The Design and Synthesis of a New Series of 1,2,3-Triazole-Cored Structures Tethering Aryl Urea and Their Highly Selective Cytotoxicity toward HepG2

Target cancer drug therapy is an alternative treatment for advanced hepatocellular carcinoma (HCC) patients. However, the treatment using approved targeted drugs has encountered a number of limitations, including the poor pharmacological properties of drugs, therapy efficiency, adverse effects, and drug resistance. As a consequence, the discovery and development of anti-HCC drug structures are therefore still in high demand. Herein, we designed and synthesized a new series of 1,2,3-triazole-cored structures incorporating aryl urea as anti-HepG2 agents. Forty-nine analogs were prepared via nucleophilic addition and copper-catalyzed azide-alkyne cycloaddition (CuAAC) with excellent yields. Significantly, almost all triazole-cored analogs exhibited less cytotoxicity toward normal cells, human embryonal lung fibroblast cell MRC-5, compared to Sorafenib and Doxorubicin. Among them, 2m’ and 2e exhibited the highest selectivity indexes (SI = 14.7 and 12.2), which were ca. 4.4- and 3.7-fold superior to that of Sorafenib (SI = 3.30) and ca. 3.8- and 3.2-fold superior to that of Doxorubicin (SI = 3.83), respectively. Additionally, excellent inhibitory activity against hepatocellular carcinoma HepG2, comparable to Sorafenib, was still maintained. A cell-cycle analysis and apoptosis induction study suggested that 2m’ and 2e likely share a similar mechanism of action to Sorafenib. Furthermore, compounds 2m’ and 2e exhibit appropriate drug-likeness, analyzed by SwissADME. With their excellent anti-HepG2 activity, improved selectivity indexes, and appropriate druggability, the triazole-cored analogs 2m’ and 2e are suggested to be promising candidates for development as targeted cancer agents and drugs used in combination therapy for the treatment of HCC.


