Design, Synthesis of Novel Tetrandrine-14-l-Amino Acid and Tetrandrine-14-l-Amino Acid-Urea Derivatives as Potential Anti-Cancer Agents.

Tetrandrine, a dibenzyltetrahydroisoquinoline alkaloid isolated from the root of the traditional Chinese medicinal plant Stephania tetrandra S. Moore, a member of the Menispermaceae, showed anti-cancer activity by inhibiting cell proliferation, preventing cell cycle progress and induction of cell death and autophagy. In this study, twelve tetrandrine-l-amino acid derivatives and twelve tetrandrine-14-l-amino acid-urea derivatives were designed and synthesized, using C14-aminotetrandrine as raw material. Then the preliminary in vitro anti-cancer activities of these derivatives against human breast cancer cell line MDA-MB-231, human leukemia cell lines HEL and K562 were evaluated. The in vitro cytotoxicity results showed that these derivatives exhibited potent inhibitory effects on cancer cell growth, and the primary structure-activity relationships were evaluated. Notably, compound 3f exhibited satisfactory anticancer activity against all three cancer cell lines, especially the HEL cell line, with the IC50 value of 0.23 µM. Further research showed that 3f could induce G1/S cycle arrest and apoptosis in a dose- and time- dependent manner on the leukemia cell line HEL. The results suggested that 3f may be used as a potential anti-cancer agent for human leukemia.


Introduction
Cancer is one of the most serious disease threats to human health worldwide. Based on the report of the International Agency for Research on Cancer (IARC), it was estimated that there were 18.1 million new cancer cases and 9.6 million cancer deaths in 2018 [1]. Cancer is the first or second leading cause of death for people under 70 years old across 91 countries at the global level [2]. Chemotherapy has one of the most important ways to fight back against cancer since the 1940s when nitrogen mustard and antifolates were introduced to treat non-Hodgkin's lymphoma and pediatric acute leukemia [3][4][5]. More than 200 chemotherapeutic drugs have been approved by the FDA for treating cancers, and 75% of them are derived from natural products [6]. Over the past decades, natural products isolated from microorganisms and plants such as doxorubicin, mitomycin C, camptothecin, vincristine, taxol and podophyllotoxin as well as their structurally modified derivatives have been used as approved chemotherapeutic drugs [7][8][9][10].
Tetrandrine (Figure 1.), a bisbenzylisoquinoline (BBI) alkaloid isolated from the dried roots o e traditional Chinese medicinal herb Stephania tetrandra S. Moore [11], has been used as tiphlogistic, antalgic, calcium channel antagonistic, anti-radical and anticancer agent [12-14 cent research indicated that the anticancer mechanism of tetrandrine was multifariou trandrine is used as a potential CDKs inhibitor that directly inhibits CDK4, CDK2-CycE to arre e cell cycle in the G1/S phase [15][16][17], and then the effects of tetrandrine on controlling th ncer-associated gene (GAGE) expression are able to activate the apoptosis and autophag thway in cancer cells [18][19][20]. Aside from the aforesaid anticancer effects, tetrandrine increase e sensibility to other chemotherapeutic drugs and reverses the MDR [21] by regulating AB nsporter activity and reversal of P-g expression [22] and inhibiting the functions of P-gp [23]. As a potential anticancer agent with multiple mechanisms of action, the structural modificatio tetrandrine is an attractive subject for many research groups. Since the 21st century, structur odifications have mainly focused on introducing halogens and alkyl groups at the C5 and C sitions of tetrandrine [24][25][26], or quaternary ammonium salts at the N2 and N1 positions [27,28 cently, our group prepared a serious of C14-amino substituted tetrandrine derivatives whic hibited satisfactory inhibitory effects on human hepatocellular carcinoma (HCC), huma kemia (HEL and K562), human breast carcinoma (MDA-MD-231), human PCa (PC3), and huma elanoma (WM9) cell lines [29][30][31]. Even though these derivatives are reported as potenti ticancer agents, their poor water solubility and low bioavailability limits their application fo veloping lead anticancer compounds [32,33].
Amino acid functional groups often used for development of antiviral, antiparasiti tibacterial and anticancer drugs [34,35], in order to improve the oral absorption, sensitivit ysiochemical property and pharmacology of drugs [36]. Further studies showed that certai ncer cells were rich in oligopeptide transporters on their cytomembrane [37,38], so the amino aci gment was promising for the improvement of the selectivity of anticancer drugs [39], such a xuridine and brivanib ( Figure 2) [40,41]. In addition to amino acid fragments, the aryl ure oiety was also proved to be good fragment for anticancer agents [42]. Based on this backgroun e have now designed and synthesized a series of tetrandrine derivatives with amino acid an ea groups at the C14-position and evaluated their in vitro anticancer bioactivity. Primary SAR an As a potential anticancer agent with multiple mechanisms of action, the structural modification of tetrandrine is an attractive subject for many research groups. Since the 21st century, structural modifications have mainly focused on introducing halogens and alkyl groups at the C 5 and C 14 positions of tetrandrine [24][25][26], or quaternary ammonium salts at the N 2 and N 1 positions [27,28]. Recently, our group prepared a serious of C 14 -amino substituted tetrandrine derivatives which exhibited satisfactory inhibitory effects on human hepatocellular carcinoma (HCC), human leukemia (HEL and K562), human breast carcinoma (MDA-MD-231), human PCa (PC3), and human melanoma (WM9) cell lines [29][30][31]. Even though these derivatives are reported as potential anticancer agents, their poor water solubility and low bioavailability limits their application for developing lead anticancer compounds [32,33].
Amino acid functional groups often used for development of antiviral, antiparasitic, antibacterial and anticancer drugs [34,35], in order to improve the oral absorption, sensitivity, physiochemical property and pharmacology of drugs [36]. Further studies showed that certain cancer cells were rich in oligopeptide transporters on their cytomembrane [37,38], so the amino acid fragment was promising for the improvement of the selectivity of anticancer drugs [39], such as floxuridine and brivanib ( Figure 2) [40,41]. In addition to amino acid fragments, the aryl urea moiety was also proved to be good fragment for anticancer agents [42]. Based on this background, we have now designed and synthesized a series of tetrandrine derivatives with amino acid and urea groups at the C 14 -position and evaluated their in vitro anticancer bioactivity. Primary SAR and mechanistic studies were also performed in this study.

