Novel Perbutyrylated Glucose Derivatives of (–)-Epigallocatechin-3-Gallate Inhibit Cancer Cells Proliferation by Decreasing Phosphorylation of the EGFR: Synthesis, Cytotoxicity, and Molecular Docking

Lung cancer is one of the most commonly occurring cancer mortality worldwide. The epidermal growth factor receptor (EGFR) plays an important role in cellular functions and has become the new promising target. Natural products and their derivatives with various structures, unique biological activities, and specific selectivity have served as lead compounds for EGFR. D-glucose and EGCG were used as starting materials. A series of glucoside derivatives of EGCG (7–12) were synthesized and evaluated for their in vitro anticancer activity against five human cancer cell lines, including HL-60, SMMC-7721, A-549, MCF-7, and SW480. In addition, we investigated the structure-activity relationship and physicochemical property–activity relationship of EGCG derivatives. Compounds 11 and 12 showed better growth inhibition than others in four cancer cell lines (HL-60, SMMC-7721, A-549, and MCF), with IC50 values in the range of 22.90–37.87 μM. Compounds 11 and 12 decreased phosphorylation of EGFR and downstream signaling protein, which also have more hydrophobic interactions than EGCG by docking study. The most active compounds 11 and 12, both having perbutyrylated glucose residue, we found that perbutyrylation of the glucose residue leads to increased cytotoxic activity and suggested that their potential as anticancer agents for further development.


Introduction
Lung cancer is one of the most commonly occurring cancer mortality worldwide, accounting for 20% of all cancer-related deaths [1]. Approximately 75-85% of lung cancers are non-small-cell lung cancer (NSCLC), including lung adenocarcinoma, squamous cell carcinoma (SCC), and large cell carcinoma (LCC) histological subtypes, and are typically diagnosed at advanced stages [2,3]. Recently, the identification of key oncogenic-driver mutations has led to the development of molecular targeted therapies for patients with NSCLC [4]. The epidermal growth factor receptor (EGFR) plays an important role in cellular functions, and approximately 50% of lung cancer patients possess EGFR overactivated [5]. Thus, EGFR tyrosine kinase has become an attractive therapeutic target for NSCLC [6][7][8]. However, the use of EGCG often has limitations such as easy oxidation [34], low bioavailability [35], and easy hydrolyses by bacterial and possibly host esterases [36]; thus, attempts to use EGCG in the treatment of human neoplasia have been mostly unsuccessful. To obtain more potent analogs and overcome these problems, many EGCG derivatives have been synthesized, including methyl-protected EGCG [37], acetyl-protected EGCG, [38], and EGCG monoester derivatives [39], and most of them exhibited more potent activity than EGCG. Recently, the glycoconjugates of small molecule anticancer drugs have become an attractive strategy in order to improve drug efficacy and pharmacokinetics, in addition to reducing side effects [40][41][42]. In our previous study, we thus reported that the synthesized glucosylated EGCG derivatives exhibited enhanced cytotoxicity and were more stable [5,43].

Chemistry
The synthesis of the (-)-epigallocatechin-3-gallate glucoside derivatives 7-12 was performed according to the reaction pathways illustrated in Schemes 1 and 2. D-glucose and EGCG were used as starting materials. 2,3,4,6-Tetra-O-butyryl-α-D-glucopyranosyl bromide 6 was prepared using the method reported in the literature [45]. EGCG (4) was allowed to react with the above compound 6 in the presence of K 2 CO 3 in acetone at 55 • C for 12 h to obtain peracetyl EGCG glucoside derivatives 7 and 8 in 25-30% yields (Scheme 1). Then, compounds 7 and 8 were treated with KOH solution in CH 3 OH at 0 • C for 72 h to yield EGCG glucoside derivatives 9 and 10 in 52-55% yields [27]. Finally, the free glucose of compounds 9 and 10 was buturylated with butyric anhydride in pyridine solution at 0 • C for 12 h to yield perbutyryl EGCG glucoside derivatives 11 and 12 in 80-82% yields (Scheme 2). All the synthesized compounds were characterized by 1 H-NMR, 13 C-NMR, electrospray ionization mass spectrometry (ESI-MS), and high-resolution mass spectrometry (HRESI-MS). In the 1 H-NMR spectra, the formation of the D-glucose residue was confirmed by the resonance of C 1 -H/C 1 -H signal (δ 4.54-4.97 ppm) with the coupling constant of the anomeric proton of D-glucose residue (J 1 /J 2 = 7.6 Hz-7.8 Hz) (Table S1). In our early study, the coupling constant of the anomeric proton of the glucose residue (J 1−2 ) is typically <4.0 Hz for the α-linkage and >7.6 Hz for the β-linkage [46].
The chemical structures of compounds were further confirmed by 2D-NMR spectra data. Based on the HMBC data ( Figure 2), the strong coupling occurred between C 1 -H (δ 4.97 ppm) and C-4" (137.3 ppm) of the D ring from the HMBC of compound 7. Similarly, coupling occurred between C 1 -H (δ 4.54 ppm) and C-4" (δ 138.5 ppm) of the D ring and also between C 1 -H (δ 4.67 ppm) of the other glucosyl residue and C-4 (δ 137.9 ppm) of the B ring from the HMBC of compound 8. The chemical shift for the proton at C 4 -H exhibited coupling with C-2, C-3; C 2 -H, C 6 "-H occurred with C-2 and C 2 "-H, C 6 "-H occurred with C-11; and C 2 -H occurred with C-11 in both compounds 7 and 8. These results indicated that the β-peracetylated glucoside (H-1 -H→C-4") had attached to the EGCG scaffold in compound 7, and two β-peracetylated glucosides (H-1 -H→C-4 and H-1 -H→C-4") had attached in compound 8.

