Natural Compounds with BMI1 Promoter Inhibitory Activity from Mammea siamensis and Andrographis paniculata

Abstract

A new coumarin derivative (1) and 30 known compounds were isolated from Mammea siamensis and Andrographis paniculata, guided by B cell-specific Moloney murine leukemia virus insertion region 1 (BMI1) promoter inhibitory activity. Among the isolated compounds, 15 compounds showed BMI1 promoter inhibitory activity, and five compounds were found to be cytotoxic. 14-Deoxy-11,12-dehydroandrographolide (18) was highly cytotoxic to DU145 cells with an IC₅₀ value of 25.4 µM. Western blotting analysis of compound 18 in DU145 cells suggested that compound 18 suppresses BMI1 expression.

Introduction

B cell-specific Moloney murine leukemia virus insertion region 1 (BMI1) is the central molecule of polycomb repressor complex 1 (PRC1), which is a protein involved in the epigenetic gene expression mechanism of cells.^1,2) BMI1 is regulated by c-Myc protein and suppresses the transcription of tumor suppressor genes such as Ink4a/ARF by epigenetic gene expression mechanisms. The Ink4a/ARF gene induces cell cycle regulators and the tumor suppressor p14ARF gene, causing cell cycle arrest or apoptosis.^3–5) BMI1 is an essential self-renewal regulator for multiple stem cell systems, such as hematopoietic, nerves, and breast stem cells, and BMI1 is also highly expressed in various tumor tissues.^6,7) Especially, BMI1 is known to be highly expressed in cancer stem cells (CSCs) that contribute to cancer recurrence and metastasis due to their resistance to chemotherapy and radiation therapy.^8–10) Therefore, compounds with BMI1 promoter inhibitory activity may lead to a decrease in cancer stem cells, providing a fundamental treatment for cancer.

In our previous study, we constructed a reporter gene assay system to evaluate the promoter activity of BMI1. The luciferase gene under the control of the BMI1 promoter was stably expressed in human fetal kidney (HEK) 293T cells, and this assay system was used to isolate natural BMI1 promoter inhibitors to obtain waricoside, digitoxigenin, somalin, and maslinic acid from Thai plants.^11,12) In addition, elaiophyllin, 2-methylelaiophyllin, and nocardamin were isolated as BMI1 promoter inhibitors.^11,13,14) Here, we search for natural compounds showing BMI1 gene expression inhibitory effect from tropical plants.

Results and Discussion

The Thailand and Bangladesh plant collection was screened for BMI1 promoter inhibitory activity, where MeOH extracts of Mammea siamensis and Andrographis paniculata were identified as potential samples. The MeOH extracts of M. siamensis (15.8 g) and A. paniculata (8.6 g) were fractionated by monitoring BMI1 promoter inhibitory activity. One new coumarin (1) (Table 1), along with 11 known coumarins (mammea E/BB cyclo F (2),¹⁵⁾ mammea E/BA cyclo F (3),¹⁵⁾ mammea B/AC cyclo F (4),¹⁶⁾ mammea A/AA cyclo F (5),¹⁶⁾ mammea E/BC cyclo D (6),¹⁷⁾ mammea A/AA cyclo D (7),¹⁸⁾ mammea A/AC cyclo F (8),¹⁸⁾ mammea E/BB cyclo D (9),¹⁹⁾ mammea B/AC cyclo D (10),²⁰⁾ mammea A/AB cyclo D (11),²¹⁾ mammea B/BC cyclo F (12)²²⁾) one known peptide, (N-benzoylphenylalaninyl-N-benzoylphenylalaninate (13))²³⁾ and four known flavonoids (amentoflavone (14),²⁴⁾ luteolin (15),²⁵⁾ naringenin (16),²⁴⁾ apigenin (17)²⁶⁾) were isolated from M. siamensis. Thirteen known labdane diterpenes (14-deoxy-11,12-dehydroandrographolide (18),²⁷⁾ 14-deoxyandrographolide (19),²⁷⁾ andrographolide (20),^27,28) isoandrographolide (21),²⁷⁾ 14-deoxy-15-methoxyandrographolide (22),²⁹⁾ neo-andrographolide (23),^27,28) 13-dien-16,15-olide(9)13,14,15,16-tetranor-ent-labd-8(17)-ene-3,12,19-triol (24),³⁰⁾ 7R-hydroxy-14-deoxyandrographolide (25),²⁹⁾ 7S-hydroxy-14-deoxyandrographolide (26),²⁹⁾ 14-deoxy-12-methoxyandrographolide (27),³¹⁾ 3-dehydrodeoxyandrographolide (28),³²⁾ 3,18,19-trihydroxy-ent-labda-8(17),13-diene-15,16-olide (29),³⁰⁾ 14-deoxy-17β-hydroxylandrographolide (30)³³⁾) and one flavonoid (andrographidine C (31))³⁴⁾ were isolated from of A. paniculata (Fig. 1). In a prior report, compounds 2 and 3 were obtained as a mixture of chemical derivatives.¹⁵⁾ In this study, compounds 2 and 3 were first isolated as natural products, and their NMR spectra were first recorded (Tables 2, 3).

