Synthesis and Antiproliferative Evaluation of 2-Deoxy-N-glycosylbenzotriazoles/imidazoles

A series of 2-deoxy-2-iodo-α-d-mannopyranosylbenzotriazoles was synthesized using the benzyl, 4,6-benzylidene and acetyl protected D-glucal in the presence of N-iodosuccinimide (NIS). Subsequent removal of the iodine at the C-2 position using tributyltin hydride under free radical conditions afforded the 2-deoxy-α-d-glucopyranosylbenzotriazoles in moderate to high yields. This method was extended to the preparation of substituted 2-deoxy-β-d-glucopyranosylimidazoles as well. The stereoselectivity of the addition reaction and the effect of the protecting group and temperature on anomer distribution of the benzotriazole series were also investigated. The anticancer properties of the newly synthesized compounds were evaluated in a series of viability studies using HeLa (human cervical adenocarcinoma), human breast and lung cancer cell lines. The N-[3,4,6-tri-O-benzyl-2-deoxy-α-d-glucopyranosyl]-1H-benzotriazole and the N-[3,4,6-tri-O-acetyl-2-deoxy-α-d-glucopyranosyl]-2H-benzotriazole were found to be the most potent cancer cell inhibitors at 20 µM concentrations across all four cell lines.


Introduction
Benzotriazoles are a versatile class of heterocycles with a wide range of applications and usages including antibacterial [1], antiviral [2] and antiparasitic [3], among others [4]. Compounds containing the benzotriazole moiety have been reported to possess powerful effects in inhibiting cell proliferation and arresting cancer development. For example, the [2-(4,5-dihydro-1H-imidazol-2-yl)-1H-benzotriazole]-dichlorocopper (II) complex has been reported to have very potent superoxide dismutase (SOD) activity inhibiting and preventing cancer cell growth [5]. Benzotriazole-containing alkanoic acids have also been demonstrated to be agonists of peroxisome proliferator-activated receptors (PPARs) and, as such, are capable of inducing apoptosis in multiple cell lines and in vivo tumors [6]. Carbohydrate-based benzotriazoles have also been found to inhibit solid tumor growth in mice. This includes the acetyl protected glucose and N-acetylglucosamine benzotriazole derivatives [7]. Substituted rybofuranosyl benzotriazoles and benzoimidazoles have also been reported to be powerful protein kinase CK2 inhibitors [8]. While the mechanism of action of the ribose derivatives is well-documented [9], little is known about the mode of action of N-glycosyl benzotriazoles.
Interest in the generation of 2-deoxy analogs of compound I arose from the abundance of reports on the biological activity displayed by 2-deoxy-D-glucose. This compound has Interest in the generation of 2-deoxy analogs of compound I arose from the abun dance of reports on the biological activity displayed by 2-deoxy-D-glucose. This com pound has been shown to significantly slow down cancer metabolism and induce cel death [10,11], yet cell viability studies using 2-deoxy-D-glucopyranosylbenzotriazoles cannot be found in the current literature. Various methods exist in the literature for the chemical synthesis of 2-deoxy sugars. Early work in this area focused on the chemica synthesis of 2-deoxy-β-glycosides [12,13]. Glycals have been used extensively as building blocks in the preparation of 2-deoxy sugars via their hydration or hydro-alkoxylation cat alyzed by methanolic hydrogen halide [14][15][16]. Other methods include the use of methane sulfonic acid as catalyst [17], enzymes [16,18], alkoxymercuration of a glycal followed by reduction with sodium borohydride [19,20] and halohydration followed by dehydrogena tion [21][22][23]. More closely related to this study is the method developed by Kashyap and co-workers via the reaction of various protected D-glucal with a mixture of tetra-n-bu tylammonium iodide with sodium periodate to afford 2-deoxyglycosides [24].
