A Diversifiable Synthetic Platform for the Discovery of New Carbasugar SGLT2 Inhibitors Using Azide–Alkyne Click Chemistry

Abstract

Sodium-glucose cotransporter 2 (SGLT2) inhibitors are clinically available to control blood glucose levels in diabetic patients via an insulin-independent mechanism. It was found that some carbasugar analogs of known SGLT2 inhibitors exert a high inhibiting ability toward SGLT2 and have a prolonged blood glucose lowering effect. In this study, we designed new candidates of carbasugar SGLT2 inhibitor that can be synthesized using copper-catalyzed azide–alkyne cycloaddition (CuAAC) into an aromatic ring, which is a part of the pharmacophore at the final stage in the synthetic protocol for the easier discovery of superior SGLT2 inhibitors. Based on the results of molecular docking studies, some selected compounds have been synthesized. Evaluation of these compounds using a cell-based assay revealed that the majority of these compounds had SGLT2 inhibitory activity in a dose-dependent manner. The SGLT2 inhibitory activity of 7b and 7c was almost equal to that of SGLT2 inhibitors in current use. Furthermore, molecular dynamics simulations also revealed that 7c is a promising novel SGLT2 inhibitor.

Introduction

Sodium-dependent glucose cotransporter 2 (SGLT2) is a transporter protein mainly expressed in the S1 segment of the proximal tubule of the kidney and where it mediates the reabsorption of the majority of glucose filtered by the kidney glomeruli.¹⁾ Tubular glucose of the remainder is reabsorbed by SGLT1, another member of the SLC5A gene family, presented in the more distal S2/S3 segment. Studies in SGLT1 and SGLT2 knockout mice have confirmed that 97% of filtered glucose is reabsorbed by SGLT2 and the remaining 3% is reabsorbed by SGLT1 under otherwise normal physiological conditions.^2,3) It has been reported that type 2 diabetes mellitus (T2DM) patients have increased SGLT2 expression and activity.⁴⁾ Inhibition of SGLT2 can thus help diabetic patients reduce blood glucose by promoting urinary glucose excretion and making it an appealing strategy for T2DM treatment via an insulin-independent mechanism.⁵⁾ Some SGLT2 inhibitors have been shown to not only lower blood glucose level and body weight but also to have cardio- and renoprotection.⁶⁾ To date, several SGLT2 inhibitors categorized under the gliflozin class, represented by canagliflozin (1), dapagliflozin (2), ipragliflozin (3), and empagliflozin (4), have been licensed for marketing. The gliflozins are based on the glycosylated diarylmethane pharmacophore (Fig. 1). It has been reported that the carbasugar approach, which involves replacing the endocyclic oxygen atom in the glucose moiety of gliflozins with a methylene group, enhances the inhibiting ability of SGLT2 and prolongs the blood glucose lowering effect.^7,8) Therefore, carbasugar analogs of gliflozins are highly promising drug candidates for the treatment of T2DM.

Fig. 1. Selected Example of SGLT2 Inhibitors

A few synthetic methods for carbasugar analogs of gliflozins have been reported.^7,8) Ohtake et al. synthesized various C-aryl carba-D-glucopyranosides 5 by coupling fully benzyl-protected cyclohexanone derivative as carbasugar residue with the corresponding lithiated diarylmethane-aglycones generated from aryl bromide with n-butyl lithium, in the same matter of conventional synthetic methodology for normal sugar-based gliflozins⁷⁾ (Chart 1A). Shing and fellow researchers successfully synthesized carbasugar analogs of dapagliflozin via a Pd-catalyzed allyl-aryl coupling reaction between allylic electrophiles as carbasugar core and the boronic acid-functionalized diarylmethane⁸⁾ (Chart 1B). However, both methods require multiple steps for the synthesis of diarylmethane-aglycone and the use of highly reactive reagents such as n-butyl lithium in activating aglycones for coupling with carbasugar parts. Additionally, in the latter case, a Pd-catalyzed allyl-aryl coupling reaction was performed for three days.

Chart 1. Synthetic Methods for Carbasugar Analogs of Gliflozins

To efficiently discover superior SGLT2 inhibitors, we set out to develop a concise synthetic methodology of carbasugar SGLT2 inhibitor candidates through copper-catalyzed azide–alkyne [3 + 2] cycloaddition (CuAAC), enabling the formation of the proximal aromatic ring as a component of diarylmethane moiety, which is a part of the pharmacophore at the final stage of the synthetic protocol as shown in Chart 1C. Wang and colleagues reported the synthesis of C-aryl β-D-glucopyranoside 6 possessing a 1,2,3-triazole core, which are gliflozin analogs with triazolylmethylaryl-aglycone instead of diarylmethane using CuAAC.⁹⁾ CuAAC did not occur in the final step in this case, since CuAAC was conducted between a fully acetyl-protected glucose analog bearing terminal alkyne moiety and the corresponding azides. The desired gliflozin analogs 6 were obtained by removal of acetyl groups after CuAAC. Most compounds 6 were reported to increase urinary glucose excretion and demonstrate inhibition of glucose transport.

In our previous studies, we have succeeded in developing versatile terminal alkyne probes and rapid ligand-free CuAAC.^10–12) In addition, we discovered terminal alkyne substrate-bearing heteroatoms such as oxygen and nitrogen atoms adjacent to the alkynes efficiently react with benzyl azide derivatives under CuAAC conditions to afford the corresponding triazole products.¹³⁾ Furthermore, the 1-hydroxyl group can also be utilized as a connection point with a tag for high-throughput synthesis and screening using DNA-encoded chemical libraries¹⁴⁾ and on-chip combinatorial libraries.¹⁵⁾ Based on these findings, we designed carbasugar analogs of gliflozins containing a 1,2,3-triazole core with a hydroxyl group at position 1 of the sugar moiety 7. Its C1 epimer 8, which could be obtained from the common precursor 9, was also prepared for the structure–activity relationship study. Herein, we report the synthesis and inhibitory activity evaluation against SGLT2 of these C-aryl carba-D-glucopyranosides bearing triazolylmethylaryl-aglycone.

Results and Discussion

Synthesis of 1-Ethynyl Carba-D-glucose

The synthesis of 1-ethynyl carba-α-D-glucose 10 and 1-ethynyl carba-β-D-glucose 11 is outlined in Chart 2. First, the known diol 17 was prepared from commercially available D-quinic acid (13).¹⁶⁾ The acid-catalyzed reaction of 13 and cyclohexanone with the removal of H₂O yielded 14. The lactone functionality of 14 was cleaved by treatment with NaOMe to afford 15. Compound 15 was oxidized with concomitant β-elimination using Dess–Martin periodinane (DMP) followed by POCl₃ to give enone 16 in 97% yield. Diisobutylaluminium hydride (DIBAL-H) was employed to reduce the ester and enone carbonyl groups in 16 to form the diol 17. Next, the two hydroxyl groups of 17 were protected by p-methoxybenzyl (PMB) groups, yielding 18. The double bond in 18 was subjected to a stereo-controlled hydroboration–oxidation sequence at the less hindered β-face,¹⁷⁾ furnishing exclusively the cyclohexane derivative 19. Acetylation of 19 with Ac₂O afforded 20 in a 97% yield, and subsequent acidic removal of the acetal in 20 gave the diol 21. Under Mitsunobu conditions, treatment of 21 with benzoic acid provided the corresponding carbaglucose analog 22 in 57% yield. It has been reported that the equatorial hydroxyl group of cis-1,2-diol in the 1,4,5,6-tetra-O-benzyl-myo-inositol is more nucleophilic than the axial one,¹⁸⁾ and thus the tetra-protected myo-inositol could be converted into various cyclitols derivatives of the scyllo-configuration by regioselective inversion using the Mitsunobu reaction.¹⁹⁾ The hydroxyl group at 1 position adjacent to methylene was suggested to favor equatorial orientation in the current conversion of 21 to 22, and thus the Mistunobu reaction occurred selectively at the position 2 in the above manner. The Dess–Martin oxidation of the hydroxyl group at position 1 proceeded nearly quantitatively, giving the corresponding cyclohexanone derivative 9. With the key intermediate 9 in hand, we next focused on the introduction of the ethynyl group at 1 position. Treatment of 9 with ethynyl magnesium bromide in tetrahydrofuran (THF) at −40 °C gave 23 and 24 in a 94% yield in a 1 : 5.9 ratio. The configurations at position 1 of these compounds were confirmed from X-ray crystallographic analysis of 24 (Supplementary Fig. S1, Table S1). Finally, deprotection of the PMB groups of 23 and 24 using 2,3-dichloro-5,6-dicyanobenzoquinone (DDQ) followed by removal of acyl groups with NaOMe afforded the 1-ethynyl carba-D-glucose 10 (α-isomer) and 11 (β-isomer), respectively.

Chart 2.

