N₅Cl [M + H]⁺, 366.1485; found, 366.1488. ¹H NMR (400 MHz, DMSO-d₆): δ 10.02 (s, 1H), 8.79–8.70 (m, 2H), 8.13 (t, J = 2.1 Hz, 1H), 7.97–7.92 (m, 2H), 7.59 (dd, J = 8.0, 2.0 Hz, 1H), 7.37 (t, J = 8.1 Hz, 1H), 7.20 (s, 1H), 7.06 (m, J = 7.9, 2.0 Hz, 1H), 3.32 (d, J = 10.6 Hz, 1H), 2.99 (d, J = 12.4 Hz, 1H), 2.95–2.81 (m, 2H), 2.56 (t, J = 10.2 Hz, 1H), 2.16–2.13 (m, 1H), 1.88–1.74 (m, 1H), 1.74–1.66 (m, 1H), 1.54 (qd, J = 12.3, 3.9 Hz, 1H). ¹³C NMR (101 MHz, DMSO-d₆): δ 171.5 (Cq), 161.0 (Cq), 159.2 (Cq), 150.5 (CH, 2C), 144.3 (Cq), 141.5 (Cq), 133.1 (Cq), 130.4 (CH), 121.7 (CH), 120.6 (CH, 2C), 119.0 (CH), 117.7 (CH), 101.4 (CH), 50.5 (CH₂), 45.83 (CH₂), 45.5 (CH), 29.54 (CH₂), 25.39 (CH₂).

N-(4-Fluorophenyl)-2-(piperidin-3-yl)-6-(pyridin-4-yl)pyrimidin-4-amine (7)

Compound 7 was purchased from Aldrich. HRMS (ESI) m/z: calcd for C₂₀H₂₁N₅F [M + H]⁺, 350.1781; found, 350.1785. ¹H NMR (600 MHz, DMSO-d₆): δ 9.81 (s, 1H), 8.75–8.72 (m, 2H), 7.97–7.93 (m, 2H), 7.77–7.72 (m, 2H), 7.22–7.17 (m, 2H), 7.12 (s, 1H), 3.30 (d, J = 9.1 Hz, 2H), 2.99 (dt, J = 12.7, 3.6 Hz, 1H), 2.91–2.82 (m, 2H), 2.58 (td, J = 11.9, 2.8 Hz, 1H), 2.15–2.09 (m, 1H), 1.81–1.66 (m, 2H), 1.55 (qt, J = 12.3, 3.8 Hz, 1H). ¹³C NMR (151 MHz, DMSO-d₆): δ 171.4 (Cq), 161.3 (Cq), 159.1, 153.8 (d, ¹J_CF = 239.8 Hz, Cq), 150.6 (CH, 2C), 144.5 (Cq), 136.2 (Cq), 121.7 (CH, 2C), 120.8 (CH, 2C), 115.5 (d, ²J_CF = 22.3 Hz, CH, 2C), 100.8 (CH), 50.1 (CH₂), 45.6 (CH₂), 45.1 (CH), 29.5 (CH₂), 25.1 (CH₂).

N-Phenyl-2-(piperidin-3-yl)-6-(pyridin-4-yl)pyrimidin-4-amine (8)

Compound 8 was purchased from Aldrich. HRMS (ESI) m/z: calcd for C₂₀H₂₂N₅ [M + H]⁺, 332.1875; found, 332.1865. ¹H NMR (400 MHz, DMSO-d₆): δ 9.80 (s, 1H), 8.77–8.73 (m, 2H), 7.97–7.93 (m, 2H), 7.77–7.73 (m, 2H), 7.40–7.34 (m, 2H), 7.18 (s, 1H), 7.05 (tt, J = 7.4, 1.2 Hz, 1H), 3.04 (d, J = 12.8 Hz, 1H), 2.93 (m, 2H), 2.64 (dd, J = 12.0, 2.4 Hz, 1H), 2.15 (d, J = 13.1 Hz, 1H), 1.86–1.68 (m, 2H), 1.64–1.58 (m, 1H) (one aliphatic signal of 1H overlaps with water).

N-(4-Chlorophenyl)-6-phenyl-2-(piperidin-4-yl)pyrimidin-4-amine (13)

Title compound 13 was prepared according to general procedure 5 using intermediate 36aa (45 mg, 0.10 mmol) and HCl (4 M in 1,4-dioxane) (0.25 mL, 1.00 mmol) in 1,4-dioxane dry (1.0 mL). Final purification by neutral alumina (DCM/MeOH from 100/0 to 90/10) afforded pure title compound 13 (15 mg, 42% yield). UPLC-MS: R_t = 1.99 min (method 1). MS (ESI) m/z: 365.5 [M + H]⁺, C₂₁H₂₂ClN₄ [M + H]⁺ calcd, 365.9. HRMS (ESI) m/z: calcd for C₂₁H₂₂N₄Cl [M + H]⁺, 365.1533; found, 365.1524. ¹H NMR (400 MHz, DMSO-d₆): δ 9.75 (s, 1H), 8.08–7.98 (m, 2H), 7.85–7.76 (m, 2H), 7.57–7.44 (m, 3H), 7.43–7.31 (m, 2H), 7.07 (s, 1H), 3.07 (d, J = 12.3 Hz, 2H), 2.86–2.80 (m, 1H), 2.66 (t, J = 11.8 Hz, 2H), 1.94 (d, J = 12.8 Hz, 2H), 1.75 (qd, J = 12.2, 4.0 Hz, 2H). ¹³C NMR (101 MHz, DMSO-d₆): δ 172.1 (Cq), 161.5 (Cq), 161.1 (Cq), 139.3 (Cq), 137.2 (Cq), 130.3 (CH), 128.9 (CH, 2C), 128.6 (CH, 2C), 126.5 (CH, 2C), 125.5 (Cq), 121.0 (CH, 2C), 99.9 (CH), 45.3 (CH₂, 2C), 44.6 (CH), 30.5 (CH₂, 2C).

N-(4-Methoxyphenyl)-6-phenyl-2-(piperidin-4-yl)pyrimidin-4-amine (14)

Title compound 14 was prepared according to general procedure 5 using intermediate 36ba (106 mg, 0.23 mmol) and HCl (4 M in 1,4-dioxane) (0.58 mL, 2.30 mmol) in 1,4-dioxane dry (2.3 mL). Final normal-phase purification (DCM/DCM/NH₃ 1 M MeOH 4:1 from 85/15 to 50/50) afforded pure title compound 14 (53 mg, yield 64%). UPLC-MS: R_t = 1.87 min (method 1). MS (ESI) m/z: 361.5 [M + H]⁺, C₂₂H₂₅N₄O [M + H]⁺ calcd, 361.5. HRMS (ESI) m/z: calcd for C₂₂H₂₅N₄O [M + H]⁺, 361.2028; found, 361.2029. ¹H NMR (400 MHz, DMSO-d₆): δ 9.44 (s, 1H), 8.14–7.89 (m, 2H), 7.64 (d, J = 8.9 Hz, 2H), 7.58–7.38 (m, 3H), 6.99 (s, 1H), 6.96–6.91 (m, 2H), 3.75 (s, 3H), 3.13 (d, J = 12.4 Hz, 2H), 2.83 (tt, J = 11.4, 3.8 Hz, 1H), 2.73 (td, J = 12.2, 2.7 Hz, 2H), 1.97 (dd, J = 12.8, 3.4 Hz, 2H), 1.81 (qd, J = 12.2, 4.0 Hz, 2H). ¹³C NMR (101 MHz, DMSO-d₆): δ 172.0 (Cq), 161.4 (Cq), 161.0 (Cq), 154.7 (Cq), 137.5 (Cq), 133.2 (Cq), 130.0 (CH), 128.7 (CH, 2C), 126.4 (CH, 2C), 121.5 (CH, 2C), 114.0 (CH, 2C), 99.0 (CH), 55.2 (CH₃), 45.2 (CH₂, 2C), 44.4 (CH), 30.4 (CH₂, 2C).

N1,N1-Dimethyl-N4-(6-phenyl-2-(piperidin-4-yl)pyrimidin-4-yl)benzene-1,4-diamine (15)

Title compound 14 was prepared according to general procedure 5 using intermediate 36ca (109 mg, 0.23 mmol) and HCl (4 M in 1,4-dioxane) (0.58 mL, 2.30 mmol) in 1,4-dioxane dry (2.3 mL). Final normal-phase purification (DCM/DCM/NH₃ 1 M MeOH 4:1 from 85/15 to 50/50) afforded pure title compound 14 (54 mg, yield 63%). UPLC-MS: R_t = 1.98 min (method 1). MS (ESI) m/z: 374.2 [M + H]⁺, C₂₃H₂₈N₅ [M + H]⁺ calcd, 374.5. HRMS (ESI) m/z: calcd for C₂₃H₂₈N₅ [M + H]⁺, 374.2345; found, 374.2341. ¹H NMR (400 MHz, DMSO-d₆): δ 9.25 (s, 1H), 8.17–7.82 (m, 2H), 7.69–7.37 (m, 5H), 6.93 (s, 1H), 6.83–6.66 (m, 2H), 3.08 (dt, J = 12.2, 3.4 Hz, 2H), 2.87 (s, 6H), 2.77 (tt, J = 11.5, 3.8 Hz, 1H), 2.66 (td, J = 12.1, 2.6 Hz, 2H), 1.93 (dd, J = 13.4, 3.5 Hz, 2H), 1.77 (qd, J = 13.0, 12.5, 4.0 Hz, 2H). ¹³C NMR (101 MHz, DMSO-d₆): δ 172.2 (Cq), 161.6 (Cq), 160.9 (Cq), 146.7 (Cq), 137.6 (Cq), 129.9 (CH), 128.7 (CH, 2C), 126.3 (CH, 2C), 121.7 (CH, 2C), 113.0 (CH, 2C), 98.4 (CH), 45.6 (CH₂, 2C), 44.9 (CH), 40.6 (CH₃, 2C), 31.0 (CH₂, 2C).

