Synthesis, Evaluation, and Mechanism Study of New Tepotinib Derivatives as Antiproliferative Agents

Inspired by the potent inhibition activity of the c-Met (mesenchymal−epithelial transition factor) inhibitor Tepotinib, a series of new Tepotinib derivatives were synthesized and evaluated for their ability to act as antiproliferative agents to find the leading compounds with good activity and limited side effects. Among them, compound 31e exhibited potent antiproliferative activity (IC50 (50% inhibitory concentration) = 0.026 μΜ) against hepatic carcinoma 97H (human liver cancer cell) cells and, importantly, had very low inhibitory activity against normal cells. A mechanism study demonstrated that 31e induced G1 phase (First growth phase or G indicating gap) arrest, inhibited the phosphorylation of c-Met and its downstream signaling component, Akt (Protein Kinase B), and also inhibited the migration of hepatic carcinoma 97H cells.


Introduction
Cancer is a major disease that seriously threatens human health. According to statistics from the World Health Organization (WHO), an estimated 9.6 million patients died from cancer in 2018 [1]. Therefore, the development of effective anti-tumor drugs is an important task for medicinal chemists. Mesenchymal−epithelial transition factor (c-Met), the high affinity receptor for hepatocyte growth factor (HGF), is a unique subfamily of receptor tyrosine kinases (RTKs) [2]. Physiologically, HGF/c-Met signaling plays important roles in cell growth, survival, motility, and morphogenesis [3]. However, aberrant c-Met activation has been observed in a wide variety of human cancers, including liver and lung cancer, as a consequence of gene amplification or rearrangement, transcriptional regulation, as well as autocrine or paracrine ligand stimulation [4][5][6][7][8][9]. Due to its status as a proto-oncogene and the correlation of its dysregulation with a poor prognosis, c-Met is becoming a promising target for cancer therapy [10]. The present c-Met kinase inhibitors are divided into three groups according to their binding modes and structural features. Type I c-Met kinase inhibitors, such as Crizotinib A, are adenosine triphosphate (ATP)-competitive inhibitors with a U-shaped binding mode at the ATP binding site (Figure 1) [11,12]. Type II c-Met kinase inhibitors are usually multi-targeted agents (for example, Cabozantinib B) ( Figure 1) [13]. Type I inhibitors are generally described as specific for c-Met kinase and more selective than class II inhibitors. However, they have limited activity against the Tyr1230His mutation that is present in certain human tumors [4]. Type III inhibitors include other atypical c-Met kinase inhibitors, such as Tivantinib C (Figure 1) [14].  [15]. We found that the optimal compound, Tepotinib D ( Figure 1), displays excellent in vitro potency and in vivo anti-tumor efficacy at low doses. According to the co-crystal structure of Tepotinib D with c-Met, it belongs to the type I c-Met kinase inhibitors. The review by Manjunath D. Ghate et al. described the recent advances of small molecule c-Met kinase inhibitors, including several pyridazinone derivatives, but the similar structures of our compounds have not been reported before [16]. Inspired by the interesting scaffold of Tepotinib, we decided to synthesize and evaluate new Tepotinib derivatives to find leading compounds with good activity and limited side effects.
Our design is shown in Figure 2 and includes the following steps: (a) Changing the position of nitrogen on the pyridazinone ring to provide pyrazin-2(1H)-one and pyrimidin-2(1H)-one derivatives for the study of the structure-activity relationships; (b) evaluating the anti-tumor activity of the derivatives with or without alkyl substitution on the methylene group; (c) examining the effect of fluorine substitution on pyridin-2(1H)-one on the anti-tumor activity; and (d) investigating the effect of side chains on the activity. Herein, we reported the synthesis, evaluation, and optimization of new Tepotinib derivatives as anti-tumor agents.

