Cyclic Sulfoxides-Garlicnins K₁, K₂, and H₁-Extracted from Allium sativum

Abstract

Newly identified cyclic sulfoxides—garlicnins K₁ (1), K₂ (2), and H₁ (3)—were isolated from the acetone extracts of the bulbs of garlic, Allium sativum. Garlicnin H₁ (3) demonstrated potential to suppress tumor cell proliferation by regulating macrophage activation. The structures of garlicnins K₁ and K₂, 3,4-dimethyl-5-allyl-tetrahydrothiophen-2-one-S-oxides, and the structure of garlicnin H₁, 3-carboxy-3-hydroxy-4-methyl-5-allylsulfoxide-tetrahydrothiophen-2-(ethane-1,2-diol)-S-oxide were characterized by spectroscopic analysis.

Onion (Allium cepa L.), garlic (Allium sativum L.), and Welsh onion (Allium fistulosum L.) belong to the genus of Allium. In particular, garlic is ranked at the top of the list of designer foods with anticancer functions recommended by the National Cancer Institute.¹⁾ In general, the various biological activities of onion and garlic can be classified into two categories: cardiovascular disease prevention and cancer prevention. The activities in the former category include the inhibition of cholesterol synthesis, platelet aggregation, and arterial smooth muscle cell proliferation as well as the anti-inflammatory, antioxidant, and hydrogen sulfide-mediated vasodilatory effects. The activities in the latter category include the effects on carcinogen metabolism (i.e., enhanced cellular glutathione synthesis that induces cell cycle arrest and apoptosis) and prevention of Helicobacter pylori infection, gastric cancer, and colorectal cancer are effects of the latter category.^2–5) The chemistry of garlic sulfides is inclusively summarized in a textbook edited by Block.⁴⁾ In the previous study, we isolated a novel stable sulfide, 5-(1-propenyl)-3,4-dimethyl-tetra-hydrothiophene-2-sulfoxide-S-oxide, from the acetone extract of A. cepa and identified its inhibitive effects on macrophage activation. We initially named this compound as onionin A,⁶⁾ and it was renamed as onionin A₁ after the identification of the other similar compounds.

So far, sulfur-containing substances, garlicnins A,⁷⁾ B₁–B₄,^8,9) C₁–C₃,^8,9) D,⁸⁾ L-1–L-4, E, and F¹⁰⁾ obtained from the acetone-soluble extract of the bulbs of A. sativum, onionins A₂ and A₃¹¹⁾ from the extract of A. cepa, and onionins A₁–A₃ from the leaf extract of A. fistulosum,¹¹⁾ were isolated and demonstrated to be highly suitable for developing natural and healthy foods that can prevent and combat a variety of diseases, particularly cancers.

In this study, the structures of three cyclic sulfoxides-garlicnins K₁ (1), K₂ (2), and H₁ (3)-isolated from the acetone-soluble extract of the bulbs of A. sativum, were characterized by using various spectroscopic techniques.

Chinese garlic bulbs (1061.0 g) were roughly chopped and blended with acetone in a high-speed mixer. Afterwards, the mixture was soaked in acetone for three days at room temperature. The filtrate was evaporated at 40°C in vacuum to produce a residue (30.1 g), which was then partitioned between ethyl acetate (EtOAc) and water. The resulting residues, EtOAc layer (Fr. 1, 4.257 g) and aqueous layer (Fr. 2, 25.012 g) were individually examined for evaluating their capabilities to inhibit macrophage activation.