Reagents and conditions: (a) Cyclohexanone, TsOH·H₂O, reflux, 94%; (b) NaOMe, MeOH, r.t., 95%; (c) i) DMP, CH₂Cl₂, r.t.; ii) POCl₃, pyridine, r.t., 97%; (d) DIBAL-H, n-hexane, THF, r.t., 90%; (e) PMBCl, NaH, TBAI, DMF, r.t., 95%; (f) i) BH₃·THF, THF, r.t.; ii) H₂O₂, NaOH, H₂O, r.t., 82%; (g) Ac₂O, pyridine, r.t., 97%; (h) AcOH, H₂O, r.t., 94%; (i) BzOH, Ph₃P, DIAD, MS4A, toluene, reflux, 57%; (j) DMP, CH₂Cl₂, r.t., 99%; (k) Ethynyl-MgBr, THF, −40 °C, 94% (23 : 24 = 1 : 5.9); (l) i) DDQ, H₂O, CH₂Cl₂, r.t., b) NaOMe, MeOH, r.t., 81% (10), quant (11).

Molecular Docking Study and Synthesis of Triazolyl-C-Aryl Carba-D-glucopyranosides 7 and 8

The cryogenic electron microscopy (cryo-EM) structure of the hSGLT2–MAP17 complex in the empagliflozin (4)-bound state with an overall resolution of 2.95 Å was recently reported.²⁰⁾ The structure reveals that 4 binds to the sugar-binding site as well as the hSGLT2 external vestibule, preventing the transport cycle from taking place. To predict the superior one among the designed compounds 7 and 8, the cryo-EM structure of the hSGLT2–MAP17 with 4 was utilized to compare the SGLT2 binding mode of compounds 7a–7e and 8a–8e to that of 4 and ipragliflozin (3), a commercially available and marketed SGLT2 inhibitor. The result of docking run was provided in Supplementary Table S2. The rank of docking score for the compounds was in the following order: 7c > 3 > 7b > 4 > 7a > 7d > 7e > 8c > 8b > 8e > 8d > 8a (Supplementary Fig. S2). The docking pose of 7c showed perfect complementarity with the cocrystallized 4. Both 7c and 4 exhibited complete alignment at the active site. The carbocyclic ring of 7c interacted similarly with the pyranose ring of 4, forming an equivalent number of hydrogen bonds with the same SGLT2 residues (Fig. 2). Interactions with C2 and C3 hydroxy groups have been identified as the most essential contributors to SGLT inhibitors’ binding.²¹⁾ Interestingly, 7c demonstrated more pronounced interactions at the C3-OH because it can function as a hydrogen bond acceptor with Asn75 and a hydrogen bond donor for the backbone of Phe98, making it more interactive than the C3-OH of 4, which contributes a hydrogen bond to the side chain of Phe98. The stereochemical configuration for C2-OH and C3-OH of 7c was comparable to that of 4 and was favorable for binding with SGLT2. The aglycone substitution at C1 promotes the inhibition of SGLT2, in which the carbocyclic ring in 7c or the pyranose sugar occupies the sugar-binding site in the carrier while the aglycone derivative extends to the external vestibule of SGLT2 (Supplementary Fig. S3). This arrangement of occupying both the sugar-binding site and the exterior vestibule causes SGLT2 inhibition and glucose transport into the cell. 7a–7c and 8a–8c were selected for synthesis based on the findings of molecular docking studies. CuAAC between 10 and 11 and benzyl azide derivatives 12a–12c smoothly proceeded to obtain these six compounds (approx. quant).

Fig. 2. The Ligand Interactions of Compounds 4 and 7c with hSGLT2

(A) The ligand interactions of 4. (B) The ligand interactions of 7c.

Biological Evaluation

The inhibitory activity of newly prepared triazolyl-C-aryl carba-D-glucopyranosides against SGLT2 was evaluated in a cell-based nonradioactive fluorescent glucose uptake assay,²²⁾ which used Human Embryonic Kidney cell 293 (HEK293) cells stably expressing hSGLT2 (Supplementary Fig. S4). The inhibitory activity of 7a–7c and 8a–8c is depicted in Fig. 3. Most of these compounds had SGLT2 inhibitory activity in a dose-dependent manner. In general, the α-isomers 7 were more active than the β-isomers 8, indicating that the α-configuration at 1 position is critical for the inhibitory activity.⁸⁾ The SGLT2 inhibitory activity of 7b and 7c bearing ethyl group and cyclopropyl groups in the para position on the distal benzene ring, respectively, was almost equal to that of 3 in current use. 7c showed slightly higher inhibitory activity than 7b. It is concidered that the reason why the concentration required for inhibition is higher than in previous reports might be due to glucose uptake by other transporters such as glucose transporter (GLUT) universally expressed in HEK293 cells.

Fig. 3. Fluorescent Glucose Uptake Assay

Molecular Dynamics Simulation Study

We next performed molecular dynamics simulation of 7c along with 3 and 4. The conformational stability of hSGLT2 and three compounds was investigated by calculating the root mean square deviation (RMSD) and root mean square fluctuations (RMSF) of the protein backbone atoms and side chains of residues, respectively (Fig. 4). Soon after the simulation began, all structures were equilibrated, and the RMSD values were stabilized. The ligand RMSD was lower than the protein RMSD in 7c as well as 3 and 4, which shows stable attraction of these compounds to the binding site during the simulation time. In all simulated complexes, the RMSF values of residues 120–131, a flexible loop at the inner side of hSGLT2, were below 2.0 Å, indicating that the simulated complexes of three compounds and hSGLT2 were stable during simulation. During molecular dynamics simulation, the number of H-bonds formed between these three compounds and hSGLT2 was traced (Supplementary Fig. S5, Table S3). The molecular dynamics simulations validated the hypothesis that 7c is a promising novel SGLT2 inhibitor by demonstrating low RMSD, quick stability, low RMSF, more than five hydrogen bonds throughout the majority of the simulation time. These findings imply that 7c may have a higher affinity for hSGLT2 than the approved SGLT2 inhibitors.

Fig. 4. RMSD of 3, 4, and 7 Bound with hSGLT2 (A–C, Respectively), and RMSF of ApohSGLT2 or Bound with 3, 4, or 7 during 100 ns MD Simulation (D)

Conclusion

In summary, we created a diversifiable synthetic platform for the discovery of new carbasugar SGLT2 inhibitors by employing CuAAC at the final stage of the synthetic protocol. Based on previous findings, we newly designed C-aryl carba-D-glucopyranosides bearing triazolylmethylaryl-aglycone 7 and its C1 epimer 8. Six compounds (7a–7c and 8a–8c) selected based on the results of molecular docking studies were synthesized and evaluated inhibitory activity against SGLT2. Among them, 7b and 7c bearing ethyl group and cyclopropyl groups in the para position, respectively, showed almost equal inhibitory activity to the approved SGLT2 inhibitors. The results of the molecular dynamics simulation study also support the potential for 7c to be further developed as novel SGLT2 inhibitors. We believe that our synthetic approach will contribute significantly to the discovery of superior SGLT2 inhibitors.

Experimental

General

All reactions were carried out under an argon atmosphere, unless otherwise noted. All reagents and solvents were purchased from commercial vendors and used without further purification, unless indicated otherwise. Pyridine was distilled over CaH₂ and stored over activated molecular sieves 4 Å (MS4A). ¹H- and ¹³C-NMR spectra were recorded on a JEOL JNM AL-400 spectrometer or JNM ECS-400 spectrometer (400 MHz for ¹H-NMR and 100 MHz for ¹³C-NMR). Chemical shifts (δ) were expressed in parts per million and internally referenced (7.26 ppm for CDCl₃, 3.30 ppm for CD₃OD or 2.49 ppm for dimethyl sulfoxide (DMSO)-d₆ for ¹H-NMR, 77.0 ppm for CDCl₃, 49.0 ppm for CD₃OD). Electrospray ionization (ESI) mass spectra were taken on a JMS T100LP instrument or a Waters Xevo Q-Tof mass spectrometer. Flash column chromatography was performed using silica gel 60N [spherical neutral (63–210 µm)] from Kanto Chemical Co., Inc. (Tokyo, Japan) or silica gel PSQ 100B (63–210 µm) from Fuji Silysia Chemical Co., Ltd. (Kasugai, Aichi, Japan).

