Key words
Colebrookea oppositifolia
- Lamiaceae - 14,15-dinor-
cis-labdane diterpene - antimycobacterium - flavonoids
Colebrookea (Lamiaceae) is a monotypic genus represented by Colebrookea oppositifolia Sm. (syn. = Colebrookea ternifolia Roxb.), commonly known as Panrasa, and is distributed in hilly parts of India and
China [1], [2]. The roots of this shrub are used for epilepsy and the leaves are applied for wound
healing and bruises [1], [2], [3]. It is used for the treatment of fractures, traumatic injuries, and rheumatoid arthritis
in China [2]. Some other traditional uses are: the decoction of its roots is given as an abortifacient;
the juice of the leaves is used to stop bleeding and as an eye and ear drop; and the
paste of the leaves is applied to toothaches and mouth and tongue sores [4]. Different extracts of this shrub are reported to exhibit antibacterial [5], [6], [7], antimycobacterial [8], antioxidant [7], and antifertility [9] activities. Acteoside, a constituent from the aerial parts, exhibited an in vitro potent synergistic fungicidal effect in combination with amphotericin B [10]. Different parts of this plant have been studied phytochemically to isolate flavonoids
[2], [11], [12], [13], [14], acteoside [10], sterols [15], and fatty compounds [11], [15].
Despite the availability of treatment, tuberculosis (TB) continues to be a deadly
disease [16], [17], [18]. We are continuously involved in the process of the isolation of novel phytochemicals
with promising anti-TB activity [19], [20], [21]. During our program for the isolation of anti-TB compounds from plants found in
Western Ghats of Maharashtra, India, a phytochemical analysis of the acetone extract
of the aerial parts of C. oppositifolia was performed. Herein we report the isolation and structure elucidation of compounds
1–12 ([Fig. 1]) and their evaluation for antimycobacterium activity against two microbial strains,
Mycobacterium tuberculosis H37Ra and Mycobacterium bovis in both active and dormant phases.
Fig. 1 Compounds 1–12 isolated from the acetone extract of the aerial parts of C. oppositifolia.
Compound 1 was obtained as yellow gum. Analysis of the 13 C NMR and DEPT-135 spectra revealed 18 resonances along with a pseudomolecular peak
[M + Na]+ at m/z 301.2135 in the high-resolution electrospray ionization mass spectrometry (HRESIMS;
Fig. 1 S, Supporting Information) and allowed for the determination of the molecular formula
as C18H30O2, corresponding to four indices of the hydrogen deficiency. The 1 H NMR data ([Table 1]) showed the presence of four tertiary methyl singlets at δ
H 0.86, 0.90, 1.06, and 2.27, and one secondary methyl at δ
H 0.72 with coupling constant 6.8 Hz. Two methine protons at δ
H 6.35 and 6.80 with coupling constant 15.9 Hz suggested the presence of a trans double bond. 13 C NMR data ([Table 1]) showed the presence of a carbonyl carbon at δ
C 197.8 and two methine carbons at δ
C 130.1 and 151.4 accounted for two indices of hydrogen deficiencies, suggesting that
1 was a bicyclic diterpenoid. The NMR data of 1 was similar to previously reported dinor-labdane diterpene, 8-hydroxy-14,15-dinor-11-labden-13-one
[22], [23], except that a tertiary carbon at δ
C 73.5 at position C-8 was replaced by a methine carbon at δ
C 34.0, and a methine carbon at δ
C 67.0 at position C-9 was replaced by the oxygenated tertiary carbon at δ
C 79.6 in compound 1. These observations established the 14,15-dinor diterpene skeleton for 1. The structure was confirmed by 2D NMR studies as follows: The methyl protons at
δ
H 0.86 and 0.90 showed the heteronuclear multiple bond correlation (HMBC; Fig. 6 S, Supporting Information) with a methylene carbon at δ
C 41.6 and a quaternary carbon at δ
C 33.4. A methine proton at δ
H 1.50 showed the HMBC correlation with a methylene carbon at δ
C 30.6 and with a methyl carbon at δ
C 17.2. Methyl protons at δ
H 0.72 and 1.06 and a methine proton at δ
H 6.35 showed the HMBC correlation with an oxygenated tertiary carbon at δ
C 79.6. This suggested the placement of an oxygenated tertiary carbon at position C-9
and a methyl (δ
H 0.72) at position C-8 unequivocally. Other HMBC correlations from protons at δ
H 6.35 and 6.80 with the ketonic carbonyl carbon at δ
C 197.8 and a proton at δ
H 6.35 with a methyl carbon at δ
C 27.9 were also observed ([Fig. 2]). The correlation spectroscopy (COSY; Fig. 7 S, Supporting Information) correlations were observed between δ
H 1.38 and 1.43, δ
H 1.43 and 1.17, δ
H 1.50 and 1.31, δ
H 1.31 and 1.52, δ
H 1.52 and 2.04, and δ
H 2.04 and 0.72, and suggested a H2-1-H2-2-H2-3 and H1-5-H2-6-H2-7-H1-8-H3-17 linkage, which confirmed the presence of a decalin ring skeleton in 1 ([Fig. 2]).
