1. Introduction
The genus
Illicium L. (Illiciaceae) consists of
ca. 40 species that form one of the earliest evolutionary branches of the angiosperms [
1]. This small taxon is represented by evergreen trees and shrubs disjunctively distributed in North America, Mexico, Peru, the West Indies and eastern Asia, with the highest concentration of species found in northern Myanmar and southern China [
1,
2]. The most well-known member of this genus is probably
Illicium verum. It serves as the source material of shikimic acid in the production of oseltamivir (Tamiflu) [
3], and its ripe pericarps (known as star anise) are widely used as a spice in many countries in Asia, in particular, China, India, and Vietnam [
4].
I. verum also has a long history of medicinal applications in China [
5]. In Mexico and the southwestern United States, its fruits are used to make herbal tea to alleviate colic of babies and stomach aches [
4,
6]. However, in recent years, intoxication cases related to the culinary and medicinal use of star anise have been reported, associating with neurological effects such as seizures, vomiting, jitteriness, rapid eye movement, and even death [
7,
8,
9]. Follow-up investigations indicated that most, if not all, of the adverse effects were caused by adulterated toxic
Illicium plants. Phytochemical and biological studies pointed to
seco-prezizaane sesquiterpenes (such as anisatin and neoanisatin) to be the toxic ingredients [
10,
11,
12]. The toxicological mechanism was elucidated to be a picrotoxin-like, non-competitive antagonism to the γ-aminobutyric acid (GABA) receptor [
13,
14,
15,
16]. However, systematic studies of the structure-toxicity relationship are limited [
17,
18,
19]. To safeguard the use of star anise and its products, studies on adulterant species of
Illicium is warranted.
Apart from the potential toxicity, some ingredients of
Illicium plants are known to display neurotrophic properties. Among others, jiadifenin, jiadifenolide, illicinin A, and 4-allyl-2,6-dimethoxy-3-(3-methylbut-2-enyl)phenol have been reported to promote neurite outgrowth in primary cultures of fetal rat cortical neurons [
20,
21,
22,
23].
Illicium plants are thus considered a potential source of neurotrophin-like natural products.
We are interested in constructing a library of secondary metabolites of Illicium plants, to identify toxic components on one hand, and search for neurotrophin-mimic natural products on the other. Several Illicium species are being investigated in our group. As part of the studies, I. lanceolatum, a toxic adulterant of Chinese star anise, was investigated for its chemical composition. This paper reports the structures of four germacrane sesquiterpenes (including three new structures), a new m-menthane monoterpene, and three other known compounds, and their biological activities in the SH-SY5Y neuroblastoma cell line.
2. Results and Discussion
From the pericarps of
I. lanceolatum, repeated open column (silica gel, RP-18, MCI, and Sephadex LH-20) and semi-preparative chromatographic separations resulted in the purification of four germacrane sesquiterpenes
1–
4, and a
m-menthane monoterpene
5 (
Figure 1), together with three other known compounds
6–
8. Germacrane
4 was a known compound, but its absolute stereochemistry was newly established in the present work.
Figure 1.
Structures of compounds 1–5.
Figure 1.
Structures of compounds 1–5.
Compound
1 was obtained as a white amorphous powder. A molecular formula of C
15H
24O
2 was determined based on the HR-ESI-MS result at
m/z 219.1733 [M−H
2O+H]
+ (calcd. 219.1743), indicating four degrees of unsaturation. The
1H-,
13C- and DEPT-NMR spectra (
Table 1 and
Table 2) indicated the presence of one methyl, eight methylenes (including three olefinic ones), three methines (including two oxygenated groups), and three quaternary olefinc carbons. The presence of three double bonds accounted for three degrees of unsaturation, the remaining one was therefore deduced arising from a ring structure in the molecule.
