Design, Synthesis and Evaluation of Novel Phorbazole C Derivatives as MNK Inhibitors through Virtual High-Throughput Screening

MNKs (mitogen-activated protein kinase-interacting protein kinases) phosphorylate eIF4E at Ser209 to control the translation of certain mRNAs and regulate the process of cell proliferation, cell migration and invasion, etc. Development of MNK inhibitors would be an effective treatment for related diseases. We used the MarineChem3D database to identify hit compounds targeting the protein MNK1 and MNK2 through high-throughput screening. Compounds from the phorbazole family showed good interactions with MNK1, and phorbazole C was selected as our hit compound. By analyzing the binding mode, we designed and synthesized 29 derivatives and evaluated their activity against MNKs, of which, six compounds showed good inhibition to MNKs. We also confirmed three interactions between this kind of compound and MNK1, which are vital for the activity. In conclusion, we report series of novel MNK inhibitors inspired from marine natural products and their relative structure–activity relationship. This will provide important information for further developing MNK inhibitors based on this kind of structure.

Given that MNKs knock out in mice did not affect the normal phenotype [24], and the phosphorylation of eIF4E is essential for tumor progression [24,25], development of MNK inhibitors would be an effective treatment for related diseases. Although the study of MNK inhibitors has lasted for more than 10 years, none have been approved for market. Inhibitors of CGP57380 [26], staurosporine [27] and cercosporamide as MNKs agents have been found to target other kinases potently [3]. Other MNK inhibitors including MNK-I1 [6], MNK-7g [28], SEL-201 [29], ETC-206 [30], DS12881479 [31], eFT508 [32] and BAY1143269 [33] ( Figure 1) have been found that inhibit the activity of MNK1 and/or MNK2 potently and selectively. Currently, BAY1143269, eFT508 and ETC-206 have entered into clinical trials for different settings [4], while BAY1143269 and ETC-206 have been withdrawn or terminated for some reasons (see clinicaltrials.gov accessed on 12 July 2020). We still need more MNK inhibitors with novel structures to interfere with MNK's pharmacologic action and push the application of MNK inhibitors in more diseases.
Mar. Drugs 2022, 20, x FOR PEER REVIEW 2 of 14 resistant estrogen receptor (ER+) breast cancer has a strong positive relationship with increased MNK phosphorylation of eIF4E [22] and rapamycin-resistant KRAS G12D colorectal cancer activated the MNK/eIF4E pathways with overexpression of c-MYC [23].
Given that MNKs knock out in mice did not affect the normal phenotype [24], and the phosphorylation of eIF4E is essential for tumor progression [24,25], development of MNK inhibitors would be an effective treatment for related diseases. Although the study of MNK inhibitors has lasted for more than 10 years, none have been approved for market. Inhibitors of CGP57380 [26], staurosporine [27] and cercosporamide as MNKs agents have been found to target other kinases potently [3]. Other MNK inhibitors including MNK-I1 [6], MNK-7g [28], SEL-201 [29], ETC-206 [30], DS12881479 [31], eFT508 [32] and BAY1143269 [33] (Figure 1) have been found that inhibit the activity of MNK1 and/or MNK2 potently and selectively. Currently, BAY1143269, eFT508 and ETC-206 have entered into clinical trials for different settings [4], while BAY1143269 and ETC-206 have been withdrawn or terminated for some reasons (see clinicaltrials.gov). We still need more MNK inhibitors with novel structures to interfere with MNK's pharmacologic action and push the application of MNK inhibitors in more diseases. Marine natural products possess rich and novel chemical structures owing to the unique ecosystem of the ocean. There are more than 30,000 marine natural small compounds that have been found, while the study of these compounds is limited. Our group participated in the construction of a 3D structure database of these marine natural products (MarineChem3D, http://mc3d.qnlm.ac/, accessed on 10 March 2020) in order to find more active compounds through virtual high-throughput screening. We used this database and docked with the protein complexes of MNK1 (PDB: 5wvd) and MNK2 (PDB: 2hw7). As a result, five hits were chosen by analyzing their interaction mode with MNK1 or MNK2. Phorbazole series compounds, which were found from Aldisa andersoni [34], showed better interaction with the MNK1 protein. We designed and synthesized series derivatives of phorbazole C and evaluated their activity against MNKs. Finally, we discussed their structure-activity relationship (SAR) and found a novel MNK inhibitor. This can be a starting compound for further exploration.