N1-(2-Cyclohexyl-6-phenylpyrimidin-4-yl)-N3,N3-dimethylbenzene-1,3-diamine (20)

Title compound 14 was prepared according to general procedure 4 using intermediate 35db (60 mg, 0.16 mmol), NH₄COOH (61 mg, 0.96 mmol), and Pd(OH)₂/C (12 mg) in MeOH dry (2.7 mL). Final normal-phase purification (Cyclohexane/EtOAc from 100/0 to 90/10) afforded pure title compound 20 (35 mg, yield 58%). UPLC-MS: R_t = 1.45 min (method 3). MS (ESI) m/z: 373.3 [M + H]⁺, C₂₄H₂₉N₄ [M + H]⁺ calcd, 373.5. HRMS (ESI) m/z: calcd for C₂₄H₂₉N₄ [M + H]⁺, 373.2392; found, 373.2393. ¹H NMR (400 MHz, DMSO-d₆) 9.40 (s, 1H), 8.06–7.95 (m, 2H), 7.56–7.44 (m, 4H), 7.12 (t, J = 8.1 Hz, 1H), 7.05 (s, 1H), 6.86 (dd, J = 7.8, 1.4 Hz, 1H), 6.40 (dd, J = 8.1, 2.4 Hz, 1H), 2.93 (s, 6H), 2.72 (tt, J = 11.6, 3.5 Hz, 1H), 2.06–1.96 (m, 2H), 1.81 (dt, J = 12.6, 3.4 Hz, 2H), 1.76–1.57 (m, 3H), 1.47–1.32 (m, 2H), 1.24 (tt, J = 12.7, 3.5 Hz, 1H). ¹³C NMR (101 MHz, DMSO-d₆): δ 173.1 (Cq), 161.4 (Cq), 161.1 (Cq), 150.9 (Cq), 141.0 (Cq), 137.5 (Cq), 130.0 (CH), 129.0 (CH), 128.8 (CH, 2C), 126.4 (CH, 2C), 107.7 (CH), 106.6 (CH), 103.9 (CH), 99.6 (CH), 46.9 (CH), 40.2 (CH₃, 2C), 31.6 (CH₂, 2C), 25.8 (CH₂), 25.8 (CH_2, 2C).

1-(4-(4-((3-(Dimethylamino)phenyl)amino)-6-phenylpyrimidin-2-yl)piperidin-1-yl)-2-methylpropan-1-one (21)

To a solution of compound 17 (30 mg, 0.08 mmol) in DMF dry (1 mL) were added isobutyric acid (8 μL, 0.09 mmol), HATU (37 mg, 0.1 mmol), and DIPEA (0.04 mL, 0.24 mmol) under argon. The reaction mixture was stirred at room temperature for 2 h; after that, water (2 mL) was added, and the aqueous layer was extracted with EtOAc (3 × 3 mL). The collected organic layer was washed with NaHCO₃ (2 mL), dried over Na₂SO₄, filtered, and concentrated under vacuum. Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 60/40) afforded pure title compound 21 (21 mg, 59% yield). UPLC-MS: R_t = 1.53 min (method 2), no ionization. HRMS (ESI) m/z: calcd for C₂₇H₃₄N₅O [M + H]⁺, 444.2763; found, 444.2765. ¹H NMR (400 MHz, CDCl₃): δ 7.99–7.94 (m, 2H), 7.46–7.42 (m, 3H), 7.24 (t, J = 8.1 Hz, 1H), 7.01 (s, 1H), 6.99 (bs, 1H), 6.80 (t, J = 2.2 Hz, 1H), 6.70 (dd, J = 7.6, 2.0 Hz, 1H), 6.56 (dd, J = 8.3, 2.5 Hz, 1H), 4.73 (d, J = 13.2 Hz, 1H), 4.06 (d, J = 13.6 Hz, 1H), 3.20 (t, J = 12.7 Hz, 1H), 3.04 (tt, J = 11.5, 3.9 Hz, 1H), 2.98 (s, 6H), 2.86 (p, J = 6.7 Hz, 1H), 2.82–2.71 (m, 1H), 2.20–2.06 (m, 2H), 1.98–2.87 (m, 2H), 1.16 (t, J = 7.8 Hz, 6H). ¹³C NMR (101 MHz, CDCl₃): δ 175.4 (Cq), 172.1 (Cq), 163.9 (Cq), 162.1 (Cq), 151.7 (Cq), 139.4 (Cq), 137.9 (Cq), 130.3 (CH), 130.1 (CH), 128.8 (CH, 2C), 127.2 (CH, 2C), 110.4 (CH), 109.2 (CH), 106.4 (CH), 97.8 (CH), 45.6 (CH₂), 45.3 (CH), 42.0 (CH₂), 40.7 (CH₃, 2C), 30.7 (CH₂, 2C), 30.3 (CH), 19.8 (CH₃, 2C).

N1-(2-(1-(2-Methoxyethyl)piperidin-4-yl)-6-phenylpyrimidin-4-yl)-N3,N3-dimethylbenzene-1,3-diamine (22)

DIPEA (0.03 mL, 0.16 mmol) and 1-bromo-3-methoxypropane (14 μL, 0.14 mmol) were added to a solution of compound 17 (50 mg, 0.13 mmol) in CH₃CN dry (1.3 mL) under argon. The reaction mixture was stirred at 60 °C for 2 days. After that, water (2 mL) was added, and the aqueous layer was extracted with EtOAc (2 mL × 3). Collected organic layers were dried over Na₂SO₄, filtered, and concentrated under vacuum. Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 40/60) afforded pure title compound 22 (22 mg, 38% yield). UPLC-MS: R_t = 2.09 min (method 1), MS (ESI) m/z: 432.3 [M + H]⁺, C₂₆H₃₄N₅O [M + H]⁺ calcd, 432.6. HRMS (ESI) m/z: calcd for C₂₆H₃₄N₅O [M + H]⁺, 432.2763; found, 432.2752. ¹H NMR (400 MHz, DMSO-d₆): δ 9.59 (bs, 1H), 8.08–7.96 (m, 2H), 7.58–7.44 (m, 3H), 7.19 (s, 1H), 7.17–7.10 (m, 2H), 7.04 (d, J = 8.1 Hz, 1H), 6.42 (dd, J = 8.3, 2.5 Hz, 1H), 3.71–3.62 (bs, 3H), 3.32 (s, 3H), 3.28–2.97 (m, 6H), 2.92 (s, 6H), 2.28–2.08 (m, 4H). ¹³C NMR (101 MHz, DMSO-d₆): δ 161.5 (Cq), 161.1 (Cq), 151.0 (Cq), 140.7 (Cq), 137.2 (Cq), 130.3 (CH), 129.2 (CH), 128.9 (CH, 2C), 126.4 (CH, 2C), 108.1 (CH), 107.1 (CH), 104.0 (CH), 100.0 (CH), 66.0 (CH₂), 58.2 (CH₃), 55.3 (CH₂, 2C, recovered from HSQC), 52.6 (CH₂, recovered from HSQC), 40.3 (CH₃, 2C), 27.6 (CH₂, 2C, recovered from HSQC).

N-(2-Methoxyphenyl)-6-phenyl-2-(piperidin-4-yl)pyrimidin-4-amine (23)

Title compound 23 was prepared according to general procedure 5 using intermediate 36ea (60 mg, 0.13 mmol) and HCl (4 M in 1,4-dioxane) (0.33 mL, 1.30 mmol) in 1,4-dioxane dry (1.3 mL). Final purification by alumina (DCM/MeOH·NH₃ 1 M from 100/0 to 95/5) afforded pure title compound 23 (25 mg, yield 53%). UPLC-MS: R_t = 1.92 min (method 1). MS (ESI) m/z: 361.3 [M + H]⁺, C₂₂H₂₅N₄O [M + H]⁺ calcd, 361.5. HRMS (ESI) m/z: calcd for C₂₂H₂₅N₄O [M + H]⁺, 361.2028; found, 361.2016. ¹H NMR (400 MHz, DMSO-d₆): δ 8.70 (s, 1H), 8.26 (d, J = 7.9 Hz, 1H), 8.08–7.99 (m, 2H), 7.57–7.45 (m, 3H), 7.30 (s, 1H), 7.11–7.01 (m, 2H), 7.00–6.96 (m, 1H), 3.88 (s, 3H), 3.09–2.98 (m, 2H), 2.76 (dt, J = 12.1, 3.0 Hz, 1H), 2.60 (tt, J = 12.1, 2.6 Hz, 2H), 1.89 (d, J = 12.8 Hz, 2H), 1.71 (qd, J = 12.2, 4.0 Hz, 2H). ¹³C NMR (101 MHz, DMSO-d₆): δ 172.4 (Cq), 161.6 (Cq), 161.2 (Cq), 149.8 (Cq), 137.5 (Cq), 130.0 (CH), 128.7 (CH, 2C), 128.5 (Cq), 126.4 (CH, 2C), 123.3 (CH), 121.7 (CH), 120.4 (CH), 111.2 (CH), 99.8 (CH), 55.7 (CH₃), 46.1 (CH₂, 2C), 45.6 (CH), 31.7 (CH₂, 2C).

N1,N1-Dimethyl-N2-(6-phenyl-2-(piperidin-4-yl)pyrimidin-4-yl)benzene-1,2-diamine (24)

Title compound 24 was prepared according to general procedure 5 using intermediate 36fa (100 mg, 0.21 mmol) and HCl (4 M in 1,4-dioxane) (0.53 mL, 2.1 mmol) in 1,4-dioxane dry (2.1 mL). Final normal-phase purification (DCM/MeOH·NH₃ 1 M from 100/0 to 95/5) afforded pure title compound 24 (13 mg, 16% yield). UPLC-MS: R_t = 2.06 min (method 1). MS (ESI) m/z: 374.3 [M + H]⁺, C₂₃H₂₈N₅ [M + H]⁺ calcd, 374.5. HRMS (ESI) m/z: calcd for C₂₃H₂₈N₅ [M + H]⁺, 374.2345; found, 374.2341.¹H NMR (400 MHz, DMSO-d₆): δ 8.64 (s, 1H), 8.25–8.14 (m, 1H), 8.10–8.00 (m, 2H), 7.55–7.44 (m, 3H), 7.36 (s, 1H), 7.18 (td, J = 7.7, 2.0 Hz, 1H), 7.07 (td, J = 7.5, 5.5 Hz, 1H), 7.03 (td, J = 7.6, 1.9 Hz, 1H), 3.00 (dt, J = 12.2, 2.0 Hz, 2H), 2.76 (tt, J = 11.4, 3.7 Hz, 1H), 2.65 (s, 6H), 2.63–2.56 (m, 2H), 1.90 (d, J = 12.7 Hz, 2H), 1.72 (qd, J = 12.2, 4.0 Hz, 2H). ¹³C NMR (151 MHz, DMSO-d₆): δ 172.2 (Cq), 161.4 (Cq), 161.3 (Cq), 144.5 (Cq), 137.5 (Cq), 133.2 (Cq), 130.1 (CH), 128.8 (2C), 126.5 (CH, 2C), 123.2 (CH), 123.1 (CH), 121.7 (CH), 119.3 (CH), 99.9 (CH), 45.7 (CH₂, 2C), 45.1 (CH), 43.8 (CH₃, 2C), 31.2 (CH₂, 2C).