Introduction
Cancer is a major public health problem in every country, as it is the leading cause of death worldwide and has an increasing incidence every year. The International Agency for Research on Cancer (IARC) reported 19,292,789 new cases of cancers and 9,958,133 deaths globally in 2020. Lung, colorectal, and liver cancers are the highest-ranked causes of death, with mortality rates of 81%, 48%, and 92%, respectively [1]. Various methods, including surgery, transplantation, radiation therapy, chemotherapy, and targeted drug therapy, have been used for the treatment of cancer [2,3], depending on the types and stages of cancers, as well as patient readiness [4][5][6][7]. In general, early-stage cancers can be cured by curative resection, radiation therapy, and transplantation, while chemotherapy and targeted drug therapy are the main options for the treatment of developed and advanced-stage cancers [8][9][10]. Targeted cancer drug therapy is an approach for the treatment of advanced cancers, which suppresses cancer cell growth mainly by the selective inhibition of enzymes and receptors in several signal transduction pathways [11]. Thus, the targeted therapy is generally effective with fewer adverse effects than chemotherapy [12,13]. Nevertheless, approved targeted cancer drugs can cause unexpected side effects and possess inappropriate pharmacological properties [14,15]. In addition, cancers can develop drug resistance, limiting the responsiveness of treatment in some patients [16,17]. Therefore, the discovery and structural developments of new, appropriate targeted cancer drugs with high safety profiles and drug-likeness must be intensively carried out.
Sorafenib (Nexavar ® ) is the first targeted cancer agent approved by the Food and Drug Administration of the United States (US FDA) for the treatment of advanced hepatocellular carcinoma (HCC) and later approved for the treatment of renal cell carcinoma (RCC) and thyroid cancer [18,19]. It suppresses cancer growth due to its multi-kinase inhibitory properties toward various receptor tyrosine kinases (RTKs) associated with cell proliferation, migration, differentiation, and angiogenesis [20]. In particular, Sorafenib is able to inhibit B-Raf [21] and VEGFR-2 [20,[22][23][24][25][26], which are overexpressed in many cancers including HCC, RCC, thyroid carcinoma, non-small lung, breast, colorectal and ovarian cancers. However, cancer therapy using Sorafenib requires a high daily dose due to its poor bioavailability of 38-49% [19] and often causes unexpected adverse effects, e.g., cardiovascular reactions (hypertension), hand-foot skin reaction (HFSR), diarrhea, renal toxicity, fatigue, etc. [27]. Moreover, several recent studies reported drug resistance to Sorafenib treatment in cancer therapy [20,28].
Recently, our group also demonstrated that the triazole-containing analogs obviously had inhibitory activity against hepatocellular carcinoma cell line Huh7 with a high selectivity index (SI). Thus, the 1,2,3-triazole structural feature had the potential to enhance the compound's safety profile, while the anti-cancer activity could potentially be preserved [4]. Our previous work showed that replacing the aryl urea of Sorafenib with a 1,2,3triazole ring resulted in a significantly reduced toxicity of the compounds, while removing an important aryl urea moiety drastically lowered the cytotoxicity toward hepatocellular carcinoma cell lines (HepG2 and Huh7) [4]. Based on this information, both the triazole part and the aryl urea moiety are retained in order to take these advantageous properties to our newly designed structures. Thus, in this new series of triazole-containing analogs 2 ( Figure 3), compared to the Sorafenib structure, the phenoxy and picolinamide portion was replaced with a 1,2,3-triazole linked to a substituted phenyl moiety, while aryl urea was maintained to preserve the anti-cancer activities. Our designed molecules were expected to target kinases and/or other proteins and mechanisms relating to cancer cell growth. We have hypothesized that the substitution of the core phenoxy ring with the heterocycle could provide interaction with targets through π − −π stacking and additional hydrogen bonding to the nitrogen atoms. Although the picolinamide part in the hinge region is highly conserved for the inhibition of kinases [55], it could tolerate substitution with heteroaromatics [7,37,38,41] or aromatics substituted with O-and N-containing functionalities, which provide important interactions, as picolinamide does in the hinge region, and they exhibit inhibitory activities toward kinases [56][57][58][59]. For example, the replacement of picolinamide with sulfonylphenyl [56], methoxyphenyl [57], cyanophenyl [58], and chlorophenyl [59] in the hinge region resulted in maintaining inhibitory activities toward cancer cell lines including HepG2, and kinases including VEGFR-2 and B-Raf. Therefore, the replacement of picolinamide moiety with the substituted phenyl ring could retain the π − −π interaction. Moreover, O-and N-containing substituents on the phenyl ring-for example, OH, OMe, CN, COOH, NH 2 , NHAc, and CONHMe-together with interactions by 1,2,3-triazole, were expected to compensate for or mimic the interaction, which the picolinamide of Sorafenib does with targets including kinases. Other substituents, such as halogens, CF 3 , and various alkyl groups could be functionalized in the analogs to explore the opportunity to inhibit kinases and/or other proteins associated with cancer development. Expectedly, it could result in potent anti-cancer activities. The background summary of this work is illustrated in Scheme 1. Based on our rationale, we aim to develop new structures containing 1,2,3-triazole for efficient anti-cancer agents with a high safety profile, using Sorafenib as a lead compound. We herein report the synthesis of a new series of 1,2,3-triazole-cored analogs tethering aryl urea and a substituted aromatic ring. The cytotoxicity of the synthesized compounds toward five cancer cell lines, including hepatocellular carcinoma HepG2, human lung carcinoma cells A549, Thai human cholangiocarcinoma HuCCA-1, T-cell acute lymphoblastic leukemia MOLT-3, and acute promyelocytic leukemia HL-60, and their structure-activity relationships (SARs) are also described. To reflect the safety profile of the analogs, the cytotoxicity toward human embryonal lung fibroblast cells MRC-5 and the selectivity index (SI) were evaluated. In addition, the cell-cycle arrest profile and apoptosis induction assay of the active analogs were studied, compared to Sorafenib. Finally, the drug-likeness was also analyzed.
The precursors and the target analogs were characterized mainly by Nuclear Magnetic Resonance Spectroscopy (NMR) using 1 H, 13 C, 19 F, and DEPT135 techniques, together with additional 2D nmR (COSY, HMQC, and HMBC) techniques in some cases. The formation of urea moiety was confirmed by the presence of two singlet peaks at δ = 6.31 and 8.48 ppm in the 1 H nmR spectrum [61], while the generation of the azide intermediates 7a-7o' and 7u' was detected as a strong peak at approximately 2200-2000 cm −1 in infrared spectra (IR) [62]. After the coupling of alkyne 5 and azides 7, a characteristic proton peak of the 1,2,3-triazole proton at approximately δ = 8.50-9.50 ppm [63] was observed, indicating the successful construction of the target triazole derivatives, which were confirmed complimentarily by High-Resolution Mass Spectrometry (HRMS).