The Discovery of Hit Compounds
Through high-throughput screening of MarineChem3D targeting to MNK1 and MNK2 proteins, we obtained five hit compounds including phorbazole A, phorbazole B, phorbazole C, phorbazole D, and 9-chloro-phorbazole D, which showed good binding to the MNK1 protein complex (PDB: 5wvd) ( Figure 2). These compounds belong to phorbazole series, consisting of three parts in the structure, a phenol motif, an oxazole motif and Marine natural products possess rich and novel chemical structures owing to the unique ecosystem of the ocean. There are more than 30,000 marine natural small compounds that have been found, while the study of these compounds is limited. Our group participated in the construction of a 3D structure database of these marine natural products (MarineChem3D, http://mc3d.qnlm.ac/, accessed on 10 March 2020) in order to find more active compounds through virtual high-throughput screening. We used this database and docked with the protein complexes of MNK1 (PDB: 5wvd) and MNK2 (PDB: 2hw7). As a result, five hits were chosen by analyzing their interaction mode with MNK1 or MNK2. Phorbazole series compounds, which were found from Aldisa andersoni [34], showed better interaction with the MNK1 protein. We designed and synthesized series derivatives of phorbazole C and evaluated their activity against MNKs. Finally, we discussed their structure-activity relationship (SAR) and found a novel MNK inhibitor. This can be a starting compound for further exploration.

The Discovery of Hit Compounds
Through high-throughput screening of MarineChem3D targeting to MNK1 and MNK2 proteins, we obtained five hit compounds including phorbazole A, phorbazole B, phorbazole C, phorbazole D, and 9-chloro-phorbazole D, which showed good binding to the MNK1 protein complex (PDB: 5wvd) ( Figure 2). These compounds belong to phorbazole series, consisting of three parts in the structure, a phenol motif, an oxazole motif and a pyrrole motif. The interaction maps showed that the hydroxyl of benzene formed a hydrogen bond with residue Asp191, and two hydrogen bonds formed between the N atom of oxazole and the NH of pyrrole and residue Leu127 (Figure 2A-E). The 9-chlorine in phorbazole A and 9-chloro-phorbazole D formed a hydrogen bond with residue Glu125 (Figure 2A,D) and the oxazole formed π-H interaction with Leu55. However, other chlorines in pyrrole exposed the outer of the binding pocket ( Figure 2B,C), and considering halogen are the characters of marine natural products, we set phorbazole C as the starting point to design our derivatives. a pyrrole motif. The interaction maps showed that the hydroxyl of benzene formed a hydrogen bond with residue Asp191, and two hydrogen bonds formed between the N atom of oxazole and the NH of pyrrole and residue Leu127 (Figure 2A-E). The 9-chlorine in phorbazole A and 9-chloro-phorbazole D formed a hydrogen bond with residue Glu125 (Figure 2A,D) and the oxazole formed π-H interaction with Leu55. However, other chlorines in pyrrole exposed the outer of the binding pocket ( Figure 2B,C), and considering halogen are the characters of marine natural products, we set phorbazole C as the starting point to design our derivatives.

Design and Synthesis of Derivatives
To confirm the role the hydroxyl group and pyrrole played in the inhibition to MNKs activity, we designed three series of compounds: (i) change the position of hydroxyl in the benzene with or without other substituents, (ii) change the pyrrole group into furan or thiophene, (iii) change the hydroxyl group into other atoms or groups (Schemes 1 and 2).