N-(3,4-Dimethoxyphenyl)-6-phenyl-2-(piperidin-4-yl)pyrimidin-4-amine (25)

Title compound 25 was prepared according to general procedure 5 using intermediate 36ga (40 mg, 0.08 mmol) and HCl (4 M in 1,4-dioxane) (0.20 mL, 0.8 mmol) in 1,4-dioxane dry (0.8 mL). Final purification by alumina (DCM/MeOH·NH₃ 1 M from 100/0 to 98/2) afforded pure title compound 25 (20 mg, 63% yield). UPLC-MS: R_t = 1.79 min (method 1). MS (ESI) m/z: 391.2 [M + H]⁺, C₂₃H₂₇N₄O₂ [M + H]⁺ calcd, 391.5. HRMS (ESI) m/z: calcd for C₂₃H₂₇N₄O₂ [M + H]⁺, 391.2134; found, 391.2136.¹H NMR (400 MHz, DMSO-d₆): δ 8.05–7.95 (m, 2H), 7.62 (s, 1H), 7.55–7.43 (m, 3H), 7.14 (dd, J = 8.6, 2.5 Hz, 1H), 6.98 (s, 1H), 6.93 (d, J = 8.7 Hz, 1H), 3.80 (s, 3H), 3.74 (s, 3H), 3.04 (dt, J = 11.9, 2.7 Hz, 2H), 2.77 (tt, J = 11.5, 3.8 Hz, 1H), 2.61 (td, J = 12.1, 2.5 Hz, 2H), 1.93 (d, J = 12.8 Hz, 2H), 1.74 (qd, J = 12.2, 4.0 Hz, 2H). ¹³C NMR (101 MHz, DMSO-d₆): δ 172.1 (Cq), 161.3 (Cq), 161.1 (Cq), 148.7 (Cq), 144.2 (Cq), 137.5 (Cq), 133.8 (Cq), 130.1 (CH), 128.8 (CH, 2C), 126.4 (CH, 2C), 112.4 (CH), 111.5 (CH), 105.2 (CH), 99.2 (CH), 55.8 (CH₃), 55.4 (CH₃), 45.3 (CH₂, 2C), 44.7 (CH), 30.8 (CH₂, 2C).

N-(3-(2-Methoxyethoxy)phenyl)-6-phenyl-2-(piperidin-4-yl)pyrimidin-4-amine (26)

Title compound 26 was prepared according to general procedure 5 using intermediate 36ha (51 mg, 0.1 mmol) and HCl (4 M in 1,4-dioxane) (0.25 mL, 1.0 mmol) in 1,4-dioxane dry (1.0 mL). Final purification by alumina (DCM/MeOH·NH₃ 1 N from 100/0 to 98/2) afforded pure title compound 26 (12 mg, 30% yield). TLC: R_f = 0.3 (DCM/MeOH·NH₃ 1 N, 98:2). HRMS (ESI) m/z: calcd for C₂₄H₂₉N₄O₂ [M + H]⁺, 405.2291; found, 405.2291.¹H NMR (400 MHz, DMSO-d₆) 9.59 (s, 1H), 8.09–7.98 (m, 2H), 7.66 (d, J = 2.3 Hz, 1H), 7.57–7.45 (m, 3H), 7.28–7.15 (m, 2H), 7.06 (s, 1H), 6.66–6.51 (m, 1H), 4.16–4.06 (m, 2H), 3.72–3.64 (m, 2H), 3.03 (dt, J = 12.2, 3.4 Hz, 2H), 2.79 (tt, J = 11.5, 3.8 Hz, 1H), 2.61 (td, J = 12.1, 2.5 Hz, 2H), 1.93 (d, J = 12.1 Hz, 2H), 1.74 (qd, J = 12.2, 4.0 Hz, 2H). ¹³C NMR (101 MHz, DMSO-d₆): δ 172.6 (Cq), 161.3 (Cq), 161.3 (Cq), 158.8 (Cq), 141.6 (Cq), 137.4 (Cq), 130.2 (CH), 129.5 (CH), 128.8 (CH, 2C), 126.5 (CH, 2C), 111.7 (CH), 108.5 (CH), 105.4 (CH), 99.8 (CH), 70.4 (CH₂), 66.9 (CH₂), 58.2 (CH₃), 46.1 (CH₂, 2C), 45.6 (CH), 31.8 (CH₂, 2C).

6-Phenyl-2-(piperidin-4-yl)-N-(3-(trifluoromethyl)phenyl)pyrimidin-4-amine (27)

Title compound 27 was prepared according to general procedure 5 using intermediate 36ia (90 mg, 0.18 mmol) and HCl (4 M in 1,4-dioxane) (0.45 mL, 1.8 mmol) in 1,4-dioxane dry (1.8 mL). Final purification by alumina (DCM/MeOH·NH₃ 1 N from 100/0 to 95/5) afforded pure title compound 27 (69 mg, 96% yield). UPLC-MS: R_t = 2.04 min (method 1). MS (ESI) m/z: 399.2 [M + H]⁺, C₂₂H₂₂F₃N₄ [M + H]⁺ calcd, 399.4. HRMS (ESI) m/z: calcd for C₂₂H₂₂N₄F₃ [M + H]⁺, 399.1797; found, 399.1801.¹H NMR (400 MHz, DMSO-d₆) 9.98 (s, 1H), 8.53 (d, J = 2.3 Hz, 1H), 8.12–7.99 (m, 2H), 7.87 (dd, J = 8.5, 2.2 Hz, 1H), 7.61–7.46 (m, 4H), 7.32 (d, J = 7.7 Hz, 1H), 7.10 (s, 1H), 3.03 (dt, J = 12.0, 3.3 Hz, 2H), 2.82 (tt, J = 11.6, 3.7 Hz, 1H), 2.62 (td, J = 12.0, 2.5 Hz, 2H), 1.97 (dd, J = 13.5, 3.6 Hz, 2H), 1.72 (qd, J = 12.1, 4.0 Hz, 2H). ¹³C NMR (101 MHz, DMSO-d₆): δ 172.5 (Cq), 161.7 (Cq), 161.0 (Cq), 141.2 (Cq), 137.2 (Cq), 130.3 (CH), 129.8 (CH), 129.5 (q, ²J_CF = 31.2 Hz, Cq), 128.8 (CH, 2C), 126.5 (CH, 2C), 124.3 (q, ¹J_CF = 276.2 Hz, Cq), 122.5 (CH), 117.8 (q, ³J_CF = 3.7 Hz, CH), 115.4 (q, ³J_CF = 4.0 Hz, CH), 100.0 (CH), 46.2 (CH₂, 2C), 45.4 (CH), 31.8 (CH₂, 2C).

N1,N1-Dimethyl-N3-(4-phenyl-6-(piperidin-4-yl)-1,3,5-triazin-2-yl)benzene-1,3-diamine (28)

Title compound 28 was prepared after two steps from intermediate 41. Step 1. The double C=C bond was reduced following general procedure 4 using intermediate 41 (70 mg, 0.15 mmol), NH₄COOH (57 mg, 0.90 mmol), and Pd(OH)₂/C (14 mg). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 75/25) afforded compound tert-butyl 4-(4-((3-(dimethylamino)phenyl)amino)-6-phenyl-1,3,5-triazin-2-yl)piperidine-1-carboxylate (49 mg, 70% yield). UPLC-MS: R_t = 2.42 min (method 2). MS (ESI) m/z: 475.4 [M + H]⁺, C₂₇H₃₅N₆O₂ [M + H]⁺ calcd, 475.6. ¹H NMR (400 MHz, CDCl₃): δ 8.54–7.47 (m, 2H), 7.57–7.51 (m, 1H), 7.49–7.43 (m, 2H), 7.35 (t, J = 2.2 Hz, 1H), 7.31 (s, 1H), 7.23 (t, J = 8.1 Hz, 1H), 6.91 (dd, J = 7.9, 2.0 Hz, 1H), 6.51 (dd, J = 8.3, 2.5 Hz, 1H), 4.21 (bs, 2H), 3.01 (s, 6H), 2.96–2.79 (m, 3H), 1.90 (qd, J = 11.5, 3.6 Hz, 2H), 1.49 (s, 9H). Step 2. The Boc-protective group was removed following general procedure 5 using tert-butyl 4-(4-((3-(dimethylamino)phenyl)amino)-6-phenyl-1,3,5-triazin-2-yl)piperidine-1-carboxylate (50 mg, 0.11 mmol) and HCl (4 M in 1,4-dioxane) (0.27 mL, 1.1 mmol) in 1,4-dioxane dry (1.1 mL). Final purification by alumina (DCM/MeOH from 100/0 to 95/5) afforded pure title compound 28 (30 mg, 73% yield). UPLC-MS: R_t = 2.10 min (method 1). MS (ESI) m/z: 375.3 [M + H]⁺, C₂₂H₂₇N₆ [M + H]⁺ calcd, 375.5. HRMS (ESI) m/z: calcd for C₂₂H₂₇N₆ [M + H]⁺, 375.2297; found, 375.2293.¹H NMR (400 MHz, DMSO-d₆): δ 9.97 (s, 1H), 8.50–8.32 (m, 2H), 7.64–7.58 (m, 1H), 7.58–7.50 (m, 2H), 7.44 (t, J = 2.2 Hz, 1H), 7.15 (t, J = 8.0 Hz, 1H), 7.07 (dt, J = 8.3, 1.2 Hz, 1H), 6.46 (dd, J = 8.0, 2.5 Hz, 1H), 3.03 (dt, J = 12.3, 3.4 Hz, 2H), 2.93 (s, 6H), 2.80–2.68 (m, 1H), 2.59 (td, J = 12.1, 2.5 Hz, 2H), 1.91 (d, J = 12.6 Hz, 2H), 1.72 (qd, J = 12.1, 4.0 Hz, 2H). ¹³C NMR (101 MHz, DMSO-d₆): δ 170.1 (Cq), 164.2 (Cq), 150.8 (Cq), 139.8 (Cq), 136.1 (Cq), 132.2 (CH), 128.9 (CH), 128.6 (CH, 2C), 128.1 (CH, 2C), 108.6 (CH), 107.5 (CH), 104.6 (CH), 45.9 (CH₂, 2C), 45.0 (CH), 40.2 (CH₃), 31.0 (CH₂, 2C).