In the
1H-
1H COSY spectrum, signals at
δH 1.98 and 2.10 (
m, H-2) displayed correlations with signals at
δH 4.24 (dd,
J = 5.2, 10.5 Hz, H-1) and
δH 2.29 (m, H-3), respectively, suggesting the presence of structural fragment
1a (
Figure 2). The correlations of
δH 1.83 (m, H-6) with
δH 4.36 (dd,
J = 3.9, 7.0 Hz, H-5) and
δH 2.68 (m, H-7), and that of
δH 1.68 (m, H-8) with
δH 2.68 (m, H-7) and
δH 2.04 (m, H-9), led to the establishment of fragment
1b (
Figure 2). In the HMBC (
Figure 3), the cross peaks between
δH 1.76 (s, H-13) and
δC 39.7 (CH, C-7),
δC 151.3 (C, C-11) and
δC 110.3 (CH
2, C-12) suggested a partial structure
1c (
Figure 2). The connection of
1a and
1c via two exocyclic double bonds was established based on the following evidence: long-range correlations between
δH 5.16 (s, H-14) and 77.2 (CH, C-1) and 29.8 (CH
2, C-9); as well as those between
δH 5.13/5.02 (both s, H-15) and 72.5 (CH, C-5) and 29.2 (CH
2, C-3). All available evidence led to the planar structure
1d (
Figure 2), belonging to germacrane sesquiterpene.
To determine the absolute configuration of C-1 and C-5, the modified Mosher ester procedure was employed [
24,
25]. Thus, treatment with (
R)- and (
S)-MTPA chlorides led to esterification of 1-OH and 5-OH, affording (
S)- and (
R)-MTPA derivatives, respectively. The
1H-NMR chemical shift differences (∆
δS-R) were observed (
1,
Figure 4). The absolute configuration of C-1 and C-5 were consequently determined to be
R and
S, respectively. In the NOESY spectrum,
δH 2.68 (m, H-7) correlated with
δH 4.24 (dd,
J = 5.2, 10.5 Hz, H-1), suggesting the
R-configuration of C-7. Thus,
1 was identified to be (1
R,5
S,7
R)-1,5-dihydroxygermacra-4(15),10(14),11(12)-triene. To the best of our knowledge, it is the first time a germacrane sesquiterpene is isolated from
Illicium plants.
Table 1.
13C-NMR spectroscopic data for compounds 1–4 (CD3OD, 100 MHz).
Table 1.
13C-NMR spectroscopic data for compounds 1–4 (CD3OD, 100 MHz).
No. | 1 (mult.) | 2 (mult.) | 3 (mult.) | 4 (mult.) |
---|
1 | 77.2 (C) | 70.6 (C) | 74.1 (C) | 75.3 (C) |
2 | 33.9 (CH2) | 33.5 (CH2) | 34.4 (CH2) | 33.5 (CH2) |
3 | 29.2 (CH2) | 26.7 (CH2) | 27.0 (CH2) | 25.8 (CH2) |
4 | 152.7 (C) | 152.2 (C) | 153.6 (C) | 151.6 (C) |
5 | 72.5 (CH) | 74.6 (CH) | 76.7 (CH) | 77.3 (CH) |
6 | 41.1 (CH2) | 40.6 (CH2) | 40.1 (CH2) | 37.8 (CH2) |
7 | 39.7 (CH) | 40.8 (CH) | 42.1 (CH) | 42.2 (CH) |
8 | 30.4 (CH2) | 30.1 (CH2) | 31.5 (CH2) | 33.4 (CH2) |
9 | 29.8 (CH2) | 35.2 (CH2) | 32.1 (CH2) | 31.4 (CH2) |
10 | 150.6 (C) | 152.1 (C) | 151.7 (C) | 150.6 (C) |
11 | 151.3 (C) | 150.7 (C) | 150.4 (C) | 150.5 (C) |
12 | 110.3 (CH2) | 110.5 (CH2) | 110.4 (CH2) | 110.5 (CH2) |
13 | 20.1 (CH3) | 20.5 (CH3) | 19.0 (CH3) | 19.4 (CH3) |
14 | 116.9 (CH2) | 113.8 (CH2) | 114.3 (CH2) | 114.4 (CH2) |
15 | 112.4 (CH2) | 111.4 (CH2) | 112.8 (CH2) | 114.9 (CH2) |
Table 2.