Chemistry
The synthetic route of tetrandrine derivatives 1a-3k is shown in Scheme 1. The mixture of concentrated nitric acid and acetic anhydride at low temperature was used as nitration reagent to obtain C14-nitro-tetrandrine selectively. The nitrotetrandrine could be restored to amino-substituted tetrandrine by using hydrazine hydrate as reducing agent in a methanol reaction medium containing palladium on carbon [43]. The C14-amino-tetrandrine was then reacted with Boc-L-amino acids in the presence of EDCI and HOBT to give tetrandrine-L-amino acid derivatives 1 in good yield. The tert-butyl carbonate groups of 1a and 1b were removed in a mixed solvent of CH2Cl2 and TFA at room temperature to obtain compounds 1k and 1l, which were then reacted with isocyanate to give tetrandrine-L-amino acid-urea derivatives 2a-3k in satisfactory yield.

Chemistry
The synthetic route of tetrandrine derivatives 1a-3k is shown in Scheme 1. The mixture of concentrated nitric acid and acetic anhydride at low temperature was used as nitration reagent to obtain C14-nitro-tetrandrine selectively. The nitrotetrandrine could be restored to amino-substituted tetrandrine by using hydrazine hydrate as reducing agent in a methanol reaction medium containing palladium on carbon [43]. The C 14 -amino-tetrandrine was then reacted with Boc-l-amino acids in the presence of EDCI and HOBT to give tetrandrine-l-amino acid derivatives 1 in good yield. The tert-butyl carbonate groups of 1a and 1b were removed in a mixed solvent of CH 2 Cl 2 and TFA at room temperature to obtain compounds 1k and 1l, which were then reacted with isocyanate to give tetrandrine-l-amino acid-urea derivatives 2a-3k in satisfactory yield.