Cytotoxicity toward Cancer Cell Lines
The cytotoxicity of the synthesized analogs toward HepG2, A549, HuCCA-1, MOLT-3, and HL-60 was evaluated compared to two positive controls, Sorafenib and Doxorubicin, by MTT or XTT assay [64]. The results revealed that the inhibitory effect of the synthesized triazole analogs was significantly more prominent toward HepG2 than the other cell lines tested. The inhibitory activities of the analogs against HepG2 are presented in Table 1, and the cytotoxicity toward other tested cancer cell lines is reported separately in the supporting information (Table S1).

Structure-Activity Relationships (SARs)
The structure-activity relationships (SARs) suggested that the substituents on the phenyl ring connecting triazole had strong influences on anti-HepG2 activities. The analogs with a substituent capable of forming hydrogen bonds to the highly conserved hinge region of kinases, such as OH, OMe, CN, COOH, NH 2 , NHAc, and CONHAc, showed moderate to low anti-HepG2 activity at 25 µM. In contrast, the analogs with a hydrophobic group such as a methyl, ethyl, isopropyl, or tert-butyl group exhibited decreasing cell viability percentages in HepG2 cases, in accordance with the increase in the size of the substituent. Especially, 2i' (R = o-iPr) and 2m' (R = p-tBu) were capable of inhibiting HepG2 with similar IC 50 values to that of Sorafenib, indicating that a bulky hydrophobic group might be an important group for the inhibition of HepG2. In spite of 2i' (R = o-iPr) and 2m' (R = p-tBu), it was found that compounds 2e (R = o-Cl) and 2y (R = p-CF 3 ), which were substituted with an electron-withdrawing group, showed similar IC 50 values against HepG2 to Sorafenib. The effect of the substituted position was not explicitly observed, nevertheless, the oand p-substituted analogs tended to possess superior anti-HepG2 activities to m-substituted analogs, according to the structure and activity of the active analogs. Based on our results, it could be deduced that the analogs with a phenyl containing a functional group capable of forming a hydrogen bond, replacing picolinamide, might not be as suitable a mimic for picolinamide as we hypothesized.

Cytotoxicity toward MRC-5 Cells and Selectivity Index (SI)
Preliminarily, the safety property of the synthetic compounds was investigated by cytotoxicity assay against MRC-5 using MTT compared to Sorafenib and Doxorubicin. It was evident that almost all synthetic analogs showed IC 50 values of more than 50 µM, which possessed approximately at least 2.5-and 22.1-fold less cytotoxic activity than Sorafenib (IC 50 = 19.7 µM) and Doxorubicin (IC 50 = 2.26 µM), respectively. The triazole analogs tended to show superior safety properties to the current approved therapeutic drugs. Thus, it could be implied that the presence of 1,2,3-triazole moiety can be one important factor in the compounds' safety profile, which agrees with previous reports [4,45,65].
In addition to the promising safety property, some triazole-containing analogs exhibited an excellent selectivity index (SI) for HepG2, superior to Sorafenib and Doxorubicin. All active analogs, 2m' (R = p-tBu), 2e (R = o-Cl), 2i' (R = o-iPr), and 2y (R = p-CF 3 ), which possessed similar inhibitory activities to Sorafenib, exhibited SI values of 14.7, 12.2, 10.1, and 9.81, respectively, which were up to 4.4-fold and 3.8-fold superior to those of Sorafenib (SI = 3.30) and Doxorubicin (SI = 3.83), respectively. The derivatives possessing SI ≥ 3.00 were considered safe and highly cancer-selective [64,66]. Moreover, the potent analogs showed significantly higher SIs than Sorafenib and Doxorubicin, implying that the synthetic analogs might be safer and, at this stage, more suitable for targeted HCC drug therapy.
Evidently, the synthetic triazole-cored analogs 2m' (R = p-tBu) and 2e (R = o-Cl) were identified as active candidates toward HepG2 with the highest SIs. These compounds were further investigated for exploring the possible mechanisms of action underlying their cytotoxic effects by the analysis of cell-cycle arrest and apoptosis induction on HepG2 cells.