1H-NMR Spectroscopic Data for Compounds 1–4 (CD3OD, 400 MHz).
Table 2.
1H-NMR Spectroscopic Data for Compounds 1–4 (CD3OD, 400 MHz).
No. | 1 [mult., J (Hz)] | 2 [mult., J (Hz)] | 3 [mult., J (Hz)] | 4 [mult., J (Hz)] |
---|
1 | 4.24 dd (5.2, 10.5) | 4.23 dd (4.5, 9.6) | 4.11 dd (4.2, 10.4) | 4.16 dd (4.1, 9.6) |
2 | 2.10 m; 1.98 m | 2.22 m; 1.95 m | 2.25 m; 1.82 m | 2.09 m; 1.56 m |
3 | 2.29 m | 2.38 dd (4.0, 11.4); 2.07 m | 2.11 (overlap) | 2.32 dd (4.0, 14.2); 2.07 m |
5 | 4.36 dd (3.9, 7.0) | 4.31 dd (3.5, 5.9) | 3.88 t (7.8) | 3.92 dd (3.9, 11.4) |
6 | 1.83 m | 1.91 ddd (2.3, 6.4, 14.2); 1.69 (overlap) | 1.67 (overlap) | 1.85 ddd (2.7, 11.3, 14.0); 1.59 m |
7 | 2.68 m | 2.42 m | 2.11 (overlap) | 2.17 m |
8 | 1.68 m | 1.74 m; 1.53 m | 1.61 m | 1.96 m; 1.64 m |
9 | 2.45 m; 2.04 m | 2.20 m; 2.13 dd (3.0, 11.2) | 2.14 (overlap) | 2.42 m; 2.05 m |
12 | 4.77 s | 4.73 s; 4.69 s | 4.66 br s | 4.69 s; 4.68 s |
13 | 1.76 s | 1.70 s | 1.67 s | 1.69 s |
14 | 5.16 s; 4.95 s | 5.19 s; 5.03 s | 5.24 s; 5.02 s | 5.18 s; 5.02 s |
15 | 5.13 s; 5.02 s | 5.05 s; 4.98 s | 5.12 s; 5.05 s | 5.02 br s |
Figure 2.
Partial structures of 1.
Figure 2.
Partial structures of 1.
Figure 3.
Selected 1H-1H COSY (—) and HMBC (→) correlations of compounds 1–3, 5.
Figure 3.
Selected 1H-1H COSY (—) and HMBC (→) correlations of compounds 1–3, 5.
Figure 4.
∆δS-R values of MTPA esters of 1–5.
Figure 4.
∆δS-R values of MTPA esters of 1–5.
Compounds
2,
3, and
4 were stereoisomers of
1. By using the same strategies as in the structural elucidation of
1, compounds
2,
3, and
4 were identified to be (1
S,5
S,7
R)-1,5-dihydroxygermacra-4(15),10(14),11(12)-triene, (1
S,5
R,7
R)-1,5-dihydroxygermacra-4(15),10(14),11(12)-triene, and (1
R,5
R,7
R)-1,5-dihydroxygermacra-4(15),10(14),11(12)-triene (
Figure 1 and
Figure 4), respectively. In the NOESY spectrum of
2,
δH 1.70 (s, H-13) correlated with
δH 4.31 (dd,
J = 3.5, 5.9, H-5), indicating the
α-orientation of H-7. For compound
3, due to the unresolved overlap of the H-7 signal in CD
3OD, NOESY spectrum was further acquired in CDCl
3 for the assignment of relative configuration of H-7. The
α-orientation of H-7 was suggested by the NOESY correlation between H-7 and 5
α-H. While the
α-orientation of H-7 in
4 was suggested by the NOESY correlation from
δH 2.17 (m, H-7) to both
δH 4.16 (dd,
J = 4.1, 9.6, H-1) and
δH 3.92 (dd,
J = 3.9, 11.4, H-5). Compound
4 is a known structure, previously reported from
Gonospermum elegans but with undetermined absolute configuration [
26].
