2020 年 68 巻 9 号 p. 903-906
Bioassay screening using Indonesian plants, such as traditional foods (vegetables, spices, and tea) and folk medicinal herbs, identified eight protein tyrosine phosphatase (PTP) 1B inhibitory and two antibacterial plants. The leaves of Syzygium polyanthum (Wight) Walp. were examined in more detail to define PTP1B inhibitory components, resulting in the isolation of a new active acylbenzene (1) along with four related congeners of 1 (2–5) and four oleanane triterpenes (6–9). The structure of 1 was elucidated as 12-oxo-12-(2,3,5-trihydroxy-4-methylphenyl)dodecanoic acid based on its spectroscopic data. The acylbenzenes 1 and 3–5 inhibited PTP1B activity with IC50 values ranging between 9.5 and 14 µM, whereas the triterpenes 7–9 also suppressed this activity with IC50 values of 3.3–5.7 µM.
Plants, microorganisms, and marine invertebrates have been a significant source of natural products with structurally and biologically diverse properties, and have contributed to the supply of various pharmaceutical applications for human health.1) In recent years, as the discovery of novel drug candidates from natural origins has been gradually declining, unutilized natural resources in developing countries have been attracting attention in this research field. North Sulawesi in Indonesia is an archipelagic region and still maintains numerous bioresources; therefore, it may be promising to access substances of interest and significance from pristine natural environments.2,3) Many types of wild plants that inhabit this area are traditionally utilized as vegetables, spices, tea, and medicinal herbs.4)
To identify useful edible plants containing bioactive constituents, thirty samples listed in Table 1 were collected around Manado city (North Sulawesi, Indonesia), and their EtOH extracts were applied to bioassay screening for protein tyrosine phosphatase (PTP) 1B inhibitory activity as well as antibacterial and antimycobacterial activities. PTP1B inhibitors are expected to be potential agents for the treatment and prevention of type 2 diabetes mellitus (T2DM) and obesity5) because this enzyme is regarded as an important negative regulator in the insulin and leptin signaling pathways.6) In the present study, the leaf extract of Syzygium polyanthum (Wight) Walp. (Myrtaceae) with potent PTP1B inhibitory effects (71% inhibition) in the in vitro enzyme assay3) was separated by bioactivity-guided purification to obtain five acylbenzene derivatives (1–5)7,8) that included one new derivative, 12-oxo-12-(2,3,5-trihydroxy-4-methylphenyl)dodecanoic acid (1), and four oleanane-type triterpenes (6–9),9) as shown in Fig. 1. We herein describe the bioactive screening results of Indonesian edible plants and the isolation, structural elucidation, and biological activities of 1–9 from S. polyanthum.
| No. | Scientific name | Part | Inhibitory ratea) | Inhibition zone (mm)b) | MIC (µg/mL) | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| PTP1B | B. subtilis | S. aureus | E. coli | M. smegmatis | M. bovis BCG | M. avium | M. intracellulare | |||
| 1 | Plumeria sp. | Leaves | 82% inhibition | —c) | — | — | >100 | >50 | >100 | >100 |
| 2 | Unidentified | Aerial | — | — | — | — | >100 | >50 | >100 | >100 |
| 3 | Unidentified | Aerial | 22% inhibition | — | — | — | >100 | >50 | >100 | >100 |
| 4 | Ruellia tuberosa | Aerial | — | — | — | — | >100 | >50 | >100 | >100 |
| 5 | Talinum paniculatum | Leaves | — | — | — | — | >100 | >50 | >100 | >100 |
| 6 | Peperomia pellucida | Aerial | — | — | — | — | >100 | >50 | >100 | >100 |
| 7 | Cordyline sp. | Leaves | 64% inhibition | — | — | — | >100 | >50 | >100 | >100 |
| 8 | Cordyline sp. | Leaves | 80% inhibition | — | — | — | >100 | >50 | >100 | >100 |
| 9 | Coleus sp. | Leaves | — | — | — | — | >100 | >50 | >100 | >100 |
| 10 | Muntingia calabura | Leaves | 76% inhibition | — | — | — | >100 | >50 | >100 | >100 |
| 11 | Cordyline sp. | Leaves | — | — | — | — | >100 | >50 | >100 | >100 |
