Identification of new diterpene esters from green Arabica coffee beans, and their platelet aggregation accelerating activities

Article abstract of DOI:10.1016/j.foodchem.2018.04.081

Eight new ent-kaurane diterpene fatty acid esters, namely caffarolides A–H (1–8), were isolated from green beans of Coffea arabica. Their chemical structures were confirmed by extensive spectroscopic analysis including 1D, 2D NMR (HSQC, HMBC, ¹

Full text of DOI:10.1016/j.foodchem.2018.04.081

Accepted Manuscript
Identification of new diterpene esters from green Arabica coffee beans, and their
platelet aggregation accelerating activities
Xia Wang, QianQian Meng, XingRong Peng, GuiLin Hu, MingHua Qiu
PII:
S0308-8146(18)30721-0
DOI:
https://doi.org/10.1016/j.foodchem.2018.04.081
FOCH 22776
Reference:
To appear in:
Food Chemistry
Received Date:
Revised Date:
Accepted Date:
15 January 2018
10 April 2018
20 April 2018
Please cite this article as: Wang, X., Meng, Q., Peng, X., Hu, G., Qiu, M., Identification of new diterpene esters
from green Arabica coffee beans, and their platelet aggregation accelerating activities, Food Chemistry (2018), doi:
https://doi.org/10.1016/j.foodchem.2018.04.081
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Identification of new diterpene esters from green Arabica coffee beans, and their
platelet aggregation accelerating activities
a,b,1
a,1
a
a,b
a,b
Xia Wang , QianQian Meng , XingRong Peng , GuiLin Hu and MingHua Qiu *
*To whom correspondence should be addressed. Tel: +86-0871-65223327. Fax:
+
86-0871-6522325. E-mail: mhchiu@mail.kib.ac.cn
a
State Key Laboratory of Phytochemistry and Plant Resources in West China, Kunming Institute
of Botany, Chinese Academy of Sciences, Kunming 650201, People’s Republic of China
b
University of the Chinese Academy of Sciences, Beijing 100049, People’s Republic China
1
These authors contributed equally to this article
1

Abstract: Eight new ent-kaurane diterpene fatty acid esters, namely caffarolides A‒H (1‒8), were
isolated from green beans of Coffea arabica. Their chemical structures were confirmed by
1
1
extensive spectroscopic analysis including 1D, 2D NMR (HSQC, HMBC, H- H COSY, and
ROESY), HRMS, IR and CD spectra and by GC-FID analysis. Interestingly, the diterpene moiety
of these new compounds first occurred in genus Coffea. All the isolates were evaluated for platelet
aggregation activity in vitro. As the results, caffarolides C, D and F (3, 4 and 6) showed induction
effect for platelet aggregation and the possible structure-activity relationships have been discussed
briefly.
Keywords: Coffea arabica, coffee lipids, diterpene esters, structure elucidation, platelet
aggregation
Abbreviations1D NMR Proton and carbon nuclear magnetic resonance
2
D NMR
Two-dimensional nuclear magnetic resonance
Heteronuclear single quantum coherence
Heteronuclear multiple bond connectivity
HSQC
HMBC
1
1
1
1
H- H COSY
H- H Homonuclear chemical shift correlated spectroscopy
ROESY
HRMS
IR
Rotating frame nuclear overhauser and exchange spectroscopy
High resolution mass spectrometry
Infrared spectroscopy
CD
Circular dichroism
GC-FID
TLC
Gas chromatography with flame ionization detector
Thin layer chromatography
Preparative thin layer chromatography
Column chromatography
P-TLC
CC
MeOH
PRP
Methanol
Platelet-rich plasma
PPP
Platelet-poor plasma
2