Table 1. ¹H- and ¹³C-NMR Spectra of 1

Position	1
	1H-NMR		13C-NMR δC (ppm)
	δH (ppm)	J (Hz)
2			159.4
3	6.22 (1H)	s	106.1
4			155.9
4a			97.2
5			161.4
6			110.2
7			163.3
8			105.2
8a			157.5
1′	6.31 (1H)	dd, J = 8.3, 3.3 Hz	72.6
2′	1.98 (1H)	m	28.2
	1.77 (1H)	m
3′	1.04 (3H)	Overlap	9.7
2″	4.93 (1H)	t, J = 9.0 Hz	93.2
3″	3.16 (2H)	Overlap	26.8
4″			71.6
5″	1.38 (3H)	s	26.0
6″	1.29 (3H)	s	24.8
1‴			206.2
2‴	3.25 (2H)	t, J = 7.4 Hz	46.6
3‴	1.77 (2H)	m	18.0
4‴	1.04 (3H)	Overlap	13.8
1″″			170.2
2″″	2.17 (3H)	s	21.0
7-OH	14.18 (1H)	s

CDCl₃, ¹H-NMR: 600 MHz, ¹³C-NMR: 150 MHz.

Fig. 1. Structure of Compounds 1–7, 9, and 18–20 from M. siamensis and A. paniculata

Table 2. ¹H- and ¹³C-NMR Spectra of 2

Position	2
	1H-NMR		13C-NMR δC (ppm)
	δH (ppm)	J (Hz)
2			159.4
3	6.23 (1H)	s	106.1
4			155.9
4a			97.2
5			161.3
6			110.3
7			163.5
8			104.9
8a			157.2
1′	6.32 (1H)	dd, J = 7.4, 4.1 Hz	72.6
2′	1.99 (1H)	m	28.2
	1.74 (1H)	m
3′	1.05 (3H)	t, J = 7.4 Hz	9.7
2″	4.94 (1H)	t, J = 9.0 Hz	93.1
3″	3.16 (2H)	Overlap	26.8
4″			71.6
5″	1.39 (3H)	s	26.0
6″	1.29 (3H)	s	24.8
1‴			210.5
2‴	3.87 (1H)	m	46.9
3‴	1.89 (1H)	m	27.1
	1.46 (1H)	m
4‴	0.98 (3H)	t, J = 7.4 Hz	11.8
5‴	1.24 (3H)	d, J = 7.4 Hz	16.6
1″″			170.2
2″″	2.17 (3H)	s	21.0
7-OH	14.21 (1H)	s

CDCl₃, ¹H-NMR: 600 MHz, ¹³C-NMR: 150 MHz.

Table 3. ¹H- and ¹³C-NMR Spectra of 3

Position	3
	1H-NMR		13C-NMR δC (ppm)
	δH (ppm)	J (Hz)
2			159.4
3	6.23 (1H)	s	106.5
4			156.1
4a			96.4
5			160.5
6			110.3
7			163.5
8			105.2
8a			157.4
1′	6.50 (1H)	dd, J = 8.3, 3.3 Hz	72.5
2′	1.90 (1H)	m	28.6
	1.75 (1H)	m
3′	1.03 (3H)	Overlap	10.1
2″	4.90 (1H)	t, J = 9.0 Hz	93.1
3″	3.19 (2H)	d, J = 9.0 Hz	26.6
4″			71.6
5″	1.41 (3H)	s	26.2
6″	1.26 (3H)	s	24.2
1‴			206.5
2‴	3.12 (2H)	d, J = 6.6 Hz	53.5
3‴	2.26 (1H)	m	25.5
4‴	1.03 (3H)	Overlap	22.6
5‴	1.03 (3H)	Overlap	22.6
1″″			170.6
2″″	2.15 (3H)	s	21.0
7-OH	14.23 (1H)	s

CDCl₃, ¹H-NMR: 600 MHz, ¹³C-NMR: 150 MHz.

