New Inhibitors of Bcr-Abl Based on 2,6,9-Trisubstituted Purine Scaffold Elicit Cytotoxicity in Chronic Myeloid Leukemia-Derived Cell Lines Sensitive and Resistant to TKIs

Bcr-Abl is an oncoprotein with aberrant tyrosine kinase activity involved in the progression of chronic myeloid leukemia (CML) and has been targeted by inhibitors such as imatinib and nilotinib. However, despite their efficacy in the treatment of CML, a mechanism of resistance to these drugs associated with mutations in the kinase region has emerged. Therefore, in this work, we report the synthesis of 14 new 2,6,9-trisubstituted purines designed from our previous Bcr-Abl inhibitors. Here, we highlight 11b, which showed higher potency against Bcr-Abl (IC50 = 0.015 μM) than imatinib and nilotinib and exerted the most potent antiproliferative properties on three CML cells harboring the Bcr-Abl rearrangement (GI50 = 0.7–1.3 μM). In addition, these purines were able to inhibit the growth of KCL22 cell lines expressing Bcr-AblT315I, Bcr-AblE255K, and Bcr-AblY253H point mutants in micromolar concentrations. Imatinib and nilotinib were ineffective in inhibiting the growth of KCL22 cells expressing Bcr-AblT315I (GI50 > 20 μM) compared to 11b–f (GI50 = 6.4–11.5 μM). Molecular docking studies explained the structure–activity relationship of these purines in Bcr-AblWT and Bcr-AblT315I. Finally, cell cycle cytometry assays and immunodetection showed that 11b arrested the cells in G1 phase, and that 11b downregulated the protein levels downstream of Bcr-Abl in these cells.


Introduction
Leukemia, a hematological malignancy, involves the uncontrolled proliferation of abnormal white blood cells.This neoplastic condition results from genetic mutations disrupting hematopoiesis and poses a significant healthcare challenge [1].Among these mutations, the BCR:ABL fusion gene stands out, a product of the combination between the Abelson (ABL) tyrosine kinase gene on chromosome 9 and the break-point cluster (BCR) gene on chromosome 22 [2].This genetic abnormality is a hallmark molecular feature of chronic myeloid leukemia (CML) and a significant contributing factor to other leukemia subtypes, such as B-cell precursor-positive acute lymphoblastic leukemia (ALL), promoting the pathogenesis of these diseases.The oncoprotein Bcr-Abl, a mutated tyrosine kinase (TK), contributes to the progression and persistence of leukemia by interacting with numerous downstream signaling pathways.These interactions lead to modifications in cellular adhesion, stimulation of mitogenic signaling, and suppression of apoptosis, ultimately resulting in the malignant transformation of hematopoietic stem cells [2][3][4].
To overcome this challenge, tyrosine kinase inhibitors (TKIs) represent the gold standard for the treatment of leukemia.The development of a small molecule with the ability to block Bcr-Abl activity dramatically impacts the disease's management.The TKI impairs the interaction of the oncoprotein with ATP, thereby blocking cell signals and, consequently, reducing cell proliferation and inducing cell death [5].Imatinib was the first TKI approved by the FDA that had efficacy in treating patients with the wild-type (WT) BCR:ABL gene (Figure 1).However, due to persistent resistance to this treatment, the need arose to develop second-generation TKIs, such as dasatinib and nilotinib, as well as third-generation TKIs like ponatinib (Figure 1) [2,5,6].
abnormal white blood cells.This neoplastic condition results from genetic mutati rupting hematopoiesis and poses a significant healthcare challenge [1].Among th tations, the BCR:ABL fusion gene stands out, a product of the combination betw Abelson (ABL) tyrosine kinase gene on chromosome 9 and the break-point cluste gene on chromosome 22 [2].This genetic abnormality is a hallmark molecular fe chronic myeloid leukemia (CML) and a significant contributing factor to other le subtypes, such as B-cell precursor-positive acute lymphoblastic leukemia (ALL), p ing the pathogenesis of these diseases.The oncoprotein Bcr-Abl, a mutated tyrosin (TK), contributes to the progression and persistence of leukemia by interacting w merous downstream signaling pathways.These interactions lead to modification lular adhesion, stimulation of mitogenic signaling, and suppression of apopto mately resulting in the malignant transformation of hematopoietic stem cells [2][3][4] To overcome this challenge, tyrosine kinase inhibitors (TKIs) represent t standard for the treatment of leukemia.The development of a small molecule w ability to block Bcr-Abl activity dramatically impacts the disease's management.impairs the interaction of the oncoprotein with ATP, thereby blocking cell sign consequently, reducing cell proliferation and inducing cell death [5].Imatinib was TKI approved by the FDA that had efficacy in treating patients with the wild-typ BCR:ABL gene (Figure 1).However, due to persistent resistance to this treatment, t arose to develop second-generation TKIs, such as dasatinib and nilotinib, as well a generation TKIs like ponatinib (Figure 1) [2,5,6].Despite the successes achieved by the existing TKI-based therapy so far, re remains an obstacle.A considerable number of patients treated with imatinib (2 and <10% of patients treated with second-generation TKIs show intrinsic or acqu sistance to treatment during the disease.The basic mechanisms of resistance can gorized into two types: Bcr-Abl-dependent and Bcr-Abl-independent mechanis latter consist mainly of increased drug efflux/decreased uptake and activation of tive onco-pathways.Bcr-Abl-dependent mechanisms primarily result from the acq of point mutations in the BCR:ABL gene, although other rarer mechanisms occur, BCR:ABL gene amplification and hyperexpression, or mutations in other cancer genes [5,7].BCR:ABL gene mutations affect the binding of TKIs to different segm the tertiary structure, such as the phosphate-binding loop (P-loop), the ATP-bindi or the activation loop (A-loop).The most prevalent mutation is T315I, which prev correct binding of the TKI to the protein and impairs the activity of imatinib an second-generation TKIs.Although more than 90 different mutations have been de the negative impact on the clinical outcome of P-loop mutations, including G250E, and E255K/V, has been widely demonstrated [3,8].Despite the successes achieved by the existing TKI-based therapy so far, resistance remains an obstacle.A considerable number of patients treated with imatinib (20-30%) and <10% of patients treated with second-generation TKIs show intrinsic or acquired resistance to treatment during the disease.The basic mechanisms of resistance can be categorized into two types: Bcr-Abl-dependent and Bcr-Abl-independent mechanisms.The latter consist mainly of increased drug efflux/decreased uptake and activation of alternative onco-pathways.Bcr-Abl-dependent mechanisms primarily result from the acquisition of point mutations in the BCR:ABL gene, although other rarer mechanisms occur, such as BCR:ABL gene amplification and hyperexpression, or mutations in other cancer-related genes [5,7].BCR:ABL gene mutations affect the binding of TKIs to different segments of the tertiary structure, such as the phosphate-binding loop (P-loop), the ATP-binding cleft, or the activation loop (A-loop).The most prevalent mutation is T315I, which prevents the correct binding of the TKI to the protein and impairs the activity of imatinib and most second-generation TKIs.Although more than 90 different mutations have been described, the negative impact on the clinical outcome of P-loop mutations, including G250E, Y253H, and E255K/V, has been widely demonstrated [3,8].
Although the significant advances achieved with drugs such as dasatinib, imatinib, and nilotinib, resistance to these agents remains a clinical challenge.For this reason, Bcr-Abl remains a highly attractive target for the development of potent and selective inhibitors that will represent an important new class of therapeutic agents for the treatment of leukemia [5].
Several new therapies for leukemia are currently in development, and some are under preclinical investigation, most of which are ATP-competitive Bcr-Abl TKIs overriding the T315I mutation, but there are very few reports evaluating the same molecule on different mutations [9,10].Since the most potent inhibitors of Bcr-Abl are those that bind directly to the ATP binding site, the purine core was chosen as the basis for designing the compounds in this study.This heterocycle has been used as a privileged scaffold in the development of new Bcr-Abl inhibitors.Azam and collaborators performed different substitutions of the 3-hydroxyphenylethyl group at N-9 and discovered that the compound AP23846 can bind within the active site of both Bcr-Abl WT and Bcr-Abl T315I [11].In this context, in 2020, our working group identified promising 2,6,9-trisubstituted purine derivatives I-III (Figure 2A), which showed enhanced inhibition of Bcr-Abl with an IC 50 of 0.040-0.090µM in an Abl kinase inhibition assay [12].In addition, compound III was more selective to Bcr-Abl than other tyrosine kinases (BTK) and a serine/threonine kinase (CDK-2).In addition, III exhibited low micromolar cytotoxicity on several leukemia cancer cell lines and decreased the phosphorylation of downstream proteins in the signaling pathways of Bcr-Abl [12].Later, in 2022, our group demonstrated that the optimal substituent at N-9 is the cyclopropylmethyl group of I-III, as evidenced in the high IC 50 values for compounds IV-VI bearing longer hydrophobic substitutions (Figure 2A) and confirmed by molecular docking, which is attributed to the size of the hydrophobic pocket in Bcr-Abl (Figure 2B) [13].
Abl remains a highly attractive target for the development of potent and selective inhi tors that will represent an important new class of therapeutic agents for the treatment leukemia [5].Several new therapies for leukemia are currently in development, and som are under preclinical investigation, most of which are ATP-competitive Bcr-Abl TKIs ov riding the T315I mutation, but there are very few reports evaluating the same molecule different mutations [9,10].Since the most potent inhibitors of Bcr-Abl are those that bi directly to the ATP binding site, the purine core was chosen as the basis for designing t compounds in this study.This heterocycle has been used as a privileged scaffold in t development of new Bcr-Abl inhibitors.Azam and collaborators performed different su stitutions of the 3-hydroxyphenylethyl group at N-9 and discovered that the compou AP23846 can bind within the active site of both Bcr-Abl WT and Bcr-Abl T315I [11].In t context, in 2020, our working group identified promising 2,6,9-trisubstituted purine d rivatives I-III (Figure 2A), which showed enhanced inhibition of Bcr-Abl with an IC50 0.040-0.090μM in an Abl kinase inhibition assay [12].In addition, compound III was mo selective to Bcr-Abl than other tyrosine kinases (BTK) and a serine/threonine kinase (CD 2).In addition, III exhibited low micromolar cytotoxicity on several leukemia cancer c lines and decreased the phosphorylation of downstream proteins in the signaling pa ways of Bcr-Abl [12].Later, in 2022, our group demonstrated that the optimal substitue at N-9 is the cyclopropylmethyl group of I-III, as evidenced in the high IC50 values compounds IV-VI bearing longer hydrophobic substitutions (Figure 2A) and confirm by molecular docking, which is attributed to the size of the hydrophobic pocket in B Abl (Figure 2B) [13].Therefore, according to these antecedents, in this work, we designed and synthesiz a series of new 2,6,9-trisubstituted purine derivatives.An inhibition study was then c ried out on Bcr-Abl WT , followed by a rigorous cytotoxic analysis on some leukemia-relat TKI-sensitive and TKI-resistant cancer cell lines that specifically contain T315I, E255K, Y253H mutated Bcr-Abl.Finally, in silico studies were performed to understand the stru ture-activity relationship of these ligands on Bcr-Abl WT and Bcr-Abl T315I .