3,4-O-Cyclohexylidenequinic Acid-1,5-lactone (14)²³⁾

Cyclohexanone (9.60 mL, 93.6 mmol) was added dropwise to a suspension of D-quinic acid (13) (12.0 g, 62.4 mmol) and p-toluenesulfonic acid monohydrate (544 mg, 3.16 mmol) in toluene (120 mL), and then the mixture was azeotoropic-refluxed with a Dean-Stark apparatus for 20 h. After the reaction mixture was cooled to room temperature (r.t.), the reaction mixture was partitioned between EtOAc and H₂O. The organic layer was washed with saturated NaHCO₃ aqueous solution, H₂O and brine, dried over Na₂SO₄, and concentrated under reduced pressure. n-Hexane was added to the residue, and the resulting precipitate was collected by filtration, washed with n-hexane to give 14 as a colorless solid (14.9 g, 94%). ¹H-NMR (CDCl₃) δ: 4.74 (1H, dd, J = 6.0 Hz, 2.5 Hz), 4.48 (1H, ddd, J = 7.9 Hz, 6.5 Hz, 3.2 Hz), 4.31 (1H, ddd, J = 6.5 Hz, 2.5 Hz, 1.3 Hz), 2.67 (1H, d, J = 11.4 Hz), 2.67 (1H, s), 2.35 (1H, ddd, J = 14.7 Hz, 7.9 Hz, 2.3 Hz), 2.29 (1H, dddd, J = 11.4 Hz, 6.0 Hz, 2.3 Hz, 1.3 Hz), 2.19 (1H, dd, J = 14.7 Hz, 3.2 Hz), 1.70–1.73 (2H, m), 1.62–1.68 (2H, m), 1.52–1.60 (4H, m), 1.37–1.43 (2H, m); ¹³C-NMR (CDCl₃) δ: 179.0, 110.6, 75.9, 71.7, 71.5, 71.0, 38.3, 36.8, 34.3, 33.6, 25.0, 23.9, 23.4; high resolution (HR)-MS (ESI) m/z: 255.1209 [M + H]⁺ (Calcd for C₁₃H₁₉O₅: 255.1232).

Methyl 3,4-O-Cyclohexylidenequinate (15)

Sodium methoxide (28% solution in MeOH, 3.08 mL, 16.0 mmol) was added dropwise to a solution of 14 (4.47 g, 17.6 mmol) in MeOH (94 mL) at 0 °C, and then the mixture was stirred at room temperature for 5 h. The reaction mixture was neutralized with Dowex 50WX8 (H⁺), and then the resin was separated. The filtrate was concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/EtOAc, 2 : 1–1 : 2) to give 15 as a yellow oil (4.75 g, 95%). ¹H-NMR (CDCl₃) δ: 4.46–4.42 (1H, m), 4.13–4.07 (1H, m), 3.96 (1H, t, J = 6.2 Hz), 3.78 (3H, s), 3.49 (1H, s), 2.37 (1H, d, J = 3.2 Hz), 2.29–2.19 (2H, m), 2.07 (1H, ddd, J = 13.7 Hz, 4.2 Hz, 1.4 Hz), 1.84 (1H, dd, J = 13.7 Hz, 10.6 Hz), 1.75–1.51 (8H, m), 1.43–1.33 (2H, m); ¹³C-NMR (CDCl₃) δ: 175.3, 109.9, 79.4, 74.0, 73.0, 68.5, 53.0, 38.9, 37.9, 34.8, 34.6, 24.9, 23.9, 23.5; HR-MS (ESI) m/z: 287.1463 [M + H]⁺ (Calcd for C₁₄H₂₃O₆: 287.1495).

Methyl 4,5-O-Cyclohexylidene-3-dehydro-4-epi-shikimate (16)

Dess–Martin periodinane (6.73 g, 15.9 mmol) was added to a solution of 15 (4.13 g, 14.4 mmol) in CH₂Cl₂ (74 mL) at 0 °C, and then the mixture was stirred at room temperature for 40 h. After saturated NaHCO₃ aqueous solution (60 mL) was added the reaction mixture at 0 °C, the mixture was partitioned between CHCl₃ and H₂O. The organic layer was washed with H₂O and brine, dried over Na₂SO₄, and concentrated under reduced pressure. After the residue was dissolved in pyridine (15 mL), phosphoryl chloride (2.8 mL, 29.8 mmol) was added to the solution at 0 °C. The resulting mixture was stirred at room temperature for 3 h. After saturated NH₄Cl aqueous solution was added the reaction mixture, the mixture was partitioned between CH₂Cl₂ and H₂O. The organic layer was washed with H₂O and brine, dried over Na₂SO₄, and concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/EtOAc, 15 : 1–5 : 1) to give 16 as a pale yellow solid (3.59 g, 97%). ¹H-NMR (CDCl₃) δ: 6.84 (1H, d, J = 2.7 Hz), 4.69 (1H, dt, J = 5.0, 1.7 Hz), 4.30 (1H, d, J = 5.0 Hz), 3.85 (3H, s), 3.24 (1H, d, J = 20.3 Hz), 2.86 (1H, ddd, J = 20.3, 5.0, 2.7 Hz), 1.69–1.35 (10H, m); ¹³C-NMR (CDCl₃) δ: 197.7, 166.2, 144.3, 131.2, 110.2, 74.7, 72.1, 52.8, 37.0, 35.2, 27.0, 26.6, 24.8, 23.7; HR-MS (ESI) m/z: 267.1214 [M + H]⁺ (Calcd for C₁₄H₁₉O₅: 267.1232).

(1R,2R,3S)-1,2-O-Cyclohexylidene-5-hydroxymethyl-4-cyclohexene-1,2,3-triol (17)

Diisobutylaluminium hydride (1.0 M solution in n-hexane, 81.0 mL, 81.0 mmol) was added dropwise to a solution of 16 (3.59 g, 13.5 mmol) in THF (79 mL) at 0 °C, and then the mixture was stirred at room temperature for 3 h. After saturated NH₄Cl aqueous solution was added the reaction mixture, the resulting precipitate was filtered off and the filtrate was partitioned between EtOAc and H₂O. The organic layer was washed with H₂O and brine, dried over Na₂SO₄, and concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/EtOAc/MeOH, 1 : 4 : 0–0 : 2 : 1) to give 17 as a colorless solid (2.90 g, 90%). ¹H-NMR (DMSO-d₆) δ: 5.50 (1H, d, J = 0.8 Hz), 4.82 (1H, d, J = 6.0 Hz), 4.70 (1H, t, J = 5.2 Hz), 4.45–4.41 (1H, m), 4.34–4.30 (1H, m), 3.96–3.92 (1H, m), 3.78 (2H, d, J = 5.2 Hz), 2.02 (1H, dd, J = 16.0, 2.0 Hz), 1.84–1.78 (1H, m), 1.51–1.22 (10H, m); ¹³C-NMR (CDCl₃) δ: 137.3, 125.6, 109.4, 76.0, 72.1, 67.0, 65.7, 35.6, 34.0, 28.7, 25.1, 23.9, 23.5; HR-MS (ESI) m/z: 263.1262 [M + Na]⁺ (Calcd for C₁₃H₂₀NaO₄: 263.1259).

(1R,2R,3S)-1,2-O-Cyclohexylidene-3-O-(4-methoxybenzyl)-5-(4-methoxybenzyl-oxy)methyl-4-cyclohexene-1,2,3-triol (18)

A solution of 17 (1.20 g, 5.0 mmol) in N,N-dimethylformamide (DMF) (10 mL) was added to a suspention of sodium hydride (60% dispersion in paraffin oil, 516 mg, 15.0 mmol) in DMF (10 mL) at 0 °C, and then the mixture was stirred at room temperature for 1 h. p-Methoxybenzyl chloride (1.67 mL, 15.0 mmol) and n-tetrabutylammonium iodide (369 mg, 1.0 mmol) were added to the reaction mixture at 0 °C, and then the mixture was stirred at room temperature for 20 h. After MeOH was added the reaction mixture, the mixture was partitioned between EtOAc and H₂O. The organic layer was washed with H₂O and brine, dried over Na₂SO₄, and concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/EtOAc, 4 : 1) to give 18 as a pale yellow oil (2.28 g, 95%). ¹H-NMR (CDCl₃) δ: 7.32 (2H, d, J = 8.4 Hz), 7.27 (2H, d, J = 8.8 Hz), 6.88 (2H, d, J = 8.4 Hz), 6.87 (2H, d, J = 8.8 Hz), 5.87 (1H, s), 4.73 (1H, d, J = 11.8 Hz), 4.62 (1H, d, J = 11.8 Hz), 4.56 (1H, ddd, J = 7.2 Hz, 3.7 Hz, 1.3 Hz), 4.51–4.48 (1H, m), 4.48 (1H, d, J = 11.2 Hz), 4.37 (1H, d, J = 11.2 Hz), 3.95 (2H, dd, J = 24.4 Hz, 12.4 Hz), 3.82–3.78 (7H, m), 2.41 (1H, dd, J = 16.0 Hz, 1.6 Hz), 1.88–1.82 (1H, m), 1.66–1.41 (8H, m), 1.33–1.24 (2H, m); ¹³C-NMR (CDCl₃) δ: 159.2, 159.1, 134.9, 130.4, 130.1, 129.5, 129.4, 125.0, 113.7, 113.7, 109.2, 74.9, 73.4, 72.9, 72.6, 71.4, 70.4, 55.2(2), 35.6, 33.7, 29.4, 25.3, 23.9, 23.6; HR-MS (ESI) m/z: 503.2427 [M + Na]⁺ (Calcd for C₂₉H₃₆NaO₆: 503.2410).