Macrophages that infiltrate cancer tissues are referred to as tumor-associated macrophages (TAMs) and are closely involved in the development of the tumor microenvironment.^12–14) TAMs are categorized as alternatively activated macrophages (M2) because of their anti-inflammatory functions.^15,16) The presence of TAMs in certain types of tumors is associated with a poor prognosis of the tumor-bearing patients.^14,17,18) It is known that the inhibition of M2-macrophage polarization can suppress tumor cell proliferation. In the human monocyte-derived macrophages (HMDM), interleukin 10 (IL-10) induces CD163 expression, a M2 macrophage marker, for 2 d. In the present study, we measured the effects of garlicnin H₁ (3) on IL10-induced CD163 expression in HMDM. As a result, garlicnin H₁ inhibited the expression of CD163 (Fig. 1), suggesting their capabilities for suppressing M2 macrophage activation. This result suggests that garlicnin H₁ (3) may demonstrate good potential to suppress tumor-cell proliferation by inhibiting the polarization of M2 alternatively activated macrophages.

Fig. 1. Effect of Garlicnin H₁ on CD163 Expression in HMDM

HMDM were incubated with galicnin H₁ (30 µM) during incubation with IL-10 (20 ng/mL) for 2 d, followed by determination of the CD163 expression by Cell-ELISA as described in Experimental.

The remaining extracts (3.95 g) were then chromatographed on silica gel under elution with CHCl₃–MeOH (200 : 1) and n-hexane–acetone (4 : 1), and three new compounds, namely, garlicnin K₁ (1, 23.5 mg), garlicnin K₂ (2, 16.2 mg), and garlicnin H₁ (3, 12.1 mg) were obtained.

Garlicnin K₁ (1) was obtained as a syrup showing [α]_D −15.0° (CHCl₃). The mass spectrum of 1 from positive mode high-resolution fast atom bombardment mass spectroscopy (HR-FAB-MS) showed a pronounced peak corresponding to [M+H]⁺ at m/z 187.0796 (C₉H₁₅O₂S). The IR spectrum of 1 showed absorption bands at 2365, 1735, and 1025 cm⁻¹ assigned to SH group, carbonyl group, and sulfoxide, respectively. The ¹H-NMR spectrum (in CDCl₃) of 1 showed signals due to two secondary methyl groups at δ 1.12 (3H, d, J=6.9 Hz), and 1.30 (3H, d, J=6.8 Hz); two methylene protons at δ 3.41 (2H, m); three methine protons at δ 2.39 (1H, ddq, J=5.1, 5.7, 6.8 Hz), 2.49 (1H, dq, J=5.1, 6.9 Hz), and 4.83 (1H, d, J=5.7 Hz); and three olefinic protons at δ 5.14 (1H, d, J=9.8 Ηz), 5.22 (1H, d, J=13.5 Hz), and 5.82 (1H, m). The ¹³C-NMR spectrum (in CDCl₃) of 1 exhibited signals due to two methyl groups at δ 13.0 and 16.3; one methylene carbon at δ 43.3; three methine carbons at δ 46.5, 49.4, and 64.1; two olefinic carbons at δ 119.3 and 133.0; and one carbonyl carbon at δ 208.0. The ¹H–¹H correlation spectroscopy (COSY) analysis spectrum showed the presence of a series of correlations between the methine proton at δ 2.49 and the methine proton at δ 2.39; between the methine proton at δ 2.39 and the methine proton at δ 4.83; and between the methylene protons at δ 3.41 and the olefinic proton at δ 5.82 and two olefinic protons at δ 5.14 and 5.22. The vicinal correlations between the methine proton at δ 2.49 and the methyl proton at δ 1.12 and between the methine proton at δ 2.39 and the methyl proton at δ 1.30 were also observed. In addition, the heteronuclear multiple-bond correlation (HMBC) spectrum indicated the correlations between the methyl proton at δ 1.12 and three carbons at δ 46.5, 49.4, and 208.0; between the methyl proton at δ 1.30 and three carbons at δ 46.5, 49.4, and 64.1; between the methine proton at δ 4.83 and the carbonyl carbon at δ 208.0; and between the methylene proton at δ 3.41 and three carbons at δ 64.1, 119.3, and 133.0. The planar structure of 1, 3,4-dimethyl-5-allyl-tetrahydrothiophen-2-oxo-S-oxide, was determined by analyzing the ¹H–¹H COSY and HMBC spectra (Fig. 1). Furthermore, the result of nuclear Overhauser effect spectroscopy (NOESY) analysis (Fig. 1) showed the their correlations between CH₃ at C-3 and H-4, between H-3 and CH₃ at C-4, and between CH₃ at C-4 and H-5.