Compound
5 was obtained as a white amorphous powder. A molecular formula of C
10H
18O
2 was determined based on the HRESIMS result at
m/z 193.1202 [M + Na]
+ (calcd. 193.1199), indicating two degrees of unsaturation. The
1H-,
13C- and DEPT-NMR spectra (
Table 3) revealed the presence of two methyls, four methylenes (including one olefinic ones), two methines (including one oxygenated group), and two quaternary carbons (including an olefinic one). All proton signals were assignable based on the gHSQC experiment (
Table 3). The
1H-
1H COSY spectrum displayed correlations between
δH 1.72 (H-2) and
δH 3.60 (d,
J = 3.16 Hz, H-3), between
δH 3.60 (H-3) and
δH 1.91 (ddd,
J = 2.8, 11.6, 14.0 Hz, H-4a) and
δH 1.63 (dt,
J = 3.4, 13.8 Hz, H-4b), between
δH 1.91 (H-4a) and
δH 2.23 (m, H-5), between
δH 2.23 (H-5) and
δH 1.53 (H-6), leading to the connections of between C
2-C
3-C
4-C
5-C
6 (
Figure 1 and
Figure 3). The HMBC correlations between
δH 1.69 (s, 10-H) and
δC 37.4 (C-5),
δC 109.0 (C-9), and
δC 149.2 (C-8), as well as correlations between
δH 4.70 (br s, H-9) and
δC 21.1 (C-10),
δC 37.4 (C-5), and
δC 149.2 (C-8), suggested that C-9 and C-10 were connected to C-5 via C-8 (
Figure 1 and
Figure 3). The remaining part of the structure was established based on the following long-range correlations:
δH 1.53 (H-6)/
δC 71.4 (C-1);
δH 1.23 (s, H-7)/
δC 71.4 (C-1) and
δC 33.7 (C-2). Two hydroxyl groups were assigned to C-1 and C-3, respectively. In the NOESY spectrum,
δH 2.23 (m, H-5) displayed correlations with
δH 3.60 (d,
J = 3.2 Hz, H-3), and
δH 1.23 (s, H-7), indicating the relative stereochemistry of 1
β,3
β-dihydroxy-(5
αH)-
m-menth-8-ene. To determine the absolute stereochemistry, modified Mosher’s ester procedure was carried out. Due to the steric hindrance on C-1, only 3-OH was esterified by (
R)- and (
S)-MTPA chlorides into (
S)- and (
R)-MTPA derivatives. The absolute configuration of C-3 was finally deduced to be
R based on the proton shift difference between (
S)- and (
R)-MTPA derivatives (
5,
Figure 4), indicating a 3
α-H. The 1
S- and 5
S-configuration were then assignable based on the NOESY results mentioned above. Consequently,
5 was determined to be (1
S,3
R,5
S)-1,3-dihydroxy-
m-menth-8-ene.
Table 3.
13C- and 1H-NMR spectroscopic data for compound 5 (CDCl3) *.
Table 3.
13C- and 1H-NMR spectroscopic data for compound 5 (CDCl3) *.
No. | δC (mult.) | δH [mult., J (Hz)] |
---|
1 | 71.4 (C) | - |
2 | 33.7 (CH2) | 1.72 (partial overlap); 1.49 (partial overlap) |
3 | 73.8 (CH) | 3.60 d (3.2) |
4 | 34.0 (CH2) | 1.91 ddd (2.8, 11.6, 14.0); 1.63 dt (3.4, 13.8) |
5 | 37.4 (CH) | 2.23 m |
6 | 26.1 (CH2) | 1.53 (partial overlap) |
7 | 26.5 (CH3) | 1.23 s |
8 | 149.2 (C) | - |
9 | 109.0 (CH2) | 4.70 br s |
10 | 21.1 (CH3) | 1.69 s |
Compounds
6,
7, and
8 were identified to be 3-hydroxyocta-1,5
E-dien-7-one [
27], 2-(4-methylphenyl)-1,2-propanediol [
28], and
trans-3,4,5-trimethoxycinnamic alcohol [
29], respectively, based on the interpretation of their NMR spectroscopic data and comparison with reported data.
