Phenylamino pyrimidine-1,2,3-triazole derivatives as analogs of imatinib: searching for novel compounds against chronic myeloid leukemia

This is an open access work under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0). Please note that the reuse, redistribution and reproduction in particular requires that the author(s) and source are credited and that individual graphics may be subject to special legal provisions. The license is subject to the Beilstein Archives terms and conditions: https://www.beilstein-archives.org/xiv/terms. The definitive version of this work can be found at https://doi.org/10.3762/bxiv.2021.44.v1 This open access document is posted as a preprint in the Beilstein Archives at https://doi.org/10.3762/bxiv.2021.44.v1 and is considered to be an early communication for feedback before peer review. Before citing this document, please check if a final, peer-reviewed version has been published.

the same binding site as imatinib, probably sharing a competitive inhibition mechanism. One compound showed the greatest interaction affinity for BCR-Abl-1 in the docking studies.

Introduction
Changes in tyrosine kinase proteins (TKPs), either by mutation or chromosomal translocation, can turn them into potent oncogenes. Continuity and increased signaling are associated with some types of neoplasms, such as chronic myeloid leukemia (CML) [1]. A CML is a neoplasm of the bone marrow that transforms normal hematopoietic progenitor cells into malignant cells [2]. This transformation is marked by the presence of an acrocentric chromosome known as Philadelphia (Ph). Ph is the result of the reciprocal chromosomal conversion between the proto-oncogene Abl1 of chromosome 9 and the BCR gene on chromosome 22 [3,4].
Research activity on compounds able to act as protein tyrosine kinase inhibitors (TKIs), which has intensified since the 1980s, has led to the identification of the (phenylamino)pyrimidine (PAP) structure [5,6]. The addition of an additional pyridine ring to PAP raised its cellular activity, producing PAPP, which, after some more chemical modifications, culminated in imatinib (IMT) [7]. PAPP has been used to develop new TKIs that are even more potent than IMT, such as nilotinib [6,7]. These drugs act as inhibitors at the ATP binding site in the inactive form of BCR-Abl-1, preventing the binding of the protein to ATP in a competitive manner and resulting in the interruption of the substrate phosphorylation process and the transduction of signals; these processes induce cell apoptosis [8,9]. As mutations in the BCR-Abl-1 enzyme domain can occur, cases of resistance have emerged in the treatment with TKIs, compromising its effectiveness [10,11]. In continuation of the work by our group to develop new imatinib analogs [12,13], in this work, we designed a series of imatinib 1,2,3-triazole analogs 1a,b and 2a-j ( Figure 1). The 1,2,3-triazoles are heterocyclic compounds, consisting of a five-membered ring, containing two carbon atoms and three nitrogen atoms [14]. The application of click chemistry, a concept developed by Sharpless and collaborators, to the Huisgen cycloaddition (a 1,3-dipolar cycloaddition reaction to obtain this heterocycle) allowed the regiospecific synthesis of the 1,4-disubstituted isomers with good yields in the presence of copper(I) salts [15][16][17]. Since then, medicinal chemists used the 1,2,3-triazole core strategically in the design of new compounds [18,19]. The explanation for this interest is associated with its resistance towards oxidation, reduction, and acidic or basic hydrolysis reactions that occur in phase I of human metabolism [20,21]. The application of this core through isosterism (classical or non-classical) in the design of new active com-pounds has shown good results [22,23]. Specifically, in the development of new inhibitors for the treatment of cancer [24], its use is associated with producing compounds able to overcoming resistance (primary or secondary) inherent to this type of treatment [25]. In this regard, Arioli and co-workers developed a series of derivatives (from PAPP) containing the 1,2,3-triazole nucleus in place of the amide group (present in the IMT structure) that showed inhibitory activity of recombinant Abl kinase equivalent to the drug of reference [26].
In this context, the intent is to replace the amide group present in IMT by the 1,2,3-triazole nucleus, similar to compound 3 described in the literature by Li and co-workers, which showed a good inhibition profile of the myelogenous leukemia K562 cell line [27].
Considering the good inhibition profile of substance 3 against K562 cells, we proposed, through the molecular simplification strategy, the derivatives 1a and 1b, which have hydroxyalkyl groups replacing the phenylpiperazine group. The hydroxymethyl and hydroxypropyl groups were chosen because they were not evaluated in the original work that led to compound 3, and, additionally, they are able to act as acceptors and donors in hydrogen bonds and thus were expected to increase the aqueous solubility of the compounds.
Additionally, Kalesh and co-workers demonstrated good results with derivative 4 which contains the amide group and the 1,2,3triazole nucleus [28]. Thus, we designed derivatives 2a-j comprising both groups, among which we included a methylene spacer in order to increase the degree of conformational freedom and to investigate its impact on the activity and interaction with the enzyme BCR-Abl-1. Regarding the phenylpiperazine group present in compound 3, we have simplified the side chain attached to the 1,2,3-triazole nucleus by including different groups, such as ester (2i,j), cycloalkanes (2e,f), a saturated heterocycle in 2g and an aromatic ring in 2h whose stereoelectronic and physicochemical characteristics vary significantly, in order to be able to carry out an initial comparison of the relationship between structure and activity.
The 1,3-dipolar cycloaddition reactions via the copper-catalyzed 1,3-dipolar cycloaddition reaction (CuAAC) of the azides 5 and 9 with suitably functionalized acetylenes 6a-j, using sodium ascorbate and copper sulfate in ACN/H 2 O 2:1 under microwave irradiation were carried out to obtain the 1,4-regioisomers of the final products 1a,b, and 2a-j, respectively, with 30-84% yields. This last step was adapted from a method already described in the literature [31].
The formation of compounds 1a,b and 2a-j was observed by the disappearance of the characteristic stretching of the azide groups at 2107 and 2103 cm −1 in the IR spectra, which are present in intermediates 5 and 9, respectively, and compounds 2a-j showed carbonyl absorption at 1671-1660 cm −1 (amide). 1 H NMR spectrum analysis showed the appearance of a single signal at 8.52 ppm and between 8.61-7.79 ppm referring to the hydrogen of the 1,2,3-triazole (C-5) for compounds 1a,b and 2a-j, respectively. In the 13 C NMR spectra, the signals in the range of 165.8-163.5 ppm are attributed to the amide carbonyl in products 2a-j.
The lower yields in the synthesis of the final products 2e-g can be associated with both the difficulty of purifying some products and the high volatility of some acetylenes used, which have been lost at the time of irradiation.
All compounds were obtained with satisfactory purity (>95%) determined by liquid chromatography or elemental analysis. Despite their purity, compounds 2b, 2d, 2e, and 2h had an absence of some signals in the 1 H NMR and 13 C NMR spectra.
For structural confirmation of these new derivatives, we also carried out an X-ray diffraction study of compound 2b. Yellow single crystals suitable for X-ray diffraction analysis were obtained by slow evaporation of a solution of 2b in dichloromethane. Based on the X-ray crystallographic analysis, the molecular structure was confirmed, as shown in Figure 2.