The molecular formula of 1 was determined to be C₂₃H₂₈O₈ from the high-resolution electrospray ionization mass spectrometry (HRESIMS) signal at m/z 455.1687 [M + Na]⁺ (Calcd for C₂₃H₂₈O₈Na 455.1682). The IR absorption spectrum suggested the presence of hydroxy (3320 cm⁻¹) and carbonyl groups (1635, 1610 cm⁻¹). The ¹H-NMR spectrum of 1 in CDCl₃ showed signals for five methyl [δ_H 1.04 (3H, overlap, H₃-4‴), 1.04 (3H, overlap, H₃-3′), 1.29 (3H, s, H₃-6″), 1.38 (3H, s, H₃-5″), and 2.17 (3H, s, H₃-2″″)], and one hydroxy group [δ_H 14.18 (1H, s, 7-OH)] (Table 1). The ¹³C-NMR spectrum of 1 in CDCl₃ contained 23 peaks, including peaks corresponding to a ketone [δ_C 206.2 (C-1‴)], aromatic carbons [δ_C 97.2 (C-4a), 161.4 (C-5), 110.2 (C-6), 163.3 (C-7), 105.2, (C-8), 157.5 (C-8a)], and an ester group [δ_C 170.2 (C-1″″)] (Table 1). The ¹H–¹H correlation spectroscopy (COSY) spectrum showed cross-peaks for H-1′ (δ_H 6.31)/H₂-2′ (δ_H 1.98, 1.77), H₂-2′/H₃-3′ (δ_H 1.04), H-2″ (δ_H 4.93)/H₂-3″ (δ_H 3.14), H₂-2‴ (δ_H 3.25)/H₂-3‴ (δ_H 1.77), and H₂-3‴/H₃-4‴. Correlations from H-3 (δ_H 6.22) to C-2 (δ_C 159.4), C-1′ (δ_C 72.6), C-4a; H₃-2″″ to C-1″″; and H-1′ (δ_H 6.31) to C-4 in the heteronuclear multiple-bond correlation spectroscopy (HMBC) revealed the structure B shown in Fig. 2. Correalations from H₂-3″ to C-5, C-6, C-7; H₃-5″,6″ to C-2″ (δ_C 93.2), C-4″ (δ_C 71.6), H₂-2‴ to C-1‴ (δ_C 206.2), and 7-OH to C-6, C-7, and C-8 revealed the partial structure A. However, no HMBC cross peaks between partial structures A and B were observed. The ¹³C-NMR chemical shifts of C4a and C8a may differ based on the connection patterns of partial structures A and B. Therefore, to determine the connection pattern of partial structures A and B, model compounds a and b were used for ¹³C-NMR chemical shift calculations (Fig. 2). Based on comparison with the calculated values, the ¹³C-NMR chemical shifts of 1 were very similar to those of model compound a (C-4a: 100.0, C-8a: 161.2). The structure of 1 was determined as shown in Fig. 2.

Fig. 2. Structural Analysis of Compound 1

Partial structures A and B were inferred based on NMR, and the structure of compound 1 was determined by comparison of the ¹³C-NMR data with that of compounds a and b predicted using Spartan18.

Among isolated compounds, compounds 4 and 5 has an asymmetric center at the C2 position of the furan ring where the C2 configuration has not been determined. Therefore, we attempted to determine the absolute configurations of 4 and 5 by comparing the experimental and calculated electronic circular dichroism (ECD) spectra. Attempted were made to determine the configuration by comparing the ECD spectra of model compounds 2S-c, 2R-c, 2S-d, and 2R-d. The ECD spectrum of 2S-c corresponded well with the experimental ECD spectrum of 4, revealing that compound 4 possesses the 2S configuration. The ECD spectrum of 2S-d corresponded well with the experimental ECD spectrum of 5, revealing that compound 5 possesses the 2S configuration (Fig. 3).

Fig. 3. ECD Spectra of Compounds 4, 5, c, and d

The isolated compounds 1–31 were evaluated using the BMI1 promoter inhibitory assay. Compounds 2, 4–12, and 18–22 showed BMI1 promoter inhibitory activity. Notably, compounds 4, 5, 7, 9, 18 and 19 showed relatively strong BMI1 promoter inhibitory activity, with IC₅₀ values of 16.1, 33.7, 24.6, 24.6, 27.6, and 27.7 µM, respectively (Fig. 4).

Fig. 4. Evaluation of BMI1 Promoter Inhibitory Activity of Compounds 4, 5, 7, 9, 18, and 19 Using BMI1/293T Cells

DMSO was used as a control (C) and c-Myc inhibitor II 10074G5 (II) was used as a positive sample.

The cytotoxicity of these compounds against cancer cell lines (DU145, Huh7, and HCT116) that are known to have high BMI1 expression was tested, and their cytotoxicity against tissue-derived cell lines (HEK293) was examined for comparison. Compounds 5, 6, 7, 18, and 20 were cytotoxic to one or more cancer cell lines. Compound 18 showed strong cytotoxicity to DU145 cells, with an IC₅₀ of 25.4 µM.

The effect of compound 18 on DU145 cells was analyzed by Western blotting. The proteins levels of DU145 treated with 18, BMI1, and c-Myc decreased, and p14ARF increased. It has been suggested that suppression of the c-Myc protein, which regulates the BMI1 promoter, leads to downregulation of BMI1, whereas the cancer suppressor p14ARF, which is regulated by BMI1, leads to apoptosis and cell cycle arrest (Fig. 5).

Fig. 5. Cytotoxicity of Compound 18 against DU145, Huh7, HCT116, and HEK293 Cells

Effect of compound 18 on BMI1, c-Myc, and p14ARF protein levels.

Conclusion

One new coumarin (1) and 15 known compounds (2–17) were isolated from M. siamensis, guided by BMI1 promoter inhibitory activity. Coumarin derivatives 2 and 4–12 suppresses the BMI1 promoter and compound 5 shows strong cytotoxicity against Huh7 and HCT116 cells. Fourteen known compounds (18–31) were identified from A. paniculata, guided by the BMI1 promoter inhibitory activity. Labdane diterpene derivatives (18–22) show BMI1 promoter inhibitory activity, and compound 18 shows strong cytotoxicity against DU145. Western blotting analysis of compound 18 using DU145 suggests that suppressing c-Myc leads to an increase in the BMI1 target protein p14ARF. There are few reports of compounds that act on BMI1 and regulate its gene expression. The search for compounds that regulate the BMI1 promoter has contributed to the discovery of compounds that are effective in controlling its expression. Moreover, the identified compounds can be applied to the analysis of epigenetic regulatory molecular mechanisms and treatment of cancer cells.