(1R,2R,3S,4R,5R)-1,2-O-Cyclohexylidene-3-O-(4-methoxybenzyl)-5-(4-methoxybenzyloxymethyl)cyclohexane-1,2,3,4-tetraol (19)

Borane-THF complex (1.0 M solution in THF, 13.3 mL, 13.3 mmol) was added to a solution of 18 (3.20 g, 6.66 mmol) in THF (71 mL) at 0 °C, and then the mixture was stirred at room temperature for 20 h. H₂O (8.04 mL), H₂O₂ (35% solution in H₂O, 21.8 mL) and 1 M NaOH (51 mL) were added to the reaction mixture at 0 °C, and then the mixture was stirred at room temperature for 3 h. After H₂O was added the reaction mixture, the mixture was partitioned between EtOAc and H₂O. The organic layer was washed with H₂O and brine, dried over Na₂SO₄, and concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/EtOAc, 3 : 1–1 : 1) to give 19 as a colorless oil (2.71 g, 82%). ¹H-NMR (CDCl₃) δ: 7.32 (2H, d, J = 8.6 Hz), 7.24 (2H, d, J = 8.8 Hz), 6.88 (2H, d, J = 8.6 Hz), 6.86 (2H, d, J = 8.8 Hz), 4.73 (1H, d, J = 12.0 Hz), 4.64 (1H, d, J = 12.0 Hz), 4.44 (2H, s), 4.31 (1H, dd, J = 4.8 Hz, 4.3 Hz), 4.15–4.08 (1H, m), 3.82–3.74 (7H, m), 3.61 (1H, dd, J = 9.2 Hz, 5.6 Hz), 3.45 (1H, dd, J = 9.2 Hz, 6.4 Hz), 3.40 (1H, dd, J = 9.4 Hz, 4.3 Hz), 2.94 (1H, s), 1.90 (1H, ddd, J = 13.8 Hz, 6.2 Hz, 3.6 Hz), 1.77–1.36 (12H, m); ¹³C-NMR (CDCl₃) δ: 159.3, 159.1, 130.2, 129.6, 129.2, 113.8, 113.7, 109.9, 80.0, 77.2, 73.6, 73.4, 72.9, 72.0, 71.4, 70.9, 55.2(2), 38.4, 38.0, 35.1, 30.7, 25.0, 24.0, 23.7; HR-MS (ESI) m/z: 521.2534 [M + Na]⁺ (Calcd for C₂₉H₃₈NaO₇: 521.2515).

(1R,2R,3S,4R,5R)-4-O-Acetyl-1,2-O-cyclohexylidene-3-O-(4-methoxybenzyl)-5-(4-methoxybenzyloxymethyl)cyclohexane-1,2,3,4-tetraol (20)

Acetic anhydride (6.48 mL, 68.5 mmol) was added to a solution of 19 (1.71 g, 3.43 mmol) in pyridine (15 mL) at 0 °C, and then the mixture was stirred at room temperature for 20 h. After saturated NaHCO₃ aqueous solution was added the reaction mixture, the mixture was partitioned between EtOAc and H₂O. The organic layer was washed with H₂O and brine, dried over Na₂SO₄, and concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/EtOAc, 3 : 1) to give 20 as a colorless oil (1.80 g, 97%). ¹H-NMR (CDCl₃) δ: 7.26 (2H, d, J = 8.6 Hz), 7.22 (2H, d, J = 8.6 Hz), 6.86 (4H, d×2, J = 8.6 Hz), 5.16 (1H, dd, J = 10.4 Hz, 9.2 Hz), 4.36 (2H, dd, J = 16.0 Hz, 11.6 Hz), 4.28 (1H, dd, J = 4.8 Hz, 4.4 Hz), 4.15–4.09 (1H, m), 3.80 (6H, s×2), 3.55 (1H, dd, J = 9.2 Hz, 4.4 Hz), 3.41 (1H, dd, J = 9.2 Hz, 4.8 Hz), 3.27 (1H, dd, J = 9.2 Hz, 7.2 Hz), 2.05–1.98 (4H, m), 1.80–1.50 (10H, m), 1.41–1.36 (2H, m); ¹³C-NMR (CDCl₃) δ: 170.3, 159.2, 159.1, 130.4, 130.2, 129.3, 129.2, 113.7, 113.6, 110.0, 77.2, 77.1, 74.2, 73.1, 72.9, 72.0, 71.5, 71.3, 55.2(2), 37.7, 35.0, 30.5, 23.9, 25.0, 23.9, 21.1; HR-MS (ESI) m/z: 563.2636 [M + Na]⁺ (Calcd for C₃₁H₄₀NaO₈: 563.2621).

(1R,2R,3S,4R,5R)-4-O-Acetyl-3-O-(4-methoxybenzyl)-5-(4-methoxybenzyloxy-methyl)cyclohexane-1,2,3,4-tetraol (21)

At 0 °C, 20 (1.65 g, 3.05 mmol) was suspended in a solution of acetic acid (8.4 mL) and H₂O (2.1 mL). The resulting mixture was stirred at room temperature for 17 h. The reaction mixture was concentrated under reduced pressure. The residue was partitioned between CH₂Cl₂ and H₂O. The organic layer was washed with H₂O and brine, dried over Na₂SO₄, and concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/EtOAc, 2 : 3) to give 21 as a colorless solid (1.31 g, 94%). ¹H-NMR (CDCl₃) δ: 7.22 (2H, d, J = 8.6 Hz), 7.20 (2H, d, J = 8.8 Hz), 6.87 (2H, d, J = 8.6 Hz), 6.85 (2H, d, J = 8.8 Hz), 5.11 (1H, t, J = 10.0 Hz), 4.56 (1H, d, J = 12.0 Hz), 4.49 (1H, d, J = 12.0 Hz), 4.36 (2H, dd, J = 20.6 Hz, 11.4 Hz), 4.14–4.11 (1H, m), 3.81 (3H, s), 3.80 (3H, s), 3.64–3.57 (1H, m), 3.38 (1H, dd, J = 9.3 Hz, 4.0 Hz), 3.34 (1H, dd, J = 9.6 Hz, 2.8 Hz), 3.26 (1H, dd, J = 9.3 Hz, 6.2 Hz), 2.52 (1H, s), 2.20 (1H, d, J = 10.4 Hz), 1.97–1.94 (4H, m), 1.79–1.64 (2H, m); ¹³C-NMR (CDCl₃) δ: 170.5, 159.3 159.1, 130.2, 129.7, 129.3, 129.3, 113.8, 113.7, 79.7, 72.9, 71.9, 71.6, 70.8, 70.2, 69.3, 55.2(2), 37.7, 30.6, 21.1; HR-MS (ESI) m/z: 460.1763 [M + K]⁺ (Calcd for C₂₅H₃₂KO₈: 460.1734).

(1R,2S,3S,4R,5R)-4-O-Acetyl-2-O-benzoyl-3-O-(4-methoxybenzyl)-5-(4-methoxy-benzyloxymethyl)cyclohexane-1,2,3,4-tetraol (22)

Benzoic acid (708 mg, 5.77 mmol), triphenylphosphine (1.50 g, 5.77 mmol) and MS4A (220 mg) were added to a suspension of 21 (442 mg, 962 µmol) in toluene (19 mL) at 0 °C. Diisopropyl azodicarboxylate (1.10 mL, 5.77 mmol) was added dropwise to the mixture at 0 °C, and then the resuting mixture was refluxed for 18 h. The reaction mixture was concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/EtOAc, 3 : 1–2 : 1) to give 22 as a colorless solid (308 mg, 57%). ¹H-NMR (CDCl₃) δ: 8.03 (2H, d, J = 7.6 Hz), 7.58 (1H, t, J = 7.6 Hz), 7.44 (2H, t, J = 7.6 Hz), 7.24 (2H, d, J = 8.4 Hz), 7.00 (2H, d, J = 8.6 Hz), 6.87 (2H, d, J = 8.4 Hz), 6.67 (2H, d, J = 8.6 Hz), 5.16 (1H, t, J = 9.4 Hz), 5.06 (1H, dd, J = 10.8 Hz, 9.6 Hz), 4.53 (2H, dd, J = 14.0 Hz, 10.8 Hz), 4.38 (2H, dd, J = 16.8 Hz, 11.2 Hz), 3.81 (3H, s), 3.83–3.75 (1H, m), 3.71–3.65 (4H, m), 3.43 (1H, dd, J = 9.2 Hz, 4.4 Hz), 3.30 (1H, dd, J = 9.2 Hz, 6.4 Hz), 2.30 (1H, d, J = 5.2 Hz), 2.26 (1H, dt, J = 13.0 Hz, 4.2 Hz), 1.98–1.88 (4H, m), 1.48 (1H, q, J = 13.0 Hz); ¹³C-NMR (CDCl₃) δ: 170.6, 166.8, 159.2, 159.0, 133.2, 130.1, 130.0, 129.8, 129.6, 129.4, 129.3, 128.4, 113.7, 113.6, 80.8, 79.3, 74.2, 74.1, 73.0, 70.8, 70.4, 55.3, 55.1, 37.8, 33.1, 20.9; HR-MS (ESI) m/z: 587.2232 [M + Na]⁺ (Calcd for C₃₂H₃₆NaO₉: 587.2257).

