Chemical and Pharmaceutical Bulletin
Online ISSN : 1347-5223
Print ISSN : 0009-2363
ISSN-L : 0009-2363
Regular Articles
Development of 1,3,6-Tribenzoylated Glucose as an Antiausterity Agent Targeting Tumor Microenvironment
Akira TakagiKazuki UsuguchiIppei TakashimaKensuke Okuda
Author information
JOURNAL FREE ACCESS FULL-TEXT HTML
Supplementary material

2023 Volume 71 Issue 8 Pages 641-649

Details
Abstract

One aspect of cancer-specific environments, nutrient starvation, is a factor in cancer cell resistance to treatment with chemotherapeutic agents and development of malignancy. Our newly synthesized novel glucose derivative β-1,3,6-O-tribenzoyl-D-glucose (3) showed preferential cytotoxicity against PANC-1 human pancreatic cancer cells as well as HT-29 human colon cancer cells depending on low nutritional environment. The amount of ester functionalization in 3 is important. None of the mono- and tetrabenzoylated D-glucose analog showed cytotoxicity, and dibenzoylated D-glucoses showed only limited cytotoxicity. Fluorescence imaging with double staining of Hoechst 33342 and propidium iodide clearly showed that 3 actually causes cell death in a nutrient deprived medium. We thus demonstrate that an inexpensive natural product, D-glucose, is a unique template for attachment of acyl moieties to target tolerance to nutrient starvation. We expect these compounds will lead to additional compounds to treat refractory cancers by diversification of chemically modified glucose.

Introduction

Cancer is one of the major causes of death, and the number of cancer patients worldwide is expected to increase year by year.1) Therefore, prevention, early detection, and early treatment of cancer are important issues for human health and life maintenance. In recent years, many cancer treatments have been developed and various standard therapies have been implemented, making more cancers potentially treatable. On the other hand, treatment for malignant cancers that do not respond to standard therapies has not yet been established. Recently, the relationship between tumor malignancy and its microenvironment has gained much attention. The tumor microenvironment is known as the cancer-specific environment widely found in solid tumors, characterized by harsh conditions with hypoxia, nutrient starvation, and acidic pH.25) Cancer cells that have adapted to such an environment often gain resistance to radiation and/or drugs, and many treatments are often ineffective.68)

In recent years, new therapeutic strategies targeting an aspect of such tumor microenvironments have been identified. The concept of “antiausterity” has focused on methods to selectively attack cancers under nutrient deprived conditions. Various antiausterity agents have been discovered from natural products and their derivatives, and it has been reported that they are selectively toxic to cancer cell lines that have acquired resistance to conventional anticancer drugs under nutrient deprived conditions.918) To accelerate an antiausterity strategy to treat malignant tumor based on a limited supply of natural products, their availability through synthetic chemistry is necessary. In addition, preparation of various derivatives also helps in identification of pharmacophores that in turn can lead to new antiausterity agents with simple structures that are easier to prepare.

Inspired by previous research on antiausterity natural products, we have achieved the first racemic total synthesis of antiausterity (−)-uvaridacol L ((−)-1) and found that (±)-1 showed similar preferential cytotoxicity against PANC-1 human pancreatic cancer cells in a nutrient deprived medium (NDM) as did the natural product (−)-119) (Fig. 1). (±)-1 also showed preferential cytotoxicity against HT-29 human colon cancer cells in a NDM, implying that (±)-1 is effective for any cancer cells that are under nutrient deprivation. We also found that one of our synthetic intermediates, mono-debenzoylated derivative ((±)-2), had diminished antiausterity property compared to (±)-1. Since three benzoyl esters and two hydroxy groups are a feature of 1, we reasoned that a molecule having these functional groups arranged similarly to the configuration of 1 could have antiausterity properties analogous to 1. Based on this hypothesis, we focused on the apparent similarity of the conformation of the core structure of (−)-1 to that of β-D-glucose as a synthetic scaffold.20) We thus felt that glucose with benzoyl groups at the β-1, 3, and 6 positions could lead to the discovery of new easily attainable antiausterity agents. Since many synthetic studies of glucose derivatives have been developed,21) we expected that a wide range of derivatives can be synthesized from D-glucose and that structure–activity relationship studies can be conducted.

Fig. 1. Structure of (–)-Uvaridacol L (1) and Its 5-Hydroxymethyl Analog ((±)-2)

Results and Discussion

Strategy and Design of Target Compounds

β-1,3,6-O-Tribenzoyl-D-glucose (3), which has a stereo structure and substitution pattern similar to the antiausterity natural product (−)-uvaridacol L ((−)-1), was designed as a candidate of a new antiausterity compound (Fig. 2). To confirm the importance of the number of benzoyl groups, we also synthesized various monobenzoylated (46), dibenzoylated (79), and tetrabenzoylated compounds (10 and 11), based on 3, respectively, and investigated the correlation between structure and activity. In addition, enantiopure (−)-1 was prepared for an exact comparison with chiral 3 that had the same stereostructure.

Fig. 2. Structure of β-1,3,6-O-Tribenzoyl-D-glucose (3), Less Benzoylated Analogs (49), and Excess Benzoylated Analogs (10 and 11)

Synthesis of Glucose Derivatives and Optical Resolution of (±)-1

The syntheses of the desired β-1,3,6-O-tribenzoyl-D-glucose (3) and the corresponding monobenzoyl, dibenzoyl, and tetrabenzoyl derivatives are shown below. The synthesis was carried out by successive conversion reactions using readily available 1,2,5,6-di-O-isopropylidene-α-D-glucofuranose (12) and α-D-glucopyranose (13) as starting materials.

The synthesis of glucose derivatives from 12 is shown in Chart 1. Benzoylation of the hydroxy group at the 3-position of 12 produced 14.22) Removal of the two acetonide protecting groups gave 3-O-benzoyl-D-glucose (5). The acetonide at 5,6-position of 14 was removed with aqueous acetic acid to afford 15,22,23) and the resultant primary hydroxy group was selectively benzoylated to give 16.24) The acetonide at 1,2-position of 16 was removed with aqueous trifluoroacetic acid (TFA) to give 3,6-O-dibenzoyl-D-glucose (9). The triol 17 was obtained by removal of the acetonides at the 5,6-position of 12,23) and 18 was synthesized by selective benzoylation of the resulting primary hydroxy group.24) The acetonide at 1,2-positions of 18 was also deprotected by using aqueous TFA to give 6-O -benzoyl-D-glucose (6).

Chart 1. Syntheses of Benzoylated Glucoses (5, 6, and 9) Starting from 1,2,5,6-Di-O-isopropylidene-α-D-glucofuranose (12)

The other synthesis of glucose derivatives from 13 is shown in Chart 2. β-1-O-Benzoyl-D-glucose (4) was obtained via β-selective Mitsunobu benzoylation25) of 13 with benzoic acid. Further benzoylation was then conducted with 2,6-lutidine as a solvent to obtain 6-selective benzoylated β-1,6-O-dibenzoyl-D-glucose (8). The reaction of 4 with 3 equivalents of benzoyl chloride gave β-1,3,6-O-tribenzoyl-D-glucose (3), β-1,2,3,6-O-tetrabenzoyl-D-glucose (10), and β-1,3,4,6-O-tetrabenzoyl-D-glucose (11) at 28, 39, and 7%, respectively. Structures of 3, 10, and 11 were verified by H–H correlation spectroscopy (COSY) NMR spectra. Protection of the 4,6-position of 4 by acetonide followed by benzoylation produced 20 which was benzoylated at the 3-position. At this step, possible benzoylation at 2-position was excluded by its H–H COSY spectrum of 20. Subsequent deprotection of the acetonide of 20 resulted in β-1,3-O-dibenzoyl-D-glucose (7).