To determine the relative configuration of the compound, aromatic solvent-induced NMR shift^19,20) was analyzed in the same way as that used for studying other cyclic garlicnins. It is known that the conformational preference of a sulfoxide is the axial form^21–23) (lower α-side). In ¹H-NMR analysis in C₆D₆, the molecule of 1 forms a collision complex¹⁹⁾ opposite to the sulfoxide, indicating that C₆D₆ is oriented towards the opposite side (upper β-side) of the sulfoxide and affects the NMR chemical shifts. As compared with the spectrum in CDCl₃, the ¹H-NMR spectrum of 1 in C₆D₆ showed that most signals were shifted upfield, including those of H-3 (−0.22), H-4 (−0.45), CH₃ at C-3 (−0.43), CH₃ at C-4 (−0.14), and H-5 (−0.18). In summary, the signals due to CH₃ at C-3, and H-4 significantly shifted towards the higher field, and the only signals that remained relatively unchanged were those of H-3, CH₃ at C-4 and H-5. Based on these results, we proposed the relative structure of 1 shown in Fig. 2. This compound may be produced via oxidation of garlicnins B groups as shown in Fig. 3.

Fig. 2. Structures of Garlicnins K₁ (1) and K₂ (2) with Key HMBC and NOESY

Fig. 3. Probable Pathway to Garlicnins K_1,2

Garlicnin K₂ (2) was obtained as a syrup showing [α]_D −8.4° (CHCl₃). This compound was regarded as a steric isomer of garlicnin K₁ (1) by their IR, HR-FAB-MS, ¹H-NMR, ¹³C-NMR, ¹H–¹H COSY, HMBC, and NOESY. The structure was determined in a similar manner as that of 1 (Fig. 2).

Garlicnin H₁ (3) was obtained as a syrup showing [α]_D −23.1° (CHCl₃). The positive mode HR-FAB-MS spectrum of 3 showed a peak corresponding to [M+Na]⁺ at m/z 349.0396 (C₁₁H₁₈O₇S₂Na). The IR spectrum of 3 showed absorption bands at 3450, 1740, and 1025 cm⁻¹ assigned to hydroxyl group, carbonyl group, and sulfoxide, respectively. The ¹H-NMR spectrum (in CDCl₃) of 3 showed the signals due to the secondary methyl group at δ 1.22 (3H, d, J=7.4 Hz); methylene group at δ 3.54 (2H, d, J=7.4 Hz); three methine protons at δ 2.76 (1H, dq, J=2.4, 7.4 Hz), 4.65 (1H, d, J=1.2 Hz), and 5.14 (1H, d, J=2.9 Hz); one oxygen-bearing methine proton at δ 4.23 (1H, ddd, J=1.2, 6.2, 10.9 Hz); two hydroxymethyl protons at δ 4.11 (1H, dd, J=6.2, 12.1 Hz), and 4.42 (1H, dd, J=10.9, 12.1 Hz); and three olefinic protons at δ 5.21 (1H, d, J=9.8 Ηz), 5.23 (1H, d, J=13.4 Hz), and 5.84 (1H, ddt, J=7.4, 9.8, 13.4 Hz). The ¹³C-NMR spectrum (in CDCl₃) exhibited the signals due to the methyl group at δ 14.3; methylene carbon at δ 42.0; three methine carbons at δ 50.1, 88.8, and 96.6; one quaternary carbon at δ 81.3; two oxygen-bearing carbons at δ 74.2 and 74.7; two olefinic carbons at δ 119.7 and 132.5; and one carboxylic carbon at δ 172.0. Moreover, the HMBC spectrum (Fig. 4) exhibited a series of correlations between the methyl proton at δ 1.22 and three carbons at δ 50.1, 81.3, and 96.6; between the methine proton at δ 4.65 and the oxygen-bearing carbon at δ 74.7 and carboxylic carbon at δ 172.0; and between the methylene protons at δ 3.54 and two olefinic carbons at δ 119.7 and 132.5. The NOESY spectrum (Fig. 4) showed the ¹H–¹H correlations between the methine proton at δ 2.76 and the methine proton at δ 5.14; between the methine proton at δ 4.65 and the oxygen-bearing proton at δ 4.23 and hydroxymethyl protons at δ 4.11 and 4.42; between the methine proton at δ 5.14 and the methylene proton at δ 3.54; between the methine proton at δ 4.65 and the methine proton at δ 5.14; and between the methylene proton at δ 3.54 and three olefinic protons at δ 5.21, 5.23, and 5.84. The planar structure of 3, 3-carboxy-3-hydroxy-4-methyl-5-allylsulfoxide-tetrahydrothiophen-2-(ethane-1,2-diol)-S-oxide, was determined by analyzing the ¹H- and ¹³C-NMR, NOESY, and HMBC spectra.