6,
7, and
8 were isolated from
Illicium plants for the first time.
Compounds 3 and 4 exhibited proliferative activity in SH-SY5Y cells at concentrations of 0.49 µM–125 µM. Compounds 3 (at 62.5 µM) and 4 (at 15.6 µM) could promote proliferation by 36.2% and 45.8% after 48-h incubation, respectively. Compounds 5–8 were inactive; 6 and 8 displayed cytotoxicity at the concentrations above 50 µM. Due to the scarcity of 1 and 2, they were not tested for the bioactivity. Efforts to obtained additional crops of these compounds are in progress.
3. Experimental
3.1. General
The FT-IR spectra (KBr) were recorded using a Thermo FT-IR Nicolet 5700. Optical rotations at sodium D line were measured with a Perkin-Elmer 241 digital polarimeter using quartz cell with a path length of 100 mm at room temperature. Concentrations (c) are given in g/100 mL. Nuclear magnetic resonance (NMR) spectra were recorded on a Bruker DPX-400 or AVANCE-400 spectrometer running at 400 MHz for 1H and 100 MHz for 13C, respectively. All chemical shifts were quoted on the δ scale in ppm using residual solvent as the internal standard (CDCl3: 7.24 ppm for 1H-NMR, 77.0 ppm for 13C-NMR; CD3OD: 3.30 ppm for 1H-NMR, 49.0 ppm for 13C-NMR; C5D5N: 8.71 ppm, 7.55 ppm, 7.19 ppm for 1H-NMR, 149.9 ppm, 135.5 ppm, 123.5 ppm for 13C-NMR). Coupling constants (J) are reported in Hz. The following abbreviations are used to indicate the multiplicity: s = singlet, d = doublet, t = triple, dd = double doublet, dt = double triplet, br = broad. Thin layer chromatography (TLC) was carried out using Merck aluminium backed sheets coated with 60F254 silica gel or 60F254 RP-silica gel. Visualization of the plates was achieved by using a UV lamp (λmax = 254 nm), and spraying a mixture of 2% p-hydroxybenzaldehyde methanolic solution and 5% sulphuric acid ethanolic solution (10:1, v/v) followed by heating. Open column chromatography was carried out on columns packed with silica gel, RP silica gel (C18) (Macherey-Nagel GmbH & Co. KG, Düren, Germany), MCI gel CPH 20 (Supelco, Sigma-Aldrich, Bellefonte, PA, USA), and Sephadex LH-20 (GE Healthcare Bio-Sciences AB, Uppsala, Sweden). For HPLC purification, a C18 semi-preparative HPLC column (Phenomenex C18 column, 250 × 10 mm, 5 μm) and a Shimadzu UFLC system were used; the UV detection wavelength and flow rate were set at UV210nm and 4 mL/min, respectively. A Shimadzu UFLC XR system coupled with a LCMS-2020 liquid chromatography mass spectrometer was used for sample analysis. All solvents used were analytical or HPLC grade. HRESIMS were measured on a Shimadzu LCMS-IT-TOF Mass Spectrometry.
3.2. Plant Material
The pericarps of Illicium lanceolatum A. C. Smith. were collected from An-hui Province, China, in October 2011, and were identified by one of the authors (Jin-Ao Duan). A voucher specimen was deposited in the Department of Medicinal Chemistry and Pharmacognosy (UICMCP001-Ilan-P), College of Pharmacy, University of Illinois at Chicago, Chicago, IL, USA.