Chemistry
The synthetic route of tetrandrine derivatives 1a-3k is shown in Scheme 1. The mixture of concentrated nitric acid and acetic anhydride at low temperature was used as nitration reagent to obtain C14-nitro-tetrandrine selectively. The nitrotetrandrine could be restored to amino-substituted tetrandrine by using hydrazine hydrate as reducing agent in a methanol reaction medium containing palladium on carbon [43]. The C14-amino-tetrandrine was then reacted with Boc-L-amino acids in the presence of EDCI and HOBT to give tetrandrine-L-amino acid derivatives 1 in good yield. The tert-butyl carbonate groups of 1a and 1b were removed in a mixed solvent of CH2Cl2 and TFA at room temperature to obtain compounds 1k and 1l, which were then reacted with isocyanate to give tetrandrine-L-amino acid-urea derivatives 2a-3k in satisfactory yield.

In Vitro Cytotoxicity Assay
Twenty-seven tetrandrine derivatives were tested for their cytotoxicity against a human leukemia cell line (HEL), K562 and a breast cancer cell line (MDA-MB-231). The IC 50 values of the tetrandrine derivatives, positive control vinblastine, the parent compounds tetrandrine and fangchinoline for 48 h were determined by the MTT assay [44], as presented in Table 1.
Compared with vinblastine, tetrandrine and fangchinoline, most of the tetrandrine derivatives showed better in vitro anti-cancer activities on all the three human cancer cell lines and the IC 50 values were as follows: 0.230-13.856 µM for HEL, 0.392-15.025 µM for K562, 0.812-9.088 µM for MDA-MB-231, respectively. Among the derivatives, six of them (1c, 1i, 3f-3i) showed better inhibitory effects on HEL cell line with IC 50 values of 0.631, 0.821, 0.230, 0.261, 0.386 and 0.940 µM, respectively. The compound 3f showed the strongest cytotoxic activity, so it was chosen for further mechanistic studies. Compared with the cytotoxicity of the tetrandrine-l-amino acid derivatives on all three cell lines, tetrandrine-l-amino acid-urea derivatives showed better anti-cancer activities. For compounds 1a-1l, when the R 1 substituents are electron-withdrawing side chains (i.e., compounds 1i, 1j), these compounds showed worse in vitro anti-cancer activities than those compounds whose R 1 substituents contained electron-donating side chains (1a, 1c, 1e). Longer branched aliphatic side chain substituents at R 1 were able to improve the inhibitory effects of the compounds (1d, 1g, 1h), as these compounds showed better activities than compound 1b, whose R 1 substituent was a methyl. The anti-cancer activities of compounds 1a and 1k didn't display prominent differences on the three cancer cell lines and the same situation happened between compounds 1b and 1l, so it followed that the tert-butyl carbonate group on the L-amino acid substituent was not essential for anti-cancer activity.
Compounds 2a-3k showed better inhibitory effects on HEL and MDA-MB-231 cell lines than K562 cell line. The change of R 1 substituent in the tetrandrine-l-amino acid-urea derivatives could influence their inhibitory effects, on account of the different R 1 substituents, compounds 2a and 3a showed prominent differences in anti-cancer activity. Compound 2a, whose R 1 substituent was benzyl, showed better activities on HEL and K562 cell line with IC 50 values of 1.171 µM and 1.616 µM, which were 2-fold and 9-fold higher than compound 3a, whose R 1 substituent was a methyl. The probable cause of the different activities between compounds 2a and 3a was the electronic effect of the R 1 substituent. The electron accepting effect of the R 2 substituent could also affect the anti-cancer activities of tetrandrine-l-amino acid-urea derivatives. When the R 2 substituent was a phenyl with electron-withdrawing groups (-F, -CF 3 , -OCF 3 , -Cl) in the para-position, the products showed increased antiproliferative activities (3f-3i).

