Pyrimidine-5-carbonitrile based potential anticancer agents as apoptosis inducers through PI3K/AKT axis inhibition in leukaemia K562

Abstract A novel series of 4-(4-Methoxyphenyl)-2-(methylthio)pyrimidine-5-carbonitrile was developed linked to an aromatic moiety via N-containing bridge and then evaluated for their cytotoxic activity against MCF-7 and K562 cell lines. Seven compounds exhibited the highest activity against both cell lines where compounds 4d and 7f were the most active against K562 cell line. Exploring their molecular mechanisms by enzyme inhibition assay on PI3Kδ/γ and AKT-1 showed that compound 7f was promising more than 4d with IC50 = 6.99 ± 0.36, 4.01 ± 0.55, and 3.36 ± 0.17 uM, respectively. Also, flowcytometric analysis revealed that 7f caused cell cycle arrest at S-phase followed by caspase 3 dependent apoptosis induction. Mechanistically, compound 7f proved to modulate the expression of PI3K, p-PI3K, AKT, p-AKT, Cyclin D1, and NFΚβ. Furthermore, in-vivo toxicity study indicated good safety profile for 7f. These findings suggest that the trimethoxy derivative 7f has strong potential as a multi-acting inhibitor on PI3K/AKT axis targeting breast cancer and leukaemia.