Experimental

General Experimental Procedure

UV spectra were obtained using a UVmini-1240 spectrophotometer (Shimadzu Corp., Kyoto, Japan), optical rotations were measured using a P-2200 polarimeter, circular dichroism was mesuared using a J-1100 CD spectrometer (JASCO), and IR spectra were acquired using an FT-IR 4700 spectrometer (JASCO, Tokyo, Japan). NMR spectra were obtained using a JEOL ECA600, ECZ600 spectrometer (Tokyo, Japan) in deuterated solvent, where the chemical shift of the solvent was used as the internal standard. HRESIMS was performed using a JEOL JMS-T100LP instrument. Low-resolution electrospray ionization mass spectra (ESI-MS) were obtained using a Shimadzu LCMS-2020 spectrometer (Kyoto, Japan). HPLC (JASCO, Tokyo, Japan) was performed using a device equipped with RI-1530 and UV-970 detectors and a PU-1580 pump. Column chromatography was performed using silica gel 60 N (Kanto Chemical Co., Inc., Tokyo, Japan), Diaion HP-20 (Mitsubishi Chemical Corp., Tokyo, Japan). COSMOSIL Cholester, COSMOSIL 5C₁₈-AR-II, COSMOSIL π-NAP (Nacalai Tesque, Inc., Kyoto, Japan), and Chiral CD-Ph (Shiseido, Inc., Tokyo, Japan) columns were used for HPLC.

Plant Materials

The leaves of M. siamensis (KKP753) were collected from Khon Kaen, Thailand. The leaves of A. paniculata (KKB239) were collected from Natore, Bangladesh. The specimens were deposited in the Department of Natural Products Chemistry, Graduate School of Pharmaceutical Sciences, Chiba University, Japan.

Extraction and Isolation from M. siamensis

Dried leaves of M. siamensis (134 g) were extracted with methanol (MeOH). The MeOH fraction (15.8 g) was resuspended in 10% MeOH and extracted three times with hexane, ethyl acetate (EtOAc), and butanol (n-BuOH). The hexane extract (3.5 g) was fractionated using HP-20 (MeOH/acetone = 1/0–0/1) to give fractions 1A–1C. Fraction 1A (MeOH/acetone = 1/0, 1.6 g) was separated using 60N silica gel column chromatography (hexane/EtOAc/MeOH, 5/1/0–0/0/1) to give 2A–2U. The combined fraction comprising 2C and 2D (hexane/EtOAc/MeOH, 10/3/0, 49.4 mg) was separated using HPLC [COSMOSIL 5C₁₈-AR-II, ϕ 10 × 250 mm, MeOH/H₂O = 16/9] to give 3A–3I. Fraction 3A was determined to be 12 (1.5 mg, t_R 11.2 min). Fraction 3B was determined to be 6 (2.2 mg, t_R 32.0 min). Fraction 3D was determined to be 9 (3.2 mg, t_R 49.6 min). Fraction 3H was determined to be 11 (2.3 mg, t_R 68.4 min). Fraction 3I was determined to be 7 (2.0 mg, t_R 72.0 min). Fraction 3E (hexane/EtOAc/MeOH, 5/3/0, 263 mg) was separated using HPLC [COSMOSIL 5C₁₈-AR-II, ϕ 10 × 250 mm MeOH/H₂O = 7/3] to give 10 (0.8 mg, t_R 29.2 min). Fraction 2E (hexane/EtOAc/MeOH, 5/3/0, 263 mg) was separated using HPLC [COSMOSIL 5C₁₈-AR-II, ϕ 10 × 250 mm MeOH/H₂O = 7/3] to give 4A–4M. Fraction 4H was determined to be 4 (2.2 mg, t_R 20.8 min). Fraction 4C was separated using HPLC [COSMOSIL π-NAP, ϕ 10 × 250 mm, MeCN/H₂O = 3/2] to give 2 (4.0 mg, t_R 21.0 min). Fraction 3M was separated using HPLC [COSMOSIL π-NAP, ϕ 10 × 250 mm, MeCN/H₂O = 3/2] to give 5 (5.4 mg, t_R 22.4 min) and 8 (8.3 mg, t_R 33.6 min). The combined fractions 2F and 2G (hexane/EtOAc/MeOH, 5/3/0, 74.5 mg) were separated using HPLC [COSMOSIL π-NAP, ϕ 10 × 250 mm, MeCN/H₂O = 23/27] to give fractions 5A–5K. Fraction 5I was separated using HPLC [Chiral CD-Ph, ϕ 4.6 × 250 mm, MeCN/H₂O = 1/1] to give 1 (2.2 mg, t_R 12.8 min). Fraction 5K was separated by HPLC [COSMOSIL Cholester, ϕ 10 × 250 mm, MeCN/H₂O = 24/25] to give 6A (2.4 mg, t_R 9.6 min). Fraction 6A was separated using HPLC [Chiral CD-Ph, ϕ 4.6 × 250 mm, MeCN/H₂O = 1/1] to give 3 (1.4 mg, t_R 16.8 min). Fraction 2J (hexane/EtOAc/MeOH, 1/1/0, 66.8 mg) was separated by HPLC [COSMOSIL Cholester, ϕ 10 × 250 mm, MeCN/H₂O = 24/25] to give 13 (2.0 mg, t_R 44.0 min). The EtOAc extract (2.7 g) was fractionated using the HP-20 column (MeOH/acetone = 1/0–0/1) to give fractions 7A–7C. Fraction 7A (MeOH/acetone = 1/0, 2.1 g) was separated using 60N silica gel column chromatography (CHCl₃/MeOH, 5/1–1/1) to give 8A–8O. Fraction 8B (CHCl₃/MeOH, 5/1, 64.1 mg) was separated using HPLC [COSMOSIL AR-II, ϕ 10 × 250 mm, MeCN/H₂O = 1/1] to give 16 (1.7 mg, t_R 16.0 min) and 17 (3.8 mg, t_R 28.0 min). Fraction 8C (CHCl₃/MeOH, 5/1, 196.1 mg) was separated using HPLC [COSMOSIL 5C₁₈-AR-II, ϕ 10 × 250 mm, MeCN/H₂O = 3/7] to give 15 (9.5 mg, t_R 14.4 min) and 14 (30.0 mg, t_R 48.0 min).