We have previously reported the synthesis of the benzyl protected N-glycosyl ben zotriazole I (Figure 1) using glycosyl halides and silver triflate (AgOTf) as starting mate rials. Compound I effectively inhibited growth of HeLa (cervical adenocarcinoma) cells (86%, 100 µ M) [25]. The goal of this study was to chemically synthesize a series of analogs of compound I and evaluate their antiproliferative and cytotoxic effects using four differ ent established human cancer cell lines (A549, HeLa, HCC827 and MDA-MB-231). Our initial efforts went into the generation of the known N-glycosyl benzotriazoles 1-4 ( Figure  1) and the study of the effects of the protecting groups of the sugar on their viability as cancer drugs. Our laboratory also became interested in the chemical synthesis of various substituted 2-deoxy-D-glucopyranosylbenzotriazole/imidazoles 5-14 ( Figure 1) and their antiproliferative properties were studied as well.

Results and Discussion
Our work began with the synthesis of compounds 1-4 using an established procedure employing commercially available acetyl protected glycosyl halides 15 and 16 (Scheme 1) as starting materials [7]. Interest in the generation of 2-deoxy analogs of compound I arose from the abundance of reports on the biological activity displayed by 2-deoxy-D-glucose. This compound has been shown to significantly slow down cancer metabolism and induce cell death [10,11], yet cell viability studies using 2-deoxy-D-glucopyranosylbenzotriazoles cannot be found in the current literature. Various methods exist in the literature for the chemical synthesis of 2-deoxy sugars. Early work in this area focused on the chemical synthesis of 2-deoxy-β-glycosides [12,13]. Glycals have been used extensively as building blocks in the preparation of 2-deoxy sugars via their hydration or hydro-alkoxylation catalyzed by methanolic hydrogen halide [14][15][16]. Other methods include the use of methane sulfonic acid as catalyst [17], enzymes [16,18], alkoxymercuration of a glycal followed by reduction with sodium borohydride [19,20] and halohydration followed by dehydrogenation [21][22][23]. More closely related to this study is the method developed by Kashyap and co-workers via the reaction of various protected D-glucal with a mixture of tetra-n-butylammonium iodide with sodium periodate to afford 2-deoxyglycosides [24].
We have previously reported the synthesis of the benzyl protected N-glycosyl benzotriazole I (Figure 1) using glycosyl halides and silver triflate (AgOTf) as starting materials. Compound I effectively inhibited growth of HeLa (cervical adenocarcinoma) cells (86%, 100 µ M) [25]. The goal of this study was to chemically synthesize a series of analogs of compound I and evaluate their antiproliferative and cytotoxic effects using four different established human cancer cell lines (A549, HeLa, HCC827 and MDA-MB-231). Our initial efforts went into the generation of the known N-glycosyl benzotriazoles 1-4 ( Figure  1) and the study of the effects of the protecting groups of the sugar on their viability as cancer drugs. Our laboratory also became interested in the chemical synthesis of various substituted 2-deoxy-D-glucopyranosylbenzotriazole/imidazoles 5-14 ( Figure 1) and their antiproliferative properties were studied as well.

Results and Discussion
Our work began with the synthesis of compounds 1-4 using an established procedure employing commercially available acetyl protected glycosyl halides 15 and 16 (Scheme 1) as starting materials [7]. Compounds 1-4 were initially tested using a range of concentrations (data not shown) in a series of viability studies using HeLa cell line. None of these compounds were capable of inhibiting cell growth, even at a concentration of 100 µM. We hypothesized that perhaps a lack of lipophilicity, especially for analogs 2 and 4 when compared to compound I (Figure 1), may have played a role in the uptake and crossing of the cell membrane in our cell assay giving rise to the poor results observed. Interestingly, this pattern was observed throughout our study, where the benzyl protected sugar outperformed the acetyl, benzylidene and fully deprotected constructs in subsequent viability studies.