Design and Synthesis of Derivatives
To confirm the role the hydroxyl group and pyrrole played in the inhibition to MNKs activity, we designed three series of compounds: i) change the position of hydroxyl in the benzene with or without other substituents, ii) change the pyrrole group into furan or thiophene, iii) change the hydroxyl group into other atoms or groups (Schemes 1 and 2). The total synthetic route of phorbazole C has been reported [35]. According to that, we synthesized compounds 1-29. The hydroxyl groups of compounds (1A, 8A-8E) were protected by chloromethyl methyl ether (MOMCl) in the presence of N,N-diisopropylethylamine (DIPEA) in dichloromethane (DCM) at room temperature. Then, compounds (2A-2F, 9A-9E) were followed by a Henry reaction with nitromethane and triethylamine (Et3N) in DMSO to yield compounds 3A-3F, 10A-10E, which were catalytically hydrogenated to the aminoethanol (4A-4F, 11A-11E) in the presence of Pd/C and hydrogen. The amide compounds (5A-5V, 12A-12G) were achieved through the reaction of aminoethanol compounds and related acids in the presence of EDCI and HOBt. The acylaminoalcohol compounds were oxidized to the ketone (6A-6V, 13A-13G) by Dess-Martin periodinane (DMP). Cyclodehydration of the acylaminoketone compounds to the oxazole (7A-7C, 14A-14G, 10, 12-29) was realized in the presence of triphenylphosphine, perchloroethane, and Et3N. The oxazole compounds were deprotected by treatment with diluted hydrochloric acid to obtain related compounds (1-9 and 11).

The Structure-Activity Relationship (SAR) of Derivatives
The activity of these compounds against MNKs was tested to assess the level of phosphorylation of eIF4E, which is the best-characterized and only in vivo validated substrate for the MNKs. We used 3T3-L1 cells treated with compounds at 10 µM for 2 h and cell lysates were analyzed by SDS-PAGE. The results showed that compounds 1, 2, 3, 4, 5, and 6 had strong inhibition of MNKs ( Figure 3A), while other compounds did not exert any inhibition of MNKs ( Figure 3B-D). We then tested compounds 1, 2, 3, 5 and 6 at a lower concentration of 3 µM, and compounds 5 and 6 were the best inhibitors among these compounds, showing complete inhibition of p-eIF4E at 3 µM ( Figure 3E). for the MNKs. We used 3T3-L1 cells treated with compounds at 10 µM for 2 h and cell lysates were analyzed by SDS-PAGE. The results showed that compounds 1, 2, 3, 4, 5, and 6 had strong inhibition of MNKs ( Figure 3A), while other compounds did not exert any inhibition of MNKs ( Figure 3B-3D). We then tested compounds 1, 2, 3, 5 and 6 at a lower concentration of 3 µM, and compounds 5 and 6 were the best inhibitors among these compounds, showing complete inhibition of p-eIF4E at 3 µM ( Figure 3E). When we kept the structure of oxazole and pyrrole, the R1 group in benzene played important roles for the activity of MNKs. The -OH group in benzene was better than other groups (-H, -F, -Cl, -Br, -OCH3) in the R1 position. For example, compound 1 was better than compounds 10, 14, 18, 22 and 26 which did not show any inhibition of MNKs. When R1 was hydrogen or R2 or R3 was -OH, compounds lost the activity against MNKs (compounds 1-6 were better than compounds 7 and 8). When R1 was -OH and other groups were R2, R3 and/or R4, this would decrease the inhibition (compounds 8 and 9 did not show inhibitory activity against MNKs). The substituent in R5 has little effect on the inhibitory activity, as in compounds 1 and 2, 5 and 6. These results indicated that -OH in R1 of compounds was vital for MNKs activity. Next, when we kept -OH in R1, pyrrole was replaced by furan and the compound lost the inhibitory activity (compounds 1 and 11), which means NH in pyrrole was necessary for MNKs inhibitory activity.
We then used the docking software (MOE) to study why these groups are important for the activity against MNKs (Figure 4). The docking map of Phorbazole C showed the hydrogen of -OH formed a hydrogen bond with the oxygen atom of Asp191. When the -OH group was replaced by halogen, this hydrogen bond could not be formed, owing to a further distance ( Figure 4A,E). The -OH at other positions of benzene was far away from Asp191, resulting in this hydrogen bond interaction being missed ( Figure 4A). When we When we kept the structure of oxazole and pyrrole, the R 1 group in benzene played important roles for the activity of MNKs. The -OH group in benzene was better than other groups (-H, -F, -Cl, -Br, -OCH 3 ) in the R 1 position. For example, compound 1 was better than compounds 10, 14, 18, 22 and 26 which did not show any inhibition of MNKs. When R 1 was hydrogen or R 2 or R 3 was -OH, compounds lost the activity against MNKs (compounds 1-6 were better than compounds 7 and 8). When R 1 was -OH and other groups were R 2 , R 3 and/or R 4 , this would decrease the inhibition (compounds 8 and 9 did not show inhibitory activity against MNKs). The substituent in R 5 has little effect on the inhibitory activity, as in compounds 1 and 2, 5 and 6. These results indicated that -OH in R 1 of compounds was vital for MNKs activity. Next, when we kept -OH in R 1 , pyrrole was replaced by furan and the compound lost the inhibitory activity (compounds 1 and 11), which means NH in pyrrole was necessary for MNKs inhibitory activity.
We then used the docking software (MOE) to study why these groups are important for the activity against MNKs (Figure 4). The docking map of Phorbazole C showed the hydrogen of -OH formed a hydrogen bond with the oxygen atom of Asp191. When the -OH group was replaced by halogen, this hydrogen bond could not be formed, owing to a further distance ( Figure 4A,E). The -OH at other positions of benzene was far away from Asp191, resulting in this hydrogen bond interaction being missed ( Figure 4A). When we introduced bulk groups in the benzene, the interaction between oxazole and Leu127 disappeared and the inhibition of MNKs was lost ( Figure 4B). When the pyrrole group was replaced by furan, the interactions with Asp191 and Leu127 were kept. However, a hydrogen bond with Leu127 was lost ( Figure 4F). All these results indicated that these three hydrogen bonds play vital roles in interacting with MNK1 to show inhibitory activity. introduced bulk groups in the benzene, the interaction between oxazole and Leu127 disappeared and the inhibition of MNKs was lost ( Figure 4B). When the pyrrole group was replaced by furan, the interactions with Asp191 and Leu127 were kept. However, a hydrogen bond with Leu127 was lost ( Figure 4F). All these results indicated that these three hydrogen bonds play vital roles in interacting with MNK1 to show inhibitory activity.