Fig. 2 Key HMBC (H→C), H1-H1COSY (−) and NOESY (↔) correlations of compound 1.
Table 1 1 H and 13 C NMR data for compound 1 in CDCl3 (500 MHz for 1 H and 125 MHz for 13 C, δ in ppm).
Position
|
δ
H
|
δ
C
|
HMBC
|
1a 1b
|
1.38 (m) 1.59 (m)
|
21.4
|
–
|
2a 2b
|
1.43 (m) 1.52 (m)
|
18.4
|
–
|
3a 3b
|
1.17 (m) 1.36 (m)
|
41.6
|
C-18, C-19
|
4
|
–
|
33.4
|
–
|
5
|
1.50 (m)
|
45.4
|
C-6, C-20
|
6a 6b
|
1.31 (m) 1.65 (m)
|
30.6
|
C-8
|
7a 7b
|
1.12 (m) 1.52 (m)
|
32.8
|
C-20
|
8
|
2.04 (m)
|
34.0
|
C-6
|
9
|
–
|
79.6
|
–
|
10
|
–
|
42.4
|
–
|
11
|
6.80 (d, J = 15.9 Hz)
|
151.4
|
C-13
|
12
|
6.35 (d, J = 15.9 Hz)
|
130.1
|
C-9, C-13, C-16
|
13
|
–
|
197.8
|
–
|
16
|
2.27
|
27.9
|
–
|
17
|
0.72 (d, J = 6.8 Hz)
|
16.2
|
C-9
|
18
|
0.90 (s)
|
33.7
|
C-3, C-4
|
19
|
0.86 (s)
|
22.0
|
C-3, C-4
|
20
|
1.06 (s)
|
17.2
|
C-5, C-7, C-9
|
Comparison of NMR values of 1 with those reported for 8-hydroxy-14,15-dinor-11-labden-13-one [22], [23] revealed a deviation in the chemical shift at the carbon 5, i.e., upfield shift
by 10 ppm at C-5 in 1. This upfield shift of C-5 (~ 10 ppm) can be explained by the cis-fused A/B ring junction [24], [25]. The fused rings assume a nonsteroidal conformation (cis A/B ring junction), as revealed by the strong nuclear overhauser effect spectroscopy
(NOESY; Fig. 8 S, Supporting Information) correlation observed between H3-20 (δ
H 1.06) and H-5 (δ
H 1.50) ([Fig. 2]). Other observed NOESY correlations were between δ
H 1.06 and 6.80 and between δ
H 1.06 and 2.04, and indicated that the presence of the side chain at position C-9
was β orientated and a methyl at position C-8 was α orientated. Thus, based on a combination of detailed analysis of the 2D NMR data
and comparison of observed and literature NMR data with the reported compound [22], [23], 1 was identified as a new natural product, (+)-14,15-dinor-9α-hydroxy-cis-labd-11(E)-en-13-one, and belongs to the rare class of 14,15-dinor-diterpenes with
a cis A/B ring junction.
Compounds 2–12 ([Fig. 1]) were identified by comparison with observed and literature NMR data, and supported
by liquid chromatography electrospray ionization mass spectrometry (LCESIMS) data.
Compounds 2–8 were identified as flavonoids alnustin (2) [26], mosloflavone (3) [27], flindulatin (4) [28], 5,6,7-trimethoxy baicalein (5) [29], tanetin (6) [30], scutellarein 4′-methyl ether (7) [31], and apigenin (8) [32]. Compound 9 was identified as a caffeic acid [33]. Compounds 10 and 11 were identified as flavonoid glycosides anisofolin A [34] and apigetrin [35], respectively. Compound 12 was identified as forsythoside A [36]. Compounds 2–4 and 6–12 are reported for the first time from the genus Colebrookea, except compound 5 was already reported from C. oppositifolia [12]. Anisofolin A (10), which was previously reported from Leucas mollissima Wall. ex Benth. (Lamiaceae) by Chinchansure and coworkers, with its antimycobacterium
activity, is also reported in this study [21].
Material and Methods
General experimental procedures, chemical, and biochemicals: Melting points were measured on Buchi B-540 instrument. Optical rotations were measured
with a JASCO P-1020 polarimeter. UV spectra were measured with a SpectraMax plus 384
Microplate Reader. The IR spectrum was measured with a Bruker ALPHA FT-IR Spectrometer.
1 H and 13 C NMR spectroscopic data were recorded on a Bruker Avance Ultra Shield NMR instrument
(1 H: 500 MHz, 13 C: 125 MHz). LCESIMS data were recorded with an API-QSTAR-PULSAR spectrometer. HRESIMS
data was recorded with a Thermo-scientific Q-Exactive spectrometer. The silica gel
(100–200 and 230–400 mesh) was purchased from Thomas Baker Pvt. Ltd., Mumbai, India.