Cell-Cycle Analysis
To further confirm the cytotoxic activity of 2m' and 2e, we sought to investigate the effect of these compounds on the cell-cycle distribution of HepG2 cancer cells. HepG2 treated with Sorafenib resulted in a decrease in cells in the G0/G1 phase and an increase in cells in the S and G2/M phases compared to the control ( Figure 4A,B), suggesting that Sorafenib induced S and G2/M phase cell-cycle arrest. The effect observed was in line with a previous report by another group [67][68][69]. The selected compounds 2m' and 2e exhibited similar cell cycle profiles compared to Sorafenib ( Figure 4C,D), indicating that the capability to induce S and G2/M phase cell-cycle arrest was recapitulated in these compounds. Given the similarity in chemical structure and biological activity of the synthesized compounds 2m' and 2e to their parent molecule Sorafenib, it is possible that these compounds share a similar mechanism of action.

Detection of Apoptosis
The induction of the apoptotic cell death of Sorafenib, 2m', and 2e on HepG2 was quantified by Annexin V binding to phosphatidylserine (PS) on the outer cell surface using Muse's Annexin V & Dead Cell Assay Kit. Based on annexin V reactivity and the intensity of the 7-AAD fluorescence, cells can be classified into four categories: dead, live, early apoptosis, and late apoptosis. HepG2 cell lines were treated for 48 h with 2.5 µM, 5.0 µM, and 10 µM of Sorafenib, 2m', and 2e, and the effects on apoptosis are shown in Figures 5 and 6. For the percentage of annexin V-PE-positive cells, the gradual increase in late stage apoptotic cells were 3.75%, 3.85%, and 6.80% for 2m', 4.60%, 4.30%, and 6.30% for 2e, and 7.60, 6.35, and 13.65% for Sorafenib at 2.5 µM, 5.0 µM, and 10 µM, respectively, with a corresponding decrease in viable cells ( Figure 5). Additionally, there was an increase in the total apoptotic cell population (early and late apoptosis) associated with higher doses of all three compounds. The average percentage of apoptotic cells after treatments at 2.5 µM, 5.0 µM, and 10.0 µM were 24.55 ± 0.28%, 35.15 ± 1.41%, and 46.45 ± 0.78% for Sorafenib; 27.10 ± 1.06%, 31.05 ± 0.04%, and 35.75 ± 0.67% for 2m'; and 27.05 ± 1.41%, 30.85 ± 3.75%, and 43.20 ± 0.42% for 2e. The basal apoptosis level in untreated cells was 22.35 ± 0.07%. Notably, there was a significant increase in total apoptosis at 10 µM for these three compounds with the highest percentage of apoptosis for Sorafenib. These results consistently indicated that 2m', 2e, and especially Sorafenib exerted their cytotoxic effects through the induction of apoptosis, and compounds 2m' and 2e induced apoptosis in a dose-dependent manner similar to Sorafenib.  To observe the cell death of analogs and Sorafenib-treated hepatocellular carcinoma cells, the morphologies of HepG2 cells were compared to those of untreated control cells by using light microscopy. Notably, the cell morphology changes (shrink and smaller in size) were observed after treating HepG2 with Sorafenib at 5.0 µM and 10.0 µM, while analog 2e maintained the morphology of intact cells after treatment at 48 h and the changes were observed when HepG2 was treated with 2m' at 10.0 µM, as presented in Figure 7. These data suggest that the characteristics of cell apoptosis morphology changes, such as the shrinkage of the cells or the loss of cell volume, were observed in a dose-dependent manner after treatment with Sorafenib and 2m', which relate to the results of the cytotoxicity and apoptosis assays.