Evaluation the Activity of the Compounds
To evaluate the effects of compounds on cell proliferation, we conducted the MTT assay in H446, Hela, MDA-MB-231 and U937 cell lines. All the compounds at 10 µM showed weak inhibition of the cell viability, including six MNK inhibitors (compounds 1-6) ( Figure 5). Among these four cell lines, compounds 1-6 showed about 50% inhibition in H446 cells. We then selected four compounds (1, 3, 5 and 6) to test their IC50 values using MTT assays. Compound 5 had a higher cytotoxic activity, with an IC50 of 6.38 µM ( Figure 6).

Evaluation the Activity of the Compounds
To evaluate the effects of compounds on cell proliferation, we conducted the MTT assay in H446, Hela, MDA-MB-231 and U937 cell lines. All the compounds at 10 µM showed weak inhibition of the cell viability, including six MNK inhibitors (compounds 1-6) ( Figure 5). Among these four cell lines, compounds 1-6 showed about 50% inhibition in H446 cells. We then selected four compounds (1, 3, 5 and 6) to test their IC 50 values using MTT assays. Compound 5 had a higher cytotoxic activity, with an IC 50 of 6.38 µM ( Figure 6).    The inhibition of MNKs and the phosphorylation of eIF4E have been found to impair the process of EMT (epithelial-to-mesenchymal transition) and metastasis in cancer cells. Therefore, we assessed the influence of compound 5 on H446 cells. We used 'scratch wound-healing assays' to analyze the effect of 5 on the migratory potential of H446 cells. In line with the previous report, we observed significantly reduced migratory potential in cells with 5 compared with DMSO at 16 h after treatment ( Figure 7A,B), and the effect was even more pronounced at 24 h ( Figure 7A,C). The inhibition of MNKs and the phosphorylation of eIF4E have been found to impair the process of EMT (epithelial-to-mesenchymal transition) and metastasis in cancer cells. Therefore, we assessed the influence of compound 5 on H446 cells. We used 'scratch wound-healing assays' to analyze the effect of 5 on the migratory potential of H446 cells. In line with the previous report, we observed significantly reduced migratory potential in cells with 5 compared with DMSO at 16 h after treatment ( Figure 7A,B), and the effect was even more pronounced at 24 h ( Figure 7A,C).