Biological assay
The compounds showed significant inhibitory activity at 10 μM but not at 1 μM (Figure 3).
Compounds 2c, 2d, and 2g were the most promising, and concentration-response curves were drawn to determine the cytotoxic concentration (CC 50 ) for 50% of both the K562 and WSS-1 cells so that the selectivity indexes (SIs) could be calculated. These data are shown in Table 1.
Even with a small number of compounds, the structure-activity relationship (SAR) shows that with the exception of compound   2h, which presents a phenyl ring as a side chain (R), derivatives with aliphatic side chains showed greater toxic activity against K562 cells (Figure 3). One hypothesis is that the derivatives that present higher lipophilicity (2c, 2d, and 2g) quantified by the octanol-water partition coefficient (cLogP, calculated by ALOGPS) as shown in Figure 3 [32], may have better passive diffusion through the membrane of the target cell, increasing its concentration in the intracellular medium and consequently its activity.
In relation to CC 50 values of the compounds selected for the cell viability test (2c, 2d, and 2g), all were less potent against K562 cells and more toxic to WSS-1 than IMT (used as a positive control), with a low SI (Table 1). According to Xie and co-workers, the high selectivity of IMT is related to an important hydrogen bond between the Glu305 residue and its amide group, which is probably absent when the synthesized compounds interact with Abl domain [33]. It is worth mentioning that compounds with low SIs can act as chemotherapeutics [34][35][36]. Despite being less potent than IMT, derivatives 2c, 2d, and 2g were capable of inhibiting K562 cells on a micromolar scale. Thus, these compounds can be used as a prototype for designing new series of substances with greater potency and less toxicity than IMT, with lesser effects for the patient.