Extraction and Isolation from A. paniculata

Dried leaves of A. paniculata (240 g) were extracted with methanol (MeOH). The MeOH fraction (8.6 g) was resuspended in 10% MeOH and extracted three times with EtOAc and n-BuOH. The EtOAc extract (4.2 g) was fractionated using HP-20 (MeOH/acetone = 1/0, 0/1) to obtain fractions 9A and 9B, respectively. Fraction 9A (MeOH/acetone = 1/0, 3.2 g) was separated using 60N silica gel column chromatography (hexane/EtOAc/MeOH, 1/1/0–0/0/1) to give 10A–10C. Fraction 10A (hexane/EtOAc/MeOH, 1/1/0, 300.2 mg) was separated using by HPLC [COSMOSIL Cholester, ϕ 10 × 250 mm, MeCN/H₂O = 3/7] to give 18 (21.7 mg, t_R 31.2 min), 19 (13.4 mg, t_R 33.6 min), and 27 (2.7 mg, t_R 38.2 min). Fraction 10C (hexane/EtOAc/MeOH, 0/0/1, 2.33 g) was separated using 60N silica gel column chromatography (CHCl₃/MeOH, 25/1–15/1) to give 11A–11H. Fraction 11B (CHCl₃/MeOH, 25/1, 50.9 mg) was separated using HPLC [COSMOSIL AR-II, ϕ 10 × 250 mm, MeOH/H₂O = 1/1] to give fractions 12A–12G. Fraction 12G was separated using HPLC [COSMOSIL Cholester, ϕ 10 × 250 mm, MeOH/H₂O = 1/1] to give 13A–13G. Fraction 13G was separated using HPLC [COSMOSIL AR-II, ϕ 10 × 250 mm, MeOH/H₂O = 11/9] to give 28 (2.6 mg, t_R 30.4 min).

Fraction 11C (CHCl₃/MeOH, 20/1, 318.3 mg) was separated using HPLC [COSMOSIL Cholester, ϕ 10 × 250 mm, MeCN/H₂O = 1/1] to give 20 (29.5 1 mg, t_R 16.4 min) and 23 (3.3 mg, t_R 30.4 min). Fraction 11C was separated using HPLC [COSMOSIL Cholester, ϕ 10 × 250 mm, MeCN/H₂O = 3/7] to give 24 (0.4 mg, t_R 15.6 min), 21 (6.5 mg, t_R 16.0 min), 29 (11.8 mg, t_R 23.6 min), and 30 (1.7 mg, t_R 28.0 min). Fraction 11D (CHCl₃/MeOH, 18/1, 709.0 mg) was separated using HPLC [COSMOSIL Cholester, ϕ 10 × 250 mm, MeOH/H₂O = 9/11] to give 22 (2.0 mg, t_R 41.6 min) to give 14A–14L. Fraction 14B was separated using HPLC [COSMOSIL Cholester, ϕ 10 × 250 mm, MeCN/H₂O = 1/3] to give 31 (4.7 mg, t_R 48.0 min). Fraction 12E (CHCl₃/MeOH, 15/1, 104.7 mg) was separated using HPLC [COSMOSIL Cholester, ϕ 10 × 250 mm, MeCN/H₂O = 1/1] to give 25 (1.9 mg, t_R 12.0 min) and 26 (2.6 mg, t_R 18.4 min).