(2R,3S,4R,5R)-4-O-Acetyl-2-O-benzoyl-3-O-(4-methoxybenzyl)-5-(4-methoxy-benzyloxymethyl)cyclohexane-2,3,4-triol-1-on (9)

Dess–Martin periodinane (275 mg, 648 µmol) was added to a solution of 22 (182 mg, 324 µmol) in CH₂Cl₂ (4.5 mL), and then the mixture was stirred at room temperature for 2 h. After saturated NaHCO₃ aqueous solution was added the reaction mixture at 0 °C, the mixture was partitioned between CHCl₃ and H₂O. The organic layer was washed with brine, dried over Na₂SO₄, and concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/EtOAc, 2 : 1) to give 9 as a colorless solid (180 mg, 99%). ¹H-NMR (CDCl₃) δ: 8.04 (2H, d, J = 7.6 Hz), 7.59 (1H, t, J = 7.6 Hz), 7.46 (2H, t, J = 7.6 Hz), 7.24 (2H, d, J = 8.8 Hz), 7.12 (2H, d, J = 8.6 Hz), 6.88 (2H, d, J = 8.8 Hz), 6.77 (2H, d, J = 8.6 Hz), 5.57 (1H, d, J = 10.8 Hz), 5.51 (1H, dd, J = 11.2 Hz, 9.6 Hz), 4.70 (1H, d, J = 11.0 Hz), 4.58 (1H, d, J = 11.0 Hz), 4.39 (2H, dd, J = 16.4 Hz, 11.6 Hz), 3.91 (1H, t, J = 9.8 Hz), 3.81 (3H, s), 3.75 (3H, s), 3.40 (2H, d, J = 4.4 Hz), 2.65 (1H, t, J = 14.0 Hz), 2.61 (1H, dd, J = 14.0 Hz, 5.2 Hz), 2.12–2.02 (1H, m), 1.96 (3H, s); ¹³C-NMR (CDCl₃) δ: 199.6, 169.8, 165.1, 159.3(2), 133.3, 129.9, 129.8, 129.7, 129.5(2), 129.2, 128.4, 113.7(2), 80.8, 79.6, 74.4, 73.0, 72.4, 68.7, 55.2, 55.2, 39.4, 37.9, 20.8; HR-MS (ESI) m/z: 587.2090 [M + Na]⁺ (Calcd for C₃₂H₃₄NaO₉: 585.2101).

(1R,2R,3S,4R,5R)-4-O-Acetyl-2-O-benzoyl-1-ethynyl-3-O-(4-methoxybenzyl)-5-(4-methoxy-benzyloxymethyl)cyclohexane-1,2,3,4-tetraol (23) and (1S,2R,3S,4R,5R)-4-O-Acetyl-2-O-benzoyl-1-ethynyl-3-O-(4-methoxybenzyl)-5-(4-methoxy-benzyloxymethyl)cyclohexane-1,2,3,4-tetraol (24)

Ethynylmagnesium bromide (0.5 M solution in THF, 6.08 mL, 3.04 mmol) was added to a solution of 9 (532 mg, 950 µmol) in THF (6 mL) at −80 °C, and then the mixture was stirred at −40 °C for 24 h. After saturated NH₄Cl aqueous solution was added to the reaction mixture at −40 °C, the mixture was partitioned between EtOAc and H₂O. The organic layer was washed with H₂O and brine, dried over Na₂SO₄, and concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/EtOAc, 4 : 1) to give a 1 : 5.6 mixtute of 23 and 24 (542 mg, 97%) as a pale yellow solid. The mixture was then separated by column chromatography on silica gel (CHCl₃/n-hexane, 50 : 1).

23: ¹H-NMR (CDCl₃) δ: 8.03 (2H, d, J = 7.4 Hz), 7.59 (1H, t, J = 7.4 Hz), 7.46 (2H, t, J = 7.4 Hz), 7.26 (2H, d, J = 8.6 Hz), 6.96 (2H, d, J = 8.4 Hz), 6.88 (2H, d, J = 8.6 Hz), 6.63 (2H, d, J = 8.4 Hz), 5.44 (1H, d, J = 10.0 Hz), 5.14 (1H, t, J = 10.0 Hz), 4.46 (2H, dd, J = 18.0 Hz, 11.2 Hz), 4.37 (2H, dd, J = 17.4 Hz, 11.2 Hz), 3.95 (1H, t, J = 10.0 Hz), 3.81 (3H, s), 3.69 (3H, s), 3.39 (1H, dd, J = 9.3 Hz, 3.8 Hz), 3.32 (1H, dd, J = 9.3 Hz, 5.6 Hz), 2.60 (1H, s), 2.36 (1H, s), 2.36–2.27 (2H, m), 1.93 (3H, s), 1.86 (1H, t, J = 14.0 Hz); ¹³C-NMR (CDCl₃) δ: 170.0, 164.9, 159.1, 158.9, 133.1, 130.1, 130.1, 129.7, 129.6, 129.4, 129.2, 128.3, 113.6, 113.5, 83.6, 79.1, 77.4, 74.3, 73.8, 73.0, 72.9, 69.6, 68.7, 55.2, 55.0, 37.2, 36.0, 20.9; HR-MS (ESI) m/z: 611.2247 [M + Na]⁺ (Calcd for C₃₄H₃₆NaO₉: 611.2257).

24: ¹H-NMR (CDCl₃) δ: 8.08 (2H, d, J = 7.4 Hz), 7.59 (1H, t, J = 7.4 Hz), 7.46 (2H, t, J = 7.4 Hz), 7.25 (2H, d, J = 8.2 Hz), 6.99 (2H, d, J = 8.6 Hz), 6.88 (2H, d, J = 8.2 Hz), 6.66 (2H, d, J = 8.6 Hz), 5.21 (1H, d, J = 9.6 Hz), 5.11 (1H, dd, J = 11.2 Hz, 9.6 Hz), 4.51 (2H, dd, J = 17.0 Hz, 11.0 Hz), 4.38 (2H, dd, J = 15.4 Hz, 11.0 Hz), 3.90 (1H, t, J = 9.6 Hz), 3.81 (3H, s), 3.71 (3H, s), 3.43 (1H, dd, J = 9.2 Hz, 4.0 Hz), 3.33 (1H, dd, J = 9.2 Hz, 5.6 Hz), 2.97 (1H, s), 2.68 (1H, s), 2.30–2.21 (2H, m), 1.92 (3H, s), 1.78 (1H, t, J = 14.0 Hz); ¹³C-NMR (CDCl₃) δ: 170.0, 166.4, 159.1, 159.0, 133.4, 130.1, 130.0, 129.9, 129.5, 129.4, 129.2, 128.4, 113.7, 113.6, 82.7, 80,6, 79.0, 75.5, 74.5, 73.8, 73.0, 70.7, 69.9, 55.2, 55.1, 38.3, 37.6, 20.9; HR-MS (ESI) m/z: 611.2265 [M + Na]⁺ (Calcd for C₃₄H₃₆NaO₉: 611.2257).

(1R,2R,3S,4R,5R)-1-Ethynyl-5-hydoxymethylcyclohexane-1,2,3,4-tetraol (1-Ethynyl-α-D-carbaglucose) (10)

H₂O (322 µL) and 2,3-dichloro-5,6-dicyano-p-benzoquinone (119 mg, 525 µmol) were added to a solution of 23 (155 mg, 262 µmol) in CH₂Cl₂ (5.3 mL), and then the mixture was stirred at room temperature for 24 h. The reaction mixture was dried over Na₂SO₄, and then concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/acetone, 3 : 1 then CHCl₃/methanol, 30 : 1) to give the corresponding triol. Sodium methoxide (25.2 mg, 460 µmol) was added to a solution of the triol in MeOH (5.5 mL), and then the mixture was stirred at room temperature for 2.5 h. The reaction mixture was neutralized with Dowex 50WX8 (H⁺), and then the resin was separated. The filtrate was washed with EtOAc. The water layer was concentrated under reduced pressure to give 10 as a colorless solid (43.5 mg, 81%). ¹H-NMR (CD₃OD) δ: 3.62–3.60 (2H, m), 3.46–3.41 (1H, m), 3.28–3.19 (2H, m), 2.74 (1H, s), 2.03 (1H, dd, J = 13.9 Hz, 3.6 Hz), 1.86–1.79 (1H, m), 1.54 (1H, t, J = 13.9 Hz); ¹³C-NMR (CD₃OD) δ: 87.6, 78.7, 76.5, 74.6, 72.5, 70.4, 63.7, 40.1, 38.9; HR-MS (ESI) m/z: 225.0721 [M + Na]⁺ (Calcd for C₉H₁₄NaO₅: 225.0739).