Molecular docking
Validation of the molecular docking protocol was performed through redocking of the IMT complexed to the BCR-Abl-1 structure (PDB code: 3PYY) [37]. Thus, the predicted mode with the lowest energy presented a MolDock value of −206.022 arbitrary units (a.u.) and a mean-square deviation of 1.68 Å, validating the molecular docking protocol with RMSD values below 2.00 Å [38].
The results, using the validated molecular docking protocol, show that 2c, 2d, and 2g interact with BCR-Abl-1 at the same binding site as IMT but show differences in the binding modes and with higher values of interaction energy. Compound 2c presented a MolDock value of −152.993 a.u. For compound 2d, the value was −152.127 a.u., and for compound 2g, it was −167.520 a.u. (Table 2).
Therefore, the computational simulations suggest that the synthesized compounds interact at the same binding site as IMT, probably sharing a competitive inhibition mechanism. The predicted affinity between IMT and this enzyme was higher than the affinities found for compounds 2c, 2d, and 2g. Among the newly synthesized derivatives, compound 2g presented the greatest interaction affinity for BCR-Abl-1 when compared to compounds 2c and 2d, which presented almost equivalent interaction energy values.
In addition, the hydrogen bond between the amino group of IMT with the Thr334 residue [32,38] (Figure 4a) was not observed for compounds 2c, 2d, and 2g, which suggests a lower binding energy of the synthesized compounds when compared to IMT ( Table 2). The lacking of that interaction prevents the pyridine ring from mimicking the ATP adenine in the hinge region [39]. As previously mentioned, the absence of the hydrogen-bond interaction between the Glu305 residue and the amide group of 2c, 2d, and 2g (Figure 4b-d) also corroborates their low SI (Table 1).

Conclusion
A new series of 1,2,3-triazole derivatives of IMT that contain the main pharmacophoric group PAPP has been synthesized and characterized. All compounds 1a,b, and 2a-j were evaluated in an in vitro cell test; these compounds included intermediate 9 and IMT (reference drug). Three of the compounds, 2c, 2d, and 2g, were found to be active, with IC 50 values between 1.0 and 7.3 μM, while the IC 50 of the reference compound was 0.08 μM. Molecular docking studies suggested that the synthesized compounds 2c, 2d, and 2g interact at the same binding site as IMT, probably sharing a competitive inhibition mechanism. In addition, among the three synthesized derivatives, compound 2g had the highest interaction affinity for BCR-Abl-1 when compared to compounds 2c and 2d, which presented almost equivalent interaction energy values. However, these results cannot be directly correlated with in vitro antimyeloproliferative assessments, which require an enzyme inhibition assay against BCR-Abl-1.

Experimental
The experimental details and analytical data for intermediates 5, 8, and 9 and the title compounds 1a,b, and 2a-j are given in Supporting Information File 1. The chemical structures of the title compounds were confirmed by 1 H and 13 C NMR spectroscopic analyses and HRMS spectrometric analyses.