Discussion
MNKs regulate cell functions by involving the process of mRNA translation and protein synthesis through the phosphorylation of eIF4E [36]. Inhibitory activity of MNKs and then the phosphorylation of eIF4E could be an effective and safe target for the treatment

Discussion
MNKs regulate cell functions by involving the process of mRNA translation and protein synthesis through the phosphorylation of eIF4E [36]. Inhibitory activity of MNKs and then the phosphorylation of eIF4E could be an effective and safe target for the treatment of cancers. MNK inhibitors have been developed for several years; however, none have been approved. Most inhibitors showed 'off target' effects, inhibiting other kinases potently [3]. Therefore, selective inhibitors with novel structure are still needed.
Marine natural products possess novel chemical structures with good bioactivity, while most compounds have not found value in medicine. The development and utilization of these compounds is necessary. MNK1/2 kinases belong to the family of Ca 2+ /calmodulindependent kinases (CaMKs) and share 80% homology with other kinases, except for a unique DFD-motif and three insertions [37]. These characteristics are the key points for the design of inhibitors. In this article, we used virtual high-throughput screening technology targeting the MNK1 and MNK2 protein complex based on the MarineChem3D database that comprises more than 30,000 marine natural products. Five compounds (phorbazole A, phorbazole B, phorbazole C, phorbazole D, and 9-chloro-phorbazole D) were selected for their good binding interactions with MNK1. We selected phorbazole C as a hit compound and designed a series of derivatives by analyzing the interaction mode between phorbazole C and MNK1. Six compounds showed good activity against MNKs and inhibited the phosphorylation of eIF4E in 3T3-L1 cells. Using the docking software MOE and pymol, we analyzed and concluded how the groups in this structure affect the MNK inhibitory activity. Firstly, the hydroxyl group at the para position of benzene formed a hydrogen bond with Asp191. Other groups or atoms (such as -F, -Br, -Cl and -OCH 3 ) in this position could decrease or lose their inhibition of MNKs owing to this important hydrogen bond not being formed. We also found that multi-substituted phenol would decrease the inhibition. Secondly, the oxazole and pyrrole also showed two hydrogen bonds' formation with Lue127 of the hinge region in the MNK1 pocket. Substituents in pyrrole also decrease the inhibitory activity of MNKs, and these two interactions seem be vital for this activity. The results showed that these hydrogen bonds keep the compound stable in the pocket, which provides an idea for further designing derivatives and selective MNK inhibitors.
Furthermore, we also found if we use the amino group (-NH 2 ) to replace the hydroxyl group at the para position of benzene, it would retain all these interactions, and may show MNK-inhibitory activity. Unfortunately, we did not obtain this compound in our synthetic route. In future, we still need to confirm this result by an appropriate route to obtain better MNK inhibitors with novel structure and good selectivity.
Several sources have reported that overexpression of MNKs and p-eIF4E were associated with the progression and development of cancer [38,39]. The loss of MNK1/2 genes did not affect cell proliferation in MEFs and KIT-mutant melanoma cells [16,25], but inhibiting the activity of MNK2 in non-small-cell lung cancer impaired the proliferation and migration [40]. In our assay, compounds 1-6 showed weak effects on cell proliferation in Hela, U937 and MDA-MB-231 cell lines, but good inhibition rate in H446 cells. Compound 5 also inhibited the migration of H446 cells significantly, which was in accordance with a previous report. Novel and potent MNK inhibitors could be designed and synthesized based on this structure.