| 12 | Unidentified | Leaves | 83% inhibition | — | — | — | >100 | >50 | >100 | >100 |
| 13 | Unidentified | Leaves | 46% inhibition | — | — | — | >100 | >50 | >100 | >100 |
| 14 | Annona muricata | Leaves | — | — | 11 | — | >100 | >50 | >100 | >100 |
| 15 | Unidentified | Leaves | — | — | — | — | >100 | >50 | >100 | >100 |
| 16 | Hibiscus rosa-sinensis | Flower | — | — | — | — | >100 | >50 | >100 | >100 |
| 17 | Unidentified | Leaves | — | — | — | — | >100 | >50 | >100 | >100 |
| 18 | Unidentified | Leaves | — | — | — | — | >100 | >50 | >100 | >100 |
| 19 | Drymoglossum piloselloides | Aerial | — | 13 | — | — | >100 | >50 | >100 | >100 |
| 20 | Abelmoschus sp. | Leaves | — | — | — | — | >100 | >50 | >100 | >100 |
| 21 | Abelmoschus sp. | Leaves | — | — | — | — | >100 | >50 | >100 | >100 |
| 22 | Carica papaya | Leaves | 51% inhibition | — | — | — | >100 | >50 | >100 | >100 |
| 23 | Clerodendrum minahassae | Leaves | — | — | — | — | >100 | >50 | >100 | >100 |
| 24 | Abelmoschus sp. | Leaves | — | — | — | — | >100 | >50 | >100 | >100 |
| 25 | Momordica charantia | Leaves | — | — | — | — | >100 | >50 | >100 | >100 |
| 26 | Diplacium esculentum | Leaves | — | — | — | — | >100 | >50 | >100 | >100 |
| 27 | Syzygium polyanthum | Leaves | 71% inhibition | — | — | — | >100 | >50 | >100 | >100 |
| 28 | Syzygium aromaticum | Fruits | 92% inhibition | — | — | — | >100 | >50 | >100 | >100 |
| 29 | Coriandrum sativum | Fruits | 77% inhibition | — | — | — | >100 | >50 | >100 | >100 |
| 30 | Myristica fragrans | Fruits | 86% inhibition | — | — | — | >100 | >50 | >100 | >100 |
a) 50 µg/mL; b) 10 µg/6 mm disk; c) Not active.
Thirty Indonesian edible plants collected at Manado and its surroundings were extracted by EtOH to evaluate their inhibitory activities against PTP1B. Screening results are summarized in Table 1.
The extracts of leaves of Plumeria sp. (No. 1), Cordyline sp. (No. 8), Muntingia calabura (No. 10), unidentified plant (No. 12), and Syzygium polyanthum (No. 27) and fruits of S. aromaticum (No. 28), Coriandrum sativum (No. 29), and Myristica fragrans (No. 30) achieved more than 70% inhibition at 50 µg/mL (Table 1). The leaf extract of another Cordyline sp. (No. 7), unidentified plant (No. 13), and Carica papaya (No. 22) exhibited moderate activity (46–64% inhibition), while aerial parts of unidentified plant (No. 3) modestly inhibited PTP1B activity (22% inhibition) at the same concentration (Table 1).
Screening extracts were also assessed for antibacterial activities against Bacillus subtilis, Staphylococcus aureus, and Escherichia coli and antimycobacterial activities against Mycobacterium smegmatis, M. bovis BCG, M. avium, and M. intracellulare using the paper disk method. The extracts of aerial parts of Drymoglossum piloselloides (No. 19) and leaves of Annona muricata (No. 14) exhibited inhibition zones at 10 µg/disk against B. subtilis (13 mm) and S. aureus (11 mm), respectively, and none of the plant extracts showed growth inhibition against four mycobacteria in the range of 50 to 100 µg/mL (Table 1). The test strains of M. smegmatis and M. bovis BCG are alternative microorganisms to detect anti-tuberculous activity10) and M. avium and M. intracellulare are pathogens that cause Mycobacterium avium complex (MAC) disease.11) Nevertheless, flower parts of Hibiscus rosa-sinensis (No. 16), which is recognized as an indigenous anti-tuberculous herb in Indonesia,12) was not active in our assays.
The leaves of S. polyanthum (No. 27), which is also known as “Daun Salam,” is a familiar herb that is used not only as a food additive and tea with flavor, but also as folk remedies toward several diseases including diabetes.13) Moreover, Saifudin et al. reported PTP1B inhibitory activity of the leaves of S. polyanthum and the active constituents.7) Therefore, we proceeded to further identify the active constituents in the leaves of S. polyanthum (No. 27) in the present study.