Preparative TLC was carried out using TLC plates supplied by Merck Ltd. MTT and rifampicin
were purchased from Sigma-Aldrich, USA. M. tuberculosis H37Ra (ATCC No. 25 177) and M. bovis (ATCC 35 734) were obtained from AstraZeneca, India.
Plant material: C. oppositifolia aerial parts were collected from Mulshi, District Pune, India on January 12, 2013
and were identified by Dr. Swati Joshi, CSIR-NCL, Pune. A herbarium (Voucher No. HVT-1)
was deposited in the Botanical Survey of India, Western Circle, Pune. The plant was
cleaned, shade dried, cut, and pulverized.
Extraction and isolation: Pulverized aerial parts, 2.9 kg, were extracted with acetone, 6 L × 3 × 14 h, at
room temperature. The acetone solubles were filtered and concentrated under reduced
pressure to yield a greenish extract (165.7 g, 5.7 % based on dry plant weight), 100 g
of which was separated by column chromatography (CC) using a gradient of acetone in
petroleum ether from 10 to 100 % as an eluent to collect 150 fractions. Fractions
showing a similar TLC pattern were combined to afford 16 fractions (COA1–COA16). The
isolation of compounds 1–12 is provided in Supporting Information.
(+)-14,15-Dinor-9α-hydroxy-cis-labd-11(E)-en-13-one (1): yellow gum; [α]D
26 + 14.8 (c 0.55, CHCl3); UV (MeOH) λ
max: 240 nm; IR (Nujol) ν
max: 3508, 2924, 1736, 1710, 1673, 1456, 1270, 986 cm− 1; 1 H NMR (CDCl3, 500 MHz) and 13 C NMR (CDCl3, 125 MHz), see [Table 1]; HRESIMS m/z: 301.2135 [M + Na]+ (calculated for C18H30O2, 278.2246).
Antimycobacterial assay: M. tuberculosis H37Ra (ATCC No. 25 177) and M. bovis (ATCC 35 734) strains were tested for their susceptibility to compounds 1–12 in active and dormant phases by using the XRMA protocol. All the experiments were
performed in triplicate, and IC50 and MIC values were calculated from their dose-response curves. The MIC was defined
as the lowest concentration of the anti-tubercular agents that prevented visible growth
with respect to the growth control [37], [38], [39]. Rifampicin was used as a positive control.
Compounds 1, 7, and 8 exhibited inhibitory activity against M. tuberculosis with IC50 values in the range of 8.1–55.0 µM (MIC 14.4–119.7 µM) in the active phase and 7.4–43.5 µM
(MIC 11.5–123.3 µM) in the dormant phase ([Table 2]). Similarly 1, 7, and 8 exhibited inhibitory activity against M. bovis with IC50 values in the range of 4.1–98.5 µM (MIC 13.7–161.0 µM) in the active phase and 4.1–111.1 µM
(MIC 13.0–166.4 µM) in the dormant phase ([Table 2]).
Table 2 Antimycobacterial activity of compounds 1, 7, and 8 against M. tuberculosis H37Ra and M. bovis.
Compound
|
Mycobacterium tuberculosis H37Ra
|
Mycobacterium bovis
|
Active
|
Dormant
|
Active
|
Dormant
|
IC50* (MIC*)
|
IC50* (MIC*)
|
IC50* (MIC*)
|
IC50* (MIC*)
|
* µM = micro molar
|
1
|
55.0 ± 0.3 (119.7 ± 0.4)
|
43.5 ± 0.2 (123.3 ± 0.1)
|
98.5 ± 0.2 (161.0 ± 0.1)
|
111.1 ± 0.4 (166.4 ± 0.3)
|
7
|
23.7 ± 0.2 (80.3 ± 0.3)
|
32.6 ± 0.3 (77.6 ± 0.4)
|
16.6 ± 0.2 (46.3 ± 0.4)
|
13.0 ± 0.4 (27.6 ± 0.1)
|
8
|
8.1 ± 0.1 (14.4 ± 0.2)
|
7.4 ± 0.2 (11.5 ± 0.4)
|
4.1 ± 0.1 (13.7 ± 0.1)
|
4.1 ± 0.3 (13.0 ± 0.3)
|
Rifampicin
|
0.0021 ± 0.0004 (0.019 ± 0.003)
|
0.021 ± 0.005 (0.031 ± 0.003)
|
0.0065 ± 0.0003 (0.034 ± 0.004)
|
0.018 ± 0.001 (0.037 ± 0.002)
|
Supporting information
HRESIMS and 1D and 2D NMR data of compound 1, dose dependence curves for compound 1, the isolation of compounds 1–12, and NMR and other characterization data for compounds 2–12 are available as Supporting Information.
Acknowledgements
We wish to thank Dr. Shanthakumari, Center for Material Characterization, National
Chemical Laboratory, Pune, for HRESIMS analysis. We are grateful to the Council of
Scientific and Industrial Research (CSIR), New Delhi, India, for a research fellowship.