Chemistry
The synthesis of target compounds 11a and 11b is summarized in Scheme 1. The conventional Suzuki coupling reaction of (3-(hydroxymethyl)phenyl)boronic acid and 2-chloro-5-fluoropyrimidine catalyzed with PdCl2(PPh3)2 gave  [15]. We found that the optimal compound, Tepotinib D ( Figure 1), displays excellent in vitro potency and in vivo anti-tumor efficacy at low doses. According to the co-crystal structure of Tepotinib D with c-Met, it belongs to the type I c-Met kinase inhibitors. The review by Manjunath D. Ghate et al. described the recent advances of small molecule c-Met kinase inhibitors, including several pyridazinone derivatives, but the similar structures of our compounds have not been reported before [16]. Inspired by the interesting scaffold of Tepotinib, we decided to synthesize and evaluate new Tepotinib derivatives to find leading compounds with good activity and limited side effects.
Our design is shown in Figure 2 and includes the following steps: (a) Changing the position of nitrogen on the pyridazinone ring to provide pyrazin-2(1H)-one and pyrimidin-2(1H)-one derivatives for the study of the structure-activity relationships; (b) evaluating the anti-tumor activity of the derivatives with or without alkyl substitution on the methylene group; (c) examining the effect of fluorine substitution on pyridin-2(1H)-one on the anti-tumor activity; and (d) investigating the effect of side chains on the activity. Herein, we reported the synthesis, evaluation, and optimization of new Tepotinib derivatives as anti-tumor agents. In 2015, Dorsch et al. reported optimized pyridazinones as being potent and selective c-Met kinase inhibitors [15]. We found that the optimal compound, Tepotinib D ( Figure 1), displays excellent in vitro potency and in vivo anti-tumor efficacy at low doses. According to the co-crystal structure of Tepotinib D with c-Met, it belongs to the type I c-Met kinase inhibitors. The review by Manjunath D. Ghate et al. described the recent advances of small molecule c-Met kinase inhibitors, including several pyridazinone derivatives, but the similar structures of our compounds have not been reported before [16]. Inspired by the interesting scaffold of Tepotinib, we decided to synthesize and evaluate new Tepotinib derivatives to find leading compounds with good activity and limited side effects.
Our design is shown in Figure 2 and includes the following steps: (a) Changing the position of nitrogen on the pyridazinone ring to provide pyrazin-2(1H)-one and pyrimidin-2(1H)-one derivatives for the study of the structure-activity relationships; (b) evaluating the anti-tumor activity of the derivatives with or without alkyl substitution on the methylene group; (c) examining the effect of fluorine substitution on pyridin-2(1H)-one on the anti-tumor activity; and (d) investigating the effect of side chains on the activity. Herein, we reported the synthesis, evaluation, and optimization of new Tepotinib derivatives as anti-tumor agents.

Chemistry
The synthesis of target compounds 11a and 11b is summarized in Scheme 1.