SDS-PAGE and Western Blot
Cell monolayers were harvested in RIPA lysis buffer with 1 mM PMSF from a commercial company (Beyotime, Beijing China). After lysis, insoluble material was removed by centrifugation at >12,000× g for 10 min at 4 • C. Protein content was determined by the BAC assay. Cell lysates were heated at 95 • C for 3 min in sample buffer (250 mM Tris-HCl, 10% (v/v) sodium dodecyl sulfate (SDS), 20% (v/v) glycerol, 0.12% (w/v) bromophenol blue) and subjected to polyacrylamide gel electrophoresis (PAGE) and electrophoretic transfer to nitrocellulose membranes. Membranes were then blocked in phosphate-buffered saline (PBS)-0.05% Tween20 containing 5% (w/v) skim milk powder for 30 min at room temperature. Membranes were probed with the indicated primary antibody overnight at 4 • C. After incubation with HRP-tagged secondary antibody, blots were visualized using Tanon-5200. Primary antibodies and secondary antibodies were from Abcam.

Scratch Wound-Healing Assay
Cells were seeded on 24-well plates. When the cells had reached 90-100% confluence, wounds to cell layers were induced by scratching with a 10 µL pipette tip. The medium was changed to remove any floating cells and the cells were then treated with compound 5 (10 µM and 3 µM) or vehicle control (DMSO) in fresh medium for indicated time. Images of wound closure were taken directly by a microscope after scratching as the pre-migration time point and at the indicated time points. Image J software was used to quantify the area of wounds (open area).

MTT Assay
Cell viability was tested by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Cells were seeded on 96-well cell culture plates with or without inhibitor at indicated concentrations in 200 µL of complete growth medium. After the indicated time, 20 µL of 5 mg/mL MTT solution was added to each well. Four hours later, 150 mL DMSO was added to each well and the cells were incubated for a further 15-30 min. Absorbance of the solubilized MTT was then measured using the SpectraMax i3 at 490 nm.

Molecular Docking
Molecular docking was performed using MOE (Molecular Operating Environment) with AMBER10: EHT forcefield. The crystal structure of MNK1 was selected and downloaded from the Protein DataBank (PDB, http://www.rcsb.org, accessed on 12 July 2020), and was used for docking. The induced-fit docking approach was applied with consideration of the side-chain flexibility of residues at the binding site. The ligand binding site was defined using the bound ligands in the crystal structures. The best scored conformation with minimum binding energy from the 30 docking conformations of the ligands was selected for analysis

General Methods for Synthetic Chemistry
All starting materials and solvents were obtained from commercial sources and used without further purification. All actions were carried out with continuous magnetic stirring in common glassware and heating of reactions was performed with an IKA®heating block. Thin-layer chromatography (TLC) was performed on precoated silica-gel 60 F254 plates (E. Merck, Shanghai, China). Column chromatography was performed on silica gel (200-300 mesh, Qingdao Marine Chemical Company, Qingdao, China). 1 H NMR and 13 C NMR spectra were obtained on a Bruker 400 NMR spectrometer ( 1 H at 400 MHz and 13 C at 100 MHz) with tetramethylsilane (Me 4 Si) as the internal standard. Chemical shifts are reported as δ values. Mass spectra were recorded on a Q-TOF Global mass spectrometer.
All the compounds were synthesized according to previous report [35]. The data of 1 H and 13 C NMR and the spectra were shown in Supplementary Materials.

General Synthesis Method for Compounds 2A and 9A-9E
Compounds 1A or 8A-8E (1 eq) and DIPEA (3 eq) were mixed in DCM, and MOMCl (3 eq) was dropwise added to the solution at 0 • C. The mixture was stirred for 8-10 h at room temperature. After the reaction was judged complete by TLC (PE:EA = 20:1), the mixture was quenched by saturated NaHCO 3 and extracted with DCM three times. The combined organic layers were washed with water and brine, dried over anhydrous MgSO 4 , filtered, and concentrated in vacuo. The crude product was purified by chromatography (PE:EA = 20:1) to yield compounds 2A and 9A-9E.

General Synthesis Method for 3A-3F and 10A-10E
A mixture of 2A-2F or 9A-9E (1 eq), nitromethane (3 eq) and Et 3 N (1 eq) in DMSO was stirred at room temperature for 12 h. The mixture was poured into a saturated aqueous NH 4 Cl solution and adjusted to pH 7 by adding hydrochloric acid. The mixture was extracted with DCM three times. The combined organic layers were washed with water and brine, dried over anhydrous MgSO 4 , filtered, and concentrated in vacuo. The crude product was purified by chromatography (PE:EA = 10:1) to yield compounds 3A-3F and 10A-10E.