3.3. Extraction and Isolation
The dried pericarps of Illicium lanceolatum (1 kg) were powdered and extracted by percolation with methanol (MeOH, 15 L), yielding 300 g of extract. The extract was applied to a silica gel flash column eluted with mixtures of petroleum ether (PE)-ethyl acetate (EA) (100:0 → 50:50), followed by dichloromethane (DCM)-MeOH (90:10, 85:15, and 0:100), to yield 40 fractions (A1–A40). The fractions after A35 (A35: elute of DCM-MeOH, 85:15) contained plenty of shikimic acid. Fraction A17 was further fractionated into 23 fractions (B1–B23) on a silica gel column eluted with mixtures of PE-EA (70:30 and 60:40). B11 was subjected to RP-18 chromatography eluted with aqueous MeOH (20% and 100%) to yield 15 fractions (C1–C15). Compound 6 (6.1 mg) was purified from fractions of C4 and C5 by Sephadex LH-20 chromatography (eluted with 30% aqueous MeOH). Fractions of C8–C14 were combined and subjected to Sephadex LH-20 chromatography (eluted with 30% aqueous MeOH) to obtain a mixture (>100 mg) of compounds 5 and 7. An aliquot of the mixture was purified by semi-preparative HPLC using 30% aqueous acetonitrile (AcCN) as mobile phase to yield compounds 5 (>50 mg), and 7 (5 mg). Fractions A20–A22 were further fractionated into 25 subfractions (D1–D25) on a MCI column eluted with aqueous methanol (10% → 100%). D14 and D15 were subjected to Sephadex LH-20 chromatography (eluted by MeOH) to yield 20 fractions (E1–E20). Compound 8 (18 mg) was purified from E13–E15 by semi-preparative HPLC separation (35% aqueous AcCN). E7 and E8 were further subjected to a silica gel column eluted with mixtures of PE-EA (80:0 → 65:35) to yield 32 fractions (F1–F32). There were three main components in fractions of F13–F16. They were purified by semi-preparative HPLC chromatography (35% aqueous MeOH) to yield compounds 1 (2.5 mg), 2 (2.2 mg). Compounds 3 (15 mg) and 4 (50 mg) were purified from F17–F26 by semi-preparative HPLC chromatography (35% aqueous MeOH).
3.4. Spectral Data
(1R,5S,7R)-1,5-Dihydroxygermacra-4(15),10(14),11(12)-triene (
1). White amorphous powder. IR (cm
−1): 3377, 2989, 1703, 1576, 1405, 1081.
−3° (
c 0.12, MeOH).
13C- and
1H-NMR spectroscopic data (CD
3OD): see
Table 1 and
Table 2, respectively. HR-ESI-MS
m/z 219.1733 [M−H
2O+H]
+ (calcd. for C
15H
23O, 219.1743),
m/z 201.1639 [M-−2H
2O+H]
+.
(1S,5S,7R)-1,5-Dihydroxygermacra-4(15),10(14),11(12)-triene (
2). Colorless oil. IR (cm
−1): 3359, 2928, 1641, 1439, 1015, 895.
−7° (
c 0.14, MeOH).
13C- and
1H-NMR spectroscopic data (CD
3OD): see
Table 1 and
Table 2, respectively. HR-ESI-MS
m/z 219.1735 [M−H
2O+H]
+ (calcd. for C
15H
23O, 219.1743), 201.1630 [M−2H
2O+H]
+.
(1S,5R,7R)-1,5-Dihydroxygermacra-4(15),10(14),11(12)-triene (
3). White amorphous powder. IR (cm
−1): 3260, 2928, 1644, 1434, 1020, 905.
−4° (
c 0.17, MeOH).
13C- and
1H-NMR spectroscopic data (CD
3OD): see
Table 1 and
Table 2, respectively.