(1S,2R,3S,4R,5R)-1-Ethynyl-5-hydoxymethylcyclohexane-1,2,3,4-tetraol (1-Ethynyl-β-D-carbaglucose) (11)

H₂O (534 µL) and 2,3-dichloro-5,6-dicyano-p-benzoquinone (247 mg, 1.09 µmol) were added to a solution of 24 (256 mg, 435 µmol) in CH₂Cl₂ (8.8 mL), and then the mixture was stirred at room temperature for 23 h. The reaction mixture was dried over Na₂SO₄, and then concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/acetone, 3 : 1 then CHCl₃/methanol, 30 : 1) to give the corresponding triol. Sodium methoxide (50.7 mg, 938 µmol) was added to a solution of the triol in MeOH (5.2 mL), and then the mixture was stirred at room temperature for 2.5 h. The reaction mixture was neutralized with Dowex 50WX8 (H⁺), and then the resin was separated. The filtrate was washed with EtOAc. The water layer was concentrated under reduced pressure to give 11 as a colorless solid (94.9 mg, quant). ¹H-NMR (CD₃OD) δ: 3.72 (1H, dd, J = 10.8 Hz, 3.6 Hz), 3.61 (1H, dd, J = 10.8 Hz, 6.0 Hz), 3.27–3.18 (2H, m), 2.92 (1H, s), 2.02 (1H, dd, J = 13.3 Hz, 3.8 Hz), 1.88–1.82 (1H, m), 1.44 (1H, t, J = 13.3 Hz); ¹³C-NMR (CD₃OD) δ: 85.1, 79.8, 77.8, 76.3, 74.6, 72.2, 63.8, 42.0, 39.1; HR-MS (ESI) m/z: 225.0725 [M + Na]⁺ (Calcd for C₉H₁₄NaO₅: 225.0739).

4-Ethylbenzylazide (12b)

Et₃N (1.01 mL, 7.26 mmol) and methanesulfonyl chloride (560 µL, 7.26 mmol) were added to a solution of 4-ethylbenzyl alcohol²⁴⁾ (761 mg, 5.59 mmol) prepared from 4-ethylbenzaldehyde in DMF (4.5 mL) at 0 °C, and the mixture was stirred at room temperature for 18 h. Et₃N (1.01 mL, 7.26 mmol) and methanesulfonyl chloride (560 µL, 7.26 mmol) were added to the reaction mixture at 0 °C, and the resulting mixture was stirred at room temperature for 24 h. After saturated NaHCO₃ aqueous solution was added the reaction mixture, the mixture was partitioned between EtOAc and H₂O. The organic layer was washed with H₂O, saturated NaHCO₃ aqueous solution and brine, dried over Na₂SO₄, and concentrated under reduced pressure. After the residue was dissolved in DMF (8.6 mL), sodium azide (1.11 g, 17.1 mmol) was added to the solution. The resulting mixture was stirred at room temperature for 15 h. After saturated NaHCO₃ aqueous solution was added the reaction mixture, the mixture was partitioned between EtOAc and H₂O. The organic layer was washed with H₂O and brine, dried over Na₂SO₄, and concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane/EtOAc, 10 : 1) to give 12b as a colorless oil (585 mg, 65%). ¹H-NMR (CDCl₃) δ: 7.30–7.26 (4H, m), 4.35 (2H, s), 2.71 (2H, q, J = 6.1 Hz), 1.30 (3H, t, J = 6.1 Hz); ¹³C-NMR (CDCl₃) δ: 144.4, 132.5, 128.3(2), 54.6, 28.5, 15.5; HR-MS (ESI) m/z: 162.1055 [M + H]⁺ (Calcd for C₉H₁₂N₃: 162.1031).

4-Cyclopropylbenzylazide (12c)

Et₃N (767 µL, 5.50 mmol) and methanesulfonyl chloride (424 µL, 5.50 mmol) were added to a solution of 4-cyclopropylbenzyl alcohol²⁵⁾ (627 mg, 4.23 mmol) prepared from 4-bromobenzyl alcohol in DMF (3.4 mL) at 0 °C, and the mixture was stirred at room temperature for 18 h. After saturated NaHCO₃ aqueous solution was added the reaction mixture, the mixture was partitioned between EtOAc and H₂O. The organic layer was washed with H₂O, saturated NaHCO₃ aqueous solution and brine, dried over Na₂SO₄, and concentrated under reduced pressure. After the residue was dissolved in DMF (7.3 mL), sodium azide (1.29 g, 19.9 mmol) was added to the solution. The resulting mixture was stirred at room temperature for 18 h. After saturated NaHCO₃ aqueous solution was added the reaction mixture, the mixture was partitioned between EtOAc and H₂O. The organic layer was washed with H₂O and brine, dried over Na₂SO₄, and concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (n-hexane) to give 12c as a colorless oil (568 mg, 77%). ¹H-NMR (CDCl₃) δ: 7.20 (2H, d, J = 8.2 Hz), 7.08 (2H, d, J = 8.2 Hz), 4.28 (2H, s), 1.94–1.87 (1H, m), 1.00–0.95 (2H, m), 0.72–0.6 (2H, m); ¹³C-NMR (CDCl₃) δ: 143.4, 135.3, 127.9, 125.6, 71.7, 15.2, 9.2; HR-MS (ESI) m/z: 174.1062 [M + H]⁺ (Calcd for C₁₀H₁₂N₃: 174.1031).

General Procedure for CuAAC Reaction between 1-Ethynyl-carbaglucose and Azide for Synthesis of 7 and 8

CuSO₄ (40 mM in H₂O, 249 µL, 10 µmol) and sodium ascorbate (100 mM in H₂O, 100 µL, 10 µmol) were added to a solution of α- or β-1-ethynyl-carbaglucose (20.2 mg, 100 µmol) and azide 12 (500 µmol) in DMF (370 µL), and then the mixture was stirred at room temperature. The reaction mixture was concentrated under reduced pressure, the resulting residue was purified by column chromatography on silica gel (CHCl₃/MeOH, 30 : 1–1 : 1) to give 7 or 8.

(1R,2R,3S,4R,5R)-1-[1-Benzyl-1H-1,2,3-triazol-4-yl]-5-hydoxymethylcyclohexane-1,2,3,4-tetraol (7a)

90%; ¹H-NMR (CD₃OD) δ: 7.85 (1H, s), 7.37–7.30 (5H, m), 5.55 (2H, s), 3.73–3.56 (3H, m), 3.39–3.29 (2H, m), 2.07–1.81 (3H, m); ¹³C-NMR (CD₃OD) δ: 155.3, 136.8, 130.0, 129.6, 129.2, 123.9, 78.1, 77.2, 74.9, 73.8, 64.2, 54.9, 40.5, 38.1; HR-MS (ESI) m/z: 358.1367 [M + Na]⁺ (Calcd for C₁₆H₂₁NaN₃O₅: 358.1379).

(1R,2R,3S,4R,5R)-1-[1-(4-Ethylbenzyl)-1H-1,2,3-triazol-4-yl]-5-hydoxymethylcyclohexane-1,2,3,4-tetraol (7b)

quant; ¹H-NMR (CD₃OD) δ: 7.80 (1H, s), 7.26 (2H, d, J = 8.2 Hz), 7.20 (2H, d, J = 8.2 Hz), 5.51 (2H, s), 3.78–3.57 (3H, m), 3.38–3.26 (2H, m), 2.62 (2H, q, J = 7.6 Hz), 2.00–1.81 (2H, m), 1.19 (3H, t, J = 7.6 Hz); ¹³C-NMR (CD₃OD) δ: 155.2, 146.1, 134.0, 129.5, 129.4, 123.7, 78.0, 77.2, 74.9, 73.7, 64.1, 54.8, 40.5, 38.1, 29.5, 16.1; HR-MS (ESI) m/z: 386.1684 [M + Na]⁺ (Calcd for C₁₈H₂₅NaN₃O₅: 386.1692).