Chemistry
All reagents and chemicals used in the chemical synthesis of the intermediates a final compounds were purchased from Sigma Aldrich (St. Louis, MO, USA).Intermedia that have been previously published were synthesized according to the reported pro dures, as indicated by the respective reference.Therefore, according to these antecedents, in this work, we designed and synthesized a series of new 2,6,9-trisubstituted purine derivatives.An inhibition study was then carried out on Bcr-Abl WT , followed by a rigorous cytotoxic analysis on some leukemia-related TKI-sensitive and TKI-resistant cancer cell lines that specifically contain T315I, E255K, or Y253H mutated Bcr-Abl.Finally, in silico studies were performed to understand the structure-activity relationship of these ligands on Bcr-Abl WT and Bcr-Abl T315I .

Chemistry
All reagents and chemicals used in the chemical synthesis of the intermediates and final compounds were purchased from Sigma Aldrich (St. Louis, MO, USA).Intermediates that have been previously published were synthesized according to the reported procedures, as indicated by the respective reference.
The melting points (mp) of all synthesized compounds were measured without correction on a Kofler Thermogerate apparatus (Reichert, Werke A.G., Wien, Vienna, Austria).The 1 H and 13 C nuclear magnetic resonance (NMR) spectra of the synthesis intermediates and final compounds were recorded on a BRUKER AVANCE III HD-400 [400 MHz ( 1 H) and 100 MHz ( 13 C)] and 200 MHz [200 MHz ( 1 H) and 50 MHz ( 13 C)] spec-trometers (Bruker, Karlsruhe, Germany), respectively.The compounds were dissolved in CDCl 3 or DMSO-d 6 with tetramethylsilane (TMS) as internal standard.In the NMR spectra, the chemical shift in each signal is given in parts per million (ppm) and, where appropriate, the coupling constants (J) are given in Hertz (Hz).The multiplicity observed in the 1 H NMR spectra for each signal is given as s (singlet), d (doublet), t (triplet), and dd (doublet doublet), respectively.High-resolution mass spectra (HRMS) or mass spectra (MS) were measured on a Q-TOF mass spectrometer (Synapt G2-Si, Waters, Milford, MA, USA) equipped with an electrospray ionization (ESI) source.Briefly, the measurement procedure consisted of injecting an acetonitrile solution of the samples directly into the ESI source using a syringe pump at a flow rate of 10 µL/min.The positive mode molecular ions were detected on the Q-TOF mass spectrometer.Reaction monitoring and verification of the purity of the synthesis products after column chromatography was performed by thin-layer chromatography (TLC) using Merck GF-254 type 60 silica gel (Merck, Burlington, VT, USA).The purity of the final compounds for biological assays was determined by TLC and HRMS spectra.