Chart 2. Syntheses of Benzoylated Glucoses (3, 4, 7, 8, 10, and 11) Starting from α-D-Glucopyranose (13)

Finally, optical resolution was performed to obtain the natural enantiomer of (±)-1. After (±)-1 was passed through chiral column, (+)-1 and (−)-1 were separated. While the literature used circular dichroism spectrum to identify the enantiomer,14) the natural enantiomer (−)-1 was determined by measuring the optical rotation.

Cell-Based Assay for Evaluation as Antiausterity Agents

Selective cytotoxicity against malignant cells that have adapted to a nutrient-deprived tumor microenvironment is the essential property of antiausterity agents. We performed a WST-8 assay to evaluate preferential cytotoxicity of the various compounds synthesized against PANC-1 cells. PANC-1 cells are known to be highly resistant to nutrient starvation and have resistance to a conventional anticancer drug. Preferential cytotoxicity among Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (+FBS), FBS free DMEM (−FBS), FBS, sodium pyruvate, and glucose free DMEM (−FBS, −SP, and −Glc), and nutrient deprived medium (NDM) was investigated for the various compounds synthesized to evaluate their property to target tumor microenvironment. Using these four media from which nutrients were removed in a stepwise manner, we could analyze their antiausterity properties linked with nutrient-deficiency level.

First, we confirmed that a conventional anticancer drug gemcitabine was ineffective for 24 h in all four media (Table 1). Then, the cytotoxicity of 3 for 24 h was evaluated under nutrient starvation. This appeared in a dose dependent manner with an IC50 in NDM of 30.2 µM while IC50s were 37.2 µM in DMEM (−FBS, −SP, & −Glc), 63.0 µM in DMEM (−FBS), and 152.9 µM in DMEM (+FBS), (Table 1, Fig. 3), correlated with nutrient deprivation status. In the case of (−)-1, values were 14.2, 24.6, 30.4, and 81.8 µM, respectively. Although 3 was slightly less potent than (−)-1, the preferential cytotoxicity of 3 was comparable to (±)-1, which was consistent with the aim of this study. It is noteworthy that simple benzoylated glucose 3, that is easily obtainable just in two steps from α-D-glucopyranose (13), has comparable activity to the less readily available natural product (−)-1. On the other hand, the various monobenzoylated (4, 5, and 6) and tetrabenzoylated compounds (10 and 11) were not toxic in any media up to 1 mM, 500 or 100 µM, respectively depending on their aqueous solubility (Table 1). The dibenzoyl forms (7, 8, and 9) were toxic only at high concentrations (Table 1, Fig. 3). These results correspond well with the fact that 2 with two benzoyl moieties was not potent compared to 1 with three benzoyl moieties.19) These results also indicate that the preferential cytotoxicity of the novel glucose derivative 3 depends on appropriate benzoylation status of D-glucose. Too much benzoylation as well as too little benzoylation degrades its potency.

Table 1. Cytotoxicity (IC50, µM) of Compounds (−)-1, 3 − 11 and Gemcitabine on PANC-1 Cells
Compound+FBS−FBS−FBS, −SP, & −GlcNDM
(−)-181.8 ± 10.030.4 ± 2.024.6 ± 5.014.2 ± 1.1
3152.9 ± 7.063.0 ± 7.037.2 ± 14.130.2 ± 5.2
4>1000>1000>1000>1000
5>1000>1000>1000>1000
6>1000>1000>1000>1000
7747.7 ± 42.9583.9 ± 28.9424.6 ± 32.8342.0 ± 128.4
8700.1 ± 96.0630.0 ± 56.7482.9 ± 117.9346.5 ± 13.8
9696.1 ± 44.5556.6 ± 91.6338.9 ± 31.2368.8 ± 62.9
10>500>100>100>100
11>500>100>100>100
Gemcitabine>1000>1000>1000>1000

Each figure indicates concentration ± standard deviation (µM) at which 50% of the cells metabolism was suppressed in each medium determined by WST-8 assay.

Fig. 3. Preferential Cytotoxic Activity of (−)-1, 3, 7, 8, and 9 against PANC-1 Cells in Each Medium Determined by WST-8 Assay

+FBS: DMEM supplemented with 10% FBS, −FBS: FBS free DMEM, −FBS, −SP, and −Glc: FBS, sodium pyruvate, and glucose free DMEM, NDM: nutrient-deprived medium. The results are expressed as the means ± standard error of the mean (n = 3–4).

Next, we targeted HT-29 cells, which are also viable under nutrient starved conditions.26) A WST-8 assay was also performed to evaluate preferential cytotoxicity in RPMI-1640 medium supplemented with 10% FBS (+FBS), FBS free RPMI-1640 medium (−FBS), FBS, sodium pyruvate, and glucose free RPMI-1640 medium (−FBS, −SP, and −Glc), and NDM of the various compounds synthesized against HT-29 cells. The IC50 trend of these compounds as well as gemcitabine and (−)-1 against HT-29 cells (Table 2, Fig. 4) was the same as that against PANC-1 cells (Table 1, Fig. 3), a result also consistent with the aim of this study.

Table 2. Cytotoxicity (IC50, µM) of Compounds (−)-1, 311 and Gemcitabine on HT-29 Cells
Compound+FBS−FBS−FBS, −SP, and −GlcNDM
(−)-166.0 ± 12.023.3 ± 9.419.2 ± 6.313.3 ± 3.7
3118.0 ± 7.033.0 ± 1.132.2 ± 1.723.3 ± 3.1
4>1000>1000>1000>1000
5>1000>1000>1000>1000
6>1000>1000>1000>1000
7>1000>1000600.5 ± 21.6686.0 ± 6.5
8>1000>1000679.4 ± 44.6412.0 ± 60.4
9>1000915.6 ± 62.0711.6 ± 112.8341.8 ± 22.7
10>500>100>100>100
11>500>100>100>100
Gemcitabine>1000>1000>1000>1000

Each figure indicates concentration ± standard deviation (µM) at which 50% of the cells metabolism was suppressed in each medium determined by WST-8 assay.

Fig. 4. Preferential Cytotoxic Activity of (−)-1, 3, 7, 8, and 9 against HT-29 Cells in Each Medium Determined by WST-8 Assay

+FBS: RPMI-1640 medium supplemented with 10% FBS, −FBS: FBS free RPMI-1640 medium, −FBS, −SP, and −Glc: FBS, sodium pyruvate, and glucose free RPMI-1640 medium, NDM: nutrient-deprived medium. The results are expressed as the means ± standard error of the mean (n = 3).