X-ray data collection and structure refinement
Single crystal X-ray data for 2b were collected on a Bruker D8 Venture diffractometer using graphite-monochromated Mo Kα radiation (λ = 0.71073 Å) at 298 K. Data collection, cell refinement, and data reduction were performed with Bruker Instrument Service v4.2.2, APEX2 [40] and SAINT [41], respectively. Absorption correction using equivalent reflections was performed with the SADABS program [35]. The structure solutions and full-matrix least-squares refinements based on F 2 were performed with the SHELX package [42-44] and were refined with fixed individual displacement parameters [Uiso\(H) = 1.2 Ueq (Csp 2 and C ar ) or 1.5 U eq (Csp 3 )] using a riding model. All nonhydrogen atoms were refined anisotropically. Some crystallization and X-ray diffraction experiments for compound 2b were performed. All samples evaluated had low scattering patterns and germinated crystals. This resulted in poor quality data, which limited the quality of the refinement. Structure illustrations were generated using ORTEP-3 for Mercury [45], and crystallographic tables were constructed using Olex2 [46]. X-ray crystallographic data in cif format are available at CCDC 2073131 can be obtained free of charge via http://www.ccdc.cam.ac.uk/data_request/cif.

Cell line and cell culture
All novel compounds synthesized (1a,b, and 2a-j) were evaluated for their activity against K562 cells (ATCC ® CRL-243TM), a CML cell line, and a WSS-1 healthy cell line (ATCC ® CRL-2029TM). The K562 strain was grown in RPMI-1640 medium according to the provider's recommendations [47,48]. Cultivation of WSS-1 cells was performed in DMEM high glucose medium (Vitrocell) according to ATCC recommendations. The media were supplemented with 10% fetal bovine serum (FBS) and 50 mg/MI of the antibiotic gentamicin, and all cell lines were grown in a cell culture bottle, which had a 0.22 µm pore membrane filter on the lid, allowing the circulation of CO 2 . WSS-1 cells grew adherently, while cells of the K562 strain grew in suspension. All strains were maintained at 35 °C, at 5% CO 2 .
Cell viability assay WSS-1 cells were incubated at 5 × 10 4 (cells per well) in black plates with a 96-well transparent bottom (Greiner Bio-One) for approximately 20 hours to allow for cell growth and adhesion. The compounds obtained were added at concentrations of 1 µM and 10 µM in DMSO and incubated for 46 hours. This assay was performed in triplicate and measured using resazurin (Sigma-Aldrich) at a concentration of 0.1 mg/mL using a FlexStation 3 microplate reader (Molecular Devices).

K562 cells assay
After plating (2 × 10 4 cell/mL), the K562 cells were incubated for one hour, and then the compounds dissolved in DMSO (Sigma-Aldrich) were added and further incubated for 47 hours. After this period, resazurin was added at a final concentration per well of 0.01 mg/mL, and the first fluorescence reading was immediately performed (λ ex = 560 nm; λ ex = 590 nm) (time zero) using FlexStation 3 microplates (Molecular Devices). After the first reading, the plate was returned to the incubator, and after one hour, the second reading was performed, completing the 48 hours of incubation with the compounds. The reference inhibitor used was IMT at fixed concentrations of 1 μM and 10 μM.

Molecular docking
The Molegro Virtual Docker v6.0 (MVD) program was used to predict the complexes and energies of interaction between the enzyme BCR-Abl 1 and inhibitors 2c, 2d, and 2g. The crystallographic structures of the enzyme BCR-Abl-1, complexed with IMT, were extracted from the Protein Data Bank (PDB code: 3PYY) [37]. The structures of the compounds were constructed and optimized by the semi-empirical Recife Model 1 (RM1) method using the Spartan 14 program (Wavefunction, Inc.).
Validation of the molecular docking protocol was performed through the redocking of IMT complexed with the BCR-Abl-1 structure (PDB code: 3PYY), free of water molecules and cofactors, using Molegro Virtual Docker 6.0 (MVD) [49]. Thus, the whole enzyme was set as the center of the search space. Ad-ditionally, the search algorithm MolDock optimizer was used with a minimum of 200 runs, and the parameter settings were as follows: population size = 300; maximum iteration = 2000; scaling factor = 0.50; offspring scheme = Scheme 1; termination scheme = variance-based; crossover rate = 0.90. Due to the stochastic nature of the algorithm search, ten independent simulations per ligand were performed to predict the binding mode. Consequently, the complexes with the lowest interaction energy were evaluated. The interactions between BCR-Abl-1 and each inhibitor were analyzed using the ligand map algorithm, a standard algorithm in the MVD program [50].