Conformational Analyses and ¹³C-NMR Calculations for Model Compounds a and b

Conformational searches were performed with Spartan 18 software (Wavefunction Inc.) using a PC (Windows 10 Home; Intel Core i7-10770; 2.90 GHz; RAM 16 GB). Stable conformers were identified using the Merck molecular force field (MMFF) method. The stable conformers suggested by MMFF were optimized using the Hartree–Fock (HF)/3-21G model, and stable conformers up to 40 kJ/mol were calculated using the ωB97X-D/6-31G* method. Stable conformers up to 15 kJ/mol were optimized using the ωB97X-D/6-31G* method. Stable conformers of up to 10 kJ/mol were optimized using the ωB97X-D/6-311 + G (2df,2p) method. The ¹³C-NMR calculation for compounds a and b was performed using the ωB97X-D/6-31G* method.

Conformational Analyses and UV/ECD Calculations for Model Compounds S-c and R-c

Conformational searches were performed with Spartan 18 software (Wavefunction Inc.) using a personal computer (PC; Windows 10 Home; Intel Core i7-10770; 2.90 GHz; RAM 16 GB). Stable conformers were identified using the MMFF method. The stable conformers suggested by MMFF were optimized using the HF/3-21G model, and stable conformers up to 40 kJ/mol were calculated using the B3LYP/6-31G method. Stable conformers with energies up to 10 kJ/mol were optimized using the ωB97X-D/6-31G* method. UV/ECD calculations for the compounds were performed using Gaussian R16W (Gaussian)³⁵⁾ on a PC (Windows 10 Education; Intel Xeon E31245; 3.30 GHz; RAM 16 GB). The dominant conformers of the compounds covered > 90% of the population, following the Boltzmann weight. For these conformers, time-dependent density functional theory (TDDFT) calculations were performed using the B3LYP/6-31G* method. The calculated rotatory strength data were converted into a line-shaped ECD curve using a Gaussian distribution function. ECD spectrum were UV corrected.

Conformational Analyses and UV/ECD Calculations for Model Compounds S-d and R-d

Conformational searches were performed with Spartan 18 software (Wavefunction Inc.) using a personal computer (PC; Windows 10 Home; Intel Core i7-10770; 2.90 GHz; RAM 16 GB). Stable conformers were identified using the MMFF method. The stable conformers suggested by MMFF were optimized using the HF/3-21G model, and stable conformers up to 40 kJ/mol were calculated using the ωB97X-D/6-31G* method. Stable conformers with energies up to 15 kJ/mol were optimized using the ωB97X-D/6-31G* method. Stable conformers with energies up to 10 kJ/mol were optimized using the ωB97X-D/6-311 + G (2df, 2p) method. UV/ECD calculations for the compounds were performed using Gaussian R16W (Gaussian)³⁵⁾ on a PC (Windows 10 Education; Intel Xeon E31245; 3.30 GHz; RAM 16 GB). The dominant conformers of the compounds covered > 90% of the population, following the Boltzmann weight. For these conformers, TDDFT calculations were performed using the BHLYP/def2tzvp method. The calculated rotatory strength data were converted into a line-shaped ECD curve using a Gaussian distribution function. ECD spectrum were UV corrected.

Cell Cultures

HEK293T, HEK293, DU145, and HCT116 cells were purchased from the American Type Culture Collection (Manassas, VA, U.S.A.), and Huh7 cells were purchased from the Health Science Research Resources Bank (Osaka, Japan). The cell lines and BMI1/293T cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum (10% FBS). The Cultures were maintained in a humidified incubator at 37 °C under 5% CO₂/95% air.

Luciferase Assay

Luciferase assay was carried out according to the procedures described in a previous paper.¹²⁾

Cell Viability Assay

The cell viability was measured by using the fluorescence microculture cytotoxicity assay (FMCA).³⁶⁾ Cytotoxicity assay was carried out according to the procedures described in a previous paper.¹²⁾ In this study, cells at a density of 1.5 × 10³ cells/well were seeded in a 96-well plate and incubated for 24 h. The plate was then treated with the compounds for 72 h. Thereafter, 200 µL of phosphate buffered saline (PBS) containing fluorescein diacetate (10 µg/mL; Wako, Osaka, Japan) was added to each well. After the plate was incubated for 1 h, the fluorescence intensity at 538 nm with excitation at 485 nm was measured using Fluoroskan Ascent (Thermo, Waltham, MA, U.S.A.).

Western Blot Analysis

DU145 (1 × 10⁶ cells) was plated in a 6-cm dish and incubated for 24 h. The plate was then incubated with the sample for 72 h. Proteins were obtained by disruption of the cells. The proteins were mixed with 5X sample buffer (20% 2-mercaptoethanol, 20% sucrose, 8% sodium dodecyl sulfate, 0.02% bromophenol blue, and 0.25 µM Tris–HCl, pH 6.8), boiled (100 °C for 3 min), loaded onto a 12.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel, and the assay was run at 30 mA. Proteins were transferred onto polyvinylidene difluoride (PVDF) membranes (Bio-Rad, Hercules, CA, U.S.A.). The membranes were blocked with 5% skimmed milk in TBST (Tris-buffered saline with Tween20) and probed with anti-BMI1 (1 : 1000, #6964, Cell Signaling, Danvers, MA, U.S.A.), anti-β-actin (1 : 4000, #A2228, Sigma, Darmstadt, Germany), anti-c-myc (1 : 200, #sc-40, Santa Cruz Biotechnology), and anti-p14ARF (1 : 200, #ab3642, abcam, U.K.) antibodies. The membrane was incubated at room temperature with horseradish peroxidase (HRP)-conjugated anti-mouse immunoglobulin G (IgG) (1 : 4000, #NA931VS, GE Healthcare, Little Chalfont, U.K.) for c-Myc and β-actin or anti-rabbit IgG (1 : 4000, #7074, Cell Signaling) for BMI1 and p14ARF. After washing with TBST, immunoreactive bands were detected using an ECL Advance Western detection system (GE Healthcare) or Immobilon Western chemiluminescent HRP substrate. Chemiluminescence signals were imaged with ChemiDoc XRS Plus (Bio-Rad, Hercules, CA, U.S.A.).