These results of WST-8 assay do not discriminate cytotoxic effects and cytostatic effects, as WST-8 assay indicates cellular metabolism.27) In terms of WST-8 assay, the response of the live cells in which metabolism is suppressed is similar to dead cells. Therefore, morphological assessment and fluorescence imaging with double staining of Hoechst 33342 and propidium iodide (PI) were performed for further evaluation of 3. Hoechst 33342 is a cell-permeable dye that penetrates both live and dead cells and emits blue fluorescence at the cell nucleus, while PI is only capable of penetrating the membrane of dead cells and emits red fluorescence at the cell nucleus. In NDM, the control HT-29 cells without 3 displayed exclusive blue fluorescence with intact cell morphology (Fig. 5: a and c), but no red fluorescence (Fig. 5: e). Treatment with 3 at 70 µM led to a significant amount of cell detachment caused by cell damages, as indicated by an increase in red fluorescence in the observed all cells, and altered cellular morphology such as cell shrinkage and plasma membrane disintegration, indicating total death of HT-29 cells within 24 h (Fig. 5: b, d, and f). These results indicate that 3 has cytotoxic activity, not just cytostatic activity, against HT-29 cells in the NDM medium.

Fig. 5. Morphology of HT-29 Cells Followed by Treatment with 3 (0 or 70 µM) for 24 h and Stained by Hoechst 33342/PI Reagents in NDM

Images were acquired under the phase-contrast and fluorescent modes. (a) Bright Field (BF) image without 3. (b) BF image with 70 µM 3. (c) Hoechst 33342 image without 3. (d) Hoechst 33342 image with 70 µM 3. (e) PI image without 3. (f) PI image with 70 µM 3. Fluorescence channel: for Hoechst 33342, λex = 385/30 nm, λem = 425/30 nm, for PI, λex = 555/30 nm, λem = 592/30 nm. Scale bar: 100 µm.

As the molecular target of (−)-1 has not been clear,14) it is increasingly important to elucidate the mechanism of action of 3 on selective cell death under nutrient deprivation in order to develop even more potent compounds. It is known to activation of Akt is correlated to adaptation to nutrient starvation of PANC-1 cells.6) Actually, some antiausterity agents, such as (−)-arctigenin and biakamide C, were found previously to inhibit the phosphorylation of Akt stimulated by nutrient deprivation in PANC-1 cells.11,13) They are supposed to inhibit the survival of PANC-1 cells under nutrient-deprived conditions through inhibition of the phosphoinositide 3-kinase (PI3K)/Akt/mammalian target of rapamycin (mTOR) signaling pathway associated with autophagy.28) The pathway possibly includes molecular targets of 3. Further investigation of the mechanism of action of 3 is underway.

Conclusion

In this study, we synthesized and evaluated novel glucose derivatives and their analogs that mimic the antiausterity natural product (−)-1. β-1,3,6-O-Tribenzoylated-D-glucose (3), the most similar to (−)-1 in terms of structure, was found to be preferentially cytotoxic to PANC-1 and HT-29 cancer cells in a NDM. On the other hand, other analogs with less benzoylation as well as excess benzoylation were less potent. Benzoylation could affect potency of action by altering the physicochemical parameters of the compound (hydrophobicity, polar surface area, etc.) or through the altered pharmacophore structure that changes interactions with the target. If physicochemical parameters are important, a tribenzoyl-D-glucose with different substitution positions would have similar activities to 3, and if not, pharmacophore affects would be the more important factor. Synthesis with superior step economy without the need for protective groups and evaluation of derivatives of 3 with various acyl groups and triacylated forms with different substitution positions of acyl groups are currently underway to find more active compounds in our laboratory.

Experimental

General Considerations

Reagents

All reactions were carried out under an ambient atmosphere in a round bottom flask containing a stir-bar with a rubber septum except as noted otherwise. Anhydrous pyridine, tetrahydrofuran (THF), dichloromethane (CH2Cl2), and N,N-dimethylformamide (DMF) were purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan) and used without further purification. All other reagents were purchased from Tokyo Chemical Industry Co. (Tokyo, Japan), Kishida Chemical Co. (Osaka, Japan), Nacalai Tesque Inc. (Kyoto, Japan), or FUJIFILM Wako Pure Chemical Corporation and used without further purification. Silicycle Inc. (Quebec, Canada) silica gel (SiliaFlash® F60, 40–63 µm, #R10030B) or Chromatorex PSQ60B (Fuji Silysia Chemical Ltd., Kasugai, Japan) was used for flash chromatography.

Analytical Methods

All reactions were monitored by TLC with E. Merck silica gel 60 F254 pre-coated plates (0.25 mm) and were visualized by UV (254 nm) and/or KMnO4 staining. IR spectra were obtained on a PerkinElmer, Inc. Spectrum One (MA, U.S.A.). 1H- and 13C-NMR spectra were recorded on a JEOL ECZ400S spectrometer (1H: 400 MHz, 13C: 100 MHz) instrument. Chemical shifts are reported in ppm relative to the residual protons of deuterated solvents (CD3OD: 3.31 ppm for 1H, CDCl3: 77.0 ppm, CD3OD: 49.0 ppm, dimethyl sulfoxide (DMSO)-d6: 39.5 for 13C) or the internal standard tetramethylsilane (CDCl3: 0.00 ppm for 1H). The mass spectra were measured on a Thermo Fisher Scientific LTQ Orbitrap Discovery. Melting points were determined with a Yanaco micro melting point apparatus MP-J3. Specific rotations were measured with a JASCO DIP-370 digital polarimeter using the sodium D line and are reported as follows: [α]tD (c = 10 mg/mL, solvent). The optical resolution was performed by Waters Breeze QS HPLC. Yields refer to isolated yields of compounds greater than 95% purity as determined by 1H-NMR analysis. All new products were characterized by 1H-, 13C-NMR, IR, and high resolution (HR)-MS. Every proton at the pyranose or furanose ring of new compounds was determined by H–H COSY. Known compounds (4,25)14,22)15,22,23)17,23) and 1824)) were synthesized according to the literature and 1H-NMR spectra are shown in this report.

Experimental Procedures for Synthesis

(3aR,5R,6S,6aR)-5-((R)-2,2-Dimethyl-1,3-dioxolan-4-yl)-2,2-dimethyltetrahydrofuro[2,3-d][1,3]dioxol-6-yl Benzoate (14)22)

To a solution of 12 (520 mg, 2.00 mmol) in dry pyridine (4.0 mL), benzoyl chloride (280 µL, 2.41 mmol) was added at 0 °C and stirred at room temperature for 20.5 h under Ar atmosphere. The reaction mixture was quenched by the addition of ice and diluted with 1 M aqueous HCl. The mixture was extracted with CHCl3 three times. The combined organic layer was dried over anhydrous Na2SO4 and evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, CHCl3, 100%) to afford 14 (588 mg, 81%) as a colorless solid.