Chemistry
The synthesis of target compounds 11a and 11b is summarized in Scheme 1. The conventional Suzuki coupling reaction of (3-(hydroxymethyl)phenyl)boronic acid and 2-chloro-5-fluoropyrimidine catalyzed with PdCl 2 (PPh 3 ) 2 gave (3-(5-fluoropyrimidin-2-yl)phenyl)methanol 5 with Chlorination of 5 in the presence of thionyl chloride provided 2-(3-(chloromethyl)phenyl)-5-fluoropyrimidine 6. On the other hand, the oxidation of ketones 1a or 1b followed by cyclization provided 2a and 2b [17], which reacted with compound 6 to give 7a and 7b. Compound 10 was synthesized by the reaction of 8 and 9. Finally, compounds 11a and 11b were prepared by the reaction of 7a, 7b, and 10 under alkaline conditions. (3-(5-fluoropyrimidin-2-yl)phenyl)methanol 5 with a yield of 68%. Chlorination of 5 in the presence of thionyl chloride provided 2-(3-(chloromethyl)phenyl)-5-fluoropyrimidine 6. On the other hand, the oxidation of ketones 1a or 1b followed by cyclization provided 2a and 2b [17], which reacted with compound 6 to give 7a and 7b. Compound 10 was synthesized by the reaction of 8 and 9. Finally, compounds 11a and 11b were prepared by the reaction of 7a, 7b, and 10 under alkaline   The synthesis of compounds 31a-g is summarized in Scheme 3. Intermediate 27 was prepared by the Suzuki coupling reaction of (3-acetylphenyl)boronic acid and 2-chloro-5-fluoropyrimidine. The reduction of compound 27 with sodium borohydride and then chlorination with thionyl chloride produced 29. In the presence of potassium carbonate, 29 reacted with compound 16 to give 30a, which reacted with the corresponding amino alcohols or amino thiols to produce target compounds 31a-f (the R and S configurations of 31e were prepared and are shown in the supporting information). The fluorine-containing target compound 31g was prepared with a similar procedure, starting from compound 14 and intermediate 30b. The synthesis of compounds 31a-g is summarized in Scheme 3. Intermediate 27 was prepared by the Suzuki coupling reaction of (3-acetylphenyl)boronic acid and 2-chloro-5-fluoropyrimidine. The reduction of compound 27 with sodium borohydride and then chlorination with thionyl chloride produced 29. In the presence of potassium carbonate, 29 reacted with compound 16 to give 30a, which reacted with the corresponding amino alcohols or amino thiols to produce target compounds 31a-f (the R and S configurations of 31e were prepared and are shown in the supporting information). The fluorine-containing target compound 31g was prepared with a similar procedure, starting from compound 14 and intermediate 30b.
The synthesis of target compounds 36a-c is listed in Scheme 4. Compound 33 was obtained by the reaction of 6 and commercially available 5-bromo-3-fluoropyridin-2(1H)-one 32. Then, the reaction with aryl boric acids 34a-c afforded the key intermediates, 35a-c, which reacted with 10 in the presence of NaOH to provide target compounds 36a-c.  The synthesis of target compounds 36a-c is listed in Scheme 4. Compound 33 was obtained by the reaction of 6 and commercially available 5-bromo-3-fluoropyridin-2(1H)-one 32. Then, the reaction with aryl boric acids 34a-c afforded the key intermediates, 35a-c, which reacted with 10 in the presence of NaOH to provide target compounds 36a-c. The synthesis of target compounds 36a-c is listed in Scheme 4. Compound 33 was obtained by the reaction of 6 and commercially available 5-bromo-3-fluoropyridin-2(1H)-one 32. Then, the reaction with aryl boric acids 34a-c afforded the key intermediates, 35a-c, which reacted with 10 in the presence of NaOH to provide target compounds 36a-c.

In Vitro Growth Inhibition of Human Cancer Cell Lines.
Hepatocellular carcinoma (HCC), a primary liver cancer with a high mortality rate, accounts for 90% of all liver cancers. It is the most common malignant tumor worldwide, especially in Asia, Africa, and southern Europe. China is the "hardest hit" by liver cancer, and the number of annual deaths from this condition in China accounts for about 50% of the total number of liver cancer deaths worldwide. C-Met overexpression, mutational activation, and amplification have been found in some types of cancer, including liver cancer [4,5]. Therefore, hepatic carcinoma 97H cells were selected for the evaluation of the anti-tumor activity of the synthesized compounds, and the results are summarized in Table 1. Most compounds displayed good antiproliferative activity. Among them, compound 31e, 5-(3-cyanophenyl)-1-(1-(3-(5-((1-methylpiperidin-4-yl)methoxy)pyrimidin-2-yl)phenyl)ethyl)pyrimi din-2(1H)-one, exhibited the best result with 26 nM of the IC50 value. Further study of the structureactivity relationship indicated the pyridazinone ring or pyrimidin-2(1H)-one ring plays an important role in antiproliferative activity. Compounds 11a and 11b, pyrazin-2(1H)-one derivatives, that were obtained by shifting the nitrogen from the C-1 to C-5 position of pyridazin-3(2H)-one, displayed poor activity (IC50 values: 9.8 μM for 11a and 9.1 μM for 11b). Interestingly, compounds 18 and 25, which were obtained by shifting the nitrogen from the C-1 to the C-4 position, resulted in significant improvement in anti-proliferative potency (IC50: 0.36-0.55 μM). Compounds 31a-g, which have a methyl substitution on the methylene carbon between pyrimidin-2(1H)-one and the benzene ring, provided 0.3606 μM to 0.026 μM of the IC50 value. This indicates that the methyl substitution at this position is favorable for antiproliferative activity. Compounds Rac-31e and (R)-31e with cyclic piperidinylmethyloxy substitution showed excellent activity (IC50 = 0.026 μM, 0.018 μM). However, the (S)-31e almost lost activity (IC50 = 1.834 μM). The replacement of the pyridazinone moiety with 3-fluoropyridin-2(1H)-one moiety afforded compounds 36a-c. Compound 36a showed the best result of the 3-fluoropyridin-2(1H)-one derivatives, indicating that the benzonitrile group is more favorable than the other two substitutions.