Introduction
Cancer is a devastating group of diseases that WHO announced it as the first or second leading cause of death worldwide in ages before 70 years. Consequently, the global burden of cancer incidence and mortality rates have unprecedented surpass according to the GLOBOCAN 2020 report. It was estimated that there were 19.3 million new cases and almost 10.0 million cancer deaths occurred in 2020 and expected to be 28.4 million cases in 2040 with a 47% rise from 2020. Thus, an effective control for cancer dissemination is an imminent need either through prevention or therapeutic intervention 1,2 .
The crucial role of the PI3K/AKT/mTOR (PAM) pathway in cell survival, proliferation, growth, apoptosis, and glycogen metabolism and its frequent activation or over expression in many types of human cancers, such as breast, ovarian, prostatic, lung, gastric, pancreatic cancers, and B-cell lymphomas has put it in the line of promising druggable targets in the war against cancer [3][4][5] . Moreover, mis-regulation or genomic alterations in PAM pathway develop a secondary resistance that menace most targeted anticancer therapies 6,7 . Hence, attention has been devoted to developing a new cancer treatment with added benefits of overcoming chemoresistance through targeting this oncogenic pathway. Phosphoinositide 3-kinases (PI3Ks) are a family of lipid kinases consisting of three major classes that phosphorylate inositol phospholipids generating the second messenger phosphatidylinositol-3,4,5-triphosphate (PIP3). Of interest is Class I which is frequently associated with cancer development containing four catalytic subunits p110a and p110b which are expressed in all tissues, p110c, and p110d which were reported to play an important role in haematological malignancies. The formed PIP3 activates the serine/threonine kinase AKT (PKB) which is considered a central node in this pathway allowing phosphorylation at Thr308 or Ser473 by phosphoinositide-dependent kinase (PDK1). Phosphorylated AKT (p-AKT) is involved in the deregulation of apoptosis, proliferation, and cell metastasis by modulating the phosphorylation of several downstream protein substrates that regulate cell growth. 8,9 One of the important downstream proteins is cyclin D1 known to promote cell proliferation through mediating cell progression from G1-phase to S-phase. AKT is known to stabilise cyclin D1 through inactive phosphorylation of GSK3Kb. This results in GSK3Kb losing its kinase activity to phosphorylate Thr286 in cyclin D1 which in turn inhibits its cytoplasmic proteasomal degradation 6 (Figure 1).
Previous studies reported that blocking AKT enzyme or inhibiting p-AKT expression only could result in compensatory resistance via reactivating PI3K or mTOR. Moreover, researchers documented that phosphorylation of AKT is a transient event that lasts for a short duration followed by re-occurring hyperphosphorylation while treatment is being continued with ATP-competitive inhibitors [10][11][12] . Whilst, targeting mTOR only by recently developed antitumor agents results in AKT activation then cuts-off the pathway feedback inhibition which attenuates the potential antitumor activity 13 . Consequently, a combination therapy or dual-acting agents on PI3K/AKT is a key approach providing broader efficacy and better patient tolerability compared to other candidates targeting only one component of this pathway 10 .
Out of the huge prevalence of heterocyclic rings in drug development, pyrimidine nucleus has crowned them as it constitutes an integral part in nucleic acid structures. In addition, it establishes a core pharmacophore in natural and synthetic drugs with diverse activity and good safety profile 14,15 . The antitumor activity of the pyrimidine scaffold is due to its ability to bind to several enzymes, receptors, and target proteins starting from 5-Fluorouracil passing by Imatinib and Methotrexate and even those entered clinical trials nowadays [16][17][18][19][20] (Figures 2 and 3).
Exemplified inhibitors for this pathway that had been developed and entered clinical trials include: A number of AKT inhibitors acting on two distinct binding sites: ATP-binding site and allosteric pocket of the autoinhibited isoform as shown in Figure  2 [18][19][20][21][22][23][24] , such as Ipatasertib [GDC0068] I 18 which is a highly selective ATP-competitive pan-Akt inhibitor in phase-III clinical trials to evaluate its effectiveness in triple-negative breast cancer [TNBC] and Capivasertib [AZD5363] II 19,20 which is also in phase-III trial to further evaluate its efficacy and safety in combination with paclitaxel in first-line treatment of patients with metastatic TNBC. Likewise, the allosteric pan-AKT inhibitor MK-2206 V 23 demonstrated activity in treating patients with relapsed acute myeloid leukaemia (AML) in phase-II clinical studies. Also, Afuresertib VI 24 exhibited a favourable safety profile and clinical activity against haematologic malignancies. Other agents acting on the PAM pathway as presented in Figure 3 [25][26][27][28] include: Idelalisib VII 25 a PI3K-d inhibitor that was approved for the treatment of different types of lymphomas. As well as compound VIII 26 which was developed by our research team having a pyrimidine nucleus known for its diverse anticancer mechanisms substituted with 6-morpholino group as Pan-PI3K inhibitor and apoptosis inducer.
In continuing our research work on developing pyrimidine derivatives acting on cell proliferation and apoptosis induction through PI3K/AKT axis inhibition, this work focussed on preparing new analogues derived from our previously reported lead compound (VIII) 26 keeping (4-(4-methoxyphenyl)-2-(methylthio)pyrimidine-5-carbonitrile) as the main core. Moreover, to study the effect of C-4 substitution on enzyme inhibition, ring variation and chain extension strategies were adopted via linking the pyrimidine core to substituted aryl moieties through different amino bridges one to four atoms in distance replacing the morpholine ring. Also, replacing indolo-hydrazino moiety at C-2 with small methylthio group to render the molecules more linear to fit with AKT pharmacophoric features 11,25 keeping the essential PI3K requirements as illustrated in Figure 4.
Since that pathway has a very obvious implication in breast cancer and leukemia 4,5 . The in-vitro cytotoxic activity of the newly synthesised compounds was assayed against breast cancer (MCF-7) and acute leukaemia (K562) cell lines. To explore the underlying molecular mechanism of these compounds; PI3Kd, PI3Kc, and AKT-1 enzyme inhibition assays were done followed by cell cycle  analysis, apoptosis induction, and caspase 3 evaluation using ELISA. In addition, western blotting assay for PI3K, p-PI3K, AKT, p-AKT, Cyclin D1, and NFKb were carried for the most active compound to see the effect on the downstream proteins of this pathway. Then, molecular simulation studies were performed to clarify the binding behaviour of the most active derivative with the nominated enzyme isoforms and to predict its pharmacokinetics. Finally, in-vivo toxicity study was done to check its safety for future development.

Chemistry
All chemicals were commercially available. Melting points were determined on Stuart apparatus and the values given are uncorrected. The IR spectra were recorded (KBr discs) on Shimadzu IR 435 spectrophotometer, College of Pharmaceutical Sciences and Drug Manufacturing, Misr University for Science and Technology, Egypt and the values were expressed wavenumber in cm À1 . The 1 H-NMR and 13 C-NMR spectral data were obtained on Bruker 400 MHz (Bruker Corp., Billerica, MA), Microanalytical Unit, Faculty of Pharmacy, Cairo University, Cairo, Egypt, using TMS as internal standard, and chemical shift values were recorded in ppm on d scale. The mass spectra and elemental analyses were performed at the Regional Centre for Mycology and Biotechnology, Al-Azahr University, Cairo, Egypt. The reactions were monitored by precoated aluminium sheets (TLC) in benzene-acetone (7:3 v/v).