Fig. 4. Structure of Garlicnins H₁ (3) with Key HMBC and NOESY

As compared with the spectrum in CDCl₃, the ¹H-NMR spectrum of 3 in C₆D₆ showed that most signals were shifted upfield, including H-2 (−0.19), H-4 (−0.21), CH₃ at C-4 (−0.43), and H-5 (−0.24). In summary, the NMR signals due to CH₃-4 significantly shifted towards the higher field; the only signals that remained relatively unchanged were those of H-2, H-4, and H-5, and the configuration at C-3 still remained elusive. Based on these results, we proposed the relative structure of 3 shown in Fig. 4. The production of garlicnin H₁ (3) might be derived from the combined compound of 1-propene sulfenic acid and pentose as shown in Fig. 5.

Fig. 5. Hypothetical Pathway to Garlicnin H₁

The undergoing in vivo experiments have already demonstrated the desired antitumor activities of these cyclic sulfoxides in mice. These results will be presented in the near future.

Experimental

General Experimental Procedures

Optical rotation was measured using a JASCO P-1020 (l=0.5) automatic digital polarimeter. The IR spectrum was measured using a Fourier Transform FT/IR-4200 spectrometer (JASCO). The ¹H- and ¹³C-NMR spectra were measured in CDCl₃ using a JEOL alpha 500 spectrometer at 500 and 125 MHz, respectively, and the chemical shifts were found to be on the δ (ppm) scale. The HR-FAB-MS were measured using a JEOL JMS-DX303HF mass spectrometer and taken in a glycerol matrix containing NaI. Column chromatography was carried out on silica gel 60 (230–400 mesh, Merck). Thin layer chromatography (TLC) was performed on silica gel plates (Kieselgel 60 F254; Merck). TLC spots were visualized under UV light (254/366 nm), sprayed with 10% H₂SO₄ and anisaldehyde, and then heated.

Plant Material

The garlic bulbs (A. sativum L. family Liliaceae) imported from China were identified by Prof. Kotaro Murakami. A voucher specimen (SBGH 13–08-16-217) was deposited in the Herbarium of the Botanical Garden at Sojo University, Kumamoto, Japan.

Extraction and Isolation

Peeled Chinese garlic bulbs (1061.0 g) of garlics were roughly chopped and homogenized in a mixer along with acetone. The mixture was subsequently soaked in acetone for three days at room temperature. The filtrate was concentrated at 40°C in vacuo to obtain a syrup residue (30.1 g), which was then partitioned between ethyl acetate (AcOEt) and water. The organic layer was taken up and evaporated under reduced pressure at 40°C to give a residue (4.257 g). The respective residues (Fr. 1: ethyl acetate residue; Fr. 2: aqueous residue) were examined for the ability to control macrophage activation.