ADP
Adenosine diphosphate
3

1. Introduction
Coffea arabica (Arabica coffee) is the most economic importance specie around 120 species
of the genus Coffea (Rubiacea), occupying 61% of the world’s coffee production (ICO, 2017).
Coffee is one of the most consumed beverages worldwide which is prepared by the ripe seeds
from the coffee plants. Chemical investigations showed that coffee is rich in bioactive compounds,
such as caffeine, trigonelline, chlorogenic acids, phenolic compounds, diterpenes and melanoidins
(
Ludwig & Crozier, 2014). They showed the activities of neuroprotective (Machado-Filho et al.,
014), anti-oxidant (Moreira & Coimbra, 2012; Zhou & Zeng, 2013), hepatoprotective (Baeza &
2
Bravo, 2015) and anti-cancer (Cavin et al., 2002). Therefore, long-time coffee consumption will
improve several related chronic diseases, such as cardiovascular diseases (Miranda & Marchioni,
2017; Ranheim & Halvorsen, 2010), , liver diseases (Muriel & Arauz, 2010; Saab & Tong, 2014),
cognition disorders (Carman et al., 2014; Nehlig et al., 2015), cancers (Bohn & Paur, 2014;
Vitaglione et al., 2012) and diabetes (Chu et al., 2011; Pan & Ho, 2016).
The chemical constituents of coffee brews are directly impacted by the chemical
compositions of green coffee beans, because green coffee beans contain all the active components
(
or their precursors) that existed in coffee brews. However, the studies that characterized the
chemical compositions of green coffee beans mainly focused on water-soluble fraction (Chu et al.,
016; Shu et al., 2014), but a few on the lipid fraction, despite that the lipid content of green
2
Arabica coffee beans reached 15 % (Kurzrock & Speer, 2001).
The main compositions of coffee lipid consist of triacylglycerols, diterpene esters, sterols,
sterol esters, and free diterpenes (Durán & Maciel, 2010; Kurzrock & Speer, 2001; Speer et al.,
2006). Among them, cafestol and kahweol are the well-known diterpenes in coffee with
4

cholesterol-rising (Urgert et al., 1995) and anti-cancers (Cavin et al., 2002). However, diterpenes
in coffee are rarely present in the form of free (0.4% of the coffee lipid), most of them are
esterified with different fatty acids (18% of the coffee lipid) (Kurzrock & Speer, 2001). Until now
only a few diterpene esters have been reported, including cafestol, kahweol, 16-O-methylcafestol
and 16-O-methylkahweol with fatty acids C16, C18, C18:1, C18:2, C20, and C22 (Speer et al., 2006).
Cafestol and kahweol palmitate are the main two diterpene esters that have been reported to
enhance glutathione S-transferase (Huber et al., 2010), inhibit angiogenesis (Moeenfard & Santos,
2016) and COX-2 (Muhammada et al., 2007) activities. The presence of diterpene eaters in coffee
brew is influenced by roasting degree and methods of preparation. In general, the content of total
diterpene esters in coffee beans decreased with an increase degree of roasting (Kurzrock & Speer,
2001; Speer et al., 2006). Meanwhile, the concentrations of total diterpene esters in different
coffee brews were in the order boiled coffee >French press coffee >espresso >filtered coffee and
instant coffee due to different contact time, grind degree, filter paper and temperature (Moeenfard
&
Alves, 2016; Ratnayake et al., 2003).
Recently, several studies found that coffee extracts can inhibit platelet aggregation in vitro
Bydlowski et al., 1987; Naito & Sumi, 2011), which is a key step in the development of
(
thrombosis and other cardiovascular diseases. Further studies demonstrated that these
water-soluble compounds like phenolic acids (Natella et al., 2008), pyridinium compounds
(
Kalaska et al., 2014) may be responsible for the antithrombotic effect of coffee extract, but the
effect of coffee lipids on platelet aggregation has been poorly studies.
Therefore, as part of our systematic chemical investigations on Coffee arabica, the aims of
the study presented here were to isolate and structurally characterize new diterpene esters from
5