The Effect of Compound 3f on Cell Proliferation
Microscopic examination was used to evaluate morphological changes within HEL cells. Cell growth curves were observed by measuring the OD value at 12, 24, 48 and 72 h using the MTT method, where the OD value is proportional to the cell viability. Compared with the control group, the microscopy examination ( Figure 3A) showed that the number of HEL cells was significantly reduced and cells had obviously died and dispersed, with the appearance of apoptotic bodies. The cell growth curve ( Figure 3B) showed that compound 3f exerted inhibitory activity on the proliferation of the HEL cell line in a time and dose dependent manner ( Figure 3).

Structure-Activity Relationship Study
Based on the MTT results, a preliminary Structure-Activity Relationship (SAR study could be performed. The substitution of L-amino acid and L-amino acid-urea, which are supposed to introduce a pivotal pharmacophore at the C14-position of tetrandrine, could enhance the anti-cancer activities of the derivatives. Compared with the cytotoxicity of the tetrandrine-L-amino acid derivatives on all three cell lines, tetrandrine-L-amino acid-urea derivatives showed better anti-cancer activities. For compounds 1a-1l, when the R1 substituents are electron-withdrawing side chains (i.e., compounds 1i, 1j), these compounds showed worse in vitro anti-cancer activities than those compounds whose R1 substituents contained electron-donating side chains (1a, 1c, 1e). Longer branched aliphatic side chain substituents at R1 were able to improve the inhibitory effects of the compounds (1d, 1g, 1h), as these compounds showed better activities than compound 1b, whose R1 substituent was a methyl. The anti-cancer activities of compounds 1a and 1k didn't display prominent differences on the three cancer cell lines and the same situation happened between compounds 1b and 1l, so it followed that the tert-butyl carbonate group on the L-amino acid substituent was not essential for anti-cancer activity.
Compounds 2a-3k showed better inhibitory effects on HEL and MDA-MB-231 cell lines than K562 cell line. The change of R1 substituent in the tetrandrine-L-amino acid-urea derivatives could influence their inhibitory effects, on account of the different R1 substituents, compounds 2a and 3a showed prominent differences in anti-cancer activity. Compound 2a, whose R1 substituent was benzyl, showed better activities on HEL and K562 cell line with IC50 values of 1.171 μM and 1.616 μM, which were 2-fold and 9-fold higher than compound 3a, whose R1 substituent was a methyl. The probable cause of the different activities between compounds 2a and 3a was the electronic effect of the R1 substituent. The electron accepting effect of the R2 substituent could also affect the anti-cancer activities of tetrandrine-L-amino acid-urea derivatives. When the R2 substituent was a phenyl with electron-withdrawing groups (-F, -CF3, -OCF3, -Cl) in the para-position, the products showed increased antiproliferative activities (3f-3i).

The Effect of Compound 3f on Cell Proliferation
Microscopic examination was used to evaluate morphological changes within HEL cells. Cell growth curves were observed by measuring the OD value at 12, 24, 48 and 72 h using the MTT method, where the OD value is proportional to the cell viability. Compared with the control group, the microscopy examination ( Figure 3A) showed that the number of HEL cells was significantly reduced and cells had obviously died and dispersed, with the appearance of apoptotic bodies. The cell growth curve ( Figure 3B) showed that compound 3f exerted inhibitory activity on the proliferation of the HEL cell line in a time and dose dependent manner ( Figure 3).

Compound 3f Induced Cell Apoptosis on HEL Cell Line
Depending on the effects of 3f on cell cycle progression, it was shown that the treatment of compound 3f led to the cell cycle arrest of the HEL cell line in the G1/S phase ( Figure 4A). Because the 3f treatment led to cellular morphological transformation and cell death, the effects of compound 3f on cell apoptosis were tested as well. Flow cytometry analysis showed that 3f treatment significantly increased the proportion of early apoptotic cells from 0.29% to 8.13%, 12.91% and 31.84%, and the proportion of late apoptotic cells was also increased from 0.09% to 1.62%, 5.98% and 15.63% after 3f treatment ( Figure 4B) in a dose dependent manner. From these results, it could be suggested that compound 3f might induce cancer cell apoptosis in a dose dependent manner.