General Synthesis Method for 4A-4F and 11A-11E
A mixture of 3A-3F or 10A-10E and moderate 10% Pd/C in methanol under a hydrogen balloon was stirred at room temperature overnight. The reaction mixture was filtered and washed with methanol. The combined filtrate was concentrated in vacuo and the crude product was used for the next step without further purification.

General Synthesis Method for 5A-5V and 12A-12G
The related carboxylic compound (1 eq) was mixed with EDCI (2 eq) and HOBt (2 eq) in DCM at room temperature for 8 h. Then, 4A-4F or 11A-11E (1 eq) was added to the mixture for another 3 h. The mixture was washed with water and extracted with DCM three times. The combined organic layers were washed with water and brine, dried over anhydrous MgSO 4 , filtered, and concentrated in vacuo. The crude product was purified by chromatography (PE:EA = 1:1) to yield compounds 5A-5V and 12A-12G.

General Synthesis Method for 6A-6V and 13A-13G
To a solution of 5A-5V or 12A-12G (1 eq) in dry DCM, Dess-Martin periodinane (1.11 eq) was added. The mixture was stirred at room temperature until the reactant disappeared and quenched by adding solution of Na 2 S 2 O 3 and saturated aqueous NaHCO 3 . The organic layer was separated and the aqueous phase was extracted with DCM three times. The combined organic layers were washed with water and brine, dried over anhydrous MgSO 4 , filtered, and concentrated in vacuo. The crude product was purified by chromatography (PE:EA = 1:1) to yield compounds 6A-6V and 13A-13G. 4.6.6. General Synthesis Method for 7A-7C, 10, 12-29 and 14A-14G The mixture of 6A-6V or 14A-14G (1 eq), thiphenylphosphine (3 eq), hexachloroethane (6 eq) and triethylamine (6 eq) in DCM was stirred for 12 h at room temperature. The mixture was washed by water and extracted by DCM three times. The combined organic layers were washed with water and brine, dried over anhydrous MgSO 4 , filtered, and concentrated in vacuo. The crude product was purified by chromatography (PE:EA=2:1) to yield compounds 7A-7C, 10, 12-29 and 14A-14G.

General Synthesis Method for 1-9 and 11
Compounds 7A-7C, 10, 12-29 or 14A-14G were dissolved in methanol and HCl (1 mL) was added dropwise. The mixture was stirred for 1 h at 50 • C. After the reaction was judged complete by TLC (PE:EA = 1:1), the mixture was concentrated in vacuo and extracted with DCM three times. The combined organic layers were washed with water and brine, dried over anhydrous MgSO 4 , filtered, and concentrated in vacuo. The crude product was purified by chromatography (PE:EA = 1:1) to yield compounds 1-9 and 11.

Statistics
All in vitro experiments were performed in triplicate, and quantitative data are shown as the average of all biological replicates. All statistical analyses were performed using GraphPad Prism 7 (Version 7.0, GraphPad Software, San Diego, CA, USA). For simple comparisons, an unpaired two-tailed Student's t test was used.

Conclusions
In conclusion, we obtained six MNK inhibitors by virtual high-throughput screening and evaluated their effect on phosphorylation of eIF4E. Among these inhibitors, compound 5 with the skeleton of oxazole and pyrrole derived from phorbazole C showed the best inhibitory activity against MNKs. Combining the interaction mode between compounds and MNK1, we showed that three hydrogen bonds (one hydrogen bond with Asp191, two hydrogen bonds with Leu127) were important for the inhibition of MNKs. In MTT assays, all compounds showed weak inhibition to cell proliferation in these cancer cell lines, and only compound 5 exerted better cytotoxic activity with an IC 50 of 6.38 µM in H446 cells. Additionally, our inhibitor (compound 5) inhibited the migration of H446 cells significantly by impairing the MNK/eIF4E axis. Further work should be carried out to exploit its pharmacology activity and to design more novel inhibitors based on this skeleton.