In the next phase of our study, we proceeded to synthesize our 2-deoxy sugar analog series starting with the benzyl protected compound 5 (Figure 1). Our laboratory has extensive experience preparing 2-deoxy-N-glycosides and has previously reported the synthesis of compound 5 [25]. Access to compounds 5-10 ( Figure 1) was achieved using 2-deoxy-2-iodo-α-mannopyranosylbenzotriazoles as precursors (Scheme 2). The benzyl protected D-glucal gave the α-mannose 1H-N-glycosylbenzotriazole as the single isomer 20 ( 3 J 1-2 = 4.0 Hz) [25], while a 13:1 (1H:2H-benzotriazole) ratio of the 2-deoxy-2-iodo-α-mannose isomer 21a and 21b was obtained for the acetyl protected sugar. This type of benzotriazole isomerization has been previously described in the literature [26]. In the case of the 4,6benzylidene protected sugar, the 2-deoxy-2-iodo-α-mannose 1H-N-glycosylbenzotriazole isomer 22a was obtained as the major product, while less than 10% of the 2-deoxy-2-iodoα-mannose 2H isomer 22b was isolated after column chromatography (Scheme 2). The addition reaction was repeated at 0 • C and room temperature and very similar product ratios were obtained for the acetyl and benzylidene protected sugar. The best yields were obtained when the reaction was refluxed for 2 h for the acetyl and 4,6-benzylidene protected D-glucal. Product ratios were determined by measuring the integration values of the anomeric protons of the crude mixture using 1 H-NMR acetone-d 6 .
Molecules 2021, 26, 3742 3 of 14 Compounds 1-4 were initially tested using a range of concentrations (data not shown) in a series of viability studies using HeLa cell line. None of these compounds were capable of inhibiting cell growth, even at a concentration of 100 µ M. We hypothesized that perhaps a lack of lipophilicity, especially for analogs 2 and 4 when compared to compound I (Figure 1), may have played a role in the uptake and crossing of the cell membrane in our cell assay giving rise to the poor results observed. Interestingly, this pattern was observed throughout our study, where the benzyl protected sugar outperformed the acetyl, benzylidene and fully deprotected constructs in subsequent viability studies.
In the next phase of our study, we proceeded to synthesize our 2-deoxy sugar analog series starting with the benzyl protected compound 5 ( Figure 1). Our laboratory has extensive experience preparing 2-deoxy-N-glycosides and has previously reported the synthesis of compound 5 [25]. Access to compounds 5-10 ( Figure 1) was achieved using 2deoxy-2-iodo-α-mannopyranosylbenzotriazoles as precursors (Scheme 2). The benzyl protected D-glucal gave the α-mannose 1H-N-glycosylbenzotriazole as the single isomer 20 ( 3 J1-2 = 4.0 Hz) [25], while a 13:1 (1H:2H-benzotriazole) ratio of the 2-deoxy-2-iodo-αmannose isomer 21a and 21b was obtained for the acetyl protected sugar. This type of benzotriazole isomerization has been previously described in the literature [26]. In the case of the 4,6-benzylidene protected sugar, the 2-deoxy-2-iodo-α-mannose 1H-N-glycosylbenzotriazole isomer 22a was obtained as the major product, while less than 10% of the 2-deoxy-2-iodo-α-mannose 2H isomer 22b was isolated after column chromatography (Scheme 2). The addition reaction was repeated at 0 °C and room temperature and very similar product ratios were obtained for the acetyl and benzylidene protected sugar. The best yields were obtained when the reaction was refluxed for 2 h for the acetyl and 4,6benzylidene protected D-glucal. Product ratios were determined by measuring the integration values of the anomeric protons of the crude mixture using 1 H-NMR acetone-d6.
It is our hypothesis that the addition reaction of the benzotriazole to the acetyl and 4,6-benzylidene protected D-glucal may follow a mechanism very similar to the well documented addition of amides to glycals using NIS as the halogen source (Scheme 3) [27][28][29]. We envision the more nucleophilic benzotriazole 23 reacting with NIS generating intermediate 25. The newly formed compound 25 delivers the iodine to the acetyl protected D-glucal 18. Although the formation of compound 27 defies the majority rule [30,31], such an intermediate has been proposed in the isolation of the α-mannose isomer when alcohols are employed as the nucleophile in the presence of NIS [32][33][34]. Nucleophilic attacks Scheme 2.
