Synthesis of new pyridothienopyrimidinone derivatives as Pim-1 inhibitors

Abstract Four series of pyridothienopyrimidin-4-one derivatives were designed and prepared to improve the pim-1 inhibitory activity of the previously reported thieno[2,3-b]pyridines. Significant improvement in the pim-1 inhibition and cytotoxic activity was achieved using structure rigidification strategy via ring closure. Six compounds (6c, 7a, 7c, 7d, 8b and 9) showed highly potent pim-1 inhibitory activity with IC50 of 4.62, 1.18, 1.38, 1.97, 8.83 and 4.18 μM, respectively. Four other compounds (6b, 6d, 7b and 8a) showed moderate pim-1 inhibition. The most active compounds were tested for their cytotoxic activity on three cell lines [MCF7, HCT116 and PC3]. Compounds 7a [the 2-(2-chlorophenyl)-2,3-dihydro derivative] and 7d [the 2-(2-(trifluoromethyl)-phenyl)-2,3-dihydro derivative] displayed the most potent cytotoxic effect on the three cell lines tested consistent with their highest estimated pim-1 IC50 values.

Most of the research published on pim inhibitors focused on pim-1 inhibitors while pim-2 is more difficult to be targeted due to its low K m for ATP (100-fold lower than that of Pim-1 and Pim-3) [13][14][15][16] .
Recently reported manuscripts indicated that pim-1 kinase plays a significant role in stem cell proliferation, self-renewal and expansion 17 . These facts encourage the use Pim-1 inhibitors as a promising targeted therapy in cancer stem cells 5,17 .
The most important feature of pim-1 kinase active site is the presence of proline base at positions 123 and 125 within the hinge region. This extends the hinge region length and moves it 4 A o to the left and thus prevents the formation of the second H-bond between the hinge backbone and the adenine moiety of ATP 23 .
Almost all the reported pim kinase inhibitors are ATP-competitive and can be classified into ATP-mimetics and non-ATP mimetics. The ATP-mimetics bind directly to the hinge region usually via H-bonding with the backbone oxygen of Glu121 and exhibited great enzyme potency but with limited or poor selectivity over other kinases 1 . On the other hand, the non-ATP mimetics bind to the ATP active site in a manner different from ATP and most of them form H-bond with Lys67 1 . Generally, they interact with the portion of the active site opposite to the hinge region and this portion differs significantly between kinases. Thus, non-ATP mimietics tend to be more selective to pim-1 enzyme and meanwhile exhibited great potency to the enzyme. Compounds I-IV ( Figure 1) are all non-ATP mimietics.
In an attempt to prepare potent pim-1 inhibitors that can be used as anticancer agents, we had recently reported the pim-1 inhibitory activity of thieno [2,3-b]pyridine derivatives V ( Figure 2) 24 as bioisosteres to benzofuran-2-carboxylic acids 25 . However, their pim-1 inhibitions as well as their cytotoxic activities were only moderate to poor. This literature recorded some of the efforts done by our lab to improve the pim-1 kinase inhibitory activity and the cytotoxic activity of thieno [2,3-b]pyridine scaffold.
Searching the literature revealed no published data on pyridothienopyrimidinones as pim-1 kinase inhibitors. However, their bioisosteres benzothienopyrimidin-4-ones VI and VII 14,26,27 and benzofuropyrimidin-4-ones VIII and IX 28,29 had been reported earlier and exhibited potent and selective pim kinase inhibition ( Figure 3). The X-ray crystallography of representatives of both series bound to pim-1 showed similar binding mode in the ATP-binding site 27,29 . In both series, the 4-carbonyl group formed H-bond to Lys67, while the group at the 8-position (Br or aryl) occupied the hydrophobic area against the hinge region. The substituent at position 2 was exposed to the solvent and thus can accommodate various groups, however, the best results were achieved with ortho-substituted phenyl or aliphatic amines like dimethylaminomethylene group 27,29 . As such, both series might act as non-ATP mimetics.
Encouraged by all the after mentioned findings, we thought of examining the pyridothienopyrimidinone scaffold as a novel class of pim-1 inhibitors. Here in, the bromine atom at position 8 was kept constant to ensure hydrophobic interaction within the pim-1 ATP active site. The carbonyl group at position 4 was also kept constant to ensure binding to Lys67. However, different substitutions were introduced at position 2 in order to establish a SAR for this new scaffold. First, the unsubstituted derivative 5 was prepared. Then, two series of 2-substituted phenyl compounds and two series of alkyl substituted derivatives were prepared. The ortho-substituted phenyl series (6a-d) was prepared to mimic compound VIII. The ortho-substitution used were chloro and flouro as halogens, OH as electron donating group and CF 3 as bulky, hydrophobic and electron withdrawing group. In the second series (7a-d), the 2,3-dihydropyrimidinones, bearing the same 2aryl groups as 6a-d, were prepared to test whether aromaticity of   the pyrimidinone ring was essential for enzyme activity or not. In the third series (8a-c), different alkyl groups with different chain lengths were prepared. In the last series (9-11), carbonyl containing alkyl groups were prepared (Schemes 1-3). Docking study of these compounds in the pim-1 active site indicated the required binding mode with high energy scores (data not shown). All the synthesized compounds were examined for their pim-1 enzyme inhibitory activity and the most active compounds were further tested for their anti-proliferative activity using three different cell lines. Up to our knowledge, this is the first published work describing the pim-1 inhibitory activity of pyridothienopyrimidinone derivatives.