(1R,2R,3S,4R,5R)-1-[1-(4-Cyclopropylbenzyl)-1H-1,2,3-triazol-4-yl]-5-hydoxymethylcyclohexane-1,2,3,4-tetraol (7c)

quant; ¹H-NMR (CD₃OD) δ: 7.77 (1H, s), 7.22 (2H, d, J = 8.4 Hz), 7.06 (2H, d, J = 8.4 Hz), 5.49 (2H, s), 3.73–3.56 (4H, m), 3.38–3.33 (1H, m), 2.05–1.80 (4H, m), 0.97–0.93 (2H, m), 0.66–0.63 (2H, m); ¹³C-NMR (CD₃OD) δ: 155.2, 146.1, 133.6, 129.4, 127.1, 123.6, 78.1, 77.2, 75.0, 73.7, 64.2, 54.7, 40.5, 38.1, 15.9, 9.8; HR-MS (ESI) m/z: 398.1675 [M + Na]⁺ (Calcd for C₁₉H₂₅NaN₃O₅: 398.1692).

(1S,2R,3S,4R,5R)-1-[1-Benzyl-1H-1,2,3-triazol-4-yl]-5-hydoxymethylcyclohexane-1,2,3,4-tetraol (8a)

quant; ¹H-NMR (CD₃OD) δ: 8.07 (1H, s), 7.38–7.30 (5H, m), 5.58 (2H, s), 3.71–3.61 (2H, m), 3.54–3.47 (1H, m), 3.36–3.25 (2H, m), 2.56–2.52 (1H, m), 1.75–1.70 (1H, m), 1.64–1.58 (1H, m); ¹³C-NMR (CD₃OD) δ: 136.9, 130.0, 129.5, 129.2, 125.8, 80.8, 77.4, 75.2, 73.3, 71.5, 64.0, 54.8, 41.5, 37.7; HR-MS (ESI) m/z: 358.1366 [M + Na]⁺ (Calcd for C₁₆H₂₁NaN₃O₅: 358.1379).

(1S,2R,3S,4R,5R)-1-[1-(4-Ethylbenzyl)-1H-1,2,3-triazol-4-yl]-5-hydoxymethylcyclohexane-1,2,3,4-tetraol (8b)

94%; ¹H-NMR (CD₃OD) δ: 8.03 (1H, s), 7.25 (2H, d, J = 8.2 Hz), 7.19 (2H, d, J = 8.2 Hz), 5.52 (2H, s), 3.70–3.61 (2H, m), 3.50 (1H, d, J = 9.6 Hz), 3.36–3.24 (2H, m), 2.62 (2H, q, J = 7.6 Hz), 2.54 (1H, dd, J = 13.2 Hz, 3.4 Hz), 1.76–1.68 (1H, m), 1.60 (1H, t, J = 13.2 Hz), 1.20 (3H, t, J = 7.6 Hz); ¹³C-NMR (CD₃OD) δ: 146.0, 134.1, 129.5, 129.3, 125.7, 80.7, 77.3, 75.2, 73.3, 71.3, 64.0, 54.6, 41.5, 37.7, 29.5, 16.1; HR-MS (ESI) m/z: 386.1673 [M + Na]⁺ (Calcd for C₁₈H₂₅NaN₃O₅: 386.1692).

(1S,2R,3S,4R,5R)-1-[1-(4-Cyclopropylbenzyl)-1H-1,2,3-triazol-4-yl]-5-hydoxymethylcyclohexane-1,2,3,4-tetraol (8c)

88%; ¹H-NMR (CD₃OD) δ: 8.02 (1H, s), 7.21 (2H, d, J = 5.0 Hz), 7.06 (2H, d, J = 5.0 Hz), 5.50 (2H, s), 3.70–3.47 (3H, m), 3.35–3.23 (2H, m), 2.56–2.52 (1H, m), 1.92–1.85 (1H, m), 1.76–1.67 (1H, m), 1.63–1.57 (1H, m), 0.97–0.93 (2H, m), 0.67–0.63 (2H, m); ¹³C-NMR (CD₃OD) δ: 146.0, 133.6, 129.2, 127.0, 125.7, 80.7, 77.3, 75.2, 73.3, 71.3, 64.0, 54.6, 41.5, 37.7, 15.9, 9.9; HR-MS (ESI) m/z: 398.1668 [M + Na]⁺ (Calcd for C₁₉H₂₅NaN₃O₅: 398.1692).

X-ray Crystallographic Structure Determination for 24

X-ray diffraction data for 24 was collected using a Rigaku Saturn 724 CCD diffractometer with Mo-Kα radiation (l = 0.71073 Å) at 123 K. Single crystals (size: 0.10 × 0.10 × 0.05 mm³) of 24 (C₃₄H₃₆O₉, Mw = 588.63) suitable for X-ray analysis were grown by the recrystallization of a solution of 24 in CH₂Cl₂/n-hexane (1 : 5) at –20 °C for approximately 2 weeks. The unit cell was monoclinic with the space group P2₁. Lattice constants with Z = 4, ρ_calcd = 1.307 g/cm³, μ = 0.094 cm⁻¹, F(000) = 1248, θ_max = 27.500° were a = 14.7668(8), b = 10.4299(6), c = 19.5056(11) Å, α = 90°, β = 95.101(5)°, γ = 90°, and V = 2992.3(3) ų. A total of 41,986 reflections were collected, of which 13,713 reflections were independent (R_int = 0.0831). The structure was refined to final R₁ = 0.0640 for 8,218 data [I > 2σ(I)] with 787 parameters and wR₂ = 0.1409 for all data, GOF = 1.000, and residual electron density max./min. = 0.282/−0.222 e·Å⁻³. The ORTEP diagram is shown in Supplementary Fig. S1, and the crystal data and structure refinement are listed in Supplementary Table S1. Data collection, cell refinement, and data reduction were conducted using the CrystalClear-SM Expert 2.0 software.²⁶⁾ The structure was solved by direct methods using the program SHELXT²⁷⁾ and refined by full-matrix least-squares methods on F² using SHELXL2014.²⁸⁾ All materials for publication were prepared by Yadokari-XG 2009 software.²⁹⁾ All non-hydrogen atoms were refined anisotropically. Tables of positional and thermal parameters, bond lengths and angles, torsion angles and Figs. may be found from the Cambridge Crystallographic Centre by referencing CCDC number 2182979.

Molecular Docking Study

Docking investigations, protein and chemical preparations were carried out in accordance with earlier descriptions.^30,31) The compounds’ 2 dimensional (2D) structures were imported and 3D optimized by LigPrep software using the OPLS2005 force field. The protein preparation wizard in the Maestro suite (Schrodinger LLC, NY, U.S.A.) was used to process and optimize the SGLT2 structure (PDB ID 7vsi). The structure was checked, protonated, optimized at physiological pH and corrected for missing atoms or side chains by using Prime software implemented in the Schrodinger suite. The OPLS2005 force field was used for energy minimization. All docking and molecular dynamics calculations in this investigation were based on the generated structure. The standard precision docking approach was used to dock all compounds (SP docking). In the docking grid, the co-crystallized ligand, empagliflozin (4), functioned as the center of a 20 Å docking box that ringed the bound ligand. Redocking of the co-crystalized ligand, 4 was employed as a quality check for docking accuracy. The finding of accurate docking of 4 at the beginning of docking operations supported the accuracy of docking protocol.

Establishment of hSGLT2 Stable Cell Line

hSGLT2 cDNA (pFN21AE5578) was purchased from Kazusa DNA Research Institute. Gene fragments for the creation of full-length hSGLT2 were synthesized on order and In-Fusion technology was used to construct a full-length hSGLT2 gene expression vector. PCR was also performed to add a Flag tag to the C-terminus of hSGLT2, which was finally assembled into the pF5A-CMV-neo vector. The construct was confirmed by sequencing. HEK293 cell line was maintained as an adherent culture in Dulbecco’s modified Eagle’s medium (D-MEM) supplemented with 10% fetal bovine serum (FBS). The cells were transfected with the plasmid pF5A-CMV-neo-SGLT2-Flag using polyethylenimine-based transfection method. One day after transfection, cells were transferred into fresh D-MEM medium containing 800 µg/mL geneticin and the medium was replaced every three days. Two weeks later, ten cell clones were separately expanded into two 10 cm dishes. Cells in dish were harvested during the exponential growth phase and frozen in 10% DMSO for storage. HEK293 cells stably overexpressing hSGLT2 (hSGLT2/HEK293) were cultured at 37 °C in a humidified atmosphere of 5% CO₂ in air in D-MEM supplemented with 10% FBS.