General procedure for the preparation of 4-(2-arylidenehy-
A mixture of hydrazinyl compound 6 (0.5 g, 1.7 mmol), glacial acetic acid (2 ml) and the appropriate aldehydes (1.7 mmol) in absolute ethanol (20 ml) was heated under reflux for 10-12 h. The precipitate was filtered and crystallised from ethanol.  13 13   dehydrogenase. Mitochondrial dehydrogenases of viable cells cleave the tetrazolium ring, giving purple formazan crystals that are dissolved in acidified isopropanol and the resulting purple solution is measured spectrophotometrically. The change in cell number is proportional to the intensity of purple colour produced and the degree of cytotoxicity caused by the test compounds. The absorbance at a wavelength of 570 nm was measured using (ROBONIK P2000 spectrophotometer) and the background absorbance of multi-well plates at 690 nm then the absorbance at 690 nm was subtracted from the 570 nm measurement. Five concentrations from each test compound ranging from (0.04 to 100 mM) were prepared and the experiments were done in triplicate. IC 50 values (the concentration required for 50% inhibition) were calculated using doseresponse curves and linear regression equation.

PI3K enzyme inhibition assay
The inhibition activities of 4d and 7f against PI3K enzyme were detected using PI3K assay kits and ADP-Glo kinase assay as a detection buffer (BPS Bioscience, San Diego, CA). According to the manufacturer's instructions, PI3K enzyme was diluted with 2.5x Kinase assay buffer to obtain 4 ng/ml. The assay was performed in 96 well plates as follows: 5 ml of PIP2 substrate was added to each well followed by 5 ul from each inhibitor except for the positive control and blanks where the inhibitor buffers were added instead. Of 5 ul of 12.5 mM ATP was added. The reactions were initiated by adding 10 ml of diluted PI3K enzyme to the positive control and inhibitors wells. Plate was carefully shaked and was incubated at 30 C for 40 min. After the 40 min incubation, 25 ml of ADP-Glo reagent were added to each well. The plate was covered with aluminium foil and incubated at room temperature for another 45 min. Of 50 ml of Kinase Detection reagent was added to each well. Plate was covered with aluminium foil and incubated at room temperature for extra 30 min. Luminescence was measured using microplate reader. Four concentrations from the two tested compounds ranged from (0.1 to 100 mM) were prepared and the experiment was repeated in triplicate. The results were expressed in the form of IC 50 values (the concentration required for 50% inhibition). IC 50 values were calculated using doseresponse curves and linear regression equation. LY294002 compound was used as the positive control. The inhibition/doseresponse plots of the tested compounds and the control are listed in the Supplementary Materials.

AKT enzyme inhibition assay
AKT activity was measured in cell lysates for the tested compounds (4d and 7f) using enzyme-linked immuno-absorbent assay (ELISA) (Abcam, Cat No. ab139436, Cambridge, UK) according to the manufacturer's protocol. In brief, cell lysates were collected after treatment with the tested compounds using lysis buffer and centrifuged at 13,000 rpm for 15 min to obtain the cytosolic fraction. Samples were soaked with the kinase diluted buffer and the reaction was initiated by adding 10 ml of diluted ATP to each well, except blanks. Plate was incubated at 30 C for up to 90 min covered with an adhesive plate sealer with gentle, shaking every 20 min. The reaction was stopped by emptying contents of each well. Of 40 ml of the phospho-specific substrate antibody was added except blank then incubated incubate at room temperature for 60 min, with shaking every 20 min. The liquid was aspirated and the wells washed three times with 1X washing buffer. After the last wash, liquid was aspirated and 40 ml of the diluted Anti-Rabbit IgG: HRP conjugate was added except blank, and incubated at room temperature for 30 min with shaking every 10 min. Plate was washed as described before. Of 60 ml of the TMB Substrate was added to each well and incubated at room temperature for 30-60 min. The reaction was stopped by adding 20 ml of the stop solution to each well. Absorbance was measured at 450 nm. Four concentrations from the tested compounds were prepared ranged from 0.1 to 100 mM and the assay was done in triplicates. The results were expressed in the form of IC 50 values (the concentration required for 50% inhibition) using doseresponse curves and linear regression equation. LY2780301 compound was used as the positive control. The inhibition/doseresponse plots of the tested compounds and the control are listed in the Supplementary Materials.