Rf: 0.70 (CHCl3); 1H-NMR (400 MHz, CDCl3) δ: 1.28 (3H, s), 1.33 (3H, s), 1.42 (3H, s), 1.56 (3H, s), 4.08–4.15 (2H, m), 4.32–4.41 (2H, m), 4.64 (1H, d, J = 3.5 Hz), 5.50 (1H, d, J = 3.0 Hz), 5.96 (1H, d, J = 3.5 Hz), 7.47 (2H, t, J = 7.5 Hz), 7.60 (1H, t, J = 7.5 Hz), 8.03 (2H, d, J = 7.5 Hz).

(3R,4S,5R,6R)-2,3,5-Trihydroxy-6-(hydroxymethyl)-tetrahydro-2H-pyran-4-yl Benzoate (5)

Compound 14 (557 mg, 1.53 mmol) was dissolved in 50% (v/v) aqueous trifluoroacetic acid (15 mL) and stirred at room temperature for 31 h. The reaction mixture was evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, CHCl3/MeOH = 4 : 1 v/v) to afford 5 (425 mg, 98%, α/β = 1.3/1) as a colorless solid.

Rf: 0.38 (CHCl3/MeOH = 5 : 1); Melting point: 116–118 °C; 1H-NMR (400 MHz, CD3OD) δ: 3.39 (1/2.3H, dd, J = 8.0, 9.5 Hz, β2), 3.42–3.45 (1/2.3H, m, β5), 3.59–3.90 (8.2/2.3H, m, α2, α4, α6, β4, β6), 3.92 (1.3/2.3H, ddd, J = 2.5, 4.5, 9.5 Hz, α5), 4.63 (1/2.3H, d, J = 8.0 Hz, β1), 5.18 (1.3/2.3H, d, J = 3.5 Hz, α1), 5.19 (1/2.3H, t, J = 9.5 Hz, β3), 5.48 (1.3/2.3H, t, J = 9.5 Hz, α3), 7.45–7.49 (4.6/2.3H, m), 7.57–7.62 (2.3/2.3H, m), 8.07–8.09 (4.6/2.3H, m); 13C-NMR (100 MHz, CD3OD) δ: 62.3, 62.5, 69.8, 69.9, 72.2, 72.9, 74.6, 77.7, 77.8, 79.7, 94.0, 98.2, 129.38, 129.41, 130.7 (2C), 131.7, 131.8, 134.0, 134.1, 167.9, 168.2; IR (KBr) cm−1: 3393, 2940, 1704, 1280, 1129, 1081, 1027, 711; HR-electrospray ionization (ESI)-MS m/z 307.0788 (Calcd for C13H16O7Na [M + Na+]: 307.0788); [α]25D +48 (c = 1.0, MeOH).

(3aR,5R,6S,6aR)-5-((R)-1,2-Dihydroxyethyl)-2,2-dimethyltetrahydrofuro[2,3-d][1,3]dioxol-6-yl Benzoate (15)22)

Compound 14 (728 mg, 2.00 mmol) was dissolved in 60% (v/v) aqueous acetic acid (4.0 mL) and stirred at room temperature for 24 h. The reaction mixture was diluted with toluene and evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, CHCl3 100% to CHCl3/MeOH = 20 : 1 v/v) to afford 15 (606 mg, 93%) as a colorless solid.

Rf: 0.45 (CHCl3/MeOH = 20 : 1); 1H-NMR (400 MHz, CDCl3); δ: 1.35 (3H, s), 1.56 (3H, s), 2.12 (1H, t, J = 6.0 Hz), 3.19 (1H, d, J = 4.0 Hz), 3.71–3.78 (2H, m), 3.85–3.92 (1H, m), 4.30 (1H, dd, J = 2.5, 8.0 Hz), 4.74 (1H, d, J = 4.0 Hz), 5.53 (1H, d, J = 2.5 Hz), 6.02 (1H, d, J = 4.0 Hz), 7.48 (2H, t, J = 8.0 Hz), 7.63 (1H, t, J = 8.0 Hz), 8.04 (2H, d, J = 8.0 Hz).

(3aR,5R,6S,6aR)-5-[(R)-2-(Benzoyloxy)-1-hydroxyethyl]-2,2-dimethyltetrahydrofuro[2,3-d][1,3]dioxol-6-yl Benzoate (16)

To a solution of 15 (324 mg, 1.00 mmol) in dry CH2Cl2 (1.0 mL) and dry pyridine (1.0 mL), benzoyl chloride (128 µL, 1.10 mmol) was added and stirred at room temperature for 15 h under Ar atmosphere. The reaction mixture was diluted with 1 M aqueous HCl. The mixture was extracted with CHCl3 three times. The combined organic layer was dried over anhydrous Na2SO4 and evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, CHCl3 100% to CHCl3/MeOH = 98 : 2 v/v) to afford 16 (368 mg, 86%) as a colorless solid.

Rf: 0.70 (CHCl3/MeOH = 98 : 2); Melting point: 43–45 °C; 1H-NMR (400 MHz, CDCl3) δ: 1.35 (3H, s), 1.55 (3H, s), 3.16 (1H, d, J = 4.0 Hz), 4.06–4.11 (1H, m, 5), 4.43 (1H, dd, J = 2.5, 10.0 Hz, 4), 4.45 (1H, dd, J = 6.0, 12.0 Hz, 6), 4.73 (1H, d, J = 3.5 Hz, 2), 4.74 (1H, dd, J = 2.5, 12.0 Hz, 6), 5.60 (1H, d, J = 2.0 Hz, 3), 6.03 (1H, d, J = 3.5 Hz, 1), 7.42 (2H, t, J = 8.0 Hz), 7.48 (2H, t, J = 8.0 Hz), 7.55 (1H, t, J = 8.0 Hz), 7.62 (1H, t, J = 8.0 Hz) 8.04–8.07 (4H, m); 13C-NMR (100 MHz, CDCl3) δ: 26.3, 26.6, 66.8, 67.2, 76.8, 79.0, 83.1, 105.0, 112.5, 128.3, 128.6, 128.8, 129.7, 129.8, 129.9, 133.1, 133.9, 166.2, 166.9; IR (KBr) cm−1: 3500, 2990, 1723, 1268, 1094, 1025, 710; HR-ESI-MS m/z 451.1359 (Calcd for C23H24O8Na [M + Na+]: 451.1363); [α]25D −29 (c = 1.0, MeOH).

[(2R,3R,4S,5R)-4-(Benzoyloxy)-3,5,6-trihydroxytetrahydro-2H-pyran-2-yl]methyl Benzoate (9)

Compound 16 (109 mg, 0.254 mmol) was dissolved in 50% (v/v) aqueous trifluoroacetic acid (2.5 mL) and stirred at room temperature for 22 h. The reaction mixture was diluted with toluene and evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, CHCl3 100% to CHCl3/MeOH = 98 : 2 v/v) to afford 9 (66.0 mg, 67%, α/β = 2/1) as a colorless solid.