Compound 1

White powder; ¹H-NMR (CDCl₃) δ: 1.04 (3H, overlap), 1.04 (3H, overlap), 1.29 (3H, s), 1.38 (3H, s), 1.77 (1H, m), 1.77 (2H, m), 1.98 (1H, m), 2.17 (3H, s), 3.16 (2H, overlap), 3.25 (2H, t, J = 7.4 Hz), 4.93 (1H, t, J = 9.0 Hz), 6.22 (1H, s), 6.31 (1H, dd, J = 8.3, 3.3 Hz), 14.18 (1H, s); ¹³C-NMR (CDCl₃) δ: 9.7, 13.8, 18.0, 21.0, 24.8, 26.0, 26.8, 28.2, 46.6, 71.6, 72.6, 93.2, 97.2, 105.2, 106.1, 110.2, 155.9, 157.5, 159.4, 161.4, 163.3, 170.2, 206.2; [α]_D²⁵ −38.4 (c 0.1, MeOH); IR (ATR) cm⁻¹: 3320, 2940, 2830, 1635, 1610, 1450, 1410, 1110, 1020, and 610; UV λ_max (MeOH) nm (log ε): 295 (4.09) and 221 (4.11); HR-ESI-MS m/z 455.1687 [M + Na]⁺ (Calcd for C₂₃H₂₈O₈Na, 455.1682).

Compound 2

White powder; [α]_D²⁶ −67.4 (c 0.1, MeOH); IR (ATR) cm⁻¹: 3320, 2940, 2830, 1650, 1450, 1020, 630, and 620; UV λ_max (MeOH) nm (log ε): 297 (1.20) and 222 (1.17); HR-ESI-MS m/z 469.1806 [M + Na]⁺ (Calcd for C₂₄H₃₀O₈Na, 469.1838); for ¹H- and ¹³C-NMR see Table 2.

Compound 3

White powder; [α]_D²⁶ −17.1 (c 0.1, MeOH); IR (ATR) cm⁻¹: 3330, 2940, 2830, 1630, 1610, 1450, 1410, 1120, and 1020; UV λ_max (MeOH) nm (log ε): 294 (1.54) and 221 (1.66); HR-ESI-MS m/z 469.1810 [M + Na]⁺ (Calcd for C₂₄H₃₀O₈Na, 469.1838); for ¹H- and ¹³C-NMR see Table 3.

Compound 4

White powder; [α]_D²⁵ −2.2 (c 0.1, MeOH); UV λ_max (MeOH) nm (log ε): 332 (0.72) and 283 (1.82); CD λ_max (MeOH) nm (Δε): 305 (2.88), 273 (5.00), 240 (0.66) and 214 (−1.16); ESI-MS m/z 375 [M + H]⁺.

Compound 5

Yellow powder; [α]_D²⁷ −13.4 (c 0.1, MeOH); UV λ_max (MeOH) nm (log ε): 343 (0.48), 281 (1.14), and 201 (1.29); CD λ_max (MeOH) nm (Δε): 344 (1.84), 285 (6.22), and 227 (−7.34); ESI-MS m/z 423 [M + H]⁺.

Acknowledgments

We thank the late Takashi Koyano (Temko Corporation) and late Thaworn Kowithayakorn (Khon Kaen University) for their collaboration in the collection and identification of plant materials. This work was supported by KAKENHI (Grant Nos. 20H03394 and 20K16024) from the Japan Society for the Promotion of Science and JST SPRING (Grant No. JPMJSP2109).