It is our hypothesis that the addition reaction of the benzotriazole to the acetyl and 4,6-benzylidene protected D-glucal may follow a mechanism very similar to the well documented addition of amides to glycals using NIS as the halogen source (Scheme 3) [27][28][29]. We envision the more nucleophilic benzotriazole 23 reacting with NIS generating intermediate 25. The newly formed compound 25 delivers the iodine to the acetyl protected D-glucal 18. Although the formation of compound 27 defies the majority rule [30,31], such an intermediate has been proposed in the isolation of the α-mannose isomer when alcohols are employed as the nucleophile in the presence of NIS [32][33][34]. Nucleophilic attacks by the benzotriazole anion from the bottom face of the three membered ring iodonium ion 27 give rise to the 2-deoxy-2-iodo-α-mannose isomers 21a and 21b.
Access to the benzyl protected 2-deoxy-β-D-glucopyranosylimidazoles 10-14 (Scheme 5) was achieved under free radical dehalogenation conditions using a procedure previously published by our laboratory [25]. We employed benzoimidazole and other commercially available substituted imidazoles. The addition reaction of the imidazole series resulted in the β-anomer 29-33 (Scheme 5) being the exclusive product in high to moderate yields. Such results can be explained based on the tendency of nitrogen glycosides to favor the equatorial and not axial position in virtue of the reverse anomeric effect [35]. Access to the benzyl protected 2-deoxy-β-D-glucopyranosylimidazoles 10-14 (Scheme 5) was achieved under free radical dehalogenation conditions using a procedure previously published by our laboratory [25]. We employed benzoimidazole and other commercially available substituted imidazoles. The addition reaction of the imidazole series resulted in the β-anomer 29-33 (Scheme 5) being the exclusive product in high to moderate yields. Such results can be explained based on the tendency of nitrogen glycosides to favor the equatorial and not axial position in virtue of the reverse anomeric effect [35]. The newly synthesized compounds 5-14 were initially tested for antiproliferative activity in a series of viability studies using HeLa cells. Those that demonstrated cytotoxicity are included in Figure 2. Cells were treated with the synthesized compounds at 10 µ M (grey bars) and 100 µ M (black bars) concentrations to determine which compounds exhibit cytotoxic effects before further evaluation.
Our results suggest that the benzyl protected 1H-2-deoxy-α-D-glucopyranosylbenzotriazole 5 was the most potent of all compounds and is capable of complete inhibition of cell viability at 100 µ M and by 92% (8% survival) at 10 µ M concentration. Our results also show a striking contrast in activity between the acetyl protected 1H (6a) and 2H-2deoxy-α-D-glucopyranosylbenzotriazole 6b. For compound 6b (0%, 100µ M) and (45%, 10 µ M) cell survival was observed, while for compound 6a (78%, 100 µ M) and (100%, 10 µ M), respectively. When comparing compound 5 and 6a (acetyl protected analogue), we can observe a significant drop off in activity even at concentrations as high as 100 µ M. As previously stated, we believe this may be attributed to the enhanced lipophilicity provided by the benzyl protecting group found in compound 5, perhaps making it more effective at crossing cell membranes. More importantly, our results show that the isomer 6b is as effective as compound 5 in killing our cancer cell lines at 100 µ M, even with the acetyl protecting group present. We also observed moderate activity for the 4,6-benzylidene protected compound 8 (54%, 100 µ M). Benzylation of the hydroxyl group at the C-3 position rendered compound 9 completely ineffective. From the glycoimidazole series, compound 10 showed the most promising results in terms of (8%, 100 µ M) survival rate. The remaining substituted imidazoles gave very poor results at our tested concentrations. It should be worth mentioning that all the fully deprotected compounds showed no activity at either concentration (results not shown).
Next, we set out to determine the cytotoxicity of compounds 5 and 6b using established lung epithelial cancer cell lines A549 and HCC827 as well as the established metastatic breast cancer cell lines MDA-MB-231 and compared them to HeLa cells (Figure 3). Compounds 5 and 6b proved to be excessively toxic at 100 µ M concentrations, resulting with a zero percent survival rate. A dose-response analysis was performed to determine the appropriate concentration of compound required to inhibit the growth of cancer cells by 50% (IC50) in HeLa cells before proceeding. The IC50 was determined to be 2.908 µ M for compound 5 and 9.940 µ M for compound 6b. DMSO was used as a control and the cells were treated similarly as in Figure 2 The newly synthesized compounds 5-14 were initially tested for antiproliferative activity in a series of viability studies using HeLa cells. Those that demonstrated cytotoxicity are included in Figure 2. Cells were treated with the synthesized compounds at 10 µM (grey bars) and 100 µM (black bars) concentrations to determine which compounds exhibit cytotoxic effects before further evaluation.