Flowcytometric analysis and apoptosis
Cell cycle analysis and apoptosis induction of the most potent compound 7f was tested against leukaemia cell line (K562).

Cell cycle analysis.
Cell cycle analysis was determined by propidium iodide (PI) flow cytometry kit/BD. K562 cells were treated by compound 7f at its IC 50 concentration. Then cells were washed in PBS, fixed in cold 70% ethanol for 30 min, washed twice in PBS, and centrifuged to remove supernatant, after that, cells were treated with 0.1 mg/ml ribonuclease, stained with 0.05 mg/ml PI. The results were analysed by flow cytometry using FACS calibre and cell cycle distribution were calculated and accomplished as reported 26,34,35 .

2.2.4.2.
Annexin-V-FITC apoptosis assay. Apoptosis induction was assayed by staining K562 cells with Annexin-V (fluorescein isothiocyanate) and the counterstaining PI using the Annexin-V-FITC/PI apoptosis detection kit (BD Biosciences, San Diego, CA) according to the manufacturer's instructions. Cells were exposed to the tested compound for 48 h. After that, the cells were collected by trypsinisation and 0.5 Â 10 6 cells and washed twice with phosphate-buffered saline (PBS) followed by staining with 5 ml Annexin-V-FITC and 5 ml PI in 1 Â binding buffer for 15 min at room temperature in the dark. FACS Calibre flow cytometer (BD Biosciences, San Diego, CA) 26,36,37 was used for analyses.

Determination of the effect of 7f on active caspase 3
The concentration of human active caspase 3 was measured using in-vitro gen (catalogue KH 01091) ELISA kit. Of 100 ml of the standard diluents buffer were added to the zero standard wells. Then 100 ml of standard and controls were added to microlitre wells. Wells were covered and incubated for 2 h at room temp. After wells were decanted, 100 ml of active caspase 3 detection antibody solution were added into each well. Plates covered and incubated for 1 h at room temp. Then wells were decanted again and 100 ml of anti-rabbit IgG HRP were added to each well. Wells were covered again and incubated for 30 min. wells were decanted and 100 ml of stabilised chromogen were added. The wells began to turn blue and then incubated for 30 min. Finally, a stop solution was added and the colour in the wells changed from blue to yellow. The plate was read using ROBONIK P2000 ELISA reader at 450 nm and standard curve was obtained 38 .

Western blot analysis
K562 cells were treated with the most promising compound (7f), for 48 h. Protein from control and treated cells was harvested and the concentration was measured via Bradford protein assay (Bio-Rad, Hercules, CA). The samples were loaded and separated by sodium dodecyl sulphate (SDS)-polyacrylamide gel electrophoresis and then transferred onto a polyvinylidene difluoride (PVDF) membrane. Western blot analysis was performed according to previously described 39 .

2.2.7.
In-vivo acute toxicity study Five female rats (180-200 g) were used in this test. The animals were fasted for 24 h prior to administration of the test compound. The compound was administered in a dose of 2 g/kg body weight, po . Animals are observed individually and the mortalities were recorded at least once during the first 30 min after dosing, periodically during the first 24 h and daily afterwards, for a total of 14 d 40 . All animals received humane care and experimental procedures were carried out in strict accordance with the health and care guidelines for experimental animals. All experimental operations performed on rats were approved by the Animal Experiment Ethics Committee of Heliopolis University (Approval number HU.REC.A0.23-2021).

Molecular modelling
Molecular modelling study was accomplished using Discovery Studio software version 4.1 (Accelrys, Inc., San Diego, CA). The target compounds were drawn on ChemDraw Professional version 16.0 (PerkinElmer Inc, Waltham, MA) and subjected to energy minimisation using CHARMm-Force Field. The X-ray crystal structure of each enzyme was retrieved from the protein data bank (http://www.rscb.org/pdb) 41,42 . First, the enzyme protein was prepared for docking by removing the unwanted chain and water molecules then protonated, and missing amino acids were fixed. Second, the minimisation step was carried after applying fixed atom constraints on heavy atoms to keep the 3D structure of the protein. After that, the binding pocket was identified using the co-crystalised ligand and the validation step was performed. Finally, the docking of the prepared new ligands into the 3D structure of the protein was carried out adopting flexible ligandrigid receptor docking in the CDOCKER protocol. The best 10 docked conformers for each ligand were retained and their CDOCKER interaction scores were studied to get the pose which has the closest binding pattern like that of the bioactive conformation of the lead molecule.