Rf: 0.47 (CHCl3/MeOH = 9 : 1); Melting point: 77–80 °C; 1H-NMR (400 MHz, CD3OD) δ: 3.46 (1/3H, dd, J = 8.0, 9.5 Hz, β2), 3.70–3.82 (6/3H, m, α2, α4, β4, β5), 4.26 (2/3H, ddd, J = 2.0, 5.0, 10.0 Hz, α5), 4.48 (1/3H, dd, J = 5.0, 12.0 Hz, β6), 4.50 (2/3H, dd, J = 5.5, 12.0 Hz, α6), 4.63 (2/3H, dd, J = 2.5, 12.0 Hz, α6), 4.67 (1/3H, dd, J = 2.0, 12.0 Hz, β6), 4.69 (1/3H, d, J = 8.0 Hz, β1), 5.19 (2/3H, d, J = 4.0 Hz, α1), 5.24 (1/3H, dd, J = 9.0, 10.0 Hz, β3), 5.52 (2/3H, t, J = 10.0 Hz α3), 7.46–7.51 (12/3H, m), 7.57–7.64 (6/3H, m), 8.04–8.11 (12/3H, m); 13C-NMR (100 MHz, CD3OD) δ: 65.1, 65.2, 70.1, 70.3, 70.7, 72.2, 74.6, 75.3, 77.5, 79.4, 94.1, 98.3, 129.42, 129.44, 129.6 (2C), 130.6 (2C), 130.8 (2C), 131.3, 131.4, 131.7, 131.8, 134.1, 134.2, 134.3 (2C), 167.83, 167.86, 167.92, 168.1; IR (KBr) cm−1: 3442, 2925, 1709, 1276, 1125, 1070, 1027, 710; HR-ESI-MS m/z 411.1054 (Calcd for C20H20O8Na [M + Na+]: 411.1050); [α]25D +76 (c = 0.5, MeOH).

(R)-1-((3aR,5R,6S,6aR)-6-Hydroxy-2,2-dimethyltetrahydrofuro[2,3-d][1,3]dioxol-5-yl)ethane-1,2-diol (17)23)

Compound 12 (520 mg, 2.00 mmol) was dissolved in 60% (v/v) aqueous acetic acid (7.0 mL) and stirred at room temperature for 17 h. The reaction mixture was diluted with toluene and evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, CHCl3/MeOH = 20 : 1 v/v) to afford 17 (435 mg, 99%) as a colorless solid.

Rf: 0.43 (CHCl3/MeOH = 9 : 1); 1H-NMR (400 MHz, CDCl3 + CD3OD) δ: 1.32 (3H, s), 1.49 (3H, s), 3.68 (1H, dd, J = 6.0, 12.0 Hz), 3.80 (1H, dd, J = 3.5, 12.0 Hz), 3.94–3.98 (1H, m), 4.03 (1H, dd, J = 2.5, 7.0 Hz), 4.30 (1H, d, J = 2.5 Hz), 4.52 (1H, d, J = 3.5 Hz), 5.94 (1H, d, J = 3.5 Hz).

(R)-2-Hydroxy-2-((3aR,5R,6S,6aR)-6-hydroxy-2,2-dimethyltetrahydrofuro[2,3-d][1,3]dioxol-5-yl)ethyl Benzoate (18)24)

To a solution of 17 (220 mg, 1.00 mmol) in dry CH2Cl2 (1.1 mL) and dry pyridine (1.1 mL), benzoyl chloride (115 µL, 0.990 mmol) was added and stirred at room temperature for 22 h under Ar atmosphere. The reaction mixture was diluted with CHCl3 and evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, CHCl3/MeOH = 20 : 1 v/v) to afford 18 (300 mg, 93%) as a colorless solid.

Rf: 0.50 (CHCl3/MeOH = 9 : 1); 1H-NMR (400 MHz, CDCl3) δ: 1.33 (3H, s), 1.49 (3H, s), 3.11 (1H, d, J = 4.0 Hz), 3.19 (1H, d, J = 3.0 Hz), 4.19 (1H, dd, J = 3.0, 6.5 Hz), 4.39 (1H, ddd, J = 3.0, 6.0, 9.2 Hz), 4.44 (1H, t, J = 3.0 Hz), 4.51 (1H, dd, J = 6.0, 12.0 Hz), 4.57 (1H, d, J = 3.5 Hz), 4.70 (1H, dd, J = 3.0, 12.0 Hz), 6.00 (1H, d, J = 3.5 Hz), 7.46 (2H, t, J = 8.0 Hz), 7.59 (1H, tt, J = 1.5, 8.0 Hz), 8.07 (2H, dd, J = 1.5, 8.0 Hz).

((2R,3S,4S,5R)-3,4,5,6-Tetrahydroxytetrahydro-2H-pyran-2-yl)methyl Benzoate (6)

Compound 18 (160 mg, 0.493 mmol) was dissolved in 50% v/v aqueous trifluoroacetic acid (5.0 mL) and stirred at room temperature for 24 h. The reaction mixture was diluted with toluene and evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, CHCl3/MeOH = 9 : 1 v/v) to afford 6 (111 mg, 79%, α/β = 1.5/1) as a colorless solid.

Rf: 0.37 (CHCl3/MeOH = 5 : 1); Melting point: 62–65 °C; 1H-NMR (400 MHz, CD3OD) δ: 3.17 (1/2.5H, dd, J = 8.0, 9.0 Hz, β2), 3.36–3.45 (5/2.5H, m, α2, α4, β3, β4), 3.60–3.64 (1/2.5H, m, β5), 3.71 (1.5/2.5H, t, J = 9.0 Hz, α3), 4.10 (1.5/2.5H, ddd, J = 2.0, 5.0, 10.0 Hz, α5), 4.42 (1/2.5H, dd, J = 5.5, 12.0 Hz, β6), 4.44 (1.5/2.5H, dd, J = 5.0, 12.0 Hz, α6), 4.52 (1/2.5H, d, J = 8.0 Hz, β1), 4.59 (1.5/2.5H, dd, J = 2.0, 12.0 Hz, α6), 4.64 (1/2.5H, dd, J = 2.0, 12.0 Hz, β6), 5.11 (1.5/2.5H, d, J = 4.0 Hz, α1), 7.47 (5/2.5H, t, J = 8.0 Hz), 7.60 (2.5/2.5H, tt, J = 1.5, 8.0 Hz), 8.02–8.04 (5/2.5H, m); 13C-NMR (100 MHz, DMSO-d6) δ: 64.59, 64.64, 69.4, 70.1, 70.5, 72.2, 72.8, 73.6, 74.7, 76.3, 92.4, 97.0, 129.0 (2C), 129.4 (2C), 129.89, 129.95, 133.59, 133.63, 166.0 (2C); IR (KBr) cm−1: 3391, 2925, 1709, 1284, 1057, 712; HR-ESI-MS m/z 307.0789 (Calcd for C13H16O7Na [M + Na+]: 307.0788); [α]25D +53 (c = 1.0, MeOH).

(2S,3R,4S,5S,6R)-3,4,5-Trihydroxy-6-(hydroxymethyl)-tetrahydro-2H-pyran-2-yl Benzoate (4)25)

Compound 13 (2.00 g, 11.1 mmol) was added in 1,4-dioxane (220 mL). After ultrasound irradiation for 15 min, benzoic acid (450 mg, 3.68 mmol) and Ph3P (5.82 g, 22.2 mmol) were added. To the mixture, diisopropyl azodicarboxylate (DIAD, 90%, 4.95 mL, 23.0 mmol) was added dropwise by syringe addition for 10 min and the mixture was stirred vigorously at room temperature for 1 h. The reaction mixture was quenched by the addition of MeOH and evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, CHCl3/MeOH = 9 : 1 to 4 : 1 v/v) to afford 4 (810 mg, 77%) as a colorless solid.