4-Phenyl-6-(piperidin-4-yl)-N-(3-(trifluoromethyl)phenyl)-1,3,5-triazin-2-amine (29)

Title compound 29 was prepared after two steps from intermediate 44. Step 1. The double C=C bond was reduced following general procedure 4 using intermediate 44 (130 mg, 0.26 mmol), NH₄COOH (98 mg, 1.57 mmol), and Pd(OH)₂/C (26 mg). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 80/20) afforded compound tert-butyl 4-(4-phenyl-6-((3-(trifluoromethyl)phenyl)amino)-1,3,5-triazin-2-yl)piperidine-1-carboxylate (113 mg, 87% yield). UPLC-MS: R_t = 2.52 min (method 2). MS (ESI) m/z: 500.3 [M + H]⁺, C₂₆H₂₉F₃N₅O₂ [M + H]⁺ calcd, 500.5. ¹H NMR (400 MHz, CDCl₃): δ 8.48 (d, J = 7.7 Hz, 2H), 8.30 (s, 1H), 7.75 (d, J = 8.1 Hz, 1H), 7.59–7.54 (m, 1H), 7.53–7.48 (m, 3H), 7.44 (s, 1H), 7.38 (d, J = 7.8 Hz, 1H), 4.22 (bs, 2H), 2.94–2.91 (m, 3H), 2.08 (d, J = 13.1 Hz, 2H), 1.95–1.82 (m, 2H), 1.49 (s, 9H). Step 2. The Boc-protective group was removed following general procedure 5 using tert-butyl 4-(4-phenyl-6-((3-(trifluoromethyl)phenyl)amino)-1,3,5-triazin-2-yl)piperidine-1-carboxylate (74 mg, 0.15 mmol) and HCl (4 M in 1,4-dioxane) (0.38 mL, 1.5 mmol) in 1,4-dioxane dry (1.5 mL). Final purification by alumina (DCM/MeOH·NH₃ 1 N from 100/0 to 90/10) afforded pure title compound 29 (43 mg, 71% yield). UPLC-MS: R_t = 2.19 min (method 1). MS (ESI) m/z: 400.2 [M + H]⁺, C₂₁H₂₁F₃N₅ [M + H]⁺ calcd, 500.5. HRMS (ESI) m/z: calcd for C₂₁H₂₁N₅F₃ [M + H]⁺, 400.1749; found, 400.1747.¹H NMR (400 MHz, DMSO-d₆): δ 10.53 (s, 1H), 8.48 (d, J = 2.6 Hz, 1H), 8.45–8.35 (m, 2H), 8.00 (dd, J = 8.1, 2.1 Hz, 1H), 7.68–7.53 (m, 4H), 7.41 (d, J = 7.7 Hz, 1H), 3.04 (dt, J = 12.3, 3.4 Hz, 2H), 2.85–2.71 (m, 1H), 2.61 (td, J = 12.1, 2.5 Hz, 2H), 1.95 (d, J = 12.5 Hz, 2H), 1.80–1.65 (m, 2H). ¹³C NMR (101 MHz, DMSO-d₆): δ 181.6 (Cq), 170.4 (Cq), 164.3 (Cq), 140.0 (Cq), 135.8 (Cq), 132.4 (CH), 129.9 (CH), 129.5 (q, ²J_CF = 32.1, Cq), 128.7 (CH, 2C), 128.2 (CH, 2C), 124.3 (q, ¹J_CF = 272.9 Hz, Cq) 123.6 (CH), 119.0 (CH), 116.4 (CH), 45.8 (CH₂, 2C), 44.9 (CH), 30.9 (CH₂, 2C).

3-(2-Methoxyethoxy)aniline (32h)

NaH (60% dispersion in mineral oil) (121.0 mg, 3.01 mmol) was added to a solution of 3-aminophenol (220.0 mg, 2.01 mmol) in THF dry (4.0 mL) under argon at room temperature. After 10 min, 1-bromo-3-methoxypropane (0.23 mL, 2.61 mmol) was slowly added, and the reaction mixture was stirred for 2 days. After that, NH₄Cl aqueous sat solution (4 mL) was slowly added, and the aqueous layer was extracted with EtOAc (5 mL × 2). Collected organic layers were dried over Na₂SO₄, filtered, and concentrated under vacuum. Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 70/30) afforded pure title compound 32h (270.0 mg, 80% yield). UPLC-MS: R_t = 1.34 min (method 1). MS (ESI) m/z: 168.1 [M + H]⁺, C₉H₁₄NO₂ [M + H]⁺ calcd, 168.2. ¹H NMR (400 MHz, CDCl₃): δ 7.05 (t, J = 7.8 Hz, 1H), 6.39–6.27 (m, 3H), 4.10–4.05 (m, 2H), 3.75–3.69 (m, 2H).

2-Chloro-N-(4-chlorophenyl)-6-phenylpyrimidin-4-amine (33a)

Title compound 33a was prepared according to general procedure 1 using intermediate 31 (245 mg, 1.09 mmol) and 4-chloroaniline 32a (151.0 mg, 1.20 mmol). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 85/15) afforded pure title compound 33a (310.0 mg, 90% yield). UPLC-MS: R_t = 1.96 min (method 2). MS (ESI) m/z: 316.1 [M + H]⁺, C₁₇H₁₁Cl₂N₃ [M + H]⁺ calcd, 317.2. ¹H NMR (400 MHz, CDCl₃): δ 7.97–7.85 (m, 2H), 7.52–7.36 (m, 5H), 7.33 (d, J = 8.8 Hz, 2H), 7.22 (s, 1H), 6.90 (s, 1H). ¹³C NMR (101 MHz, CDCl₃): δ 166.6 (Cq), 163.2 (Cq), 161.2 (Cq), 136.1 (Cq, 2C), 131.2 (CH), 131.2 (Cq), 129.9 (CH, 2C), 129.0 (CH, 2C), 127.3 (CH, 2C), 124.5 (CH, 2C), 98.0 (CH).

2-Chloro-N-(4-methoxyphenyl)-6-phenylpyrimidin-4-amine (33b)

Title compound 33b was prepared according to general procedure 2 using intermediate 31 (100 mg, 0.44 mmol) and p-methoxyaniline 32b (55.2 mg, 0.44 mmol). Final normal-phase purification (cyclohexane/TBME from 95/5 to 75/25) afforded pure title compound 33b (86.8 mg, 63% yield). UPLC-MS: R_t = 1.39 min (method 2); MS (ESI) m/z: 312.1 [M – H]⁺, [M – H]⁺ calcd: 312.1. 1H NMR (400 MHz, DMSO-d₆): δ 9.91 (s, 1H), 8.06–7.84 (m, 2H), 7.65–7.37 (m, 5H), 7.06 (s, 1H), 7.02–6.92 (m, 2H), 3.76 (s, 3H).

N1-(2-Chloro-6-phenylpyrimidin-4-yl)-N4,N4-dimethylbenzene-1,4-diamine (33c)

Titled compound 33c was prepared according to general procedure 2 using intermediate 31 (300 mg, 1.33 mmol) and N1,N1-dimethylbenzene-1,4-diamine 32c (191.1 mg, 1.33 mmol). Final normal-phase purification (cyclohexane/TBME from 100/0 to 80/20) afforded pure title compound 33c (272 mg, 63% yield). UPLC-MS: R_t = 1.56 min (method 2); MS (ESI) m/z: 325.1 [M + H]⁺, [M + H]⁺ calcd: 325.1. ¹H NMR (400 MHz, DMSO-d₆): δ 9.76 (s, 1H), 7.93 (dd, J = 6.7, 3.0 Hz, 2H), 7.52 (dd, J = 4.6, 2.4 Hz, 3H), 7.36 (s, 2H), 7.00 (s, 1H), 6.87–6.64 (m, 2H), 2.89 (s, 6H).

N1-(2-Chloro-6-phenylpyrimidin-4-yl)-N3,N3-dimethylbenzene-1,3-diamine (33d)

Titled compound 33d was prepared according to general procedure 2 using intermediate 31 (300 mg, 1.13 mmol) and N1,N1-dimethylbenzene-1,3-diamine 32d (181.5 mg, 1.33 mmol). Final normal-phase purification (cyclohexane/TBME from 100/0 to 80/20) afforded pure title compound 33d (246 mg, 67% yield). UPLC-MS: R_t = 1.73 min (method 2); MS (ESI) m/z: 325.1 [M + H]⁺, [M + H]⁺ calcd: 325.1. 1H NMR (400 MHz, DMSO-d₆): δ 9.93 (s, 1H), 8.14–7.82 (m, 2H), 7.60–7.47 (m, 3H), 7.29–7.09 (m, 2H), 7.04 (s, 1H), 6.99–6.85 (m, 1H), 6.51 (ddd, J = 8.4, 2.5, 0.8 Hz, 1H), 2.92 (s, 6H).

2-Chloro-N-(2-methoxyphenyl)-6-phenylpyrimidin-4-amine (33e)

Title compound 33e was prepared according to general procedure 1 using intermediate 31 (180 mg, 0.80 mmol) and 2-methoxyaniline 32e (0.07 mL, 0.55 mmol). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 85/15) afforded pure title compound 33e (230 mg, 91% yield). UPLC-MS: R_t = 1.57 min (method 2). MS (ESI) m/z: 312.1 [M + H]⁺, C₁₇H₁₅ClN₃O [M + H]⁺ calcd, 312.8. ¹H NMR (400 MHz, CDCl₃): δ 7.96 (m, 2H), 7.91 (d, J = 7.9 Hz, 1H), 7.51–7.41 (m, 3H), 7.32 (s, 1H), 7.15 (td, J = 7.8, 1.6 Hz, 1H), 7.04 (td, J = 7.7, 1.4 Hz, 1H), 7.00 (s, 1H), 6.97 (dd, J = 8.1, 1.4 Hz, 1H), 3.90 (s, 3H). ¹³C NMR (101 MHz, CDCl₃): δ 166.0 (Cq), 162.6 (Cq), 161.2 (Cq), 150.3 (Cq), 136.4 (Cq), 131.0 (Cq), 128.9 (CH, 2C), 127.3 (CH, 2C), 125.1 (CH), 121.6 (CH), 121.2 (CH), 111.1 (CH), 99.0 (CH), 55.9 (CH₃).

N1-(2-Chloro-6-phenylpyrimidin-4-yl)-N2,N2-dimethylbenzene-1,2-diamine (33f)

Title compound 33f was prepared according to general procedure 1 using intermediate 31 (112 mg, 0.50 mmol) and N1,N1-dimethylbenzene-1,2-diamine 32f (0.10 mL, 0.55 mmol). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 80/20) afforded pure title compound 33f (149 mg, 92% yield). UPLC-MS: R_t = 1.91 min (method 2). MS (ESI) m/z: 323.3 [M – H]⁻, C₁₇H₁₇ClN₄ [M – H]⁻ calcd, 323.8. ¹H NMR (400 MHz, CDCl₃): δ 8.03–7.96 (m, 2H), 7.95–7.92 (m, 2H), 7.47 (m, 3H), 7.22–7.10 (m, 3H), 7.09 (s, 1H), 2.68 (s, 6H). ¹³C NMR (101 MHz, CDCl₃): δ 166.0 (Cq), 162.2 (Cq), 161.3 (Cq), 144.9 (Cq), 136.4 (Cq), 132.8 (Cq), 131.0 (CH), 128.9 (CH, 2C), 127.3 (CH, 2C), 124.6 (CH), 124.5 (CH), 120.5 (CH), 120.3 (CH), 99.1 (CH), 44.5 (CH₃, 2C).

2-Chloro-N-(3,4-dimethoxyphenyl)-6-phenylpyrimidin-4-amine (33g)

Title compound 33g was prepared according to general procedure 1 using intermediate 31 (107 mg, 0.47 mmol) and 3,4-dimethoxyaniline 32g (64 mg, 0.41 mmol). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 70/30) afforded pure title compound 33f (149 mg, 92% yield). UPLC-MS: R_t = 1.54 min (method 2). MS (ESI) m/z: 340.2 [M – H]⁻, C₁₇H₁₆ClN₃O₂ [M – H]⁻ calcd, 340.8. ¹H NMR (400 MHz, CDCl₃): δ 7.97–7.83 (m, 2H), 7.49–7.38 (m, 3H), 7.07 (s, 1H), 6.93–6.88 (m, 3H), 6.81 (s, 1H), 3.93 (s, 3H), 3.89 (s, 3H).