The AcOEt residue (3.95 g) was repeatedly chromatographed on silica gel with CHCl₃–methanol (200 : 1) and n-hexane–acetone (4 : 1) to obtain garlicnins K₁ (1, 23.5 mg), K₂ (2, 16.2 mg), and H₁ (3, 12.1 mg) together with sixteen compounds previously reported garlicnins A⁷⁾ (48.2 mg), B₁⁸⁾ (52.0 mg), B₂⁹⁾ (47.2 mg), B₃⁹⁾ (9.8 mg), B₄⁹⁾ (9.3 mg), C₁⁸⁾ (16.4 mg), C₂⁹⁾ (13.4 mg), C₃⁹⁾ (14.6 mg), D⁸⁾ (42.0 mg), (E)-ajoene⁴⁾ (79.7 mg), L-1¹⁰⁾ (47.2 mg), L-2¹⁰⁾ (9.8 mg), L-3¹⁰⁾ (9.3 mg), L-4¹⁰⁾ (13.4 mg), E¹⁰⁾ (14.6 mg), and F¹⁰⁾ (15.2 mg).

Determination of the Inhibitory Effect of Fractions on CD163 Expression

Human monocyte-derived macrophages (5×10⁴ cells per well of a 96-well plate) were incubated with fractions (100 µg/mL) for 24 h after treatment with IL-10 (20 nM) for 2 d, followed by the determination of CD163 expression by cell enzyme-linked immunosorbent assay (cell-ELISA).

Cell-ELISA

Expression of CD163 on human monocyte-derived macrophages was evaluated using a cell-ELISA procedure, as described previously.²⁴⁾ Briefly, each well of a 96-well plate was blocked with Block Ace, and washed 3 times with phosphate buffered saline containing 0.05% Tween 20 (washing buffer). The wells were incubated with anti-CD163 antibody AM3K (2 µg/mL) and dissolved in washing buffer for 1 h. The wells were then washed with washing buffer 3 times and reacted with HRP-conjugated anti-mouse IgG antibody followed by reaction with Ultrasensitive TMB (Moss, Inc., Pasadena, MD, U.S.A.). The reaction was terminated by the addition of 1 M sulfuric acid, and the absorbance at 450 nm was then read on a micro-ELISA plate reader.

Statistics

All data are representative 2 or 3 independent experiments. Data are expressed as means (SD). Mann–Whitney’s U-test was used for 2-group comparison. A value of p<0.05 was considered statistically significant.

Garlicnin K₁ (1)

Syrup, [α]_D −15.0° (c=0.50, CHCl₃). Positive HR-FAB-MS: [M+H]⁺ at m/z 187.0796 (Calcd for C₉H₁₅O₂S, 187.0793). IR ν_max (KBr): 2365 (SH group), 1735 (carbonyl), and 1025 (sulfoxide) cm⁻¹.

¹H-NMR spectrum (in CDCl₃) δ: 1.12 (3H, d, J=6.9 Hz, CH₃ at C-3), 1.30 (3H, d, J=6.8 Hz, CH₃ at C-4), 3.41 (2H, m, H₂-1′), 2.39 (1H, ddq, J=5.1, 5.7, 6.8 Hz, H-4), 2.49 (1H, dq, J=5.1, 6.9 Hz, H-3), 4.83 (1H, d, J=5.7 Hz, H-5), 5.14 (1H, d, J=9.8 Hz, Ha-3′), 5.22 (1H, d, J=13.5 Hz, Hb-3′), 5.82 (1H, m, H-2′).

¹H-NMR spectrum (in C₆D₆) δ: 0.69 (3H, d, J=7.0 Hz, CH₃ at C-3), 1.16 (3H, d, J=6.9 Hz, CH₃ at C-4), 1.94 (1H, ddq, J=5.2, 5.6, 6.9 Hz, H-4), 2.27 (1H, dq, J=5.2, 7.0 Hz, H-3), 4.65 (1H, d, J=5.6 Hz, H-5).

¹³C-NMR spectrum (in CDCl₃) δ: 13.0 (CH₃ at C-3), 16.3 (CH₃ at C-4), 43.3 (C-1′), 46.5 (C-4), 49.4 (C-3), 64.1 (C-5), 119.3 (C-3′), 133.0 (C-2′), 208.0 (C-2).