Compounds
28.4 ± 1.00 37.9 ± 0.12 32.1 ± 0. 10  As shown in Table 1, the majority of glucoside derivatives of EGCG display weak cytotoxicity against all five cancer cell lines, and most of them show higher potency than the control EGCG. Among the glucoside derivatives of EGCG, compounds 11 and 12 showed better growth inhibition than others in four cancer cell lines (HL-60, SMMC-7721, A-549, and MCF), with IC 50 values in the range of 22.90-37.87 µM. Interestingly, all glucoside derivatives of EGCG showed good cytotoxicity against A-549 cells except compound 10 (having IC 50 > 40 µM). Previously, we reported that the cytotoxic activity of glucosylated EGCG derivatives with a free glucose residue mostly shows weak activity against breast cancer (MCF and MDA-MB-231 cells) [27]. However, our findings indicate that EGCG derivatives with a perbutyrylated glucose residue (11 and 12) lead to increased activity rather than a free glucose residue (11 vs. 9 and 12 vs. 10).
Cancer chemotherapy is often associated with low/non-selectivity of cancer drugs, which attack cancer cells and normal cells, leading to serious side effects. EGCG derivatives 11 and 12 were tested for their growth inhibitory effects on normal human bronchial epithelial cells (BEAS-2B) ( Table 1). The IC 50 values of compounds 11 and 12 (all having IC 50 > 40 µM), indicated that glucoside derivatives of EGCG showed moderated selectivity toward cancer cells.

EGCG Derivatives Inhibited Phosphorylation of EGFR and Downstream Signaling Protein in A-549 Cells
To investigate whether the effects of EGCG derivatives 11 and 12 might involve EGFR signaling, we assessed the expression of several key regulators that function within the EGFR signaling pathway. The results showed that treatment with EGCG slightly exhibited P-EGFR, P-AKT, and P-Erk ( Figure 3A). Compounds 11 and 12 further exhibited the phosphorylation of EGFR and downstream signaling protein, while the total protein level of EGFR remained unchanged ( Figure 3B,C). As shown in Figure 3D,E, it can be observed that EGCG, 11 and 12 can reduce the expression of phosphorylated EGFR, ERK, and AKT proteins. In summary, EGFR signaling may play a significant role in the effects of EGCG derivatives 11 and 12 in A-549 cells.

Molecular Docking Analysis
To further investigate the potential binding between EGFR and the compounds, molecular docking was performed. The binding modes of compounds 4, 11, and 12 at the ATP-binding pocket of EGFR (PDB code: 2ITY) were shown in Figure 4. As shown in Figure 4, compounds 11 and 12 could be docked into the active site of EGFR, the butyryl group occupies the ATP-binding pocket. The detailed docking results were presented in Table S2. It was observed that the binding energies between EGFR and glucoside derivatives of EGCG 11 and 12 are weak, compared with EGCG. The docking energy for compounds 4, 11, and 12 were −5.18 kcal/mol, 0.47 kcal/mol, and 1.81 kcal/mol, respectively. The complex structures for compounds 11 and 12 were presented in Figure 5.
The key residues were labeled and the important molecular interaction including the hydrogen bonds (black dotted lines), hydrophobic interactions (red dotted lines), π-stacking (gray dotted lines), and salt bridges (yellow dotted lines) were summarized in Table 2.
As shown in Figure 5, the molecules in complex with EGFR were analyzed in Figure 5A,B. It was observed that there are three hydrophobic interactions between compound 11 and three residues of EGFR (PHE723, ALA743, LEU844), five hydrogen bonds between 11 and three residues of EGFR (LEU718, MET793, ASN842), and two salt bridges between 11 and two residues of EGFR (LYS745, ARG841). There are seven hydrophobic interactions between compound 12 and seven residues of EGFR (ALA722, PHE723, VAL726, LYS745, MET793, ARG841, and VAL876), three hydrogen bonds between 12 and three residues of EGFR (GLU758, ASP855, and VAL876), one π-stacking and a salt bridge between 12 and the residue of EGFR (LYS745 and ARG841, respectively). Similarly, EGCG was docked into the EGFR are shown in Figure S1, which result in ten hydrogen bonds, one hydrophobic interaction, one π-stacking and one salt bridge, respectively. The presence of more hydrophobic interactions in compounds 11 and 12 seems the key factor for their high activity.