Compound 3f Induced Cell Apoptosis on HEL Cell Line
Depending on the effects of 3f on cell cycle progression, it was shown that the treatment of compound 3f led to the cell cycle arrest of the HEL cell line in the G1/S phase ( Figure. 4A). Because the 3f treatment led to cellular morphological transformation and cell death, the effects of compound 3f on cell apoptosis were tested as well. Flow cytometry analysis showed that 3f treatment significantly increased the proportion of early apoptotic cells from 0.29% to 8.13%, 12.91% and 31.84%, and the proportion of late apoptotic cells was also increased from 0.09% to 1.62%, 5.98% and 15.63% after 3f treatment ( Figure. 4B) in a dose dependent manner. From these results, it could be suggested that compound 3f might induce cancer cell apoptosis in a dose dependent manner.

Instruments and Materials
Tetrandrine was obtained with purity ≥ 98%. The reagents and solvents were purchased from Adamas (Shanghai, China), J&K Chemical (Chengdu, China), Energy Chemical (Shanghai, China) and other local commercial dealers. All the reagents and solvents were commercially analytical or guaranteed purity products and used without further purification. Column chromatography was performed on silica gel (Qingdao Haiyang Chemical, Qingdao, China 200-300 mesh) using the indicated eluents. Thin-layer (0.25 mm, GF254) chromatography was carried out on silica gel plates (Qingdao Haiyang Chemical, Qingdao, China). 1 H-NMR spectra were recorded on 600 MHz (Bruker, Boston, MA, USA) and 400 MHz (Varian, Palo Alto, CA, USA) spectrometers in appropriate solvents using TMS as internal standard or the solvent signals as secondary standards and the chemical shifts are shown in δ scales. Multiplicities of NMR signals are designated as s (singlet), d (doublet), t (triplet), br (broad), and m (multiplet, for unresolved lines). 13 C-NMR spectra were recorded at 150 and 100 MHz. High-resolution mass spectra were obtained by using an ESI-QTOF mass spectrometer (Bruker, Beijing, China). All the NMR spectra can be found in Supplementary Materials (Figures S1-S52). Melting points (uncorrected) were determined on a WRX-4 micro melting point apparatus (Tansoole, Shanghai, China).  To a mixture of Tet-NO 2 (400.0 mg, 0.60 mmol) and palladium on carbon (5%, 40 mg) were added analytical methanol (20 mL) and hydrazine hydrate (85%, 0.18 mL, 4.80 mmol). The mixture was stirred at 65 • C for about 4 h before it was filtered by celite under reduced pressure. The filter was quenched with saturated sodium chloride solution, extracted with DCM (5 × 20 mL), dried over anhydrous sodium sulfate and filtered. The solvent was removed under reduced pressure. The crude product was recrystallized from cyclohexane and acetone (2/7, v/v) to give Tet-NH 2

General Procedure for the Preparation of Compounds 1a-1k
To a mixture of Tet-NH 2 (100 mg, 0.16 mmol), HOBT (8.47 mg, 0.63 mmol), EDCI (27.3 mg, 0.17 mmol) and Boc-l-amino acid (0.17 mmol, 1.1 eq) was added DCM (2.0 mL) under the protection of argon atmosphere, and stirred at room temperature for 1.5 to 3 h. The reaction mixture was quenched with saturated aqueous solution of sodium bicarbonate, extracted with DCM (3 × 10 mL), dried over anhydrous sodium sulfate and filtered. The solvent was removed under reduced pressure, and the residue was purified by silica gel chromatography eluated with DCM/MeOH (40/1 v/v, 0.5% TEA) to afford compounds 1a-1k.

General Procedure for the Preparation of Compounds 1k and 1l
Trifluoroacetic acid (0.1 mL) was slowly added dropwise to a solution of 1k (160 mg, 0.16 mmol) or 1l (130 mg, 0.16 mmol) in DCM (2 mL) at 0 • C. After 10 min, the reaction mixture was warmed up to room temperature, and stirred for 0.5 to 1.5 h before the reaction finished. The reaction mixture was quenched with saturated aqueous solution of sodium bicarbonate, extracted with DCM (3 × 10 mL), dried over anhydrous sodium sulfate and filtered. The solvent was removed under reduced pressure. The residue was purified by silica gel chromatography from DCM/MeOH (30/1 v/v, 0.5 % TEA) to afford the pure compounds 1k and 1l.