Fluorescent SGLT2 Assay

hSGLT2/HEK293 were washed twice with the pretreatment buffer (10 mM N-(2-hydroxyethyl)piperazine-N′-2-ethanesulfonic acid (HEPES), 140 mM choline chloride, 2 mM KCl, 1 mM CaCl₂, and 1 mM MgCl₂, pH 7.4, pH was adjusted with Tris), then incubated with the uptake buffer (10 mM HEPES, 140 mM NaCl, 2 mM KCl, 1 mM CaCl₂, and 1 mM MgCl₂, pH 7.4, pH was adjusted with Tris). The cells were incubated with 25 mM 1-azido-1-deoxy-D-glucose (25) in the uptake buffer in the presence of the indicated amount of 3, 7 or 8 in DMSO or DMSO as a control at 37 °C for 2 h. After removing the culture medium, the cells were washed twice with ice-cold washing buffer (the pretreatment buffer containing 100 µM phlorizin) to remove the un-transported 25. Cells were then treated with sodium dodecyl sulfate (SDS) lysis buffer containing CuAAC-activated fluorogenic alkyne probe and CuAAC reagents (100 µM 4-ethynyl-N-ethyl-1,8-naphthalimide, 250 µM tris[(1-ethoxycarbonylmethyl-1H-1,2,3-triazol-4yl)methyl]-amine, 2.5 mM CuSO₄·5H₂O, 5 mM sodium ascorbate, 0.1% SDS in phosphate buffered saline (PBS)) and incubated at 37 °C for 18 h to afford the desired fluorescence adduct. After a certain amount of each well was centrifuged (15000 rpm × 10 min, 24 °C), the fluorescence intensity of the generated fluorescently labeled triazolyl glycan was measured by a multimode plate reader (TriStar LB941, Berthold Technologies) with an excitation/emission wavelength of 355 nm/460 nm. The percent of inhibition was calculated by comparing fluorescence intensity. The results were confirmed by at least three independent experiments with two cultures each and expressed as an average of mean ± standard deviation (S.D.) from three experiments.

Molecular Dynamics Simulation Study

Molecular dynamics simulations lasting 100 ns were carried out using the Desmond software, Schrödinger LLC (New York, U.S.A.). The system builder tool was used to set up each system. The protein was introduced into a DPPC membrane at 325 K. The hSGLT2 structures embedded in the membrane were solvated in a cubic box of water of the orthorhombic solvent model (transferable intermolecular interaction potential 3 points; TIP3P). The simulation made use of the OPLS 2005 force field. Counter ions were used to neutralize the models. One hundred fifty millimolar NaCl was used to mimic physiological circumstances. One atm pressure and 300 K temperature were maintained for the whole simulation in the isothermal-isobaric ensemble. To minimize the systems, 5000 steepest decline steps were followed by progressive heating from 0 to 300 K. For five ns of temperature relaxation, the Nosé-Hoover Chain thermostat method^32–34) and the Martyna-Tobias-Klein barostat method³⁵⁾ for 5 ns of pressure relaxation were used. Coordinates were collected every 100 ps to generate trajectories of 1000 frames. RMSD, RMSF, hydrogen bonding between proteins and bonds distances were studied in the trajectory.

Acknowledgments

This work was supported in part by financial support from Suzuken Memorial Foundation. We also acknowledge the Division of Instrumental Analysis and the Division of Genomics Research, Life Science Research Center, Gifu University, for maintaining the instruments used in this study.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

References

• 1) Kanai Y., Lee W. S., You G., Brown D., Hediger M. A., J. Clin. Invest., 93, 397–404 (1994).
• 2) Gorboulev V., Schürmann A., Vallon V., et al., Diabetes, 61, 187–196 (2012).
• 3) Rieg T., Masuda T., Gerasimova M., Mayoux E., Platt K., Powell D. R., Thomson S. C., Koepsell H., Vallon V., Am. J. Physiol. Renal Physiol., 306, F188–F193 (2014).
• 4) López G. P., Albarrán O. G., Megías M. C., Nefrologia, 30, 618–625 (2010).
• 5) Chao E. C., Henry R. R., Nat. Rev. Drug Discov., 9, 551–559 (2010).
• 6) Ni L., Yuan C., Chen G., Zhang C., Wu X., Cardiovasc. Diabetol., 19, 98 (2020).
• 7) Ohtake Y., Sato T., Matsuoka H., Kobayashi T., Nishimoto M., Taka N., Takano K., Yamamoto K., Ohmori M., Higuchi T., Murakata M., Morikawa K., Shimma N., Suzuki M., Hagita H., Ozawa K., Yamaguchi K., Kato M., Ikeda S., Bioorg. Med. Chem., 20, 4117–4127 (2012).
• 8) Ng W.-L., Li H.-C., Lau K.-M., Chan A. K. N., Lau C. B.-S., Shing T. K. M., Sci. Rep., 7, 5581 (2017).
• 9) Li L.-T., Zhou L.-F., Li Y.-J., Huang J., Liu R.-H., Wang B., Wang P., Bioorg. Med. Chem. Lett., 22, 642–644 (2012).
• 10) Shiraishi T., Kitamura Y., Ueno Y., Kitade Y., Chem. Commun., 47, 2691–2693 (2011).
• 11) Kitamura Y., Edayoshi K., Kitade Y., Tetrahedron Lett., 53, 6987–6989 (2012).
• 12) Ren Q., Tsunaba K., Kitamura Y., Nakashima R., Shibata A., Ikeda M., Kitade Y., Bioorg. Med. Chem. Lett., 24, 1519–1522 (2014).
• 13) Kitamura Y., Taniguchi K., Maegawa T., Monguchi Y., Kitade Y., Sajiki H., Heterocycles, 77, 521–532 (2009).
• 14) Satz A. L., Brunschweiger A., Flanagan M. E., Gloger A., Hansen N. J. V., Kuai L., Kunig V. B. K., Lu X., Madsen D., Marcaurelle L. A., Mulrooney C., O’Donovan G., Sakata S., Scheuermann J., Nat. Rev. Methods Primers, 2, 3 (2022).
• 15) Benz M., Asperger A., Hamester M., Welle A., Heissler S., Levkin P. A., Nat. Commun., 11, 5391 (2020).
• 16) Lesuisse D., Berchtold G. A., J. Org. Chem., 50, 888–890 (1985).
• 17) Shing T. K. M., Tang Y., J. Chem. Soc. Chem. Commun., 748–749 (1990).
• 18) Angyal S. J., Tate M. E., J. Chem. Soc., 6949–6955 (1965).
• 19) Guidot J. P., Gall T. L., Tetrahedron Lett., 34, 4647–4650 (1993).
• 20) Niu Y., Liu R., Guan C., Zhang Y., Chen Z., Hoerer S., Nar H., Chen L., Nature (London), 601, 280–284 (2022).
• 21) Bhattacharya S., Rathore A., Parwani D., Mallick C., Asati V., Agarwal S., Rajoriya V., Das R., Kashaw S. K., Eur. J. Med. Chem., 204, 112523 (2020).
• 22) Wu S.-H., Yao C.-H., Hsieh C.-J., Liu Y.-W., Chao Y.-S., Song J.-S., Lee J.-C., Eur. J. Pharm. Sci., 74, 40–44 (2015).
• 23) Arthurs C. L., Morris G. A., Piacenti M., Pritchard R. G., Stratford I. J., Tatic T., Whitehead R. C., Williams K. F., Wind N. S., Tetrahedron, 66, 9049–9060 (2010).
• 24) Giraud F., Loge C., Pagniez F., Crepin D., Barres S., Picot C., Le Pape P., Le Borgne M., J. Enzyme Inhib. Med. Chem., 24, 1067–1075 (2009).
• 25) Yao C.-H., Song J.-S., Chen C.-T., Yeh T.-K., Hsieh T.-C., Wu S.-H., Huang C.-Y., Huang Y.-L., Wang M.-H., Liu Y.-W., Tsai C.-H., Kumar C. R., Lee J.-C., Eur. J. Med. Chem., 55, 32–38 (2012).
• 26) CrystalClear-SM Expert 2.0: Rigaku Corporation, Tokyo, Japan, 2011.
• 27) Sheldrick G. M., Acta Crystallogr. A, 71, 3–8 (2015).
• 28) Sheldrick G. M., Acta Crystallogr. C, 71, 3–8 (2015).
• 29) Kabuto C., Akine S., Nemoto T., Kwon E., J. Cryst. Soc. Jpn., 51, 218–224 (2009).
• 30) Burayk S., Oh-hashi K., Kandeel M., Pharmaceuticals, 15, 282 (2022).
• 31) Kandeel M., Kim J., Fayez M., Kitade Y., Kwon H.-J., PeerJ, 10, e12929 (2022).
• 32) Nosé S., J. Chem. Phys., 81, 511–519 (1984).
• 33) Hoover W. G., Phys. Rev. A, 31, 1695–1697 (1985).
• 34) Evans D. J., Holian B. L., J. Chem. Phys., 83, 4069–4074 (1985).
• 35) Martyna G. J., Tuckerman M. E., Tobias D. J., Klein M. L., Mol. Phys., 87, 1117–1157 (1996).