General Procedure for the Synthesis of Intermediates 2a-c
The synthesis of intermediates 2a-c was carried out according to the procedure described above, which consisted of adding to a solution of 2,6-dichloro-9H-purine (1, 2.0 g, 1.058 mol) the corresponding alkyl halides (1.58 mol) in the presence of K 2 CO 3 (3.36mg, 3.174 mol) and dissolved in DMF (20 mL).The mixture was stirred at room temperature for 12 h.The suspension was then filtered and concentrated on a rotary evaporator.This reaction crude was purified by silica gel chromatography using a mixture of EtOAc/hexane (1:1) as mobile phase to give the pure products 2a-c.
The yields of the compounds obtained after purification were for the following: The analytical data for 2a-c agreed with the literature [12].

General Procedure for the Synthesis of Intermediates 3a-i
The synthesis of intermediates 3a-i was carried out according to the procedure described above, which consisted of adding the corresponding anilines (0.411 mmol) to a solution of 2a-c (0.411 mmol) in the presence of the base N,N-diisopropylethylamine (DIPEA, 0.15 mL, 0.822 mmol), all dissolved in n-butanol (20 mL).This mixture was stirred at 110 • C for 12 h.After cooling to room temperature, the reaction mixture was concentrated on a rotary evaporator.The solid was then purified by silica gel chromatography using a mobile phase mixture of hexane:AcOEt (70:30) to give products 3a-i.
The yields of the compounds obtained after purification were for the following: The analytical data for 3a-c are in accordance with the literature [12].

General Procedures for the Synthesis of Intermediates 5 and 9a-b
To a solution of 1-(chloromethyl)-4-nitrobenzene 4 or 4-fluoro-nitrobenzene 8 (7.1 mmol), 1-methylpiperazine or 2-hidroxyethyl piperazine (7.1 mmol), and K 2 CO 3 (2226 mg, 21 mmol) in DMF (20 mL) were added, and the mixture was stirred at room temperature for 6 h.Then, the reaction mixture was filtered and evaporated.The mixture was extracted with EtOAc, and the organic layer was dried by anhydrous Na 2 SO 4 and concentrated to give products 5 and 9a-b.To a solution of 5 or 9a-b (1.35 mmol), Pd-C (30 mg) in ethanol (20 mL) with hydrogen atmosphere was added, and the mixture was stirred at room temperature for 4-6 h.After the completion of the reaction, the reaction mixture was filtered in celite and concentrated to give 6 and 10a-b. 4

General Procedure for the Synthesis of Intermediates 12a-c
To a solution of 2a (100 mg, 0.411 mmol) with a solution 25% of ammonium hydroxide (154 mg, 0.411 mmol) in dioxane (20 mL) for 12a, or benzylamine/cyclohexylamine (0.411 mmol) and DIPEA (0.15 mL, 0.822 mmol) in n-butanol (20 mL) for 12b-c, were added and these mixtures were stirred at 100 • C for 12 h.After cooling to room temperature, the reaction mixtures were concentrated on a rotary evaporator.The respective solid was then purified by silica gel chromatography using a mobile phase mixture of acetone/dichloromethane (1:1) to give product 12a, and a mobile phase mixture of acetone/dichloromethane (1:1) to give product 12b-c.11a-i, and 13a-c A solution consisting of 3a-i or 12a-c (0.519 mmol) and 6 or 10a-b (0.57 mmol) in dioxane (2 mL), Pd(OAc) 2 (24 mg, 0.1 mmol), Xanthphos (120 mg, 0. 2 mmol) and 2 M K 2 CO 3 (aq) (1 mL, 0.2 mmol) were added.These reaction mixtures were stirred at 100 • C for 12-24 h.After cooling to room temperature, these mixtures were filtered on celite, and the filtrates were diluted with dichloromethane.These residues were purified by silica gel chromatography using a methanol/dichloromethane (1:9) mobile phase to give the respective final product.The Rf values of each compound were calculated using the above mobile phase.

Docking
For the 14 final products proposed as inhibitors of these kinases, their three-dimensional structures were constructed using OECHEM and then the respective protonation states were adjusted at pH 7.2 using FixpKa from the QUACPAC package.Conformers for these ligands were generated using OMEGA v. 4.1.2.0 software.The crystal structures of Bcr-Abl WT and Bcr-Abl T315I kinases were downloaded from the RCSB PDB Protein Data Bank (6BL8 and 4TWP, respectively).These crystals were stripped of any water molecules present, as well as any ions and cofactors present.Hydrogen atoms and protons were then added, and partial charges assigned (according to the protonation state at physiological pH).The kinases were prepared using the Chimera USCF program [14].A local minimization was then performed to eliminate possible bad contacts.The minimization was carried out in the presence of certain constraints to keep the kinase conformation very similar to that observed in the experimental model.Molecular docking studies were carried out using Glide [15] and docking runs were performed using the standard and extra precision scoring approaches (SP and XP scoring) implemented in the software [16].The best poses for each ligand were optimized using Prime, and the binding energy was estimated by considering the solvation energies of the interacting molecules in addition to the molecular mechanics (MM) energies.The contribution of polar solvation energies was calculated using the generalized Born implicit solvent (GBI) model, while the non-polar contribution of solvation energy depended on the solvent-accessible area (SA) [17].The interactions of the active site amino acid residues with the ligands were identified using a 6 Å radius around the docked position as a reference.Images were obtained using PyMol v.2.5.2 software.

Cytotoxicity Assay
Cytotoxicity was determined using resazurin, an indicator dye to measure oxidationreduction reactions occurring in the mitochondria of live cells.The cells were seeded into 96-well plates and treated with compounds for 72 h (six different doses of each, in triplicate).After treatment, resazurin (Merck, Darmstadt, Germany) solution was added for 4 h, and the fluorescence of formed resorufin corresponding to live cell quantity was measured at 544 nm/590 nm (excitation/emission) using a Fluoroskan Ascent microplate reader (Labsystems, Vantaa, Finland).The GI 50 value, the drug concentration lethal to 50% of the cells, was calculated from the dose-response curves in Origin 6.0 software.

Flow Cytometry
The cell lines were seeded into 96-well plates and then treated with compounds for 24 h.After treatment, the cells were directly stained by adding 5 × staining solution (17 mM trisodium citrate dihydrate, 0.5% IGEPAL ® CA-630, 7.5 mM spermine tetrahydrochloride, 2.5 mM Tris; pH 7.6 containing 50 µg/mL propidium iodide).DNA content was measured by flow cytometry using a 488 nm laser (BD FACS Verse with software BD FACSuite™, version 1.0.6.).Cell cycle distribution was analyzed using ModFit LT 5.0.9 (Verity Software House, Augusta, ME, USA).