In Vitro Growth Inhibition of Human Cancer Cell Lines.
Hepatocellular carcinoma (HCC), a primary liver cancer with a high mortality rate, accounts for 90% of all liver cancers. It is the most common malignant tumor worldwide, especially in Asia, Africa, and southern Europe. China is the "hardest hit" by liver cancer, and the number of annual deaths from this condition in China accounts for about 50% of the total number of liver cancer deaths worldwide. C-Met overexpression, mutational activation, and amplification have been found in some types of cancer, including liver cancer [4,5]. Therefore, hepatic carcinoma 97H cells were selected for the evaluation of the anti-tumor activity of the synthesized compounds, and the results are summarized in Table 1. Most compounds displayed good antiproliferative activity. Among them, compound 31e, 5-(3-cyanophenyl)-1-(1-(3-(5-((1-methylpiperidin-4-yl)methoxy)pyrimidin-2-yl)phenyl)ethyl)pyrimidin -2(1H)-one, exhibited the best result with 26 nM of the IC 50 value. Further study of the structure-activity relationship indicated the pyridazinone ring or pyrimidin-2(1H)-one ring plays an important role in antiproliferative activity. Compounds 11a and 11b, pyrazin-2(1H)-one derivatives, that were obtained by shifting the nitrogen from the C-1 to C-5 position of pyridazin-3(2H)-one, displayed poor activity (IC 50 values: 9.8 µM for 11a and 9.1 µM for 11b). Interestingly, compounds 18 and 25, which were obtained by shifting the nitrogen from the C-1 to the C-4 position, resulted in significant improvement in anti-proliferative potency (IC 50 : 0.36-0.55 µM). Compounds 31a-g, which have a methyl substitution on the methylene carbon between pyrimidin-2(1H)-one and the benzene ring, provided 0.3606 µM to 0.026 µM of the IC 50 value. This indicates that the methyl substitution at this position is favorable for antiproliferative activity. Compounds Rac-31e and (R)-31e with cyclic piperidinylmethyloxy substitution showed excellent activity (IC 50 = 0.026 µM, 0.018 µM). However, the (S)-31e almost lost activity (IC 50 = 1.834 µM). The replacement of the pyridazinone moiety with 3-fluoropyridin-2(1H)-one moiety afforded compounds 36a-c. Compound 36a showed the best result of the 3-fluoropyridin-2(1H)-one derivatives, indicating that the benzonitrile group is more favorable than the other two substitutions.  To evaluate the ability of compound 31e to disrupt the regulated cell cycle distribution, we performed flow cytometry analysis to determine the arrest effect of 31e on the G1 transition, a rigorously regulated process in the cell cycle. The results in Figure 3 show that 31e can arrest cells in the G1 phase in a dose-dependent manner. When hepatic carcinoma 97H cells were treated with 10 nM of 31e for 24 h, the population of cells in the G1 phase (83.84%) increased compared to that of the vehicle (DMSO, 70.42%), along with concomitant losses in the G2/M phase. This phenomenon was more obvious at a concentration of 100 nM. A population of up to 90.88% cells was found in the G1 phase. Correspondingly, a population of only 5.78% cells was found in the G2/M phase. To evaluate the ability of compound 31e to disrupt the regulated cell cycle distribution, we performed flow cytometry analysis to determine the arrest effect of 31e on the G1 transition, a rigorously regulated process in the cell cycle. The results in Figure 3 show that 31e can arrest cells in the G1 phase in a dose-dependent manner. When hepatic carcinoma 97H cells were treated with 10 nM of 31e for 24 h, the population of cells in the G1 phase (83.84%) increased compared to that of the vehicle (DMSO, 70.42%), along with concomitant losses in the G2/M phase. This phenomenon was more obvious at a concentration of 100 nM. A population of up to 90.88% cells was found in the G1 phase. Correspondingly, a population of only 5.78% cells was found in the G2/M phase.