2-Chloro-N-(3-(2-methoxyethoxy)phenyl)-6-phenylpyrimidin-4-amine (33h)

Title compound 33h was prepared according to general procedure 1 using intermediate 31 (85 mg, 0.38 mmol) and 3,4-dimethoxyaniline 32h (68 mg, 0.41 mmol). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 65/35) afforded pure title compound 33h (108 mg, 80% yield). UPLC-MS: R_t = 1.36 min (method 2). MS (ESI) m/z: 356.2 [M + H]⁺, C₁₉H₁₉ClN₃O₂ [M + H]⁺ calcd, 356.8. ¹H NMR (400 MHz, CDCl₃): δ 7.97–7.88 (m, 2H), 7.50–7.40 (m, 3H), 7.32 (t, J = 8.1 Hz, 1H), 7.15 (s, 1H), 7.00 (s, 1H), 6.97 (t, J = 2.3 Hz, 1H), 6.93 (dd, J = 7.8, 1.7 Hz, 1H), 6.82 (ddd, J = 8.4, 2.5 Hz, 1H), 4.16–4.13 (m, 2H), 3.79–3.74 (m, 2H), 3.45 (s, 3H).

2-Chloro-6-phenyl-N-(3-(trifluoromethyl)phenyl)pyrimidin-4-amine (33i)

Title compound 33i was prepared according to general procedure 1 using intermediate 31 (180 mg, 0.80 mmol) and 3,4-dimethoxyaniline 32i (0.11 mL, 0.88 mmol). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 90/10) afforded pure title compound 33i (200 mg, 71% yield). UPLC-MS: R_t = 1.87 min (method 2). MS (ESI) m/z: 348.3 [M – H]⁻, C₁₇H₁₀ClF₃N₃ [M – H]⁻ calcd, 348.7. ¹H NMR (400 MHz, CDCl₃): δ 8.00–7.89 (m, 2H), 7.70–7.64 (m, 2H), 7.56 (s, 1H), 7.52–7.42 (m, 4H), 7.09 (s, 1H), 6.96 (s, 1H).

tert-Butyl 4-(4-((4-Chlorophenyl)amino)-6-phenylpyrimidin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (35aa)

Title compound 35aa was prepared according to general procedure 3 using intermediate 33a (310 mg, 0.98 mmol) and boronic ester 34a (368 mg, 1.19 mmol). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 85/15) afforded pure title compound 35aa (380 mg, 83% yield). UPLC-MS: R_t = 2.50 min (method 2). MS (ESI) m/z: 461.6 [M – H]⁻, C₂₆H₂₆ClN₄O₂ [M – H]⁻ calcd, 462.0. ¹H NMR (400 MHz, DMSO-d₆): δ 9.77 (s, 1H), 8.15–8.03 (m, 2H), 7.80 (d, J = 8.5 Hz, 2H), 7.56–7.50 (m, 3H), 7.41 (d, J = 8.5 Hz, 2H), 7.20 (s, 1H), 7.10 (s, 1H), 4.13 (s, 2H), 3.57 (t, J = 5.0 Hz, 2H), 2.69 (s, 2H), 1.44 (s, 9H).

tert-Butyl 4-(4-((4-Methoxyphenyl)amino)-6-phenylpyrimidin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (35ba)

Title compound 35ba was prepared according to general procedure 3 using intermediate 33b (80 mg, 0.26 mmol), boronic ester 34a (98.1 mg, 0.31 mmol), Pd(dppf)Cl₂·DCM (42 mg, 0.05 mmol), and K₂CO₃ (2 M)_aq (0.39 mL, 0.78 mmol) in 1,4-dioxane dry (2.6 mL). Final normal-phase purification (cyclohexane/EtOAc from 95/5 to 75/25) afforded pure title compound 35ba (104.7 mg, 89% yield). UPLC-MS: R_t = 2.24 min (method 2). MS (ESI) m/z: 459.3 [M + H]⁺, C₂₇H₃₁N₄O₃ [M + H]⁺ calcd, 459.6. ¹H NMR (400 MHz, DMSO-d₆): δ 9.44 (s, 1H), 8.12–8.02 (m, 2H), 7.68–7.60 (m, 2H), 7.58–7.46 (m, 3H), 7.17 (s, 1H), 7.01 (s, 1H), 6.99–6.92 (m, 2H), 4.12 (s, 2H), 3.75 (s, 3H), 3.56 (t, J = 5.7 Hz, 2H), 2.68 (q, J = 4.0, 3.4 Hz, 2H), 1.44 (s, 9H).

tert-Butyl 4-(4-((4-(Dimethylamino)phenyl)amino)-6-phenylpyrimidin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (35ca)

Title compound 35ca was prepared according to general procedure 3 using intermediate 33c (100 mg, 0.35 mmol), boronic ester 34a (117.8 mg, 0.37 mmol), Pd(dppf)Cl₂·DCM (57 mg, 0.07 mmol), and K₂CO₃ (2 M)_aq (0.52 mL, 1.05 mmol) in 1,4-dioxane dry (3.5 mL). Final normal-phase purification (cyclohexane/EtOAc from 95/5 to 75/25) afforded pure title compound 35ca (107.4 mg, 65% yield). UPLC-MS: R_t = 2.58 min (method 2). MS (ESI) m/z: 472.3 [M + H]⁺, C₂₈H₃₄N₅O₂ [M + H]⁺ calcd, 472.6. ¹H NMR (400 MHz, DMSO-d₆): δ 9.28 (s, 1H), 8.12–7.96 (m, 2H), 7.64–7.42 (m, 5H), 7.15 (s, 1H), 6.96 (s, 1H), 6.86–6.69 (m, 2H), 4.11 (d, J = 4.0 Hz, 2H), 3.55 (t, J = 5.7 Hz, 2H), 2.87 (s, 6H), 2.67 (d, J = 7.0 Hz, 2H), 1.44 (s, 9H).

tert-Butyl 4-(4-((3-(Dimethylamino)phenyl)amino)-6-phenylpyrimidin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (35da)

Title compound 35da was prepared according to general procedure 3 using intermediate 33d (120 mg, 0.37 mmol), boronic ester 34a (129.5 mg, 0.41 mmol), Pd(dppf)Cl₂·DCM (60 mg, 0.07 mmol), and K₂CO₃ (2 M)_aq (0.55 mL, 1.11 mmol) in 1,4-dioxane dry (3.7 mL). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 80/20) afforded pure title compound 35da (155.1 mg, 89% yield). UPLC-MS: R_t = 2.58 min (method 2). MS (ESI) m/z: 472.3 [M + H]⁺, C₂₈H₃₄N₅O₂ [M + H]⁺ calcd, 472.6. ¹H NMR (400 MHz, DMSO-d₆): δ 9.44 (s, 1H), 8.17–7.95 (m, 2H), 7.67–7.41 (m, 3H), 7.31 (s, 1H), 7.21 (s, 1H), 7.14 (t, J = 8.1 Hz, 1H), 7.11 (s, 1H), 7.03–6.93 (m, 1H), 6.42 (ddd, J = 8.3, 2.7, 0.8 Hz, 1H), 4.11 (s, 2H), 3.56 (t, J = 5.6 Hz, 2H), 2.93 (s, 6H), 2.81–2.62 (m, 2H), 1.44 (s, 9H).

N1-(2-(Cyclohex-1-en-1-yl)-6-phenylpyrimidin-4-yl)-N3,N3-dimethylbenzene-1,3-diamine (35db)

Title compound 35db was prepared according to general procedure 3 using intermediate 33d (70 mg, 0.22 mmol), boronic ester 34b (67 mg, 0.32 mmol), Pd(dppf)Cl₂·DCM (36 mg, 0.04 mmol), and K₂CO₃ (2 M)_aq (0.33 mL, 0.66 mmol) in 1,4-dioxane dry (2.2 mL). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 90/10) afforded pure title compound 35db (65 mg, 80% yield). UPLC-MS: R_t = 1.55 min (method 3). MS (ESI) m/z: 371.2 [M + H]⁺, C₂₄H₂₇N₄ [M + H]⁺ calcd, 370.5. ¹H NMR (400 MHz, CDCl3): δ 8.07–8.00 (m, 2H), 7.51–7.42 (m, 3H), 7.35 (s, 1H), 7.24 (d, J = 8.0 Hz, 1H), 7.02 (d, J = 1.1 Hz, 1H), 6.84 (s, 1H), 6.70 (dd, J = 7.7, 2.0 Hz, 1H), 6.56 (d, J = 8.3 Hz, 1H), 2.99 (s, 6H), 2.71–2.67 (m, 2H), 2.36–2.32 (m, 2H), 1.87–1.75 (m, 2H), 1.75–1.66 (m, 2H).

tert-Butyl 4-(4-((2-Methoxyphenyl)amino)-6-phenylpyrimidin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (35ea)

Title compound 35ea was prepared according to general procedure 3 using intermediate 33e (190 mg, 0.57 mmol), boronic ester 34a (211 mg, 0.68 mmol), Pd(dppf)Cl₂·DCM (93 mg, 0.11 mmol), and K₂CO₃ (2 M)_aq (0.86 mL, 1.7 mmol) in 1,4-dioxane dry (5.7 mL). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 85/15) afforded pure title compound 35ea (230 mg, 88% yield). UPLC-MS: R_t = 2.47 min (method 2). MS (ESI) m/z: 459.3 [M + H]⁺, C₂₇H₃₁N₄O₃ [M + H]⁺ calcd, 458.6. ¹H NMR (400 MHz, CDCl₃): δ 8.26 (dd, J = 7.0, 2.5 Hz, 1H), 8.10–8.00 (m, 2H), 7.48–7.44 (m, 3H), 7.24 (bs, 2H), 7.07–7.03 (m, 2H), 6.96–6.93 (m, 2H), 4.21 (q, J = 3.0 Hz, 2H), 3.91 (s, 3H), 3.67 (t, J = 5.8 Hz, 2H), 2.84 (bs, 2H), 1.51 (s, 9H).

tert-Butyl 4-(4-((2-(Dimethylamino)phenyl)amino)-6-phenylpyrimidin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (35fa)

Title compound 35fa was prepared according to general procedure 3 using intermediate 33f (150 mg, 0.46 mmol), boronic ester 34a (171 mg, 0.55 mmol), Pd(dppf)Cl₂·DCM (75 mg, 0.09 mmol), and K₂CO₃ (2 M)_aq (0.69 mL, 1.38 mmol) in 1,4-dioxane dry (4.6 mL). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 95/5) afforded pure title compound 35fa (134.7 mg, 62% yield). UPLC-MS: R_t = 1.54 min (method 3). MS (ESI) m/z: 472.3 [M + H]⁺, C₂₈H₃₄N₅O₂ [M + H]⁺ calcd, 472.6. ¹H NMR (400 MHz, CDCl₃): δ 8.28 (d, J = 8.0 Hz, 1H), 8.14–8.04 (m, 2H), 7.85 (s, 1H), 7.50–7.43 (m, 3H), 7.20 (dd, J = 7.8, 1.5 Hz, 1H), 7.16 (dd, J = 7.8, 1.6 Hz, 1H), 7.06 (td, J = 7.7, 1.5 Hz, 1H), 7.00 (s, 1H), 4.22 (q, J = 3.0 Hz, 2H), 3.68 (t, J = 5.7 Hz, 2H), 2.86 (bs, 2H), 2.70 (s, 6H), 1.51 (s, 9H).