Garlicnin K₂ (2)

Syrup, [α]_D −8.4° (c=0.50, CHCl₃). Positive HR-FAB-MS: [M+H]⁺ at m/z 187.0790 (Calcd for C₉H₁₅O₂S, 187.0793). IR ν_max (KBr): 2363 (SH group), 1735 (carbonyl), 1023 (sulfoxide) cm⁻¹.

¹H-NMR spectrum (in CDCl₃) δ: 1.18 (3H, d, J=6.9 Hz, CH₃ at C-3), 1.25 (3H, d, J=6.8 Hz, CH₃ at C-4), 3.41 (2H, m, H₂-1′), 2.14 (1H, ddq, J=4.6, 6.8, 10.3 Hz, H-4), 2.29 (1H, dq, J=4.6, 6.9 Hz, H-3), 4.40 (1H, d, J=10.3 Hz, H-5), 5.14 (1H, d, J=9.8 Ηz, Ha-3′), 5.22 (1H, d, J=13.5 Hz, Hb-3′), 5.82 (1H, m, H-2′).

¹H-NMR spectrum (in C₆D₆) δ: 0.95 (3H, d, J=6.8 Hz, CH₃ at C-3), 1.11 (3H, d, J=6.8 Hz, CH₃ at C-4), 1.71 (1H, ddq, J=4.6, 6.3, 10.3 Hz, H-4), 1.87 (1H, dq, J=4.6, 6.9 Hz, H-3), 4.02 (1H, d, J=10.3 Hz, H-5).

¹³C-NMR spectrum (in CDCl₃) δ: 13.7 (CH₃ at C-3), 16.0 (CH₃ at C-4), 43.1 (C-1′), 46.9 (C-4), 54.9 (C-3), 63.1 (C-5), 119.5 (C-3′), 132.6 (C-2′). 207.0 (C-2).

Garlicnin H₁ (3)

Syrup, [α]_D −23.1° (c=0.50, CHCl₃), Positive HR-FAB-MS: [M+Na]⁺ at m/z 349.0396 (Calcd for C₁₁H₁₈O₇S₂Na, 349.0392). IR ν_max (KBr): 3450 (hydroxyl), 1740 (carboxyl), 1025 (sulfoxide) cm⁻¹.

¹H-NMR spectrum (in CDCl₃) δ: 1.22 (3H, d, J=7.4 Hz, CH₃ at C-4), 2.76 (1H, dq, J=2.4, 7.4 Hz, H-4), 3.54 (2H, d, J=7.4 Hz, H₂-2″), 4.11 (1H, dd, J=6.2, 12.1 Hz, Ha-2′), 4.23 (1H, ddd, J=1.2, 6.2, 10.9 Hz, H-1′), 4.42 (1H, dd, J=10.9, 12.1 Hz, Hb-2′), 4.65 (1H, d, J=1.2 Hz, H-2), 5.14 (1H, d, J=2.9 Hz, H-5), 5.21 (1H, d, J=9.8 Ηz, Ha-4″), 5.23 (1H, d, J=13.4 Hz, Hb-4″), 5.84 (1H, ddt, J=7.4, 9.8, 13.4 Hz, H-3″).

¹H-NMR spectrum (in C₆D₆) δ: 0.79 (3H, d, J=7.4 Hz, H₃ at C-4), 2.55 (1H, dq, J=2.4, 7.4 Hz, H-4), 4.46 (1H, d, J=1.2 Hz, H-2), 4.90 (1H, d, J=2.9 Hz, H-5).

¹³C-NMR (CDCl₃) δ: 14.3 (CH₃ at C-4), 42.0 (C-2″), 50.1 (C-4), 74.2 (C-2′), 74.7 (C-1′), 81.3 (C-3), 88.8 (C-2), 96.6 (C-5), 119.7 (C-4″), 132.5 (C-3″), 172.0 (COOH at C-3).