1H spectroscopic data (CDCl
3, 400 Hz):
δH 5.24 (1H, s, H-14a), 5.03 (1H, s, H-14b); 5.16 (1H, s, H-15a), 5.06 (1H, s, H-15b); 4.65 (1H, s, H-12a), 4.64 (1H, s, H-12b); 4.16 (1H, dd,
J = 4.3, 10.2 Hz, H-1); 3.95 (1H, dd,
J = 5.0, 10.7 Hz, H-5); 2.27 (1H, m, H-2a), 1.83 (1H, m, H-2b); 2.13 (4H, overlapped, H-3 and H-9); 2.06 (1H, t,
J = 5.7 Hz, H-7); 1.73 (1H, t,
J = 5.2 Hz, H-6a), 1.59 (1H, partially overlapped, H-6b); 1.58 (2H, partially overlapped, H-8a; 1.65 (3H, s, H-13). HR-ESI-MS
m/z 219.1734 [M−H
2O+H]
+ (calcd. for C
15H
23O, 219.1743), 201.1642 [M−2H
2O+H]
+, 237.1820 [M+H]
+.
(1R,5R,7R)-1,5-Dihydroxygermacra-4(15),10(14),11(12)-triene (
4). Colorless oil. IR (cm
−1): 3368, 2927, 1642, 1449, 1012, 893.
+11° (
c 0.16, MeOH).
13C- and
1H-NMR spectroscopic data (CD
3OD): see
Table 1 and
Table 2, respectively. HR-ESI-MS
m/z 219.1740 [M−H
2O+H]
+ (calcd. for C
15H
23O, 219.1743), 201.1650 [M−2H
2O+H]
+.
(1S,3R,5S)-1,3-Dihydroxy-m-menth-8-ene (
5). White amorphous powder.
13C- and
1H-NMR spectroscopic data (CDCl
3): see
Table 3. HR-ESI-MS
m/z 193.1202 [M+Na]
+ (calcd. for C
10H
18O
2Na, 193.1199).
3.5. Preparation of the (R)- and (S)-MTPA Ester Derivatives of 1–5
In these experiments, (
R)- and (
S)-MTPA chloride was used to react with each compound to yield its (
S)- and (
R)-MTPA derivatives, respectively [
25]. Two aliquots of compound (0.5–1.0 mg each) were transferred into two NMR tubes and dried overnight in a desiccator with P
2O
5 inside. After successive addition of 6 μL of (
R)- or (
S)-MTPA chloride and 600 μL of pyridine-
d5, the NMR tubes were sealed immediately, and shaken vigorously. The tubes were then kept in desiccator overnight until the reaction was complete [
30]. The
1H-NMR spectra of the final (
R)- and (
S)-MTPA derivatives were recorded, and the chemical shifts were assigned based on the
1H-
1H COSY NMR experiments. In case that signals could not be unambiguously assigned, gHSQC and gHMBC experiments were carried out. The ∆
δS-R values were calculated [
24,
25].
3.6. In Vitro Assay on SH-SY5Y
The SH-SY5Y cells were maintained in the Opti-MEM with 10% FBS, 100 U/mL penicillin and 100 µg/mL streptomycin. The cells (5 × 104 or 1 × 105/well in 1 mL growth medium) were incubated with various concentrations of compounds in 24-well culture plates. After 48-h incubation, 20 µL MTT (3-(4,5-dimethylthiazol-2yl)-2.5-diphenyltetrazolium bromide, 5 mg/mL in PBS) were added to the each well. The supernatant were removed after further 4 h incubation. The formazan in each well were dissolved in 300 µL isopropanol with 4 mM HCl and 0.1% Nondet P-40. The absorbance was read at 590 nm with a reference filter of 620 nm by using microplate reader (Infinite M200 Pro, Tecan, San Jose, CA, USA). The cells without treatment were as vehicle control. The cells were treated by corresponding concentration of DMSO as control. The percentage of growth promotion was calculated using the following formula: % cell promotion = 100 × (OD590nm test compound − OD590nm control)/OD590nm control. Results were expressed as the mean of at least three independent experiments.