As previously stated, we believe this may be attributed to the enhanced lipophilicity provided by the benzyl protecting group found in compound 5, perhaps making it more effective at crossing cell membranes. More importantly, our results show that the isomer 6b is as effective as compound 5 in killing our cancer cell lines at 100 µM, even with the acetyl protecting group present. We also observed moderate activity for the 4,6-benzylidene protected compound 8 (54%, 100 µM). Benzylation of the hydroxyl group at the C-3 position rendered compound 9 completely ineffective. From the glycoimidazole series, compound 10 showed the most promising results in terms of (8%, 100 µM) survival rate. The remaining substituted imidazoles gave very poor results at our tested concentrations. It should be worth mentioning that all the fully deprotected compounds showed no activity at either concentration (results not shown).
Next, we set out to determine the cytotoxicity of compounds 5 and 6b using established lung epithelial cancer cell lines A549 and HCC827 as well as the established metastatic breast cancer cell lines MDA-MB-231 and compared them to HeLa cells (Figure 3). Compounds 5 and 6b proved to be excessively toxic at 100 µM concentrations, resulting with a zero percent survival rate. A dose-response analysis was performed to determine the appropriate concentration of compound required to inhibit the growth of cancer cells by 50% (IC 50 ) in HeLa cells before proceeding. The IC 50 was determined to be 2.908 µM for compound 5 and 9.940 µM for compound 6b. DMSO was used as a control and the cells were treated similarly as in Figure 2, this time using 20 µM, 10 µM and 1 µM to determine their potency at lesser concentrations in other established cancer cell lines. On average, there was less than 5% survival rate across all cell lines tested at 20 µM for compound 5 and less than 24% for compound 6b. The effectiveness of both compounds 5 and 6b decreased as concentrations were lowered to 10 µM and 1 µM.

Chemistry
All 1 H and 13 C NMR spectra were recorded on an ECS 400 MHz JEOL spectrometer. 1 H NMR spectra were recorded in acetone-d6 and are referenced to residual (CH3)2CO at δ = 2.04 ppm and the 13 C NMR spectra are referenced to the peak at δ = 205 ppm. 1 H NMR spectra recorded in CDCl3 are referenced to residual CHCl3 at δ = 7.24 ppm and 13 C NMR spectra are referenced to the central peak of CDCl3 at δ = 77.0 ppm (See Supplementary Materials). Assignments were made by standard gCOSY and gHSQC experiments. HRMS data were obtained with a Bruker Ultraflex MALDI-TOF mass spectrometer. Tri-O-benzyl-D-glucal, tri-O-acetyl-D-glucal and N-iodosuccinimide (NIS) were purchased from Aldrich. Heterocycles were purchased from Aldrich Chemical Co. 4,6-O-Benzylidene-D-glucal was purchased from Carbosynth Ltd. All column chromatography was performed on silica gel 60 (EM Science, 70-230 mesh). Reactions were monitored by TLC on Kieselgel 60 F254 (EM Science) and the compounds were detected by examination under UV light and by charring with 10% sulfuric acid in methanol. Solvents were removed under reduced pressure at <40 °C. CH3CH2CN, CH2Cl2 was distilled from CaH2 and stored over molecular sieves (3 Å ).