Design and Synthesis
In the design of our compounds, we considered our previous results and the chemical structures of the purine-based kinase inhibitors (Figure 3).Firstly, the aminophenyl fragment on C-6 was mainly substituted with one or two fluorine atoms in meta and/or para positions, like II or III (Figure 2).In addition, an amino (-NH 2 ), cyclohexylamino (-NH-Cy), or benzyl (-NH-Bz) group was substituted on C-6.This modification led us to validate the hypothesis that to inhibit Bcr-Abl, a hydrogen bond in the binding pocket had to be between an aminophenyl fragment and Met318 (Figure 3).Secondly, to confirm the evidence found in our previous work that the optimal hydrophobic moiety on N-9 is a cyclopropylmethyl group (I-VI, Figure 2); modifications in the volume and length of this alkyl group were considered.To complete this design, the N-methyl-arylpiperazine fragment on C-2 was conserved in some of these new purine derivatives, and in others, (i) the methyl group was replaced by a hydroxyethyl group mimicking the moiety present in dasatinib, and this portion was oriented towards the exposed solvent region, or (ii) a methylene group was incorporated between the piperazine and aromatic rings to explore the influence of the flexibility of this fragment on Bcr-Abl activity, mimicking imatinib or ponatinib.
dence found in our previous work that the optimal hydrophobic moiety on N-9 is a cyclopropylmethyl group (I-VI, Figure 2); modifications in the volume and length of this alkyl group were considered.To complete this design, the N-methyl-arylpiperazine fragment on C-2 was conserved in some of these new purine derivatives, and in others, (i) the methyl group was replaced by a hydroxyethyl group mimicking the moiety present in dasatinib, and this portion was oriented towards the exposed solvent region, or (ii) a methylene group was incorporated between the piperazine and aromatic rings to explore the influence of the flexibility of this fragment on Bcr-Abl activity, mimicking imatinib or ponatinib.The synthesis of these new 2,6,9-trisubstituted purines 7a-b, 11a-i, and 13a-c was carried out by short, simple, and efficient synthetic methods described by our group and shown in Scheme 1 [12,13,19].We obtained 14 compounds using 2,6-dichloropurine (1) as a starting material.The first step was the alkylation of 1 with the respective alkyl halides under basic conditions to give 2a-c [12,13,19].To obtain 7a or 7b, later, a regioselective nucleophilic substitution (SNAr) at position C-6 with 3-floroaniline or 3,4-difluoroaniline, using n-butanol as the solvent and N,N-diisopropylethylamine as the base, in reflux for 12 h gave compounds 3a-b.Subsequently, a Buchwald-Hartwig C-N coupling reaction at C-2 of 3a-b with 4-((4-methylpiperazin-1-yl)methyl)aniline (6) catalyzed by palladium (II) afforded the purine derivatives 7a-b in 67-68% yields.Compound 6 was previously synthesized from 1-(chloromethyl)-4-nitrobenzene (4) in two synthetic steps, as shown in Scheme 1.The synthesis of 11a-i considered a similar route to that of 7a-b and our previously reported methodology [12,13], the difference being that in this case, we used N-(2hydroxyethyl)piperazine and 4-fluoro-nitrobenzene (8) as starting materials to quantitatively obtain the required aniline derivative 10a.Finally, the purine derivatives 13a-c were synthesized following the same logic of the previous syntheses but using amines or ammonia in the second step (nucleophilic substitution at C-6, 80-90% yield) and the N-methyl-piperazinyl-aniline 10b for the C-N coupling reaction on C-2 with moderate yields (54-63%).All compounds were purified by column chromatography, and their structures were established based on their spectral properties ( 1 H NMR and 13 C NMR; see the Materials and Methods section and Supplementary Materials).The synthesis of these new 2,6,9-trisubstituted purines 7a-b, 11a-i, and 13a-c was carried out by short, simple, and efficient synthetic methods described by our group and shown in Scheme 1 [12,13,19].We obtained 14 compounds using 2,6-dichloropurine (1) as a starting material.The first step was the alkylation of 1 with the respective alkyl halides under basic conditions to give 2a-c [12,13,19].To obtain 7a or 7b, later, a regioselective nucleophilic substitution (S N Ar) at position C-6 with 3-floroaniline or 3,4-difluoroaniline, using n-butanol as the solvent and N,N-diisopropylethylamine as the base, in reflux for 12 h gave compounds 3a-b.Subsequently, a Buchwald-Hartwig C-N coupling reaction at C-2 of 3a-b with 4-((4-methylpiperazin-1-yl)methyl)aniline (6) catalyzed by palladium (II) afforded the purine derivatives 7a-b in 67-68% yields.Compound 6 was previously synthesized from 1-(chloromethyl)-4-nitrobenzene (4) in two synthetic steps, as shown in Scheme 1.The synthesis of 11a-i considered a similar route to that of 7a-b and our previously reported methodology [12,13], the difference being that in this case, we used N-(2-hydroxyethyl)piperazine and 4-fluoro-nitrobenzene (8) as starting materials to quantitatively obtain the required aniline derivative 10a.Finally, the purine derivatives 13a-c were synthesized following the same logic of the previous syntheses but using amines or ammonia in the second step (nucleophilic substitution at C-6, 80-90% yield) and the N-methyl-piperazinyl-aniline 10b for the C-N coupling reaction on C-2 with moderate yields (54-63%).All compounds were purified by column chromatography, and their structures were established based on their spectral properties ( 1 H NMR and 13 C NMR; see the Materials and Methods section and Supplementary Materials).

Kinase Inhibition and Structure-Activity Relationship
In accordance with the goal of this work to discover new Bcr-Abl inhibitors, the next step was to screen all the synthesized compounds for the inhibitory activity on this kinase.Considering the results shown in Table 1, there are some compounds with better activity than the reference drugs, imatinib and nilotinib, such as 11b and 11c, which elicited lower IC50 values (0.327 and 0.047 μM vs. 0.015 and 0.020 μM, respectively).
On the other hand, from a chemical point of view, a structure-activity relationship (SAR) can be established for this target: 1.The incorporation of a methylene linker between the piperazine and phenylamino ring at C-6 diminished the inhibitory potency on Bcr-Abl in 7a and 7b compared to their analogues II and III (0.180 and 0.225 μM vs. 0.040 μM, respectively).This indicates that the flexibility of this moiety is disadvantageous for the Bcr-Abl activity.

Kinase Inhibition and Structure-Activity Relationship
In accordance with the goal of this work to discover new Bcr-Abl inhibitors, the next step was to screen all the synthesized compounds for the inhibitory activity on this kinase.Considering the results shown in Table 1, there are some compounds with better activity than the reference drugs, imatinib and nilotinib, such as 11b and 11c, which elicited lower IC 50 values (0.327 and 0.047 µM vs. 0.015 and 0.020 µM, respectively).
On the other hand, from a chemical point of view, a structure-activity relationship (SAR) can be established for this target: 1.
The incorporation of a methylene linker between the piperazine and phenylamino ring at C-6 diminished the inhibitory potency on Bcr-Abl in 7a and 7b compared to their analogues II and III (0.180 and 0.225 µM vs. 0.040 µM, respectively).This indicates that the flexibility of this moiety is disadvantageous for the Bcr-Abl activity.