Cytotoxicity Assay
The effects of the (-)-epigallocatechin-3-gallate glucoside derivatives on the survival of cancer cells were determined using the MTT assay. Five human cancer cell lines (HL-60, SMMC-7721, A-549, MCF-7, and SW480) were used in the cytotoxicity assay. All the cancer cells were seeded in 96-well plates and then each tumor cell line was exposed to the test compound at various concentrations in triplicate for 48 h. After the incubation, MTT (100 µg) was added to each well, and the incubation continued for 4 h at 37 • C. After removal of the culture medium, the produced MTT formazan crystals were dissolved with 150 µL DMSO and measured at 492 nm using a microplate reader. The percentage of inhibition was calculated as follows: inhibition ratio (IR, %) = (1 − OD(sample)/OD(control)) × 100%. The experiments were carried out in triplicate, and the IC 50 (the concentration of drug that inhibits cell growth by 50%) values were determined.

Western Blotting
A-549 cells were lysed in RIPA buffer containing PMSF (protease and phosphatase inhibitors) and quantified via BCA protein assay. Proteins separated on 8% SDS-PAGE electrophoresis and then blotted onto polyvinyl difluoride membranes. After the membranes were blocked with BSA for 1 h, the expression of various proteins was detected using primary and secondary antibodies conjugated with horseradish peroxidase. HRP was detected using the Prolight HRP Chemiluminescent Kit (Tiangen Biotech, Beijing, China) and FluorChem E Synthem (ProteinSimple, Santa Clara, CA, USA).

Docking Studies
The X-ray crystal structure of EGCG (PDB code: 2ITY) was retrieved from Protein Data Bank (www.rcsb.org (accessed on 17 February 2020)). Ligand docking was carried out by applying the Lamarckian Genetic Algorithm implemented in AutoDock 4.2. The modeled EGCG structure was employed as the receptor to docking with these small molecules. DiscoveryStudio 4.0 software was used for the preparation of ligand and receptor. Autodock Tools 1.5.6 was used for grid and docking according to the literature [48] The grid size was set to 60 Å, 60 Å, and 60 Å along the X-, Y-and Z-axis to recognize the binding site. The default 0.375 Å spacing was adopted for the grid box. The number of GA runs was set to 50, and the maximum number of evals (medium) was set to 5,000,000 on AutoGrid 4.2.6 and AutoDock 4.2.6. In the selected cluster, conformations with the lowest binding energy and RMSD (<2.0 Å) were finally chosen to analyze the receptor-ligand interaction. Other miscellaneous parameters were assigned to the default values obtained from the AutoDock 4.2.6.

Conclusions
In conclusion, we synthesized a series of glucoside derivatives of EGCG (7-12) and evaluated for their in vitro anticancer activity against five human cancer cell lines, including HL-60, SMMC-7721, A-549, MCF-7, and SW480. All these derivatives display different levels of anticancer activity, which can be affected by the nature of substituents on the glucose residue. Among them, compounds 11 and 12 showed better growth inhibition than others in four cancer cell lines (HL-60, SMMC-7721, A-549, and MCF), with IC 50 values in the range of 22.90-37.87 µM. EGCG derivatives 11 and 12 also display moderate selectivity toward cancer cells over normal human bronchial epithelial cells (BEAS-2B). The docking study indicated that the presence of more hydrophobic interactions in compounds 11 and 12. In addition, compounds 11 and 12 decreased phosphorylation of EGFR and downstream signaling protein. The most active compounds 11 and 12, both having perbutyrylated glucose residue, led us to conclude that perbutyrylation of the glucose residue causes increased cytotoxic activity, which is indicative of their potential as anticancer agents for further development.
Supplementary Materials: The following are available online, Table S1: The characteristic 1 H-NMR and 13 C-NMR of compounds 7-12, Table S2: The energies of the binding between EGFR and compounds 4, 11, and 12, Figure S1: The details of the binding between EGFR and EGCG (4)

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.