Rf: 0.40 (CHCl3/MeOH = 4 : 1); 1H-NMR (400 MHz, CD3OD) δ: 3.62 (1H, dd, J = 4.0, 10.0 Hz), 3.71–3.77 (3H, m), 3.86 (1H, dd, J = 8.0, 10.0 Hz), 3.93 (1H, d, J = 4.0 Hz), 5.70 (1H, d, J = 8.0 Hz), 7.49 (2H, t, J = 8.0 Hz), 7.63 (1H, tt, J = 1.5, 8.0 Hz), 8.09–8.11 (2H, m).

[(2R,3S,4S,5R,6S)-6-(Benzoyloxy)-3,4,5-trihydroxytetrahydro-2H-pyran-2-yl]methyl Benzoate (8)

To a solution of 4 (284 mg, 1.00 mmol) in 2,6-lutidine (10 mL), benzoyl chloride (175 µL, 1.51 mmol) was added and stirred at room temperature for 22 h under Ar atmosphere. The reaction mixture was quenched by the addition of MeOH and diluted with toluene. The resulting mixture was evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, CHCl3, 100% to CHCl3/MeOH = 20 : 1 v/v) to afford 8 (244 mg, 63%) as a colorless solid.

Rf: 0.30 (CHCl3/MeOH = 9 : 1); Melting point: 74–76 °C; 1H-NMR (400 MHz, CDCl3 + CD3OD) δ: 3.54 (1H, t, J = 9.5 Hz, β4), 3.63–3.70 (2H, m, β2, β3), 3.81 (1H, ddd, J = 2.0, 5.0, 9.5 Hz, β5), 4.58 (1H, dd, J = 2.0, 12.0 Hz, β6), 4.66 (1H, dd, J = 5.0, 12.0 Hz, β6), 5.83 (1H, d, J = 7.0 Hz, β1), 7.43–7.48 (4H, m), 7.58 (1H, t, J = 8.0 Hz), 7.59 (1H, t, J = 7.0 Hz), 8.06 (2H, d, J = 8.0 Hz), 8.12 (2H, d, J = 7.0 Hz); 13C-NMR (100 MHz, CDCl3) δ: 63.6, 69.8, 72.7, 74.8, 76.1, 94.4, 128.35, 128.37, 128.9, 129.3, 129.9, 130.1, 133.3, 133.6, 165.2, 167.4; IR (KBr) cm−1: 3422, 2913, 1724, 1280, 1069, 707; HR-ESI-MS m/z 411.1054 (Calcd for C20H20O8Na [M + Na+]: 411.1050); [α]25D –9 (c = 1.0, MeOH).

Procedure for Synthesis of Tri- and Tetrabenzoyl β-D-Glucose

To a solution of 4 (580 mg, 2.04 mmol) in dry pyridine (20 mL), benzoyl chloride (710 µL, 6.11 mmol) was added and stirred at room temperature for 17 h under Ar atmosphere. The reaction mixture was quenched by the addition of H2O and diluted with toluene. The residue was purified by column chromatography (SiO2, CHCl3 100%) to afford 3 (286 mg, 28%), 10 (470 mg, 39%) as colorless solids, and a mixture including 11 (355 mg). The mixture was purified by column chromatography (SiO2, CHCl3/MeOH = 100 : 1 v/v) to afford 11 (87.0 mg, 7%) as a colorless solid.

(2S,3R,4S,5R,6R)-6-[(Benzoyloxy)methyl]-3,5-dihydroxytetrahydro-2H-pyran-2,4-diyl Dibenzoate (3)

Rf: 0.70 (CHCl3/MeOH = 9 : 1); Melting point: 70–72 °C; 1H-NMR (400 MHz, CDCl3) δ: 2.90 (1H, br s), 3.49 (1H, br s), 3.86 (1H, t, J = 10.0 Hz, β4), 3.94 (1H, ddd, J = 2.0, 3.5, 10.0 Hz, β5), 4.02 (1H, dd, J = 8.0, 9.0 Hz, β2), 4.58 (1H, dd, J = 2.0, 12.0 Hz, β6), 4.84 (1H, dd, J = 4.0, 12.0 Hz, β6), 5.34 (1H, t, J = 9.0 Hz, β3), 5.99 (1H, d, J = 8.0 Hz, β1), 7.43–7.48 (6H, m), 7.56–7.63 (3H, m), 8.07–8.11 (6H, m); 13C-NMR (100 MHz, CDCl3) δ: 63.2, 68.5, 71.8, 75.3, 78.7, 94.7, 128.4, 128.46, 128.51, 128.9, 129.1, 129.4, 130.01, 130.04, 130.1, 133.4, 133.70, 133.72, 164.9, 167.3, 167.9; IR (KBr) cm−1: 3455, 2913, 1720, 1452, 1272, 1068, 709; HR-ESI-MS m/z 515.1309 (Calcd for C27H24O9Na [M + Na+]: 515.1313); [α]25D +25 (c = 1.0, MeOH).

(2S,3R,4S,5R,6R)-6-[(Benzoyloxy)methyl]-5-hydroxytetrahydro-2H-pyran-2,3,4-triyl Tribenzoate (10)

Rf: 0.67 (CHCl3/MeOH = 98 : 2); Melting point: 138–141 °C; 1H-NMR (400 MHz, CDCl3) δ: 3.59 (1H, d, J = 4.0 Hz), 3.99 (1H, dt, J = 4.0, 10.0 Hz, β4), 4.04 (1H, ddd, J = 2.0, 3.0, 10.0 Hz, β5), 4.62 (1H, dd, J = 2.0, 12.0 Hz, β6), 4.90 (1H, dd, J = 3.0, 12.0 Hz, β6), 5.64 (1H, t, J = 10.0 Hz, β3), 5.75 (1H, dd, J = 8.0, 10.0 Hz, β2), 6.19 (1H, d, J = 8.0 Hz, β1), 7.23–7.61 (12H, m), 7.91 (2H, d, J = 8.0 Hz), 7.99 (2H, d, J = 8.0 Hz), 8.02 (2H, d, J = 8.0 Hz), 8.10 (2H, d, J = 8.0 Hz); 13C-NMR (100 MHz, CDCl3) δ: 62.9, 69.0, 70.5, 75.5, 75.9, 92.7, 128.40, 128.44 (2C), 128.48, 128.5, 128.7, 128.8, 129.3, 129.8, 129.96, 130.00, 130.1, 133.4 (2C), 133.6, 133.8, 164.7, 165.3, 167.0, 167.3; IR (KBr) cm−1: 3483, 2965, 1735, 1451, 1268, 1068, 709; HR-ESI-MS m/z 619.1571 (Calcd for C34H28O10Na [M + Na+]: 619.1575); [α]25D +25 (c = 0.75, MeOH).