Conflict of Interest

The authors declare no conflict of interest.

References

• 1) Theisen C., J. Natl. Cancer Inst., 93, 1049–1050 (2001).
• 2) Rose P., Whiteman M., Moore P. K., Zhu Y. Z., Nat. Prod. Rep., 22, 351–368 (2005).
• 3) Amagase H., J. Nutr., 136 (Suppl.), 716S–725S (2006).
• 4) Block E., “Garlic and Other Alliums: The Lore and Science,” The Royal Society of Chemistry, U.K., 2010.
• 5) Dewick P. M., “Medicinal Natural Products a Biosynthetic Approach,” 2nd ed., John Willey & Sons, U.K., 2002.
• 6) El-Aasr M., Fujiwara Y., Takeya M., Ikeda T., Tsukamoto S., Ono M., Nakano D., Okawa M., Kinjo J., Yoshimitsu H., Nohara T., J. Nat. Prod., 73, 1306–1308 (2010).
• 7) El-Aasr M., Fujiwara Y., Takeya M., Ono M., Nakano D., Okawa M., Kinjo J., Ikeda T., Miyashita H., Yoshimitsu H., Nohara T., Chem. Pharm. Bull., 59, 1340–1343 (2011).
• 8) Nohara T., Kiyota Y., Sakamoto T., Manabe H., Ono M., Ikeda T., Fujiwara Y., Nakano D., Kinjo J., Chem. Pharm. Bull., 60, 747–751 (2012).
• 9) Nohara T., Fujiwara Y., Ikeda T., Murakami K., Ono M., Nakano D., Kinjo J., Chem. Pharm. Bull., 61, 695–699 (2013).
• 10) Nohara T., Fujiwara Y., Ikeda T., Yamaguchi K., Manabe H., Murakami K., Ono M., Nakano D., Kinjo J., Chem. Pharm. Bull., 62, 477–482 (2014).
• 11) Nohara T., Fujiwara Y., Kudo R., Yamaguchi K., Ikeda T., Murakami K., Ono M., Kajimoto T., Takeya M., Chem. Pharm. Bull., 62, 1142–1145 (2014).
• 12) Mantovani A., Schioppa T., Biswas S. K., Marchesi F., Allavena P., Sica A., Tumor, 89, 459–468 (2003).
• 13) Sica A., Schioppa T., Mantovani A., Allavena P., Eur. J. Cancer, 42, 717–727 (2006).
• 14) Lewis C. E., Pollard J. W., Cancer Res., 66, 605–612 (2006).
• 15) Hagemann T., Biswas S. K., Lawrence T., Sica A., Lewis C. E., Blood, 113, 3139–3146 (2009).
• 16) Gordon S., Nat. Rev. Immunol., 3, 23–35 (2003).
• 17) Joyce J. A., Pollard J. W., Nat. Rev. Cancer, 9, 239–252 (2009).
• 18) Sica A., Larghi P., Mancino A., Rubino L., Porta C., Totaro M. G., Rimoldi M., Biswas S. K., Allavena P., Mantovani A., Semin. Cancer Biol., 18, 349–355 (2008).
• 19) Juaristi E., Cruz-Sanchez J. S., Petson A., Glass R. S., Tetrahedron, 44, 5653–5660 (1988).
• 20) Ronayne J., Williams D. H., J. Chem. Soc. B, 1967, 540–546 (1967).
• 21) Johnson C. R., McCants D., J. Am. Chem. Soc., 86, 2935–2936 (1964).
• 22) Martin J. C., Uebel J. J., J. Am. Chem. Soc., 86, 2936–2937 (1964).
• 23) Lambert J. B., Keske R. G., J. Org. Chem., 31, 3429–3431 (1966).
• 24) Komohara Y., Hirahara J., Horikawa T., Kawamura K., Kiyota E., Sakashita N., Araki N., Takeya M., J. Histochem. Cytochem., 54, 763–771 (2006).