2.
The substitution of methyl by a hydroxymethyl group in the N-piperazine ring increased the inhibition of Bcr-Abl activity, this behavior being observed in 11a-c compared to I-III.The IC 50 values diminished from 0.090 to 0.037 µM (for I to 11a), from 0.045 to 0.015 µM (for II to 11b), and from 0.040 to 0.020 µM (for III to 11c).Interestingly, this fragment is present in dasatinib, and according to our previous docking results, this fragment is in the solvent-exposed region.This effect is so significant that even compounds with more voluminous hydrophobic moieties increased their inhibitory activity (11f vs. IV and 11i vs. VI).

3.
The fact that the optimal substitution for alkyl chains at N-9 is by the cyclopropylmethyl group is confirmed, as can be seen by the comparison of compounds 11a-c with their respective analogues 11d-i.This result support the hypothesis about the size of the hydrophobic region in Bcr-Abl compared to other kinases, such as BTK or FLT3 [13].

4.
These results also prove that the substitution of hydrogens by fluorine atoms in meta and para positions of the 6-aminophenyl fragment is beneficial for the inhibition of this kinase.

5.
Finally, it is also demonstrated that it must be a 6-aminophenyl fragment and not just a substituent on the purine that possesses the amino group, because according to the IC 50 values, any option other than the benzene ring drastically decreases the kinase activity (13a-c, with IC 50 values between 0.845 µM and 11.06 µM).This would indicate that not only the hydrogen bond donor group is a key to the binding site, but there are also important steric and electronic effects to consider.

Molecular Docking Studies for Bcr-Abl WT
Docking calculations were performed to understand the differences in activity of these purine derivatives on Bcr-Abl WT .The molecular docking protocol was first validated by performing self-docking of the co-crystallized ligand for the purvalanol B-Bcr-Abl (PDB ID: 6BL8) [20].The docking protocol showed that the program succeeds in reproducing the main interactions between the co-crystallized ligand and their respective proteins (Figure S1).The complexes with the best docking poses were further energy minimized and the interaction energy was estimated using the MM/GBSA method.The results of the binding energies for all compounds are shown in Table S1.According to these results, a trend was observed in the affinity energies obtained by different calculation methods, such as RDock, XP Score (Glide), and MM-GBSA, and the IC 50 values on Bcr-Abl WT .These computer programs allowed for efficient discrimination between the most and least active compounds.For example, the ∆G binding values were −74.32 and −48.37 kcal/mol for 11c (IC 50 = 0.020 µM) and for 13a (IC 50 = 3.96 µM).The correlation of the biological activity is shown in Figure S2.
In addition, the binding interactions, and the orientation patterns for the most and the least active compounds (11a-c and 13a-b) are determinants for Bcr-Abl inhibition.As we previously reported [13], a network of multiple interactions was observed between 11a-c and the binding pocket of this kinase (Figure 4A-C), considering a double hydrogen bridge between the N-7/NH-phenyl amino group and Met318, which also allowed for a series of hydrophobic interactions with Val299, Ala380, Phe382, and Leu370 in the hydrophobic cavity.However, the increase in the inhibitory activity of 11a-c compared to I-III could be attributed to the new hydrogen bonds formed between the hydroxymethyl fragment of these purines with Asp363, Asp367, and Asn368 (red color in Figure 4) in the solventexposed region.This evidence confirms our hypothesis mentioned in the design section.Similarly, these mentioned interactions could also explain the increase in potency of 11i (Figure 4E) compared to VI (Figure 2).

pharmacophore.
Finally, according to these in silico results, the optimal size of the of alkyl substitutions at N-9 would be a cyclopropylmethyl group instead of n-hexyl for Bcr-Abl WT .However, an isopentyl group on N-9 could establish hydrophobic interactions with T315 (Fig- ure 4E, green region).Likewise, the substitution of hydrogens atoms by fluorine atoms of the aminophenyl ring at C-6 is beneficial for Bcr-Abl inhibition.On the other hand, this docking study can explain the low activity of 7a-b and 13 a-c.For 7a-b, the incorporation of a methylene linker between the piperazine and the phenylamino ring at C-2 prevents the formation of the pattern interactions observed for 11a-c in the red region (Figure 4D).Similarly, for 13a-c, the absence of the phenylamino moiety at C-6 prevents the formation of interactions with Leu248 (Figure 4F) and eventually with Phe317 for 13a and 13c, confirming the importance of this fragment as part of the purine pharmacophore.
Finally, according to these in silico results, the optimal size of the of alkyl substitutions at N-9 would be a cyclopropylmethyl group instead of n-hexyl for Bcr-Abl WT .However, an isopentyl group on N-9 could establish hydrophobic interactions with T315 (Figure 4E, green region).Likewise, the substitution of hydrogens atoms by fluorine atoms of the aminophenyl ring at C-6 is beneficial for Bcr-Abl inhibition.