(2S,3R,4R,5R,6R)-6-[(Benzoyloxy)methyl]-3-hydroxytetrahydro-2H-pyran-2,4,5-triyl Tribenzoate (11)

Rf: 0.65 (CHCl3/MeOH = 98 : 2); Melting point: 179–182 °C; 1H-NMR (400 MHz, CDCl3) δ: 2.98 (1H, d, J = 4.0 Hz), 4.10 (1H, ddd, J = 4.0, 8.0, 9.0 Hz, β2), 4.30 (1H, ddd, J = 3.0, 5.0, 10.0 Hz, β5), 4.47 (1H, dd, J = 5.0, 12.0 Hz, β6), 4.62 (1H, dd, J = 3.0, 12.0 Hz, β6), 5.62 (1H, t, J = 9.0 Hz, β3), 5.75 (1H, t, J = 10.0 Hz, β4), 6.09 (1H, d, J = 8.0 Hz, β1), 7.34–7.41 (6H, m), 7.47 (2H, t, J = 8.0 Hz), 7.51–7.63 (3H, m), 7.61 (1H, t, J = 8.0 Hz), 7.92 (2H, d, J = 8.0 Hz), 7.99 (2H, d, J = 8.0 Hz), 8.01 (2H, d, J = 8.0 Hz), 8.12 (2H, d, J = 8.0 Hz); 13C-NMR (100 MHz, CDCl3) δ: 62.8, 68.6, 72.1, 72.9, 76.3, 94.6, 128.3, 128.45 (2C), 128.48, 128.70, 128.75, 128.8, 129.5, 129.8 (2C), 130.0, 130.2, 133.0, 133.5, 133.6, 133.8, 164.8, 165.2, 166.1, 167.1; IR (KBr) cm−1: 3528, 3061, 1728, 1452, 1279, 1066, 619; HR-ESI-MS m/z 619.1572 (Calcd for C34H28O10Na [M + Na+]: 619.1575); [α]25D +10 (c = 0.5, CHCl3).

(4aR,6S,7R,8R,8aS)-7,8-Dihydroxy-2,2-dimethylhexahydropyrano[3,2-d][1,3]dioxin-6-yl Benzoate (19)

To a solution of 4 (264 mg, 0.929 mmol) and p-toluenesulfonic acid monohydrate (3.5 mg, 0.018 mmol) in dry DMF (2.5 mL), 2-methoxy propene (100 µL, 1.04 mmol) was added and stirred at room temperature for 16.5 h under Ar atmosphere. NaHCO3 (84 mg, 1.00 mmol) was added and stirred at the same temperature for 1.5 h. The resulting mixture was diluted and filtrated with toluene. The filtrate was evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, CHCl3/MeOH = 20 : 1 v/v) to afford 19 (258 mg, 86%) as a colorless solid.

Rf: 0.50 (CHCl3/MeOH = 20 : 1); Melting point: 178–180 °C; 1H-NMR (400 MHz, CDCl3) δ: 1.46 (3H, s), 1.53 (3H, s), 2.62 (1H, d, J = 2.5 Hz), 2.83 (1H, d, J = 1.0 Hz), 3.50 (1H, ddd, J = 5.5, 10.0, 10.0 Hz, β5), 3.63–3.67 (1H, m, β4), 3.76–3.83 (3H, m, β2, β3, β6), 3.98 (1H, dd, J = 5.5, 10.0 Hz, β6), 5.86 (1H, d, J = 8.0 Hz, β1), 7.46 (2H, t, J = 8.0 Hz), 7.61 (1H, tt, J = 1.5, 8.0 Hz), 8.08–8.11 (2H, m); 13C-NMR (100 MHz, CDCl3) δ: 19.1, 28.9, 61.8, 68.1, 72.8, 73.7, 74.1, 94.7, 99.9, 128.5, 128.8, 130.1, 133.8, 164.9; IR (KBr) cm−1: 3551, 3450, 2929, 1723, 1272, 1078, 1027, 842, 707; HR-ESI-MS m/z 347.1100 (Calcd for C16H20O7Na [M + Na+]: 347.1101); [α]25D −42 (c = 0.5, MeOH).

(4aR,6S,7R,8R,8aR)-7-Hydroxy-2,2-dimethylhexahydropyrano[3,2-d][1,3]dioxine-6,8-diyl Dibenzoate (20)

To a solution of 19 (130 mg, 0.401 mmol) in dry pyridine (4.0 mL), benzoyl chloride (50 µL, 0.43 mmol) was added and stirred at room temperature for 23.5 h. The reaction mixture was quenched by the addition of MeOH. The resulting mixture was diluted with toluene and evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, hexane/EtOAc = 3 : 1 v/v) to afford 20 (57.0 mg, 33%) as a colorless solid.

Rf: 0.42 (CHCl3/MeOH = 98 : 2); Melting point: 165–168 °C; 1H-NMR (400 MHz, CDCl3) δ: 1.40 (3H, s), 1.52 (3H, s), 3.14 (1H, s), 3.61–3.68 (1H, m, β5), 3.83 (1H, t, J = 9.5 Hz, β6), 3.95–4.07 (3H, m, β2, β4, β6), 5.32 (1H, t, J = 9.5 Hz, β3), 5.98 (1H, d, J = 8.0 Hz, β1), 7.46 (2H, t, J = 8.0 Hz), 7.48 (2H, t, J = 8.0 Hz), 7.58–7.63 (2H, m), 8.09 (4H, t, J = 8.0 Hz); 13C-NMR (100 MHz, CDCl3) δ: 19.0, 28.9, 62.0, 68.1, 70.9, 73.1, 76.1, 95.2, 99.9, 128.5 (2C), 128.9, 129.4, 129.97, 130.10, 133.5, 133.8, 164.8, 167.5; IR (KBr) cm−1: 3419, 2997, 1743, 1718, 1270, 1093, 1068, 855, 715; HR-ESI-MS m/z 451.1366 (Calcd for C23H24O8Na [M + Na+]: 451.1363); [α]25D −22 (c = 0.5, MeOH).

(2S,3R,4S,5R,6R)-3,5-Dihydroxy-6-(hydroxymethyl)-tetrahydro-2H-pyran-2,4-diyl Dibenzoate (7)

Compound 20 (57 mg, 0.13 mmol) was dissolved in CH2Cl2/trifluoroacetic acid/H2O (100 : 10 : 1 v/v/v, 3.0 mL) and stirred at room temperature for 1 h. The reaction mixture was evaporated under reduced pressure. The residue was purified by column chromatography (SiO2, hexane/EtOAc = 1 : 1 v/v) to afford 7 (21.5 mg, 42%) as a colorless solid.