In-vitro cytotoxic activity assay against breast cancer cell line (MCF-7) and leukaemia cell line (K562)
The newly synthesised compounds 4a-e, 6, 5a-d, and 7a-f were assessed for their cytotoxic activity against MCF-7 and K562 cell lines with the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) colorimetric assay 32,33 using Staurosporine as the positive control. The cytotoxic efficacy of these compounds was expressed as IC 50 value representing the concentration of the compound required to produce a 50% inhibition of cell growth after 48 h of incubation with the tested compounds as shown in Table 1.
The IC 50 values revealed that most of the newly developed pyrimidine derivatives demonstrated high cytotoxic activity on both cell lines compared to the reference compound Staurosporine In the aniline series (5a-d), the activity was decreased compared to the piperazine series. The two promising compounds in this group are the 4-chloro and 4-bromo-aniline derivatives 5c and 5d with IC 50 value of 5.82 ± 0.28 and 5.52 ± 0.26 uM on K562 cell line, respectively. These results display the importance of the presence of an aromatic ring substituted with a halogen.
Moreover, by increasing the linker length to be three atoms utilising the hydrazono linker between the aromatic moiety and the pyrimidine-5-carbonitrile scaffold in 7a-f series enhanced the cytotoxic activity again especially on leukaemia K562 cell line. Compound 7f manifested excellent cytotoxic activity with IC 50 value of 3.22 ± 0.18 uM on MCF-7 and 2.62 ± 0.12 uM on K562 cell lines. Also, compounds 7c and 7e exhibited remarkable cytotoxic activity with IC 50 values of 3.76 ± 0.18 and 5.36 ± 0.25 uM, respectively, on the K562 cell line indicating the positive effect of electron withdrawing groups on the cytotoxic activity.
In conclusion, it was found that Schiff's derivatives 7a-f and piperazine derivatives 4a-e series have better cytotoxic activity over aniline derivatives 5a-d which may be attributed to the presence of a linker (3 or 4 atoms in length) between the pyrimidine scaffold and the aromatic terminal that allow better interaction with their targets.
Alongside, the cytotoxicity of the most active compounds 4d and 7f were evaluated on a normal fibroblast cell line (WI-38) to judge their safety profile 43 . The results shown in Table 2 clarified that both candidate 4d and 7f have better selective toxicity towards cancer cells with selectivity index (SI ¼ 109.72 and 13.29, respectively) than the reference drug Staurosporine with SI ¼ 1.97. This suggested that the new compounds exhibited a good safety profile.

PI3K/AKT enzyme inhibition assays
In order to investigate the molecular mechanism by which the most active compounds 4d and 7f produce their potential antiproliferative action, enzyme inhibition assays were done on PI3Kc, PI3Kd, and AKT-1 isoforms known for their effects on cell survival and proliferation. Compound 7f exhibited a significant inhibitory activity value on the three tested enzymes compared to the reference compound LY294002 as depicted in Table 3 with p < .01, respectively. Compound 7f was more active to PI3Kd isoform than  PI3Kc isoform which is known for its overexpression in leukaemia, in addition to its excellent inhibitory activity on AKT enzyme. These results support the objectives of designing dual PI3K/AKT inhibitors to overcome the compensatory resistance developed with single enzyme-acting inhibitors. For compound 4d, it gave less inhibitory activity on PI3K isoforms and moderate inhibition on AKT-1. Thus, compound 7f was chosen to further explore its effects on apoptosis as an endpoint for inhibiting this signal transduction axis.

Cell cycle analysis and apoptosis induction
As reported in the literature, inhibition of PI3K/AKT pathway affects the cell survival and proliferation 8,9 . Thus, to investigate the effect of compound 7f in PI3K/Akt dual inhibition on cell cycle progression, cell cycle analysis was performed by flowcytometric analysis as reported 34,35 on K562 cells treated with compound 7f. The outcomes presented in Table 4 indicated the interference with the cell cycle distribution, following 7f treatment, where there was a marked increase in the cells in pre-G1 phase by nearly 17-fold as compared to the control. In addition, it showed an increase in S-phase by almost 1.3-fold as displayed in Figure 5. These results demonstrated the antiproliferative effect of compound 7f as evidenced by promoting S-phase cell cycle arrest and apoptosis induction.
Consequently, the pro-apoptotic activity of compound 7f was explored 36,37 as shown in Table 5. Data of the current research highlighted that treating Leukaemia K562 cells for 48 h with the compound 7f caused an increase in the percentage of Annexin-V positive apoptotic cells both in the early and late stage which represents a total significant 17-fold increase compared to the control. This finding suggests that the reported cytotoxic activity of the trimethoxy derivative 7f is attributed to apoptosis induction; as demonstrated in Figure 6.