General notes
Griffin apparatus was used to determine the melting points and they were uncorrected. Shimadzu IR 435 spectrophotometer recorded the IR spectra and the values were represented in cm À1 . Bruker 400 MHz and 100 MHz spectrophotometer recorded 1 H NMR and 13 C NMR spectra, respectively. TMS was used as an internal standard and chemical shifts were recorded in ppm on d scale. Both IR and NMR spectra were carried out at Faculty of Pharmacy, Cairo University, Cairo, Egypt. The electron impact (EI) mass spectra were recorded on Thermo Scientific ISQLT single quadrapole mass spectrometer. Both mass spectra and elemental analyses were carried out at the regional center for mycology and biotechnology, Al-Azhar University, Cairo, Egypt. Analytical thin layer chromatography (TLC) on silica gel plates containing UV indicator was employed routinely to follow the course of the reactions and to check the purity of the products. All reagents and solvents were purified and dried by standard techniques.   General procedure for the synthesis of 8-bromo-2- pyridine-2-carboxamide 4 (0.6 g, 2 mmol), the appropriate 2-substituted benzaldehyde (2 mmol) and conc. HCl (0.5 mL) in DMF (5 mL) was heated under reflux for 12 h. The reaction was cooled, and the solid formed was filtered, dried and crystallized from the suitable solvent.  General procedure for the synthesis of 8-bromo-2- pyridine-2-carboxamide 4 (0.6 g, 2 mmol) and the appropriate 2-substituted benzaldehyde (2 mmol) in glacial acetic acid (5 mL) was heated under reflux for 10 h. The reaction was cooled, and the solid formed was filtered, dried and crystallized from acetic acid.  under reflux for 5 h. The reaction was cooled, poured onto icecold water (50 mL) and left overnight. The solid formed was filtered, dried and crystallized from the suitable solvent.

Materials and methods
The kinase inhibitory activity of the synthesized compounds was determined using the Kinexus compound profiling service, Canada. All the compounds were tested for their inhibitory activity against pim-1 kinase at 50 lM. The kinase used was cloned, expressed and purified using proprietary methods. Quality control testing is routinely performed to ensure compliance to acceptable standards. 33 P-ATP was purchased from PerkinElmer. All other materials were of standard laboratory grade.

Pim-1 kinase protein assay
The protein kinase target profiling was executed via employing a radioisotope assay format. All the assays were performed in a prepared radioactive working area. The protein kinase profiling assays were performed at room temperature for 20-30 min in a final volume of 25 lL according to the following assay reaction components: Component 1; 5 ll of diluted active protein kinase ($10-50 nM final concentration in the assay). Component 2; 5 ll of stock solution of substrate. Component 3; 5 ll of kinase assay buffer. Component 4; 5 ll of the test compound, Staurosporine at 1 lM or 10% DMSO. Component 5; 5 ll of 33 P-ATP (250 lM stock solution, 0.8 lCi).
The assay was initiated by the addition of 33 P-ATP, followed by incubation at ambient temperature for 30 min. The assay was then terminated by spotting 10 lL of the reaction mixture onto a Multiscreen phosphocellulose P81 plate. The Multiscreen phosphocellulose P81 plate was washed three times for approximately 15 min each in a 1% phosphoric acid solution. The radioactivity on the P81 plate was counted in the presence of scintillation fluid in a Trilux scintillation counter. Blank control was set up that included all the assay components except the addition of the appropriate substrate (replaced with equal volume of assay dilution buffer). The corrected activity for protein kinase target was determined by removing the blank control value. The results were displayed in terms of percent inhibition and IC 50 for the most active compounds. Table 1 and Figure 4 showed the obtained results.