tert-Butyl 4-(4-((3,4-Dimethoxyphenyl)amino)-6-phenylpyrimidin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (35ga)

Title compound 35ga was prepared according to general procedure 3 using intermediate 33g (100 mg, 0.29 mmol), boronic ester 34a (108 mg, 0.35 mmol), Pd(dppf)Cl₂·DCM (47 mg, 0.06 mmol), and K₂CO₃ (2 M)_aq (0.43 mL, 0.87 mmol) in 1,4-dioxane dry (2.9 mL). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 95/5) afforded pure title compound 35ga (85 mg, 60% yield). UPLC-MS: R_t = 2.05 min (method 2). MS (ESI) m/z: 489.3 [M + H]⁺, C₂₈H₃₃N₄O₄ [M + H]⁺ calcd, 489.6. ¹H NMR (400 MHz, CDCl₃): δ 8.03–7.95 (m, 2H), 7.50–7.42 (m, 3H), 7.23 (s, 1H), 7.05 (s, 1H), 6.95–6.89 (m, 2H), 6.83 (bs, 1H), 4.18 (s, 2H), 3.91 (s, 3H), 3.87 (s, 3H), 3.64 (s, 2H), 2.80 (s, 2H), 1.49 (s, 9H).

tert-Butyl 4-(4-((3-(2-methoxyethoxy)phenyl)amino)-6-phenylpyrimidin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (35ha)

Title compound 35ha was prepared according to general procedure 3 using intermediate 33h (70 mg, 0.20 mmol), boronic ester 34a (74 mg, 0.24 mmol), Pd(dppf)Cl₂·DCM (33 mg, 0.04 mmol), and K₂CO₃ (2 M)_aq (0.30 mL, 0.60 mmol) in 1,4-dioxane dry (2.0 mL). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 85/15) afforded pure title compound 35ha (70 mg, 70% yield). UPLC-MS: R_t = 2.11 min (method 2). MS (ESI) m/z: 503.3 [M + H]⁺, C₂₉H₃₅N₄O₄ [M + H]⁺ calcd, 502.6. ¹H NMR (400 MHz, CDCl₃): δ 7.96–7.88 (m, 2H), 7.51–7.40 (m, 3H), 7.32 (td, J = 8.1, 1.1 Hz, 1H), 7.25 (bs, 1H), 7.19 (s, 1H), 7.00 (d, J = 0.9 Hz, 1H), 6.97 (t, J = 2.3 Hz, 1H), 6.93 (dd, J = 7.9, 1.5 Hz, 1H), 6.84–6.81 (m, 1H), 4.17–4.12 (m, 2H), 3.81–3.72 (m, 2H), 3.45 (s, 3H).

tert-Butyl 4-(4-Phenyl-6-((3-(trifluoromethyl)phenyl)amino)pyrimidin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (35ia)

Title compound 35ia was prepared according to general procedure 3 using intermediate 33i (185 mg, 0.53 mmol), boronic ester 34a (197 mg, 0.64 mmol), Pd(dppf)Cl₂·DCM (86 mg, 0.10 mmol), and K₂CO₃ (2 M)_aq (0.79 mL, 1.59 mmol) in 1,4-dioxane dry (5.3 mL). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 85/15) afforded pure title compound 35ia (220 mg, 84% yield). UPLC-MS: R_t = 2.59 min (method 2). MS (ESI) m/z: 495.4 [M + H]⁺, C₂₇H₂₈F₃N₄O₂ [M + H]⁺ calcd, 496.5. ¹H NMR (400 MHz, CDCl₃): δ 8.09 (s, 1H), 8.07–7.99 (m, 2H), 7.62 (d, J = 8.1 Hz, 1H), 7.51–7.43 (m, 4H), 7.36 (d, J = 7.7 Hz, 1H), 7.26 (bs, 1H), 6.91 (s, 2H), 4.21 (q, J = 3.9 Hz, 2H), 3.67 (t, J = 5.7 Hz, 2H), 2.82 (b, 2H), 1.51 (s, 9H).

tert-Butyl 4-(4-((4-chlorophenyl)amino)-6-phenylpyrimidin-2-yl)piperidine-1-carboxylate (36aa)

Et₃SiH (1.38 mL, 8.66 mmol) and Pd/C (200 mg) were added to a solution of compound 35aa (200 mg, 0.43 mmol) in a 1:1:1 mixture of toluene/EtOAc/EtOH (6 mL) at 10 °C under argon. The reaction mixture was stirred for 30 min and filtered through a Celite pad. The residue was concentrated under vacuum and subjected to the first trituration with a 1:1 mixture of DCM/EtOAc, the resulting solid was filtered, and the mother liquor was concentrated under vacuum. Final normal-phase purification (toluene/EtOAc from 100/0 to 85/15) afforded compound 36aa (45 mg, 23% yield) impure of 3% of dechlorinated byproduct. UPLC-MS: R_t = 2.37 min (method 2). MS (ESI) m/z: 463.4 [M – H]⁻, C₂₆H₂₈ClN₄O₂ [M – H]⁻ calcd, 464.0. ¹H NMR (400 MHz, CDCl₃): δ 1H NMR (400 MHz, CDCl₃): δ 8.01–7.92 (m, 2H), 7.49–7.43 (m, 3H), 7.43–7.39 (m, 2H), 7.37–7.32 (m, 2H), 6.88 (s, 2H), 4.22 (bs, 2H), 2.95 (tt, J = 11.5, 3.7 Hz, 1H), 2.93–2.88 (m, 2H), 2.06 (d, J = 11.2 Hz, 2H), 1.89 (qd, J = 12.2, 4.3 Hz, 2H), 1.49 (s, 9H).

tert-Butyl 4-(4-((4-Methoxyphenyl)amino)-6-phenylpyrimidin-2-yl)piperidine-1-carboxylate (36ba)

Title compound 36ba was prepared according to general procedure 4 using intermediate 35ba (120 mg), NH₄CO₂H (65 mg, 1 mmol), and Pd(OH)₂/C (46 mg) in MeOH dry (6.2 mL). The crude was used as such without further purification. UPLC-MS: R_t = 1.94 min (method 2). MS (ESI) m/z: 461.2 [M + H]⁺, C₂₇H₃₃N₄O₃ [M + H]⁺ calcd, 461.6.

tert-Butyl 4-(4-((4-(Dimethylamino)phenyl)amino)-6-phenylpyrimidin-2-yl)piperidine-1-carboxylate (36ca)

Title compound 36ca was prepared according to general procedure 4 using intermediate 35ca (120 mg, 0.26 mmol), NH₄CO₂H (65 mg, 1 mmol), and Pd(OH)₂/C (46 mg) in MeOH dry (6.5 mL). The crude was used as such without further purification. UPLC-MS: R_t = 2.23 min (method 2). MS (ESI) m/z: 472.6 [M – H]⁻, C₂₈H₃₄N₅O₂ [M – H]⁻ calcd, 472.3.

tert-Butyl 4-(4-((2-Methoxyphenyl)amino)-6-phenylpyrimidin-2-yl)piperidine-1-carboxylate (36ea)

Title compound 36ea was prepared according to general procedure 4 using intermediate 35ea (230 mg, 0.50 mmol), NH₄CO₂H (189 mg, 3.00 mmol), and Pd(OH)₂/C (46 mg) in MeOH dry (8 mL). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 70/30) afforded pure title compound 36ea (191 mg, 83% yield). UPLC-MS: R_t = 2.30 min (method 2). MS (ESI) m/z: 459.5 [M – H]⁻, C₂₇H₃₁N₄O₃ [M – H]⁻ calcd, 459.6. ¹H NMR (400 MHz, CDCl₃): δ 8.19 (dd, J = 7.6, 1.9 Hz, 1H), 8.04–7.96 (m, 2H), 7.50–7.40 (m, 3H), 7.21 (s, 1H), 7.08 (td, J = 7.5, 1.9 Hz, 1H), 7.03 (td, J = 7.5, 1.8 Hz, 1H), 6.97–6.91 (m, 2H), 4.23 (bs, 2H), 2.99 (tt, J = 11.5, 3.8 Hz, 1H), 2.96–2.86 (m, 2H), 2.09 (d, J = 13.1 Hz, 2H), 1.89 (qd, J = 12.1, 4.0 Hz, 2H), 1.49 (s, 9H).

tert-Butyl 4-(4-((2-(Dimethylamino)phenyl)amino)-6-phenylpyrimidin-2-yl)piperidine-1-carboxylate (36fa)

Title compound 36fa was prepared according to general procedure 4 using intermediate 35fa (135 mg, 0.29 mmol), NH₄CO₂H (110 mg, 1.74 mmol), and Pd(OH)₂/C (27 mg) in MeOH dry (4.8 mL). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 85/15) afforded pure title compound 36fa (110 mg, 80% yield). UPLC-MS: R_t = 1.35 min (method 3). MS (ESI) m/z: 472.4 [M – H]⁻, C₂₈H₃₄N₅O₂ [M – H]⁻ calcd, 472.6. ¹H NMR (400 MHz, CDCl₃): δ 8.22 (dd, J = 8.0, 1.5 Hz, 1H), 8.07–7.99 (m, 2H), 7.83 (s, 1H), 7.48–7.43 (m, 3H), 7.19 (dd, J = 7.8, 1.5 Hz, 1H), 7.16 (td, J = 7.8, 1.6 Hz, 1H), 7.05 (td, J = 7.6, 1.5 Hz, 1H), 6.99 (s, 1H), 4.24 (bs, 2H), 2.99 (tt, J = 11.5, 3.8 Hz, 1H), 2.95–2.85 (m, 2H), 2.69 (s, 6H), 2.10 (d, J = 13.0 Hz, 2H), 1.95 (qd, J = 12.3, 4.3 Hz, 2H), 1.50 (s, 9H).

tert-Butyl 4-(6-((3,4-Dimethoxyphenyl)amino)-4-phenyl-4,5-dihydropyrimidin-2-yl)piperidine-1-carboxylate (36ga)

Title compound 36ga was prepared according to general procedure 4 using intermediate 35ga (80 mg, 0.16 mmol), NH₄CO₂H (61 mg, 0.96 mmol), and Pd(OH)₂/C (20 mg) in MeOH dry (2.7 mL). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 45/55) afforded pure title compound 36ga (41 mg, 52% yield). UPLC-MS: R_t = 1.88 min (method 2). MS (ESI) m/z: 491.4 [M + H]⁺, C₂₈H₃₇N₄O₄ [M + H]⁺ calcd, 491.6. NMR (400 MHz, CDCl₃): δ 7.95–7.93 (dd, J = 6.7, 3.0 Hz, 2H), 7.44–7.43 (m, 3H), 7.03 (s, 1H), 6.90 (bs, 3H), 6.80 (s, 1H), 4.22 (bs, 2H), 3.92 (s, 3H), 3.89 (s, 3H), 2.98–2.83 (m, 3H), 2.05 (d, J = 12.7 Hz, 2H), 1.91 (qd, J = 12.2, 4.2 Hz, 2H), 1.48 (s, 9H).