Caspase 3 enzyme assay using ELISA in K562
PI3K/AKT inhibitors are known to potentiate the intrinsic apoptotic pathway through increasing the levels of active caspase 3 38 . To further investigate the effect of the novel studied compound on caspase 3 level, K562 cells were treated with 7f at its previously reported IC 50 value , where it showed significant increase in active caspase 3 level by approximately 5-fold (p < .001) as compared to the control cells. The reported result of 7f compound was nearly the same as that of the reference compound staurosporine (protein kinase inhibitor) as shown in Figure 7.
3.2.5. Western blot assay for the down-stream proteins AKT regulates cell proliferation by stabilising cyclin D1 through GSK3b inactivation. Activated AKT phosphorylates serine 9 of GSK3b to inactivate its kinase activity on threonine 286 of cyclin D1, which then blocks the nuclear export and the cytoplasmic proteasomal degradation of cyclin D1 44,45 . AKT can also phosphorylate and activate IjB kinase (IKK), a kinase that induces  Figure 5. The distribution of cells in cell cycle phases after treatment with the tested compound 7f. degradation of the NF-jB inhibitor, IjB39. Degradation of IjB releases NF-jB from the cytoplasm, allowing nuclear translocation, and activation of the oncogenic genes 46,47 . Western blot assay was conducted adopting the reported procedures 39 to investigate the PI3K/AKT dual inhibitory activity of the most active compound 7f on the protein expression levels of the down-stream biomarkers PI3K, AKT, and their phosphorylated forms in addition to Cyclin D1 and NF-jB in K256 cell line. The given results depicted in Table 6 and Figure 8 show a marked reduction in the expression of the investigated proteins indicating that 7f effectively inhibits cell growth and proliferation through attenuating PI3K/AKT axis activity.

In-vivo acute toxicity study
The toxicological profile of the most promising synthesised anticancer compound 7f was evaluated in female rats using the reported method in OECD guidelines 40 . Data from this study showed that the median lethal dose (LD 50 ) is greater than the test dose (2000 mg/kg) which proved that our new compound 7f is non-toxic and is well tolerated by experimental animals.      newly synthesised compound 7f on AKT-1, the binding patterns were simulated in the ATP-binding site and the allosteric site. For ATP competitive binding inhibition, the pdb code 4ekl 21,41,48 was used co-crystalised with the reference compound Ipatasertib which has the typical pharmacophore of AKT inhibitors: 1) it has a heteroaromatic moiety that form hydrogen bond in the Table 7. CDOCKER scores of the most active compound 7f in Kcal/mol in AKT-1 (ATP-binding site). hinge region with Ala230 and Glu228. 2) Aa linker/spacer connected to a basic amine interacting with Glu234 and Glu278 in the acidic rich pocket, and 3) a substituted aromatic ring that can fit into a hydrophobic groove formed by Gly159, Lys158, Lys179, and Leu181 under the P-loop.  Figure 11. The 2D/3D interaction diagrams of: a) the co-crystalised ligand; b) the reference compound (OR4); c) 7f binding in PI3Kc.
The 2D and 3D interaction diagrams in Figure 9 revealed similar orientation to the reference compound inside the ATP-active site where the aromatic 4-methoxyphenyl was inserted inside the hinge region (Ala230), the trimethoxyphenyl moiety interacts with Lys178 in the hydrophobic pocket, and the pyrimidine interacts with Glu234 in the acidic region. The docking scores are shown in Table 7. 3.3.1.2. AKT-docking (allosteric binding site). Furthermore, the most active compound 7f was docked into the allosteric binding site (pdb code:4ejn) 42 of AKT-1 to study different inhibition modes. The allosteric non-ATP dependent pocket is formed by the interaction of the PH domain and the KD in AKT-1 leading to an auto-inhibited conformation. In this autoinhibited conformation, the ATP binding site is occluded by nonpolar clusters (Ile84) forming a hydrophobic lock that prevents ATP and ATP-competitive inhibitors from binding. The main interactions in this binding site are a) The core nucleus imidazopyridine is stacked hydrophobically between Trp80 embedded in the PH domain, Val270, and Leu264; b) hydrogen bond interaction with N-pyridine ring and Asp292; c) hydrophobic interactions with Tyr272; d) the amide linker helps direct the terminal aromatic ring towards the hydrophobic lock.
Through studying the binding pattern of the most active compound 7f within the AKT-1 allosteric site represented in Figure 10 and the scores in Table 8, it was found that it has a similar binding mode like the reference compound (OR4). The core pyrimidine is stacked hydrophobically between the key amino acids Val270, Leu264, and Trp80. Also, the hydrazono linker at C-6 helps orienting the terminal phenyl ring towards the hydrophobic lock containing ILe84 similar to the amide linker of the reference ligand, and the trimethoxy phenyl interacts with Arg273 similar to the m-fluorophenyl of the reference compound. This suggests that its mechanism of inhibiting AKT-1 isoform might be through noncompetitive allosteric binding.
3.3.1.3. PI3K-c docking. The docking simulation of compound 7f inside the PI3K-c ATP-binding site (pdb code: 3r7r) 49 revealed that the binding patterns are similar to the reference compounds with the key amino acids Val882 in the hinge region and Lys833 in the P-loop as illustrated in Figure 11. Also, the docking scores are comparable to the co-crystalised ligand and reference compound O92 as presented in Table 9. These simulation studies imply that the newly synthesised compound inhibits PI3K-c through competing with ATP and AKT-1 through acting on its allosteric binding site.