Physicochemical Properties and Lipinski's Rule of Five
In order to evaluate the druggability of the potent triazole-containing analogs 2m' and 2e compared to Sorafenib, the physicochemical properties and analysis by Lipinski's rule of five were conducted using the SwissADME website service [70]. All the calculated parameters are presented in Table 2. The results revealed that the synthetic compound 2m' exhibited inferior physicochemical properties to Sorafenib. Although 2m' had a similar molecular weight (MW), number of hydrogen bonds (nHBA and nHBD), and number of rotatable bonds (nRB) to Sorafenib, it exhibited inappropriate properties including higher lipophilicity (Clog P) and lower water solubility (Log S). Moreover, the violation of Lipinski's rule was detected in the case of 2m', as its Clog P (4.74) was greater than or equal to the reported requirement (4.15) [71]. On the other hand, 2e showed superior drug-likeness to 2m' and Sorafenib. Analog 2e possessed superior properties including lower molecular weight, more hydrophilicity, higher water solubility, and a smaller number of rotatable bonds, although 2e had less hydrogen bonding positions (nHBA and nHBD) than Sorafenib. Additionally, compound 2e possessed a topological polar surface area (TPSA) value of 71.84 Å 2 , which was slightly above the range attributed to most successful drugs (≤60-70 Å 2 ) [72]. This value was much lower than that of Sorafenib (92.35 Å 2 ), suggesting that 2e tended to exhibit greater cell membrane permeability than Sorafenib [73]. According to the physicochemical parameters, 2e exhibited appropriate druggability and conformed the Lipinski's guidelines, thus tending to be a promising drug candidate for a targeted liver cancer agent.
The signals were reported as follows: chemical shifts, multiplicity, and coupling constant. The multiplicities were given as s = singlet, d = doublet, t = triplet, q = quartet, dd = doublet of doublet, dt = doublet of triplet, td = triplet of doublet, tt = triplet of triplet, ddd = doublet of doublet of doublet, dddd = doublet of doublet of doublet of doublet, m = multiplet, br = broad. High-Resolution Mass Spectrometry (HRMS) was performed using an ESI and APCI ionization technique on a Bruker Daltonics MicroTOF spectrometer. Melting points were measured on a Büchi Melting Point B-545 apparatus.

General Procedure for the Preparation of Sorafenib Derivatives 2p'-2r'
Triazole-cored derivatives 2p'-2r' were prepared according to the procedure described previously [90]. A stirred solution of tin (II) chloride dihydrate (208 mg, 0.92 mmol, 4.00 eq) in conc. HCl (1.0 mL) was stirred at 0 • C for 5 min and then the nitrobenzene 2c', 2d', or 2e' (100 mg, 0.23 mmol, 1.00 eq) was added. The reaction mixture was stirred at 65 • C for 3 h. The resulting solution was then cooled to room temperature, diluted with water (30 mL), basified to pH 8 by using an aqueous sodium hydrogen carbonate solution, and extracted with ethyl acetate (3 × 50 mL). The combined organic phase was dried over anhydrous sodium sulfate and filtered. The solvent was removed under reduced pressure and the crude product was purified by column chromatography (silica gel, ethyl acetate in hexane) to obtain compound 2p', 2q', or 2r'.

Cell-Cycle Analysis
HepG2 cells were seeded into 6-well plates at a density of 2 × 10 5 cells/well and incubated at 37 • C in the presence of 5% CO 2 for 24 h to allow attachment. Afterward, cells were treated with complete media containing Sorafenib, compounds 2m' and 2e at 2 µM, or DMSO as vehicle control for 72 h. The concentration of DMSO was kept at 0.5% v/v for all conditions. Cells were washed with PBS before they were harvested; 1 × 10 5 Cells were collected and fixed with 70% ice-cold ethanol, and then washed twice with ice-cold PBS. Afterward, the cells were treated with 100 g/mL of DNase-free RNase A (Sigma-Aldrich, St. Louis, MO, USA) in PBS containing 0.1% v/v Triton-X 100 (Sigma-Aldrich, St. Louis, MO, USA) for 5 min at room temperature and then stained with 20 g/mL propidium iodide (Life Technologies, Carlsbad, CA, USA) in PBS containing 0.1% v/v Triton-X 100 for 15 min at room temperature while protected from light. Cell-cycle distribution was then analyzed with a flow cytometer (Attune NxT, Thermo Fischer Scientific, Waltham, MA, USA). Data were analyzed with Attune NxT Software (Thermo Fischer Scientific, Waltham, MA, USA). All experiments were performed in triplicate. Student's t-test was used for statistical analysis; p < 0.05 was considered statistically significant [96].