Cytotoxic Studies
First, all tested compounds were screened for cytotoxic activity against cell lines with Bcr-Abl rearrangement, such as KBM5 (chronic myeloid leukemia), BV173 (B-cell precursor leukemia), K562 (chronic myelogenous leukemia), as well as against non-neoplastic cells HEK-293T (Table 2).Overall, the cytotoxicity of the 2,6,9-trisubstituted purines was heterogeneous depending on the cell type, as well as the assayed compounds.K562 was the most sensitive cell line (all compounds with GI 50 values between 0.7 and 6.3 µM, except 13a with GI 50 > 100 µM), while KCL22 was the least sensitive (only six compounds with GI 50 values < 6.5 µM).Among the compounds tested, 11a-c proved to be the most potent, with GI 50 values not exceeding 3 µM in the Bcr-Abl rearranged cell lines tested.Interestingly, their cytotoxicity was in most cases more pronounced than their template analogues I-III, indicating the importance of the N-hydroxyethyl group in their structure.Likewise, similarly to the Bcr-Abl inhibition, the aminophenyl fragment at C-6 was pivotal for the antiproliferative effect demonstrated by comparing 11a with 13a-c, which were less cytotoxic in all cell lines.In general, the cytotoxicity of the purine derivatives was consistent with enhanced Abl inhibition, especially in the K562 and KCL22 cell lines, with similar structure-activity relationships to those described above.Compound 11b was found to be the most potent derivative with GI 50 values of 1.3, 1.5, and 0.7 µM for KBM5, BV173, and K562 cell lines.However, none of the synthetized compounds showed better performance than TKIs.a GI 50 values were determined in at least three independent experiments using the resazurin assay.
To directly compare the sensitivity of synthesized compounds to the most common Bcr-Abl mutations, they were evaluated on a panel of KCL22 subclones expressing different Bcr-Abl point mutants, including B8 (T315I, 100%), F4 (E255K, 100%), and B10 (Y253H, 50%) [18].The sensitivity to these compounds was evaluated by cytotoxicity assays, and these results were expressed as GI 50 values (Table 3).The growth of KCL22 cells harboring WT or mutant Bcr-Abl was inhibited at different potencies depending on the purine derivative.In particular, the WT cell line was sensitive to all derivatives with GI 50 values in the micromolar range except 13a-c and 11c.From a chemical point of view, this confirms that the aminophenyl fragment on C-6 is pivotal for the antiproliferative effect, which was demonstrated by comparing 13a-c with its respective analogues.Compared to newly synthesized compounds, template molecules II and III showed only weak antiproliferative effects overall and the incorporation of a methylene linker in 7a and 7b did not lead to any major improvements.Among all compounds tested, 11b and 11c exhibited the strongest antiproliferative properties on Bcr-Abl WT cells (GI 50 = 2.0 and 2.9 µM, respectively), consistent with enhanced Abl inhibition.However, none of the synthesized derivatives overcame the potency of the reference drugs, imatinib (GI 50 = 0.482 µM) and nilotinib (GI 50 = 0.019 µM).Interestingly, both standards were ineffective in inhibiting the growth of B8 cells carrying the T315I mutation (GI 50 > 20 µM), whereas compounds 11c-f and 11h reached single-digit micromolar GI 50 values.It should be recalled that the T315I mutation is known to be responsible for resistance to nilotinib [21] and imatinib [22], and thus this result agrees with this clinical evidence.Such antiproliferative effect of 11c-f (GI 50 = 6.4-7.2 µM) suggests that possible inhibition of Bcr-Abl T315I is favored by the presence of an isopentyl group at N-9, rather than that observed for Bcr-Abl WT , where the most optimal substitution was cy-clopropyl group.This challenging hypothesis will be discussed in detail in the next section using in silico studies.In addition, when comparing the GI 50 values obtained in F4 and B10 mutated cell lines, the synthesized derivatives 11b-f and I showed promising antiproliferative effects in both mutated cell lines (GI 50 = 4.1-5.7 µM and 2.6-7.8µM, respectively), overcoming the potency of imatinib, which only weakly inhibited the proliferation of the F4 mutant cell line (GI 50 = 9.6 µM), with B10 cells being resistant (GI 50 > 20 µM).Consequently, 11b, as the compound with the best overall potency, and 11e, with approximately equal antiproliferative activity in the KCL22-derived cell lines tested (including the resistant B8 mutant model), were selected for detailed experiments to clarify their cellular effects.a GI 50 values were determined in at least three independent experiments using the resazurin assay.

In Silico Studies for Bcr-Abl T315I
To explain the observed results on the antiproliferative effect of these purines between KCL22 WT and KCL22 B8 cells, we assume that they are related to the presence of Bcr-Abl in its WT or T315I mutant state, as well as to the degree of inhibition of these purines.Therefore, to validate this hypothesis, we carried out molecular docking studies on the Bcr-Abl T315I (PDB ID: 4TWP [23]).Results showed a correlation between the binding affinity energies and the GI 50 values shown in Table 3 (Figure S3), highlighting the positioning of the most potent antiproliferative compounds, 11b-f, on KCL22 B8 cells (Table 4).According to Table 4, some purine derivatives with isopentyl group at N-9 (11d-f) had higher ∆G binding than those with cyclopropylmethyl group (11b-c), which could be related to the level of possible inhibition on Bcr-Abl T315I and could indicate the role of this fragment.Comparing the pattern of interactions to stabilize the ligand-Bcr-Abl T315I complex with the ligand-Bcr-Abl WT complex, it is observed that some of them are conserved in both models, such as the essential hydrogen bonds with Met318 with the purine ring and Asp363 with the hydroxyethyl group (Figure 5A-E).However, some new interactions and conformations were observed in Bcr-Abl T315I .In the hydrophobic pocket, 11c-e showed interactions with the residues Ile315, Ala269, and Lys271.Interestingly, Ile315 is the point of mutation and Ala269 and Lys271 are residues present in the deepest hydrophobic pocket and contribute significantly to the improved affinity energy, as shown in Table 4 (∆G Lipo).Moreover, a new hydrogen bonding interaction between the aminophenyl group at C-2 of the purine ring with Tyr253 was observed for 11b-d and 11f, and an electrostatic interaction between the protonated piperazine fragment moiety and Asp381 was also observed for these purines, the latter being interactions that strongly contribute to ligandkinase stabilization (Table 4, ∆G Coul.).On the other hand, the eventual most active compounds against Bcr-Abl T315I , 11e and 11f, showed interaction profiles similar to that of compound 11c (Figure 5B,D,E).However, the incorporation of one or two fluorine atoms in the phenyl of these compounds enhances the hydrophobic interactions with Phe317 and Leu248.This effect is clearly reflected in the calculated van der Waals energies values (Table 4, ∆GvdW).An important aspect is the conformation adopted by these ligands during their interaction with Bcr-Abl WT , as opposed to Bcr-Abl T315I .This is because, we hypothesized, that the presence of T315I drastically changes the conformation adopted by the ligands, being a "V" type shape for Bcr-Abl WT and a "T" type shape for Bcr-Abl T315I (Figure 6).In the "T" form, the ligands showed a better fit, aided by the larger size of the hydrophobic region, which easily accommodates the N-9 isopentyl group, as well as a better pose for the phenylamino-N-hydroxyethyl-piperazine fragment at C-2.An important aspect is the conformation adopted by these ligands during their interaction with Bcr-Abl WT , as opposed to Bcr-Abl T315I .This is because, we hypothesized, that the presence of T315I drastically changes the conformation adopted by the ligands, being a "V" type shape for Bcr-Abl WT and a "T" type shape for Bcr-Abl T315I (Figure 6).In the "T" form, the ligands showed a better fit, aided by the larger size of the hydrophobic region, which easily accommodates the N-9 isopentyl group, as well as a better pose for the phenylamino-N-hydroxyethyl-piperazine fragment at C-2.
a "V" type shape for Bcr-Abl WT and a "T" type shape for Bcr-Abl T315I (Figure 6 form, the ligands showed a better fit, aided by the larger size of the hydroph which easily accommodates the N-9 isopentyl group, as well as a better pose nylamino-N-hydroxyethyl-piperazine fragment at C-2.