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

References

• 1) Avgustinova A., Benitah S. A., Nat. Rev. Mol. Cell Biol., 17, 643–658 (2016).
• 2) Gray F., Cho J. H., Shukla S., He S., Harris A., Boytsov B., Jaremko L., Jaremko M., Demeler B., Lawlor R. E., Grembecka J., Cierpicki T., Nat. Commun., 7, 13343 (2016).
• 3) Lin X., Ojo D., Wei F., Wong F., Gu Y., Tang D., Biomolecules, 5, 3396–3415 (2015).
• 4) Avgustinova A., Benitah S. A., Nat. Rev. Mol. Cell Biol., 17, 643–658 (2016).
• 5) Zhou M., Xu Q., Huang D., Luo L., Biomedical Reports, 14, 52 (2021).
• 6) Chen D., Wu M., Li Y., Chang I., Yuan Q., Ekimyan-Salvo M., Deng P., Yu B., Yu Y., Dong J., Szymanski M. J., Ramadoss S., Li J., Wang C., Cell Stem Cell, 20, 621–634.e6 (2017).
• 7) Yang D., Liu H. Q., Yang Z., Fan D., Tang Q. Z., eBioMedicine, 63, 103193 (2021).
• 8) Kreso A., van Galen P., Pedley N. M., Lima-Fernandes E., Frelin C., Davis T., Cao L., Baiazitov R., Du W., Sydorenko N., Moon Y. C., Gibson L., Wang Y., Leung C., Iscove N. N., Arrowsmith C. H., Szentgyorgyi E., Gallinger S., Dick J. E., O’Brien C. A., Nat. Med., 20, 29–36 (2014).
• 9) Siddique H. R., Saleem M., Stem Cells Int., 30, 372–378 (2012).
• 10) Aponte M. P., Caicedo A., Stem Cells Int., 2017, 5619472 (2017).
• 11) Ishibashi M., J. Antibiot., 74, 629–638 (2021).
• 12) Kaneta Y., Arai M. A., Ishikawa N., Toume K., Koyano T., Kowithayakorn T., Chiba T., Iwama A., Ishibashi M., J. Nat. Prod., 80, 1853–1859 (2017).
• 13) Yokoyama Y., Arai M. A., Hara Y., Ishibashi M., Bioorg. Med. Chem., 27, 2998–3003 (2019).
• 14. Yokoyama Y., Arai M. A., Hara Y., Ishibashi M., Nat. Prod. Commun., 14, 1934578X19866583 (2019).
• 15) Crombie L., Games D. E., Haskins N. J., Reed G. F., J. C. S. Perkin, I, 2255–2260 (1972).
• 16) Prachyawarakorn V., Mahidol C., Ruchirawat S., Pharm. Biol., 38 (Suppl 1), 58–62 (2000).
• 17) Mahidol C., Kaweetripob W., Prawat H., Ruchirawat S., J. Nat. Prod., 65, 757–760 (2002).
• 18) Verotta L., Lovaglio E., Vidari G., Finzi P. V., Neric G. M., Raimondi A., Parapini S., Taramelli D., Riva A., Bombardelli E., Phytochemistry, 65, 2867–2879 (2004).
• 19) Morel C., Guileta D., Oger M. J., Séraphin D., Sévenetb T., Wiartc C., Hadic A., Richommea P., Bruneton J., Phytochemistry, 50, 1243–1247 (1999).
• 20) Wirongrong K., Chulabhorn M., Hunsa P., Somsak R., Pharm. Biol., 38, 55–57 (2000).
• 21) Prachyawarakorn V., Mahidol C., Ruchirawat S., Pharm. Biol., 38 (Suppl 1), 58–62 (2000).
• 22) Crombie L., Games D. E., Haskins N. J., Reed G. F., Finnegan R. A., Merkel K. E., Tetrahedron, 11, 3975–3978 (1970).
• 23) Catalána A. C., Heluania S. C., Kotowicza C., Gedrisb E. T., Herz W., Phytochemistry, 64, 625–629 (2003).
• 24) Markham K. R., Sheppard C., Geiger H., Phytochemistry, 26, 3335–3337 (1987).
• 25) Park Y., Moon B. H., Lee E., Lee Y., Yoon Y., Ahn J. H., Lim Y., Magn. Reson. Chem., 45, 674–679 (2007).
• 26) Christophoridou S., Dais P., Anal. Chim. Acta, 633, 283–292 (2009).
• 27) Matuda T., Kuroyanagi M., Sugiyama S., Umehara K., Ueno A., Nishi K., Chem. Pharm. Bull., 42, 1216–1225 (1994).
• 28) Sun X., Yan H., Zhang Y., Wang X., Qin D., Yu J., Molecules, 24, 620 (2019).
• 29) Chen L., Zhu H., Wang R., Zhou K., Jing Y., Qiu F., J. Nat. Prod., 71, 852–855 (2008).
• 30) Chen L., Qiu F., Wei H., Qu G., Yao X., Helv. Chim. Acta, 89, 2654–2664 (2006).
• 31) Gan L., Zheng Y., Deng L., Sun P., Ye J., Wei X., Liu F., Yu L., Ye W., Fan C., Liu J., Zhang W., Molecules, 24, 2726 (2019).
• 32) Cava M. P., Chan W. R., Stein R. P., Willis C. R., Tetrahedron, 21, 2617–2632 (1965).
• 33) Shen Y., Li L., Xiao W., Xu G., Lin Z., Zhao Q., Sun H., J. Nat. Prod., 69, 319–322 (2006).
• 34) Kuroyanagi M., Sato M., Ueno A., Nishi K., Chem. Pharm. Bull., 35, 4429–4435 (1987).
• 35) Gaussian 16, Revision B.01, Frisch M. J., Trucks G. W., Schlegel H. B., et al., Gaussian, Inc., Wallingford CT, 2016.
• 36) Lindhagen E., Nygren P., Larsson R., Nat. Protoc., 3, 1364–1369 (2008).