Detection of Apoptosis
The detection of apoptosis was performed by the Muse ® Annexin V & Dead Cell Kit according to the manufacturer's protocol (Millipore, Billerica, MA, USA). HepG2 cells at 5 × 10 5 cells/mL in a completed DMEM medium were seeded into 24-well plates and incubated at 37 • C in the presence of 5% CO 2 for 24 h. The medium was removed from the plates, followed by treatment with Sorafenib, 2m', and 2e at concentrations of 1.25 µM, 2.5 µM, 5.0 µM, and 10.0 µM for 48 h. After incubation, the cells were washed with 300 µL PBS/well and removed from plates by Trypsinization (300 µL of trypsin/well). An amount of 100 µL of HepG2 cells in suspension and 100 µL of Muse Annexin V & Dead Cell reagent were added to a 1.5 mL tube. The apoptosis was measured using the Muse cell analyzer and Muse analysis software (Millipore, Billerica, MA, USA). Cells were classified into four groups: live (Annexin V− and 7-AAD−), early apoptosis (Annexin V+ and 7-AAD−), late apoptosis (Annexin V+ and 7-AAD+), and dead or necrotic (Annexin V− and 7-AAD+). The apoptosis experiment was performed in duplicate.

Physicochemical Property Methodology
All drug-likeness properties and Lipinski's rule of five were obtained by using Swis-sADME website services [70].

Conclusions
A new series of 1,2,3-triazole-cored analogs, in which the core phenoxy ring and picolinamide ring of Sorafenib were replaced with 1,2,3-triazole linking a substituted phenyl ring, were synthesized successfully via nucleophilic addition and 1,3-dipolar cycloaddition and evaluated for their in vitro anti-cancer activity against five different cancer cell lines. The synthetic triazole-cored analogs exhibited inhibitory activities toward HepG2 dominantly over other cancer cell lines. Analogs 2m' (R = p-tBu) and 2e (R = o-Cl) exhibited similar anti-HepG2 properties to that of Sorafenib, but were less active than a chemotherapy drug, Doxorubicin. In addition, 2m' (R = p-tBu) and 2e (R = o-Cl) showed 4.4-and 3.7-fold superior SIs to that of Sorafenib and were 3.8-and 3.2-fold superior to that of Doxorubicin. Disappointingly, the analogs with a functional group capable of forming hydrogen bonds in the hinge region did not show potent anti-HepG2 activity, as expected. For cell-cycle analysis on HepG2, both compounds caused an increased number of cells in the S and G2/M phases, similar to those of Sorafenib, suggesting that they could share a similar mechanism of action. The induction of apoptotic cell death was observed in the treated HepG2 with 2m' and 2e in a dose-dependent manner at 48 h, similar to that of the Sorafenib case. The cytotoxic effect of the candidates might be due to the inhibition of kinases and/or other proteins involving apoptosis cell death and/or even affecting other mechanisms of action. Further investigations should be carried out.
Evidently, the replacement of the core phenoxy and picolinamide ring of Sorafenib with a 1,2,3-triazole ring linking an orthoand para-substituted phenyl ring with electronwithdrawing and bulky alkyl groups could still maintain anti-HepG2 activity. In addition, the presence of 1,2,3-triazole was proven to promote the compounds' SI, which agrees with a previous report [4]. Therefore, the 1,2,3-triazole linking a substituted benzene was an interesting structural feature for selective anti-HepG2 agents with a high safety profile. Furthermore, these two compounds (2m' and 2e) displayed good physicochemical profiles, especially compound 2e, which possessed several parameters superior to those of Sorafenib. Therefore, this study identified promising candidates for further development as targeted HCC drugs and drugs used in combination therapy with other anti-HCC drugs in the clinic.