Cell Lines and Cell Culture
Human leukemic cell lines (HEL and K562) and breast cell line MDA-MB-231 were obtained from the University of Toronto (Toronto, ON, Canada). Cells cultured in RPMI (HEL and K562) or DMEM (MDA-MB-231) medium (high glucose) supplemented with 5% fetal bovine serum FBS (HyClone, GE Healthcare, Sydney, Australia) and maintained in a humidified incubator of 5% CO 2 at 37 • C. When the growing cells reached approximately 70-90% confluence, they were treated with 3f.

In Vitro Cytotoxicity Assay
The cells were cultured in 96-wells plates as density of 1 × 10 4 /well. The plates were incubated for 12 h to allow cell to adapt growing circumstance before the test compounds were added. After the adding of compounds in different doses, the cells were incubated for another two days. Then, each well was added with 20 µL diphenyltetrazolium bromide (MTT) and incubated for 4 h, medium removed and 200 µL of dimethyl sulfoxide (DMSO) was added. The IC 50 was detected by measuring the absorbance at 490 nm on a plate reader (BioTek, Winooski, VT, USA). All experiments were in triplicates and repeated at least three times.

Cell Growth Curve Assay
The compound 3f was prepared to original solution (20 µM) by DMSO and stored at −20 • C. The human leukemic cell line HEL was cultured in 96-wells plates as density of 1 × 10 4 /well. The plates were incubated for 12 h to allow cells to adapt growing circumstance. Cells then treated with 3f for 12 h, 24 h, 48 h and 72 h. The cell viability was measured by the MTT method.

Apoptosis Analysis by Annexin V and Propidium Iodide STAINING
HEL cells (3 × 10 5 /well) were cultured in 6 well-plates and treated with 3f or DMSO as a vehicle control for 24 h. The treated cells were gathered and washed with cold PBS for three times, then redistributed in binding buffer and stained with annexin V and PI, according to manufacturer instruction (BD Biosciences, Franklin Lakes, NJ, USA). Apoptotic cells were analyzed by flow cytometer (ACEA Biosciences Inc, San Diego, CA, USA).

Cell Cycle Analysis by Flow Cytometry
HEL cells (3 × 10 5 /well) were cultured in 6 well-plates and treated with 3f or DMSO. The treated cells were collected and washed with cold PBS, then dealt with iced 70% ethanol and stored at 4 • C overnight. After that, the cells were centrifuged and washed with PBS for three times, then redistributed in PBS (0.5 mL) containing 100 µg/mL RNase and 50 µg/mL PI. After it was let sit for 1 h in the dark at 37 • C, the cellular DNA content was analyzed by flow cytometry.

Statistical Analysis
The experimental data for all in vitro anticancer experiments were repeated in triplicates at least in three independent times. The t-test was used to determine statistical differences between treated groups and controls, and P < 0.05** was considered statistically significant. The values were presented as mean ± SD of the number of experiments. The significance level was calculated using one-way analysis of variance to assess the differences between experimental groups.

Conclusions
In conclusion, twenty-four tetrandrine derivatives were designed and synthesized. All the derivatives were obtained efficiently under mild reaction conditions. The anti-cancer activity tests of these derivatives against the HEL, K562 and MDA-MB-231 cell lines showed that they exhibited better inhibitory effects than the original compound tetrandrine and the positive control vinblastine. Among these derivatives, compounds 3f and 3g showed the strongest cytotoxic effect against the HEL cell line, with IC 50 values of 0.23 µM and 0.26 µM, which were 85-fold and 24-fold lower than those of tetrandrine, and 65-fold and 36-fold lower than those of vinblastine. Meanwhile, the preliminary mechanistic study results exhibited that compound 3f could induce cell cycle arrest in the G1/S phase of the HEL cell line. 3f could also induced HEL cell death through apoptosis. The results thus showed that 3f could be a potential agent for the treatment of leukemia, but further mechanistic and toxicologic researches should be performed to confirm this.