In vitro cytotoxic activity
Cell culture Cancer cells from different cancer cell lines were purchased from American type Cell Culture collection (ATCC, Manassas, VA). The cell lines used in this study were human breast adenocarcinoma (MCF7), human colon adenocarcinoma (HCT116) and human prostate cancer cells (PC3). The cell lines were grown on the appropriate growth medium Dulbecco's modified Eagle's medium (DMEM) or Roswell Park Memorial Institute medium (RPMI 1640) supplemented with 100 mg/mL of streptomycin, 100 units/mL of penicillin and 10% of heat-inactivated fetal bovine serum in a humidified, 5% (v/v) CO 2 atmosphere at 37 C.

Cytotoxicity assay by 3-[4,5-dimethylthiazole-2-yl]-2,5-diphenyltetrazolium bromide (MTT)
Exponentially growing cells from different cancer cell lines were trypsinized, counted and seeded at the appropriate densities (2000-10 000 cells/0.33 cm 2 well) into 96-well microtiter plates. The cells were incubated in a humidified atmosphere at 37 C for 24 h. Then, the cells were exposed to different concentrations of compounds 6c, 7a, 7c, 7d, 8b and 9 (0.1, 10, 100 and 1000 lM) for 72 h. The viability of the treated cells was determined using MTT technique. The media were removed; cells were incubated with 200 lL of 5% MTT solution/well (Sigma Aldrich, St. Louis, MO) and were allowed to metabolize the dye into colored-insoluble formazan crystals for 2 h. The remaining MTT solution was discarded from the wells and the formazan crystals were dissolved in 200 lL/ well-acidified isopropanol for 30 min, covered with aluminum foil and with continuous shaking using a MaxQ 2000 plate shaker (Thermo Fisher Scientific Inc, MI) at room temperature. The absorbance was measured at 570 nm using a Stat FaxR 4200 plate reader (Awareness Technology, Inc., Palm City, FL). The cell viability were expressed as percentage of control and the concentration that induces 50% of maximum inhibition of cell proliferation (IC 50 ) was determined for each compound using Graph Pad Prism version 5 software (Graph Pad software Inc, CA) 32,33 . The results are shown in Table 2 and represented graphically in Figure  5.

Chemistry
The synthesis of the target compounds was outlined in Schemes 1-3.
On the other hand, reacting compound 4 with 2-substituted benzaldehydes in acetic acid gave 2,3-dihydropyridothienopyrimidin-4-ones 7a-d. The formation of the dihydro products was confirmed by the spectral data. Thus, the IR spectra of compounds 7a-d showed two NH bands at 3433-3275 cm À1 and one C ¼ O band at 1662-1635 cm À1 . A bright evidence was obtained from the 1 H NMR spectra of compounds 7a-d that revealed the presence of two exchangeable doublet signals at d 6.77-7.18 ppm and d 7.99-8.37 ppm corresponding to two NH protons as well as a doublet of doublets signal at d 6.06-6.16 ppm corresponding to CH-2 proton. Moreover, the 13 C NMR spectra of compounds 7a,b showed a signal at d 61.8-64.8 ppm corresponding to CH-2 carbon together with the signals of the introduced aromatic carbons and the C ¼ O carbons.
Reacting the starting compound 4 with different acid anhydrides afforded 2-methyl, 2-trifluoromethyl and 2-ethylpyridothienopyrimidinones 8a-c. The IR spectroscopy confirmed the structures of 8a-c through the appearance of NH band at 3421-3197 cm À1 and C ¼ O band at 1662-1670 cm À1 . Meanwhile, their 1 H NMR spectra showed an NH exchangeable singlet signal at d 12.78-12.95 ppm. The 13 C NMR spectrum of compound 8b showed a signal at d 172.4 ppm corresponding to C ¼ O carbon.
The reaction of the amino amide derivative 4 and ethyl acetoacetate or diethyl malonate gave the 2-(2-oxopropyl) derivative 9 and the acetate derivative 10, respectively. The IR spectrum of both compounds showed two carbonyl bands at 1685 and 1639 cm À1 in case of compound 9 and 1747, 1658 cm À1 in case of compound 10. The 1 H NMR spectrum of compound 9 revealed the appearance of two singlet signals at d 2.28 ppm and d 3.96 ppm corresponding to the 2-oxopropyl protons as well as an exchangeable singlet signal at d 12.84 ppm indicating the cyclization of the pyrimidine ring. Whilst, the 1 H NMR spectrum of compound 10 showed a singlet signal at d 3.86 ppm assigned to the methylene proton beside triplet and quartet signals assigned to the ethyl protons.
Alkaline hydrolysis of the ester group of compound 10 using 10% alc. KOH resulted in the formation of the acetic acid derivative 11. Its IR spectrum indicated the absence of the ester C ¼ O at 1747 cm À1 and the appearance of acidic C ¼ O at 1720 cm À1 . While, its 1 H NMR spectrum revealed the absence of the characteristic triplet and quartet signals of the ester group.