Chemistry
All 1 H and 13 C NMR spectra were recorded on an ECS 400 MHz JEOL spectrometer. 1 H NMR spectra were recorded in acetone-d 6 and are referenced to residual (CH 3 ) 2 CO at δ = 2.04 ppm and the 13 C NMR spectra are referenced to the peak at δ = 205 ppm. 1 H NMR spectra recorded in CDCl 3 are referenced to residual CHCl 3 at δ = 7.24 ppm and 13 C NMR spectra are referenced to the central peak of CDCl 3 at δ = 77.0 ppm (See Supplementary Materials). Assignments were made by standard gCOSY and gHSQC experiments. HRMS data were obtained with a Bruker Ultraflex MALDI-TOF mass spectrometer. Tri-O-benzyl-D-glucal, tri-O-acetyl-D-glucal and N-iodosuccinimide (NIS) were purchased from Aldrich. Heterocycles were purchased from Aldrich Chemical Co. 4,6-O-Benzylidene-D-glucal was purchased from Carbosynth Ltd. All column chromatography was performed on silica gel 60 (EM Science, 70-230 mesh). Reactions were monitored by TLC on Kieselgel 60 F254 (EM Science) and the compounds were detected by examination under UV light and by charring with 10% sulfuric acid in methanol. Solvents were removed under reduced pressure at <40 • C. CH 3 CH 2 CN, CH 2 Cl 2 was distilled from CaH 2 and stored over molecular sieves (3 Å).

General Synthetic Procedures
Synthesis of compound 1-4 was performed as previously described [25]. Synthesis of the N-[2-deoxy-2-iodo-α-D-mannopyranosyl]-1H-benzotriazole series of compounds 20, 21a, 21b, 22a and 22b was performed as follows. D-glucal 17, 18 and 19 (1.83 mmol) was diluted in freshly distilled propionitrile (5 mL) followed by the addition of benzotriazole (3.67 mmol). NIS (2.74 mmol) was added to the reaction mixture and refluxed for 2 h. The reaction was cooled to room temperature and quenched with deionized water. The crude mixture was diluted in dichloromethane (50 mL) and washed with saturated Na 2 S 2 O 3 solution (100 mL) and (3 × 100 mL) deionized water. The organic layer was dried over MgSO 4 , filtered and the solvent was removed under reduced pressure. The crude mixture was purified by flash column chromatography on silica gel (hexane/ethyl acetate eluent mixture).
Preparation of 2-deoxy-α-D-glycosylbenzotriazole series 5-9 was performed as follows. Compound 20, 21a, 21b and 22a (1.38 mmol) was diluted in toluene (5 mL). Tributyltin hydride (2.76 mmol) was added dropwise followed by AIBN (1% by weight) and the resulting mixture was refluxed for 1 h. The reaction was allowed to cool down to room temperature and diluted in dichloromethane (50 mL). The organic layer was washed with deionized water (3 × 100 mL) and dried with MgSO 4 , followed by the subsequent removal of the solvent under reduced pressure. The crude mixture was purified by flash column chromatography on silica gel (hexane/ethyl acetate eluent mixture).  13 13 13 6 Hz, J 2,2 = 2.8 Hz, H-2), 2.25 (dd, 1H, J 2,1 = 11.6 Hz, J 2,2 = 3. 8 Hz, H-2 ) ppm. 13  N-[2-deoxy-β-D-glucopyranosyl]-1H-benzoimidazole (11). The crude mixture (0.75 g) was purified by flash column chromatography on silica gel (CHCl 3 /CH 3 OH, 9:1 v/v) to give a cancer cell lines tested against in this study. Compound 5 showed an average of less than 5% survival rate at 20 µM concentrations, while the acetyl protected compound 6b was less than 24% for all four cell lines. The IC 50 was found to be 2.908 µM and 9.940 µM for compounds 5 and 6b, respectively. When examining the nature of the different protecting groups for the 1H-2-deoxy-D-glucopyranosylbenzotriazole series, the more lipophilic benzyl ethers outperformed the acetyl and 4,6-benzylidene counterparts in cell viability studies. This pattern was also observed for the glucose derivatives 1-4 when compared to compound I. For the 2-deoxy-N-glycoimidazole series, only the benzimidazole containing compound 10 showed potency at 100 µM concentration in our cell viability studies. The cytotoxic screening of compounds 5 and 6b using a broader cell panel is currently underway, as is their optimization and an investigation into their mode of action, with results forthcoming in future publications.