Inhibition of the Migration of Hepatic Carcinoma 97H Cells with Compound 31e
Tumor cell migration is one of the major causes of death in cancer patients. To evaluate the inhibitory ability of 31e against tumor cell migration, hepatic carcinoma 97H cells (5 × 10 4 cells per well) were seeded in six-well plates with or without the presence of 31e and cultured as confluent monolayers. The results shown in Figure 4 demonstrate that 31e can inhibit tumor cells' migration. At a concentration of 10 nM, 31e displayed moderate inhibition of migration. Migration was significantly suppressed when 31e was used at a dosage of 50 nM. Tumor cells' migration was completely suppressed with 100 nM of 31e.  (B) Quantitative analysis of the percentage of cells in each cell cycle phase was analyzed by EXPO32 ADC analysis software. The data were presented as the mean ± SEM *P < 0.05, significantly different compared with the control by t test.

Inhibition of the Migration of Hepatic Carcinoma 97H Cells with Compound 31e
Tumor cell migration is one of the major causes of death in cancer patients. To evaluate the inhibitory ability of 31e against tumor cell migration, hepatic carcinoma 97H cells (5 × 10 4 cells per well) were seeded in six-well plates with or without the presence of 31e and cultured as confluent monolayers. The results shown in Figure 4 demonstrate that 31e can inhibit tumor cells' migration. At a concentration of 10 nM, 31e displayed moderate inhibition of migration. Migration was significantly suppressed when 31e was used at a dosage of 50 nM. Tumor cells' migration was completely suppressed with 100 nM of 31e.

Effect of Compound 31e on Cell Cycle Progression
To evaluate the ability of compound 31e to disrupt the regulated cell cycle distribution, we performed flow cytometry analysis to determine the arrest effect of 31e on the G1 transition, a rigorously regulated process in the cell cycle. The results in Figure 3 show that 31e can arrest cells in the G1 phase in a dose-dependent manner. When hepatic carcinoma 97H cells were treated with 10 nM of 31e for 24 h, the population of cells in the G1 phase (83.84%) increased compared to that of the vehicle (DMSO, 70.42%), along with concomitant losses in the G2/M phase. This phenomenon was more obvious at a concentration of 100 nM. A population of up to 90.88% cells was found in the G1 phase. Correspondingly, a population of only 5.78% cells was found in the G2/M phase.

Inhibition of the Migration of Hepatic Carcinoma 97H Cells with Compound 31e
Tumor cell migration is one of the major causes of death in cancer patients. To evaluate the inhibitory ability of 31e against tumor cell migration, hepatic carcinoma 97H cells (5 × 10 4 cells per well) were seeded in six-well plates with or without the presence of 31e and cultured as confluent monolayers. The results shown in Figure 4 demonstrate that 31e can inhibit tumor cells' migration. At a concentration of 10 nM, 31e displayed moderate inhibition of migration. Migration was significantly suppressed when 31e was used at a dosage of 50 nM. Tumor cells' migration was completely suppressed with 100 nM of 31e.

Inhibition of Phosphorylation of c-Met and its Downstream Signaling Component, Akt, with Compound 31e
To further study the antiproliferative mechanism of compound 31e, we analyzed its effects on the phosphorylation of c-Met and its downstream signaling component, Akt, in hepatic carcinoma 97H cells, which overexpress c-Met. As shown in Figure 5A, when hepatic carcinoma 97H cells were treated with compound 31e at a concentration of 10 or 50 nM for 1 h, the phosphorylation level of Akt and c-Met was suppressed in a dose-dependent manner. When 100 nM of compound 31e was used, the phosphorylation of both c-Met and Akt was completely inhibited. When 97H cells were treated at the same concentration of compound 31e for 2 h, the inhibition of c-Met and Akt phosphorylation was more obvious (Figure 5B). for 36 h were photographed under a phase contrast microscopy (magnification: 4 × objective). All results were expressed as the mean ± SD of at least three independent experiments.