Cell Cycle Cytometry Analysis and Immunodetection
To gain insight into the mechanism of action of the selected compounds we performed cell cycle cytometry analysis of KCL22-derived cells treated f ures 7 and S4).An increasing concentration of 11b resulted in a significant

Cell Cycle Cytometry Analysis and Immunodetection
To gain insight into the mechanism of action of the selected compounds 11b and 11e, we performed cell cycle cytometry analysis of KCL22-derived cells treated for 24 h (Figures 7 and S4).An increasing concentration of 11b resulted in a significant block of WT cells in G1 phase, with an increase in the percentage of dead cells at 5 µM concentration, which could correspond to Bcr-Abl inhibition, and was comparable to the effect of imatinib.Interestingly, 11b treatment also arrested F4 and B10 mutant models in G1 phase, in contrast to imatinib treatment, which is consistent with previously shown cytotoxicity assay data.However, the KCL22 B8 cell line containing Bcr-Abl T315I was resistant to treatment with both selected compounds, including 11e, which had previously shown promising antiproliferative properties in all KCL22 models (GI 50 = 5.6-6.9µM).However, this was clearly not reflected in the cell cycle distribution of any of the KCL22-derived models and, as expected, no effect was also observed when treated with imatinib and nilotinib.These results suggest that the compound 11e probably exerts its antiproliferative effect in KCL22 models through mechanisms other than cell cycle blockade.
To determine whether the antiproliferative effect of 11b was dependent on the inhibition of Bcr-Abl, we immunodetected changes in the expression or phosphorylation of proteins known to be involved in the regulation of proliferation and the Bcr-Abl signaling pathway.The results obtained in 11b-sensitive WT, F4 and B10 KCL22 cell lines were compared with those obtained after imatinib treatment (Figure 8).Notably, 11b downregulated the levels of p-Crkl and p-STAT5, downstream proteins of Bcr-Abl, in the WT KCL22 cell line, but to a lesser extent than imatinib.We also observed cleavage of PARP and a decrease of cyclin A levels, both of which are associated with G1 cell cycle arrest and an increase in cell death.These cellular effects of 11b were significantly pronounced in both mutant cell lines, F4 and B10, especially observed in the decrease in STAT5 phosphorylation and cleavage of PARP.In contrast, imatinib was less effective in the F4 cells and B10 cells were not affected in any of proteins studied.
promising antiproliferative properties in all KCL22 models (GI50 = 5.6-6.9µM).However, this was clearly not reflected in the cell cycle distribution of any of the KCL22-derived models and, as expected, no effect was also observed when treated with imatinib and nilotinib.These results suggest that the compound 11e probably exerts its antiproliferative effect in KCL22 models through mechanisms other than cell cycle blockade.To determine whether the antiproliferative effect of 11b was dependent on the inhibition of Bcr-Abl, we immunodetected changes in the expression or phosphorylation of proteins known to be involved in the regulation of proliferation and the Bcr-Abl signaling pathway.The results obtained in 11b-sensitive WT, F4 and B10 KCL22 cell lines were compared with those obtained after imatinib treatment (Figure 8).Notably, 11b downregulated the levels of p-Crkl and p-STAT5, downstream proteins of Bcr-Abl, in the WT KCL22 cell line, but to a lesser extent than imatinib.We also observed cleavage of PARP and a decrease of cyclin A levels, both of which are associated with G1 cell cycle arrest and an increase in cell death.These cellular effects of 11b were significantly pronounced in both mutant cell lines, F4 and B10, especially observed in the decrease in STAT5 phosphorylation and cleavage of PARP.In contrast, imatinib was less effective in the F4 cells and B10 cells were not affected in any of proteins studied.

Calculated Physicochemical Properties and ADME Parameters
Finally, it should be remembered that both pharmacological properties and pha cokinetic profiles are important in the drug discovery and development process therefore necessary to be able to predict or determine the latter properties, which rel administration, distribution, metabolism, and excretion (ADME), and to take them account during the optimization of a bioactive compound until it becomes a succ candidate for preclinical studies [24].The free online platform SwissA (http://www.swissadme.ch/index.php,accessed on 29 January 2024) was used to d mine the physicochemical properties of compound 11b based on Lipinski's rule shown in Table 5, 11b fulfils the criteria for good permeability and bioavailability on the values of molecular weight (MW), hydrogen bond donor (HBD), hydrogen acceptor (HBA), and cLogP [25].Furthermore, according to Veber's rules, 11b has a logical polar surface area (TPSA) and number of rotatable bonds (NRB) values less 140 Å 2 and ≤10 NRB (Table 5) [26].All these values would indicate that 11b would h high ability to penetrate cell membranes and a good oral absorption according t Lipinski and Veber rules.In addition, the SwissADME platform provides a bioavaila radar plot that considers the following parameters: flexibility (FLEX), lipophilicity (L solubility (INSOLU), size (SIZE), polarity (POLAR), and saturation (INSATU), and parameters are within the desired range (pink region), good oral absorption is exp for this compound.Figure 9 shows that these criteria are met for 11b.

Calculated Physicochemical Properties and ADME Parameters
Finally, it should be remembered that both pharmacological properties and pharmacokinetic profiles are important in the drug discovery and development process.It is therefore necessary to be able to predict or determine the latter properties, which relate to administration, distribution, metabolism, and excretion (ADME), and to take them into account during the optimization of a bioactive compound until it becomes a successful candidate for preclinical studies [24].The free online platform SwissADME (http://www.swissadme.ch/index.php,accessed on 29 January 2024) was used to determine the physicochemical properties of compound 11b based on Lipinski's rules.As shown in Table 5, 11b fulfils the criteria for good permeability and bioavailability based on the values of molecular weight (MW), hydrogen bond donor (HBD), hydrogen bond acceptor (HBA), and cLogP [25].Furthermore, according to Veber's rules, 11b has a topological polar surface area (TPSA) and number of rotatable bonds (NRB) values less than 140 Å 2 and ≤10 NRB (Table 5) [26].All these values would indicate that 11b would have a high ability to penetrate cell membranes and a good oral absorption according to the Lipinski and Veber rules.In addition, the SwissADME platform provides a bioavailability radar plot that considers the following parameters: flexibility (FLEX), lipophilicity (LIPO), solubility (INSOLU), size (SIZE), polarity (POLAR), and saturation (INSATU), and if all parameters are within the desired range (pink region), good oral absorption is expected for this compound.Figure 9 shows that these criteria are met for 11b.

Conclusions
Continuing with our search for new Bcr-Abl inhibitors based on puri this work, we report the design and synthesis of 14 new 2,6,9-trisustituted tives.The biological results showed that the incorporation of N-hydroxyeth linked to a phenylamino fragment at C-2 of the purine ring increased the in ity on Abl as well as the antiproliferative properties on three CML cell lines Bcr-Abl rearrangement.Notably, 11b was the most potent inhibitor of Bcrderivatives as well as compared to nilotinib, with an IC50 = 0.015 µM.In showed high cytotoxicity in CML cell lines with a GI50 values between 0. However, the most interesting results were obtained when these purines w in a panel of KCL22 subclones expressing different Bcr-Abl point mutant Abl T315I ) and F4 (Bcr-Abl E255K ) cells, 11b-f were more effective than imatinib in inhibiting the growth of these cells, and more effective than imatinib for Abl Y253H ).Molecular docking studies identified the main interactions of thes the binding sites of Bcr-Abl WT and Bcr-Abl T315I and helped to show the diff tency against both kinases, confirming our hypothesis about the relevance o ments decorating the purine ring.In addition, a possible different binding Abl WT and Bcr-Abl T315I was proposed.Furthermore, 11b was able to inhibit and B10 cells and to arrest the cell cycle in the G1 phase, which correlated proliferative effects.Finally, 11b was predicted to have a good pharmacokin oral administration.Our results showed that 11b can be considered as a pr pound for the development of new drugs for CML treatment.