tert-Butyl 4-(4-((3-(2-Methoxyethoxy)phenyl)amino)-6-phenylpyrimidin-2-yl)piperidine-1-carboxylate (36ha)

Title compound 36ha was prepared according to general procedure 4 using intermediate 35ha (100 mg, 0.20 mmol), NH₄CO₂H (76 mg, 1.20 mmol), and Pd(OH)₂/C (20 mg) in MeOH dry (3.3 mL). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 70/30) afforded pure title compound 36ha (65 mg, 64% yield). TLC: R_f = 0.3 (75/25 cyclohexane/EtOAc). ¹H NMR (400 MHz, CDCl₃): δ 1H NMR (400 MHz, CDCl₃): δ 7.99–7.95 (m, 2H), 7.49–7.42 (m, 3H), 7.29 (td, J = 8.2, 2.0 Hz, 1H), 7.08–7.06 (m, 1H), 7.05–6.93 (m, 3H), 6.80–6.71 (m, 1H), 4.31–4.18 (m, 2H), 4.17–4.12 (m, 2H), 3.78–3.75 (m, 2H), 3.45 (s, 3H), 2.98–2.93 (m, 3H), 2.06 (d, J = 12.6 Hz, 2H), 1.90 (qd, J = 12.3, 4.3 Hz, 3H), 1.48 (s, 9H).

tert-Butyl 4-(4-Phenyl-6-((3-(trifluoromethyl)phenyl)amino)pyrimidin-2-yl)piperidine-1-carboxylate (36ia)

Title compound 36ia was prepared according to general procedure 4 using intermediate 35ia (220 mg, 0.44 mmol), NH₄CO₂H (167 mg, 2.64 mmol), and Pd(OH)₂/C (44 mg) in MeOH dry (7.3 mL). Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 75/25) afforded pure title compound 36ia (189 mg, 86% yield). UPLC-MS: R_t = 2.44 min (method 2). MS (ESI) m/z: 497.4 [M – H]⁻, C₂₇H₂₈F₃N₄O₂ [M – H]⁻ calcd, 497.5. ¹H NMR (400 MHz, CDCl₃): δ 8.03–7.93 (m, 3H), 7.65 (d, J = 8.2 Hz, 1H), 7.53–7.41 (m, 4H), 7.37 (d, J = 7.7 Hz, 1H), 6.91 (s, 2H), 4.18 (bs, 2H), 2.99 (tt, J = 11.5, 3.8 Hz, 1H), 2.96–2.91 (m, 2H), 2.10 (d, J = 13.0 Hz, 2H), 1.90 (qd, J = 11.8, 4.3 Hz, 2H), 1.49 (s, 9H).

2,4-Dichloro-6-phenyl-1,3,5-triazine (38)

Phenyl boronic acid (44 mg, 0.36 mmol), Pd(PPh₃)₂Cl₂ (12.7 mg, 0.02 mmol), and K₂CO₃ (200 mg, 1.44 mmol) were added to a solution of cyanuric chloride 37 (100 mg, 0.54 mmol) in toluene (1.8 mL) under argon. The reaction mixture stirred at 70 °C for 2 h. After that, the reaction mixture was diluted with EtOAc (2 mL) and water (3 mL), and the organic layer was divided, dried over Na₂SO₄, filtered, and concentrated under vacuum. Purification by silica (cyclohexane/EtOAc from 100 to 98/2) afforded pure compound 38 as a white solid (42 mg, 51% yield). UPLC-MS: R_t = 1.48 min (method 2), no ionization. ¹H NMR (400 MHz, CDCl₃): δ 8.55–8.46 (m, 2H), 7.66 (tt, J = 7.4, 1.4 Hz, 1H), 7.56–7.51 (m, 2H).

(1-(tert-Butoxycarbonyl)-1,2,3,6-tetrahydropyridin-4-yl)boronic Acid (39)

Boronic ester 34a (155 mg, 0.50 mmol), NaIO₄ (385 mg, 1.80 mmol), and NH₄OAc (139 mg, 1.8 mmol) were suspended in an acetone/water 1:1 mixture (5 mL). The reaction mixture was stirred for 24 h at room temperature. The resulting precipitate was filtered, and acetone was concentrated under vacuum. The aqueous solution was extracted with Et₂O (4 × 3 mL). Collected organic layers were dried over Na₂SO₄, filtered, and concentrated under vacuum. The resulting crude was used as such without further purification. UPLC-MS: R_t = 1.62 min (method 2). MS (ESI) m/z: 172.0 [M + H]⁺, C₁₀H₁₉BNO₄ [M + H]⁺ calcd, 172.1. ¹H NMR (400 MHz, CDCl₃): δ 6.43 (bs, 1H), 4.09–3.89 (m, 2H), 3.46 (q, J = 6.2 Hz, 2H), 2.34–2.26 (n, 2H), 1.47 (s, 9H). ¹H NMR analysis taken at different times (0, 18 h) showed an equilibrium between two species.

tert-Butyl 4-(4-Chloro-6-phenyl-1,3,5-triazin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (40)

K₂CO₃ (2 M)_aq (1.1 mL, 2.21 mmol), PdCl₂(PPh₃)₂ (15.5 mg, 0.02 mmol), and boronic acid 39 (250 mg, 1.11 mmol) were added to a degassed solution of compound 38 (250 mg, 1.11 mmol) in toluene dry (5 mL) under argon. The reaction mixture was stirred at 80 °C for 2 h. After that, water (3 mL), NaCl sat sol (2 mL), and DCM (5 mL) were added, and the collected organic layer was dried over Na₂SO₄, filtered, and concentrated under vacuum. Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 90/10) afforded pure title compound 40 (295 mg, 71% yield). ¹H NMR (400 MHz, CDCl₃): δ 8.57–8.46 (m, 2H), 7.62 (bs, 1H), 7.61 (tt, J = 7.3, 1.4 Hz, 1H), 7.53–7.49 (m, 2H), 4.33–4.17 (m, 2H), 3.65 (q, J = 5.7 Hz, 2H), 2.76 (bs, 2H), 1.50 (s, 9H).

tert-Butyl 4-(4-((3-(Dimethylamino)phenyl)amino)-6-phenyl-1,3,5-triazin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (41a) and tert-Butyl 4-(4-((3-(Dimethylamino)phenyl)amino)-6-phenyl-1,3,5-triazin-2-yl)-3,4-dihydropyridine-1(2H)-carboxylate (41b)

Aniline 32d (84 mg, 0.40 mmol) and DIPEA (0.38 mL, 2.15 mmol) were added to a solution of compound 40 (100 mg, 0.27 mmol) under argon. The reaction mixture was stirred for 3 h at 80 °C until complete consumption of the starting material. After that, the crude was concentrated under vacuum, the residue was risen with water (2 mL) and DCM (3 mL), and the organic layer was collected through a phase separator and concentrated under vacuum. Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 70/30) afforded a mixture of 41a and 41b (with isomerized double C=C bond) in a ratio of 1:0.4 (120 mg, 94% combined yield). ¹H NMR of major isomer 41a (400 MHz, CDCl₃): δ 8.51 (d, J = 7.7 Hz, 2H), 7.58–7.43 (m, 4H), 7.34 (s, 1H), 7.31 (s, 1H), 7.23 (d, J = 8.1 Hz, 1H), 6.94 (dd, J = 7.9, 2.0 Hz, 1H), 6.52 (dd, J = 8.3, 2.5 Hz, 1H), 4.22 (q, J = 3.8 Hz, 2H), 3.65 (t, J = 5.7 Hz, 2H), 3.01 (s, 6H), 2.77 (bs, 2H), 1.51 (s, 9H). ¹H NMR of minor isomer 41b (400 MHz, CDCl₃): δ 8.49 (d, J = 7.7 Hz, 2H), 7.57–7.52 (m, 1H), 7.52–7.45 (m, 2H), 7.39 (bs, 1H), 7.31 (bs, 1H), 7.23 (t, J = 8.1 Hz, 1H), 7.07 (d, J = 8.4 Hz, 1H), 6.94 (d, J = 8.3 Hz, 1H), 6.88 (d, J = 7.8 Hz, 1H), 6.52 (dd, J = 8.3, 2.5 Hz, 1H), 5.26–5.16 (m, 1H), 3.92–3.84 (m, 1H), 3.64 (bs, 2H), 3.01 (s, 6H), 2.36–2.27 (m, 2H), 1.52 (s, 9H).

4,6-Dichloro-N-(3-(trifluoromethyl)phenyl)-1,3,5-triazin-2-amine (42)

Aniline 32i (0.07 mL, 0.55 mmol) and DIPEA (0.11 mL, 0.61 mmol) were added to a solution of cyanuric chloride 37 (100 mg, 0.55 mmol) in DCM dry (2 mL) at 0 °C under argon. After 1 h, water (2 mL) and DCM (2 mL) were added, and the organic layer was collected through a phase separator and concentrated under vacuum. Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 80/20) afforded compound 42 (41 mg, 24% yield). UPLC-MS: R_t = 1.25 min (method 2). MS (ESI) m/z: 307.0/309.0 [M – H]⁻, C₁₀H₄Cl₂F₃N₄ [M – H]⁻ calculated, 308.1. ¹H NMR (400 MHz, CDCl₃): δ 7.84 (s, 1H), 7.78 (d, J = 8.0, 1H), 7.68 (bs, 1H) 7.55 (t, J = 7.9 Hz, 1H), 7.48 (d, J = 7.8 Hz, 1H).

tert-Butyl 4-(4-Chloro-6-((3-(trifluoromethyl)phenyl)amino)-1,3,5-triazin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (43)

K₂CO₃ (2 M)_aq (0.16 mL, 0.32 mmol) was added to a degassed mixture of compound 42 (50 mg, 0.16 mmol), boronic acid 39 (44 mg, 0.19 mmol), and PdCl₂(PPh₃)₂ (6 mg, 0.008 mmol) in toluene dry (1 mL) under argon. The reaction mixture was stirred at 90 °C for 2 h. After that, water (2 mL) was added, and the aqueous layer was extracted with EtOAc (3 × 2 mL). Collected organic layers were dried over Na₂SO₄, filtered, and concentrated under vacuum. Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 90/10) afforded compound 43 (40 mg, 55% yield). R_f = 0.40 (8/2 cyclohexane/EtOAc). ¹H NMR (400 MHz, CDCl₃): δ 8.17 (bs, 1H), 7.63 (bs, 1H), 7.55–7.31 (m, 4H), 4.27–4.16 (m, 2H), 3.61 (t, J = 5.7 Hz, 2H), 2.63 (bs, 2H), 1.49 (s, 9H).

tert-Butyl 4-(4-Phenyl-6-((3-(trifluoromethyl)phenyl)amino)-1,3,5-triazin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (44)

K₂CO₃ (2 M)_aq (0.60 mL, 0.12 mmol) was added to a degassed mixture of compound 43 (36 mg, 0.08 mmol), phenylboronic acid (12 mg, 0.10 mmol), and PdCl₂(PPh₃)₂ (3 mg, 0.004 mmol) in toluene dry (0.4 mL) under argon. The reaction mixture was stirred at 120 °C for 2 h. After that, water (2 mL) was added, and the aqueous layer was extracted with EtOAc (3 × 2 mL). Collected organic layers were dried over Na₂SO₄, filtered, and concentrated under vacuum. Final normal-phase purification (cyclohexane/EtOAc from 100/0 to 90/10) afforded compound 44 (15 mg, 38% yield). R_f = 0.50 (75/25 cyclohexane/EtOAc). ¹H NMR (400 MHz, CDCl₃): δ 8.54–8.46 (m, 2H), 8.35 (s, 1H), 7.71 (d, J = 8.1 Hz, 1H), 7.63 (s, 1H), 7.60–7.54 (m, 1H), 7.54–7.44 (m, 4H), 7.38 (d, J = 7.9 Hz, 1H), 4.24 (q, J = 3.6 Hz, 2H), 3.66 (t, J = 5.8 Hz, 2H), 2.76 (bs, 2H), 1.51 (s, 9H).