The EtOH extract from the leaves of S. polyanthum was separated into ten fractions using an octadecyl silica (ODS) column. Bioactive fractions were subsequently purified by preparative HPLC (ODS) to afford compounds 1 (2.9 mg), 2 (4.5 mg), 3 (2.7 mg), 4 (1.4 mg), 5 (3.8 mg), 6 (3.9 mg), 7 (1.2 mg), 8 (2.9 mg), and 9 (7.1 mg). Compounds 2–9 were assigned as 10-oxo-10-(2,3,5-trihydroxy-4-methylphenyl)decanoic acid,7) 1-(2,3,5-trihydroxy-4-methylphenyl)hexane-1-one,8) 1-(2,3,5-trihydroxy-4-methylphenyl)octane-1-one,8) 1-(2,3,5-trihydroxy-4-methylphenyl)decane-1-one,8) oleanolic acid,9) ursolic acid,9) arjunolic acid,9) and asiatic acid,9) respectively, by comparing their spectroscopic data with reported values (Fig. 1).
The UV absorption of 1 at 291 nm in CH3OH suggested the presence of a phenyl moiety, and the IR spectrum of 1 indicated the presence of hydroxy and carbonyl groups by bands at 3411 and 1684 cm−1, respectively. The 1H-NMR spectrum of 1 showed an aromatic proton (δH 5.88), aryl methyl proton (δ 1.90), and some methylene protons (δ 3.01, 2.27, 1.64, 1.58, and 1.31), and 17 carbon signals in the 13C-NMR spectrum were classified into two carbonyl carbons corresponding to ketone and carboxyl groups (δC 207.6 and 177.8), three sp2 oxygenated quaternary carbons (δ 165.0, 163.8, and 161.3), two sp2 quaternary carbons (δ 105.2 and 103.6), one sp2 methine carbon (δ 94.8), eight sp3 methylene carbons (δ 44.9, 35.0, 30.7, 30.6, 30.4, 30.2, 26.4, and 26.1), and one methyl carbon (δ 7.3) from the analysis of DEPT135 and heteronuclear multiple quantum coherence (HMQC) spectra. These spectroscopic data of 1 were identical to those of 2,7) suggesting that the molecular structure of 1 possesses a benzene skeleton with a long acyl chain similar to that of 2.
High resolution-electron ionization (HR-EI)MS of 1 gave its molecular formula as C19H28O6 (m/z 352.1873 [M]+, Δ–1.3 mmu), which was suggested to have two more methylene units (–CH2CH2–; 28 Da) than that of 2. Additionally, the integral value of the methylene proton at δH 1.31 (approximately 12H) for 1 was larger than that for 2 (approximately 8H) in their 1H-NMR spectra. These marked differences between 1 and 2 demonstrated that compound 1 was a derivative of 2 with a longer acyl chain due to the presence of C2, which was confirmed by the analysis of correlation spectroscopy (COSY) and heteronuclear multiple bond connectivity (HMBC) data (Fig. 2) and named 12-oxo-12-(2,3,5-trihydroxy-4-methylphenyl)dodecanoic acid (Fig. 1).
The PTP1B inhibitory effects of all isolated compounds, except for 6, were examined using an enzyme assay.3) Compound 6, oleanolic acid, is a typical PTP1B inhibitor,14) and, thus, authentic oleanolic acid was used as a positive control in the present study. The IC50 values of 1–5, 7–9, and the positive control (oleanolic acid) are listed in Table 2. Compound 1 inhibited PTP1B activity with an IC50 value of 9.6 µM, while 2 exhibited inhibitory activity of 35% at 31 µM and 3–5 showed IC50 values of 14, 9.5, and 9.6 µM, respectively. The inhibitory activities of 2 and 5 with a C10 long chain revealed that the 10′-methyl group in 5 was more favorable for this activity than the 1′-carboxylic acid moiety in 2. Moreover, the longer acyl chain appeared to be more suitable for inhibitory activity. Although the IC50 value of 2 was reported as 13.1 µM in the previous study,7) our assay method gave the weaker inhibitory activity (35% inhibition at 31 µM). The oleanane triterpenes 7–9 showed the lowest IC50 values (3.3–4.3 µM) in the present study, and we already reported the PTP1B inhibitory activities of the same type of triterpenes from Lantana camara,2) which is used to control blood glucose levels in T2DM similar to S. polyanthum.13) Consequently, the evaluation of medicinal plants containing triterpenes in an in vivo mouse model is ongoing, and the results obtained will be described elsewhere.