Drug-likeness and molecular properties calculations
Since the biological activity of a given drug is a function of its physicochemical properties as they affect its ability to reach its biological target in reasonable concentration, it was important to determine these ADMET parameters for the newly discovered pyrimidine derivative 7f using Discovery Studio version 3.0 suite. The presented data in Table 10 demonstrated that 7f obeys Lipinski's rule of five which gives an indication about its drug-likeness. It has a Log P of 5 that may reflect on the low solubility, absorption levels and enhances the blood-brain barrier penetration. Additionally, it has plasma proteins binding ability due to its lipophilicity which will affect its duration of action, half-life, and elimination. Also, it was estimated that it is not a substrate for CYP2D6 therefore there will probably not be drug-drug interactions. Finally, the AMES and hepatotoxicity prediction suggested its safety 50,51 .

Conclusions
Innovative dual-acting PI3K/AKT inhibitors bearing 4-(4-methoxyphenyl)pyrimidine were synthesised and evaluated for their antiproliferative activity. All target compounds were screened for their cytotoxicity against breast cancer (MCF-7) and leukaemia (K562) cell lines. They demonstrated promising activities especially compounds 4d and 7f that exhibited excellent activities on both cell lines. The derivative 7f showed a strong effect in inhibiting PI3K and AKT enzymes as well as decreasing their protein expression. Furthermore, compound 7f caused cell cycle arrest at S phase leading to induction of apoptosis in leukaemia (k562) cells through caspase 3 activation by 5-fold compared to the control. It also showed a multi-protein targeting through preventing the downstream expression of Cyclin D1and NF-Kb, in addition to the non-toxic profile through the in-vivo toxicity study and ADMET prediction. Simulation studies proved that the design strategy adopted was efficient to develop dual enzyme inhibitor where the PI3K-c inhibition via ATP-competitive inhibition and AKT-1 through acting on the allosteric binding site. In conclusion, the newly synthesised pyrimidine compounds are effective anticancer leads through inhibition of PI3K/AKT axis and caspase 3 dependent apoptosis induction with a potential for advanced optimisation in future work.

Statistical analysis
Data are reported as means ± SD. Statistical analysis was performed using the GraphPad Prism version (GraphPad Software, La Jolla, CA). A p value less than .05 was considered statistically significant.  False False ÃÃ The extra modifications: 10 or less rotatable bonds and polar surface area (PSA) equal to or less than 140 Å 2 . ÃÃÃ a Lipophilicity descriptor; b polar surface area; c solubility level (0 ¼ extremely low, 1 ¼ very low but soluble, 2 ¼ low, 3 ¼ good, and 4 ¼ optimal); d absorption level (0 ¼ good, 1 ¼ moderate, 2 ¼ low, and 3 ¼ very low).

Animal use
All experimental operations performed on rats were approved by the Animal Experiment Ethics Committee of Heliopolis University (Approval number HU.REC.A0.23-2021).

Disclosure statement
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.