In Vitro and In Vivo Experiments

Cell Viability

Cell viability assay was performed as previously described.¹² Cancer cell lines were cultured according to the method described in Jahid et al., 2022.¹²

In Vivo Efficacy

In vivo efficacy test was performed as described previously with a slight modification.¹² After tumors reached an initial size range of 150–250 mm³, mice were administrated 10 mg/kg ARN25062, ARN24928, or vehicle via tail vein injection for 1–2 weeks daily. Tumors were measured every other day with a caliper, and tumor volume was calculated ((length × width × width)/2). At the end of 2 weeks, tumors were extracted and measured to determine volume (length × width × height). GraphPad Prism9 software was used to generate line and bar graphs and determine two-way ANOVA and two-tailed T-test. Equal numbers of males and females were used in the experiments.

Computational Methods

Molecular Docking

In order to predict and evaluate the interaction between CDC42 and ARN24928/ARN25062 compounds, we first performed molecular docking. As a receptor structure, we employed the CDC42 protein in complex with the CRIB domain of PAK6 (PDB code 2ODB, resolution of 2.4 Å) where we formerly identified a previously unappreciated allosteric pocket at the protein–protein interface.¹² The structure was refined by using the Protein Preparation Wizard³⁰ workflow implemented in Maestro Release 2021-3. Specifically, hydrogen atoms were added, and charges and protonation states were assigned, titrating the protein at physiologic pH. The steric clashes were relieved by performing a small number of minimization steps, until the rmsd of the non-hydrogen atoms reached 0.30 Å. The formerly identified pocket was used to center the grid. Precisely, the cubic grid box of 26 × 26 × 26 ų was centered on the previously identified hit compound of the CDC42-ligand complex.¹² The compound was prepared using LigPrep software implemented in Maestro. First, we added hydrogens and generated ionization states at pH 7.4 ± 0.5. Then, we generated tautomers and all stereochemical isomers. Finally, we used Glide³¹⁻³³ to perform the molecular docking, using Extra Precision and retaining maximum 20 poses. The best docking pose was chosen for further investigation through classical MD simulation.

MD Simulations

MD simulations were performed on the protein–ligand complexes obtained from our docking calculations. The GTP substrate as well as the catalytic Mg²⁺ ion in the active site of the proteins were considered. The systems were hydrated with a 14 Å layer of TIP3P water molecules³⁴ from the protein center. The coordinates of the water molecules at the catalytic center were taken from the PDB X-ray structure 2ODB. Sodium ions were added to neutralize the charge of the systems. The final models are enclosed in a box of ∼89 × 89 × 89 ų, containing ∼18,800 water molecules, resulting in ∼59,000 atoms for each system. The AMBER-ff14SB force field³⁵ was used for the parametrization of the protein. The parameters for the ligands ARN24928 and ARN25062 were determined via Hartree–Fock calculation, with the 6-31G* basis set, convergence criterium SCF = tight after structure optimization (DFT B3LYP functional; 6-31G* basis set). The Merz-Singh-Kollman scheme³⁶ was used for the atomic charge assignment. The GTP and the Mg²⁺ were parametrized according to Meagher et al. and Allner et al. respectively.^37,38 Joung–Chetham parameters were used for monovalent ions.³⁹ All MD simulations were performed with Amber,⁴⁰ and all the systems were the object of the following equilibration protocol. To relax the water molecule and the ions, we performed an energy minimization, imposing a harmonic potential of 300 kcal/mol Ų on the backbone, the GTP, and the docked compounds. Then, two consecutive MD simulations in NVT and NPT ensembles (1 and 10 ns, respectively) were carried out, imposing the previous positional restraints. To relax the solute, two additional energy minimization steps were performed, imposing positional restraints of 20 kcal/mol Ų and without any restraints, respectively. Such minimized systems were heated up to 303 K with four consecutive MD simulations in NVT (∼0.1 ns, 100 K) and NPT ensembles (∼0.1 ns, 100 K; ∼0.1 ns, 200 K; and ∼0.2 ns, 303 K), imposing the previous positional restraints of 20 kcal/mol Ų. We used the Andersen-like temperature-coupling scheme,⁴¹ while pressure control was achieved with the Monte Carlo barostat at a reference pressure of 1 atm. Long-range electrostatics were treated with the particle mesh Ewald method. We performed an additional MD simulation (∼1.5 ns) in the NPT ensemble at 303 K without any restraint to relax the system at such temperature. Finally, multiple replicas of 500 ns were performed in the NPT ensemble for each system with an integration time step of 2 fs.

Biophysical Methods

His-CDC42 Production and Purification

The His-CDC42 wild type (amino acids Ile4-Pro182) was expressed from the pET28a+ vector in Escherichia coli BL21 (DE3) cells and purified, GppNHp or GDP-bound, as previously described.¹²

Binding Check of Hit Derivatives by Microscale Thermophoresis

MicroScale Thermophoresis experiments were performed according to the NanoTemper Technologies protocols in a Monolith NT.115 Pico (Pico Red/Nano Blue—NanoTemper Technologies). His-CDC42 was RED–NHS-labeled and used at a concentration of 10 nM. Compound concentration was 50 μM throughout all the experiments. DMSO was also constant across samples at 0.5% v/v. Solutions were prepared in 100 mM Trizma base (Sigma) pH 7.5, 40 mM NaCl, 0.05% v/v Tween 20 or PBS, 0.05% v/v Tween 20 and incubated 5 min before loading on Premium Capillaries and analysis. Binding was detected at 24 °C, MST power high, and 20% light-emitting diode power. The MST traces were recorded as follows: 3 s MST power off, 20 s MST power on, and 1 s MST power off. The difference in normalized fluorescence (ΔF_norm [‰] = F_hot/F_cold) between the protein/compound sample and a protein-only sample at 1.5–2.5 s is calculated and plotted through MO.Affinity analysis v2.3 (NanoTemper Technologies) and GraphPad Prism 8.0.0 (GraphPad Software, San Diego, California USA). The signal-to-noise ratio and response amplitude were used to evaluate the quality of the binding data according to NanoPedia instructions (NanoTemper Technologies). Only a signal-to-noise ratio of more than 5 and a response amplitude of more than 1.5 were considered acceptable, while a signal-to-noise of more than 12 was considered excellent.

NMR Confirmation of Target Engagement

All the NMR experiments were recorded at 25 °C using a NMR Bruker 600 MHz NEO equipped with a 5 mm CryoProbe QCI ¹H/¹⁹F−¹³C/¹⁵N−D quadruple resonance, shielded z-gradient coil, and an automatic sample changer SampleJet system with temperature control. For all samples, a 1D ¹H NMR experiment was recorded, and the water suppression was obtained using the standard NOESY (nuclear Overhauser effect spectroscopy) presat Bruker pulse sequence, with 64 k data points, a spectral width (sw) of 30 ppm, 64 scans, acquisition time (aq) of 1.835 s, a relaxation delay (d1) of 4 s, and a mixing time of 10 ms. ¹H T₂ filter experiments were recorded using the CPMG⁴² spin-echo train sequence with a total echo time to 750 ms consisting of 150 repetitions with a τ time of 5 m and a 180° pulse of approximately 27 μs, 128 scans, d1 4 s, sw 30 ppm, aq 1.835 s. The WaterLOGSY experiments were achieved with a 7.5 ms-long 180° Gaussian-shaped pulse, aq 0.852 s, mixing time of 1.7 s, relaxation delay of 2 s, 512 scans. ¹⁹F T₂ filter experiments were recorded using the CPMG spin-echo scheme with a 50 ms time interval between the 180 pulses and different total lengths (200 and 400 ms, respectively), 64 scans, sw 40 ppm, aq 0.72 s and d1 5 s. The data were multiplied with an exponential window function with 1 Hz line broadening prior to Fourier transformation for ¹H 1D, ¹H T₂, and ¹⁹F T₂ filter experiments and 2 Hz line broadening for the WaterLOGSY experiments. The solubility of the compounds were evaluated by ¹H 1D experiments and aggregation by WaterLOGSY, testing the compounds in the binding assay buffer at theoretical concentrations of 10, 50, and 100 μM in the presence of 400 μM 4-trifluoromethyl benzoic acid (internal reference).

To evaluate the stability and the aggregation state of both activated and inactive forms of His-CDC42, a mixture of 35 fluorinated fragments (soluble and not aggregating) was tested at 20 μM in the absence and in the presence of the proteins just after purification (t₀) and 24 h later (t₁) by ¹⁹F T₂ filter experiments.

For the binding experiments, all the compounds were tested at 50 μM in 20 mM Trizma base (Sigma) pH 7.5, 40 mM NaCl, 5 mM MgCl₂, 5 μM ethylenediaminetetraacetic acid, 10% D₂O (for the lock signal) and in the presence of 2 μM GppNHp or GDP and 2 μM His-CDC42 (loaded with GDP) or His-CDC42 (loaded with GppNHp). The total amount of DMSO-d₆ in all sample was 1%. All fluorine chemical shifts were referred to the CFCl₃ signal in water.

Glossary

Abbreviations

DCM

dichloromethane

DMF

N,N-dimethylformamide

DMSO

dimethyl sulfoxide

EtOAc

ethyl acetate

MST

microscale thermophoresis

THF

tetrahydrofuran

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jmedchem.3c00276.

• ¹H, ¹³C, and ¹⁹F NMR spectral data and chromatography analysis of key compounds; MST and NMR analyses for binding evaluation; molecular docking of compounds 17, 27, and 28; procedures for aqueous kinetic solubility, in vitro metabolic stability, and in vitro plasmatic stability; and PK studies (PDF)

• Molecular formula strings (CSV)

Author Contributions

^# N.B., L.M.V., and I.M.A. contributed equally.

The authors declare the following competing financial interest(s): A.K.G., M.D.V. and J.O. are co-inventors on patents related related to this work. A.K.G and M.D.V are also founders of Alyra Therapeutics, a startup company that is investigating the use of these compounds for the treatment of human diseases.