Rf: 0.45 (CHCl3/MeOH = 9 : 1); Melting point: 91–94 °C; 1H-NMR (400 MHz, CD3OD) δ: 3.62 (1H, ddd, J = 2.5, 5.0, 10.0 Hz, β5), 3.74–3.80 (2H, m, β4, β6), 3.83 (1H, dd, J = 8.0, 10.0 Hz, β2), 3.89 (1H, dd, J = 2.5, 12.0 Hz, β6), 5.34 (1H, t, J = 10.0 Hz, β3), 5.88 (1H, d, J = 8.0 Hz, β1), 7.49 (2H, t, J = 8.0 Hz), 7.50 (2H, t, J = 8.0 Hz), 7.59–7.66 (2H, m), 8.09–8.12 (4H, m); 13C-NMR (100 MHz, CD3OD) δ: 61.9, 69.2, 72.4, 78.8, 79.5, 96.1, 129.5, 129.7, 130.7, 130.8, 130.9, 131.7, 134.2, 134.8, 166.5, 167.8; IR (KBr) cm−1: 3426, 2928, 1721, 1452, 1278, 1069, 710; HR-ESI-MS m/z 411.1053 (Calcd for C20H20O8Na [M + Na+]: 411.1050); [α]25D +18 (c = 0.5, MeOH).

Optical Resolution of (±)-1

The solution of (±)-119) (10 mg/mL in acetonitrile, 100 µL) was separated using YMC CHIRAL ART Cellulose-SC (5 µm, 4.6 × 250 mm, mobile phase: acetonitrile, flow rate: 1.0 mL/min) to (+)-1 (7.2 min) and (−)-1 (11.2 min).

(1S,2S,3R,6R)-5-[(Benzoyloxy)methyl]-2,6-dihydroxycycl-ohex-4-ene-1,3-diyl dibenzoate ((–)-1) [α]25D −90 (c = 0.3, CHCl3).

Procedure of Biological Assay

Cell Cultures

Human pancreatic cancer cell line PANC-1 was cultured in low glucose DMEM (Nacalai Tesque Inc., #08458-45) supplemented with heat-inactivated 10% fetal bovine serum (FBS, PAN Biotech GmbH, Lot: P180803)). Human colon cancer cell line HT-29 was cultured in RPMI-1640 medium (Nacalai Tesque Inc., #30264-85) supplemented with heat-inactivated 10% FBS (Gibco, lot: 42Q4173K). Every medium was supplemented with penicillin G potassium (50 units/mL, Meiji Seika Pharma Co., #45397387), streptomycin sulfate (50 µg/mL, Meiji Seika Pharma Co., #45844665), and kanamycin sulfate (50 µg/mL, Meiji Seika Pharma Co., #46204854), and the cultured cells were maintained in a humidified atmosphere of 5% CO2 and 95% air at 37 °C. A subculture was performed once or twice per week from subconfluent cultures using a trypsin–ethylenediaminetetraacetic acid (EDTA) solution (10 times diluted FUJIFILM Wako Pure Chemical Corporation, #208-17251). Glucose and sodium pyruvate free DMEM (#09891-25), glucose and sodium pyruvate free RPMI-1640 (#09892-25) were purchased from Nacalai Tesque Inc. Nutrient deprived medium (NDM) to mimic tumor microenvironment is as follows: 25 mM N-(2-hydroxyethyl)piperazine-N′-2-ethanesulfonic acid (HEPES) supplemented with 6.4 g/L NaCl, 700 mg/L NaHCO3, 400 mg/L KCl, 265 mg/L CaCl2·2H2O, 200 mg/L MgSO4·7H2O, 109 mg/L NaH2PO4, 0.1 mg/L Fe(NO3)·9H2O, 15 mg/L phenol red, 40 mL/L MEM vitamin solution (100×) (Gibco).

Assay for Growth Inhibitory Activity under Nutrient Deprived Conditions

PANC-1 cells in DMEM with 10% FBS were seeded into each well of 96-well plates (2.0 × 104 cells/well/100 µL, cell culture 96-well plate, flat bottom (TPP Techno Plastic Products AG, #92696) then incubated for 24 h in a humidified atmosphere of 5% CO2 and 95% air at 37 °C. After removal of the medium, the cells in each well were rinsed with 100 µL of phosphate buffered saline (PBS(−)). Then, the plates were incubated in either (a) DMEM with 10% FBS (+FBS), (b) FBS-free DMEM (−FBS), (c) FBS, sodium pyruvate and glucose-free DMEM (−FBS, −SP, and −Glc), or (d) NDM with the test compounds (1% DMSO) for 24 h. After the incubation, 10% WST-8 cell counting kit solution (Kishida Chemical Co., #260-96160) in DMEM with 10% FBS (100 µL) was added to the each well. After 2–4 h incubation, each absorbance at 450 nm (Abs450) to quantify metabolite formazan and at 650 nm (Abs650) as background was measured (Molecular Devices Inc. SpectraMax iD5 multiplate reader). Cell viability was calculated from the mean values of two wells by using the following equation:

  

Each experiment was performed in triplicate at least. The WST-8 assay on HT-29 cells were also performed according to the method on PANC-1 cells except that the medium was changed to RPMI-1640 medium.

Morphological Assessment of Preferential Cytotoxicity by Fluorescence Imaging

For the morphological analysis, HT-29 cells were seeded in 35 mm dishes (4.0 × 105 cells/dish/2.0 mL RPMI-1640 medium with 10% FBS, TPP tissue culture dish, #93040) then incubated for 24 h. After removal of the medium, the cells were washed with 1.0 mL of PBS(−) twice. Then, the dishes were incubated in NDM with 3 (1% DMSO, 0 or 70 µM) for further 24 h. The medium was changed to 1.0 mL of Hoechst 33342/propidium iodide (PI) reagents (5.0 µg/mL Hoechst 33342 (FUJIFILM Wako Pure Chemical Corporation) and 5.0 µg/mL propidium iodide (FUJIFILM Wako Pure Chemical Corporation) in PBS(−)) and incubated for 5 min under dark in a humidified atmosphere of 5% CO2 and 95% air at 37 °C. After removal of the medium, the cells were washed with PBS(−) (1.0 mL) three times, then the morphology of the cells was captured in PBS(−) (1.0 mL) using Axio observer 7 inverted microscope (Carl Zeiss AG, 20× objective (numerical aperture: 0.4) with Colibri7 LED (λex = 385/30 nm, λem = 425/30 nm for Hoechst 33342, λex = 555/30 nm, λem = 592/30 nm for PI) and Prime BSI sCMOS camera (Teledyne Photometrics) under phase contrast and fluorescent mode.

Acknowledgments

This work was supported by JSPS KAKENHI Grant Number 21K06525 (A.T.), 19K07035 (K.O.), Chugai Foundation for Innovative Drug Discovery Science (A.T.), SHIONO WELLNESS FOUNDATION (A.T.), Koyanagi foundation (K.O.), and the Vehicle Racing Commemorative Foundation Grant Number 6110, 6231, 6297 (K.O.). We appreciate Ms. Kyoko Tasaka for some compounds preparation, Ms. Chiemi Tsuda, Ms. Yuka Iio, and Ms. Nana Urakami (Laboratory of Bioorganic & Natural Products Chemistry, Kobe Pharmaceutical University) for performing some biological experiments, and Dr. Kenneth L. Kirk (National Institute of Diabetes and Kidney Diseases, NIH) for helpful comments on the manuscript.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials. Copy of 1H-NMR, 13C-NMR, and COSY spectra of the compounds can be found in supplementary materials.

References
 
© 2023 The Pharmaceutical Society of Japan
feedback
Top