Pim-1 kinase inhibitory activity
All the compounds were tested for their ability to inhibit pim-1 kinase at 50 lM using the Kinexus compound profiling service, Canada and the results (in terms of percentage inhibition and IC 50 for the active compounds) were displayed in Table 1 and represented graphically in Figure 4.
SAR study of the pyrido[3',2':4,5]thieno[3,2-d]pyrimidin-4(1H)ones as pim-1 inhibitors indicated the following points: The 2-unsubstituted derivative 5 showed poor enzyme inhibitory activity (37%). Regarding the 2-(2-substituted phenyl) series (6a-d), it was found that the 2-chlorophenyl derivative 6a gave poor inhibition of pim-1 kinase (36%) with IC 50 >100 lM. However, other members of the same series showed moderate to potent   activity (54-85%). The 2-hydroxyphenyl derivative 6c showed the highest pim-1 inhibitory activity in this series (85% inhibition of pim-1 kinase with IC 50 4.62 lM). The 2-(2-substituted phenyl)-2,3-dihydro series (7a-d) afforded the most potent pim-1 inhibitors in this study. Members belonging to this series showed pim-1 inhibition in the range of 66-98% with IC 50 values in the range of 1.18-1.97 lM. The 2chlorophenyl derivative 7a showed the highest pim-1 kinase inhibition with IC 50 of 1.18 lM (compare with 6a). Thus, it seemed that aromaticity of the pyrimidine ring was not essential for pim-1 inhibition. Further study of the effect of substitution at meta or para positions of the phenyl ring is still needed The 2-alkyl derivatives 8a-c exhibited great variability in their activities as pim-1 inhibitors. Thus, while the 2-methyl derivative 8a showed moderate pim-1 inhibition (51%), its replacement with 2-triflouromethyl group in 8b enhanced the activity significantly (96% inhibition and IC 50 of 8.83 lM). On the other hand, increasing the chain length into 2-ethyl group (compound 8c) reduced the enzyme inhibition greatly (23%). Regarding the carbonyl containing alkyl series 9-11, it was found that the oxopropyl derivative 9 showed potent pim-1 inhibitory activity (89% with IC 50 of 4.18 lM). Nevertheless, the ethyl acetate derivative 10 and its acid derivative 11 gave poor pim-1 inhibition.

In vitro cytotoxic activity
The most active pim-1 inhibitors in this study work, namely, compounds 6c, 7a, 7c, 7d, 8b and 9 were screened for their cytotoxic activity against three cell lines using MTT method 32,33 . The cell lines examined were the human breast adenocarcinoma (MCF7), the human colon adenocarcinoma (HCT116) and the human prostate cancer cells (PC3). The results in terms of IC 50 in lM are given in Table 2 and represented graphically in Figure 5.
From the results, it can be concluded that MCF7 and HCT116 cell lines were more sensitive to the action of the compounds than PC3 cell line.
Both compounds 6c and 7c exhibited moderate cytotoxic effect on all the cell lines tested, whilst compound 9 displayed weak cytotoxic activity on the three cell lines.
Again, the results obtained here were better than those obtained with thieno[2,3-b]pyridine derivatives.

Conclusion
Structure rigidification via ring closure proved to be a successful strategy to improve the pim-1 inhibitory activity as well as the cytotoxic activity of thieno [2,3-b]pyridines. In the present work, four series of pyridothienopyrimidin-4-one derivatives were designed and prepared as pim-1 inhibitors. While only one thieno [2,3-b]pyridine derivative displayed potent pim-1 inhibition with IC 50 of 12.71 lM, six pyridothienopyrimidin-4-ones (6c, 7a, 7c, 7d, 8b and 9) showed highly potent pim-1 inhibitory activity with IC 50 of 4.62, 1.18, 1.38, 1.97, 8.83 and 4.18 lM, respectively. SAR study of pyridothienopyrimidin-4-ones indicated that the 2-(2-substituted phenyl)-2,3-dihydro series 7a-d afforded the most potent pim-1 inhibitors. The most active compounds were tested for their cytotoxic activity on three cell lines [MCF7, HCT116 and PC3]. Compounds 7a [the 2-(2-chlorophenyl)-2,3-dihydro derivative] and 7d [the 2-(2-(trifluoromethyl)-phenyl)-2,3-dihydro derivative] exhibited the most potent cytotoxic activity on the three cell lines tested. A significant improvement of the cytotoxic activity was also noticed relative to the precursors thieno[2,3-b]pyridine derivatives. The results of the cytotoxicity were in good agreement with the pim-1 IC 50 values. Further work on pyridothienopyrimidin-4ones is still needed to obtain more potent pim-1 inhibitors and to improve the physicochemical properties of these derivatives.