| Compound | (IC50, µM) |
|---|---|
| PTP1B | |
| 1 | 9.6 |
| 2 | 35% inhibition at 31 µM |
| 3 | 14 |
| 4 | 9.5 |
| 5 | 9.6 |
| 7 | 4.3 |
| 8 | 3.3 |
| 9 | 5.7 |
| Oleanolic acida) | 1.1 |
a) Positive control.14)
In conclusion, we screened thirty Indonesian plant extracts for PTP1B inhibitory, antibacterial, and antimycobacterial activities, identified eight plant extracts that potently inhibited PTP1B, and isolated a new PTP1B inhibitor, 12-oxo-12-(2,3,5-trihydroxy-4-methylphenyl)dodecanoic acid (1), from the leaves of S. polyanthum (No. 27). Continuous efforts to identify and isolate the active compounds of the other selected extracts are currently under way.
The edible plants used in the present study were collected in the area of Manado, North Sulawesi, Indonesia. Voucher specimens were preserved in the University of Pembangunan Indonesia (Manado, Indonesia) and North Sulawesi Research and Development Agency (Manado, Indonesia). Each plant sample (12.5–750 mg) was cut into small pieces and extracted with EtOH (approximately 200 mL). After filtration, the solution was concentrated in vacuo to give a crude extract. The extracts obtained were dissolved in CH3OH at a concentration of 5 mg/mL and applied to bioactive screening assays.
Isolation of New Compound 1The leaf parts of S. polyanthum were collected at Manado in Indonesia. A voucher specimen was deposited at the Faculty of Nursing, University of Pembangunan Indonesia and North Sulawesi Research and Development Agency as 19F12.
The plant (215 g, dry weight) was cut into small pieces and exhaustively extracted with EtOH. The extract (8.7 g), after evaporation, was divided into ten fractions (Frs. 1–10) with an ODS column (100 g) by stepwise elution with CH3OH in H2O. Fractions 2–5 eluted with 30, 50, and 70% CH3OH were combined (112 mg) and purified by HPLC [column; PEGASIL ODS SP100 (Senshu Scientific. Co., Ltd., Tokyo, Japan), i.d. 10 × 250 mm; mobile phase, 40% CH3OH in H2O containing 0.05% TFA; detection, UV at 210 nm; flow rate, 2.0 mL/min] to give compound 1 (2.9 mg, tR = 30.3 min).
12-Oxo-12-(2,3,5-trihydroxy-4-methylphenyl)dodecanoic acid (1): Pale yellow oil; 1H-NMR (CDCl3) δ: 5.88 (1H, s, H-6), 3.01 (2H, t, J = 7.3 Hz, H2-11′), 2.27 (2H, br s, H2-2′), 1.90 (3H, s, 4-CH3), 1.64 (2H, t, br s, H2-10′), 1.58 (2H, br s, H2-3′), 1.31 (10H, br s, H2-4′–H2-9′); 13C-NMR (CDCl3) δ: 207.6 (s, C-12′), 177.8 (s, C-1′), 165.0 (s, C-3), 163.8 (s, C-5), 161.3 (s, C-2), 105.2 (s, C-1), 103.6 (s, C-4), 94.8 (d, C-6), 44.9 (t, C-11′), 35.0 (t, C-2′), 30.7, 30.6, 30.4, 30.2, 26.4, 26.1 (t, C-3′–C-10′), 7.3 (q, 4-CH3); IR (KBr) cm−1: 3411, 2931, 2855, 1684, 1626, 1435, 1207; UV λmax (CH3OH) nm (log ε): 223 (3.6), 291 (3.8); HR-EI-MS m/z: 352.1873 (Calcd for C19H28O6: 352.1886); EI-MS m/z: 352 [M]+.
This work was supported in part by the Kanae Foundation for the Promotion of Medical Science to H. Y. and the Grant for Basic Science Research Projects from The Sumitomo Foundation to H. Y. and by the Directorate for Research and Community Service-Directorate General of Research and Development Strengthening-Ministry of Research, Technology, and Higher Education of the Republic of Indonesia to K. M. M. We express our thanks to Mr. T. Matsuki and S. Sato of Tohoku Medical and Pharmaceutical University for measuring mass spectra.
The authors declare no conflict of interest.
The online version of this article contains supplementary materials.