Supplementary Materials:
The following supporting information can be d www.mdpi.com/xxx/s1, 1 H-, 13 C-, and 19 F-NMR and HRMS of selected compounds ing affinity scores (kcal/mol) of the purine derivatives with RDOCK, XP Score (∆Gbind and their respective energy contributions) in Bcr-Abl WT .Figure S1.Co-crys

Conclusions
Continuing with our search for new Bcr-Abl inhibitors based on purine scaffold, in this work, we report the design and synthesis of 14 new 2,6,9-trisustituted purine derivatives.The biological results showed that the incorporation of N-hydroxyethyl-piperazine linked to a phenylamino fragment at C-2 of the purine ring increased the inhibitory activity on Abl as well as the antiproliferative properties on three CML cell lines harboring the Bcr-Abl rearrangement.Notably, 11b was the most potent inhibitor of Bcr-Abl among all derivatives as well as compared to nilotinib, with an IC 50 = 0.015 µM.In addition, 11b showed high cytotoxicity in CML cell lines with a GI 50 values between 0.7 and 1.3 µM.However, the most interesting results were obtained when these purines were evaluated in a panel of KCL22 subclones expressing different Bcr-Abl point mutants.For B8 (Bcr-Abl T315I ) and F4 (Bcr-Abl E255K ) cells, 11b-f were more effective than imatinib and nilotinib in inhibiting the growth of these cells, and more effective than imatinib for B10 cells (Bcr-Abl Y253H ).Molecular docking studies identified the main interactions of these inhibitors at the binding sites of Bcr-Abl WT and Bcr-Abl T315I and helped to show the difference in potency against both kinases, confirming our hypothesis about the relevance of certain fragments decorating the purine ring.In addition, a possible different binding mode in Bcr-Abl WT and Bcr-Abl T315I was proposed.Furthermore, 11b was able to inhibit Bcr-Abl in F4 and B10 cells and to arrest the cell cycle in the G1 phase, which correlated with the antiproliferative effects.Finally, 11b was predicted to have a good pharmacokinetic profile for oral administration.Our results showed that 11b can be considered as a promising compound for the development of new drugs for CML treatment.

Figure 1 .
Figure 1.Chemical structures of Bcr-Abl inhibitors approved by the FDA for TKI-based th

Figure 1 .
Figure 1.Chemical structures of Bcr-Abl inhibitors approved by the FDA for TKI-based therapy.

Figure 2 .
Figure 2. Purine derivatives developed previously by our group as Bcr-Abl inhibitors: (A) Chemi structures and their IC50 values.(B) Predicted binding mode of purines substituted with the cyc propylmethyl group at N-9 to Bcr-Abl [13].

Figure 2 .
Figure 2. Purine derivatives developed previously by our group as Bcr-Abl inhibitors: (A) Chemical structures and their IC 50 values.(B) Predicted binding mode of purines substituted with the cyclopropylmethyl group at N-9 to Bcr-Abl [13].

D325T315Figure 3 .
Figure 3. Design of new purine derivatives as potential inhibitors of Bcr-Abl.

Figure 3 .
Figure 3. Design of new purine derivatives as potential inhibitors of Bcr-Abl.

Figure 4 .
Figure 4. Graphical representation of the poses obtained by molecular docking of the most active compounds, 11a (A), 11b (B), and 11c (C), and the less active compounds, 7a (D), 11i (E), and 13b (F), into the binding site of Bcr-Abl (PDB ID: 6BL8).The red region is occupied when piperazine has a hydroxyl group and the green region is highlighted by Thr315 in the hydrophobic pocket.

Figure 4 .
Figure 4. Graphical representation of the poses obtained by molecular docking of the most active compounds, 11a (A), 11b (B), and 11c (C), and the less active compounds, 7a (D), 11i (E), and 13b (F), into the binding site of Bcr-Abl (PDB ID: 6BL8).The red region is occupied when piperazine has a hydroxyl group and the green region is highlighted by Thr315 in the hydrophobic pocket.

Figure 6 .
Figure 6.Proposed conformations of II at the binding site of Bcr-Abl WT and of 11e in

Figure 6 .
Figure 6.Proposed conformations of II at the binding site of Bcr-Abl WT and of 11e in Bcr-Abl T315I .

Figure 7 .
Figure 7. Effects of compounds 11b and 11e on the phases of cell cycle (designated as % of G1, S, and G2) and apoptosis (sub-G1) compared to imatinib and nilotinib after 24 h treatment.

Figure 7 .Figure 8 .
Figure 7. Effects of compounds 11b and 11e on the phases of cell cycle (designated as % of G1, S, and G2) and apoptosis (sub-G1) compared to imatinib and nilotinib after 24 h treatment.Pharmaceutics 2024, 16, x FOR PEER REVIEW 21

Figure 8 .
Figure 8. Effects of compounds 11b and imatinib on the expression or phosphorylation of proteins involved in the proliferation and Bcr-Abl pathways after 24 h treatment.

Figure 9 .
Figure 9. Radar plot for the bioavailability of 11b.The pink area indicates the range of optimal values for each property related to oral bioavailability.The predicted properties are shown as red lines.

Table 1 .
Inhibition of recombinant Abl1 kinase in vitro by tested compounds.

Table 1 .
Inhibition of recombinant Abl1 kinase in vitro by tested compounds.

Table 1 .
Inhibition of recombinant Abl1 kinase in vitro by tested compounds.

Table 1 .
Inhibition of recombinant Abl1 kinase in vitro by tested compounds.

Table 1 .
Inhibition of recombinant Abl1 kinase in vitro by tested compounds.

Table 1 .
Inhibition of recombinant Abl1 kinase in vitro by tested compounds.

Table 1 .
Inhibition of recombinant Abl1 kinase in vitro by tested compounds.

Table 1 .
Inhibition of recombinant Abl1 kinase in vitro by tested compounds.

Table 1 .
Inhibition of recombinant Abl1 kinase in vitro by tested compounds.

Table 1 .
Inhibition of recombinant Abl1 kinase in vitro by tested compounds.
[12]50 values were determined in at least three independent experiments.bValuesreportedin[12].aIC50 values were determined in at least three independent experiments.bValuesreportedin[12].aIC50 values were determined in at least three independent experiments.bValuesreportedin[12].aIC50 values were determined in at least three independent experiments.bValuesreportedin[12].aIC50 values were determined in at least three independent experiments.bValuesreportedin[12].aIC50 values were determined in at least three independent experiments.bValuesreportedin[12].aIC50 values were determined in at least three independent experiments.bValuesreportedin[12].aIC50 values were determined in at least three independent experiments.bValuesreportedin[12].aIC50 values were determined in at least three independent experiments.bValuesreportedin[12].aIC50 values were determined in at least three independent experiments.bValuesreportedin[12].aIC50 values were determined in at least three independent experiments.bValuesreportedin[12].aIC50 values were determined in at least three independent experiments.bValuesreportedin[12].aIC 50 values were determined in at least three independent experiments.bValuesreported in[12].

Table 2 .
In vitro cytotoxicity of tested compounds on cancer cell lines and non-neoplastic cells HEK-293T.