Inhibition of Phosphorylation of c-Met and its Downstream Signaling Component, Akt, with Compound 31e
To further study the antiproliferative mechanism of compound 31e, we analyzed its effects on the phosphorylation of c-Met and its downstream signaling component, Akt, in hepatic carcinoma 97H cells, which overexpress c-Met. As shown in Figure 5A, when hepatic carcinoma 97H cells were treated with compound 31e at a concentration of 10 or 50 nM for 1 h, the phosphorylation level of Akt and c-Met was suppressed in a dose-dependent manner. When 100 nM of compound 31e was used, the phosphorylation of both c-Met and Akt was completely inhibited. When 97H cells were treated at the same concentration of compound 31e for 2 h, the inhibition of c-Met and Akt phosphorylation was more obvious ( Figure 5B).

Selectivity of 31e towards Other Cancer Cells and Normal Human Cells
To evaluate the selectivity of 31e towards other cancer cells and normal human cells, the antiproliferative activities against Pc9 (human non-small cell lung cancer cells), Hela (human epithelial cervical cancer cells), SJSA1 (human osteosarcoma cells), LO2 (human normal cells), and HLF (human embryonic lung fibroblast cells) were examined. As shown in Table 2, compound 31e barely inhibited the cancer cells (Pc9, Hela, SJSA1) and normal cells (LO2, HLF), demonstrating that it exclusively inhibits 97H cancer cells. The reason for this remains to be further investigated.

Chemistry
1 H-NMR and 13 C-NMR spectra were recorded on a Bruker AVANCE 400 or 500 spectrometer (Karlsruhe, Germany). Chemical shifts of protons are reported in parts per million downfield from

Selectivity of 31e towards Other Cancer Cells and Normal Human Cells
To evaluate the selectivity of 31e towards other cancer cells and normal human cells, the antiproliferative activities against Pc9 (human non-small cell lung cancer cells), Hela (human epithelial cervical cancer cells), SJSA1 (human osteosarcoma cells), LO2 (human normal cells), and HLF (human embryonic lung fibroblast cells) were examined. As shown in Table 2, compound 31e barely inhibited the cancer cells (Pc9, Hela, SJSA1) and normal cells (LO2, HLF), demonstrating that it exclusively inhibits 97H cancer cells. The reason for this remains to be further investigated.

Synthesis of 5-(3-cyanophenyl)-1-(3-(5-((3-(dimethylamino)propyl)thio)pyrimidin-2-yl)benzyl)
pyrazin-2(1H)-one (11a). In a 200 mL two-necked round-bottomed flask provided with a magnetic stirrer and condenser, 50 mmol of thiourea 9 were dissolved in 60 mL of absolute ethanol. Compound 8 (3.161 mg, 20 mmol) was added in one portion in the thiourea solution. After heating at reflux for 12 h, the reaction mixture was concentrated by rotary evaporation under reduced pressure. The desired compound 10 was obtained as a white solid and was further used without purification [18]. NaOH (135 mg, 3.375 mmol) in 0.5 mL water was added to a mixture of 7a (287.25 mg, 0.75 mmol) and 10 (444 mg) in DMF (5 mL) under nitrogen, and the mixture was stirred at room temperature for 15 min. Then, the mixture was stirred at 60 • C under nitrogen for 8 h. The reaction mixture was allowed to cool to room temperature. The aqueous phase was extracted with dichloromethane (30 mL × 3). The combined organic layer was washed with H 2 O (15 mL) and brine (10 mL), and then dried over anhydrous Na 2 SO 4 , filtered, and evaporated in vacuo. The residue was purified by flash chromatography over silica gel (DCM/MeOH = 40:1-10:1) to give