coffee lipid fraction and to evaluate their activities of platelet aggregation.
2. Material and Methods
2.1 General
A Jasco P-1020 polarimeter (Jasco, Japan) was used to obtain optical rotations. Ultraviolet
spectra were measured by UV-2401 PC spectrophotometers (Shimadzu, Japan). A Bruker
Tensor-27 instrument (Bruker, German) was used for recording infrared spectra by using KBr
pellets and HRMS data were measured by an API QSTAR Pulsar spectrometer (Waters, UK). The
Bruker AM-400, and DRX-600 instruments (Bruker, Zurich, Switzerland) with transcranial
1
13
magnetic stimulation (TMS) were used to detect H and C NMR spectra. Circular dichroism
spectra were scanned by a Chirascan spectropolarimeter (Applied Photophysics, UK).
Semi-preparative HPLC was performed on an Agilent HP1100 or 1260 series instrument with a
UV L-2400 detector (Agilent, USA) and an ZORBAX SB C-18 column (5 μm ,9.4 mm × 250 mm,
wavelength detection at 220 and 290 nm). The analysis of fatty acid methyl esters was
acomplished by an Agilent 6890N gas chromatograph (Agilent Technologies, Germany) equipped
with FID detector. The green coffee beans were ground into 50-300 mesh powder by YB-2500B
grinder (Yunbang, Yongkang, China).
2.2 Chemical and reagents section
TLC detection was performed on TLC plates (200-250 μm thickness, F254 Si gel 60,
Qingdao Marine Chemical, Inc., China). The ordinary column chromatographic materials include
Lichroprep RP-18 (40-63 μm, Fuji, Japan), Sephadex LH-20 (20-150 μm, Pharmacia, USA) and
Silical gel (200-300 mesh, Qingdao Marine Chemical, Inc., China). The industrial-grade methanol,
chloroform, ethyl acetate, acetone, petroleum ether were purchased from Tianjing Chemical
6

Reagents Co. (Tianjing, China). The analytical-grade acetonitrile, tetrahydrofuran, hydrochloric
acid, sodium hydroxide, methyl palmitate (16:0), methyl linoleate (18:2), methyl oleate (18:1),
methyl nonadecadienoate (19:2), and methyl eicosadienoate (20:2) were purchased from Aladdin
Industrial Corporation (Shanghai, China).
2.3 Plant Material.
The green coffee beans of Coffea arabica cultivated in Yunnan province (P. R. China), with a
subtropical monsoon climate at the location of 22°68ʹ north altitude and 100.94ʹ east longitude,
were harvested in June 2014. The material was authenticated by Ming-Hua Qiu, Kunming
Institute of Botany, Chinese Academy of Sciences. A specimen was deposited in State Key
Laboratory of Phytochemistry and Plant Resources in West China, Kunming Institute of Botany,
Chinese Academy of Sciences.
2.4 Extraction and isolation of the lipid fraction of green Arabica coffee beans.
The powder of dried coffee green beans (50-300 mesh, 10 kg) was soaked in acetone for four
days at room temperature, then the acetone extract was evaporated under reduced pressure. The
residue (963 g) was dissolved with CHCl , filtered with a Buchner funnel and recrystallized for
3
removing caffeine. Then, the CHCl
3
layer (63 g) was subjected to Silica gel column
chromatography (CC , 20.0 × 120 cm) and eluted in a step gradient manner with petroleum
ether/acetone (20:1, 5:1, 1:1, 0/100, v/v) to yield four fractions (Fr.): Fr. 1 (16 g), Fr. 2 (9 g), Fr. 3
(
5 g), and Fr. 4 (30 g).
Fr. 1 (16 g) was then further separated on a RP C-18 (15.0 × 100 cm) column and eluted in a
2
gradient of MeOH/H O (55→80%, v/v) to yield three sub-fractions (Fr. 1-1‒1-3). Fr. 1-1 (5 g) was
subjected to Silica gel CC (5.0 × 70 cm), eluted with a CHCl
7
3
/MeOH (80:1→20:1, v/v) gradient

system to afford four minor fractions (Fr. 1-1-1−1-1-4) on the basis of TLC analysis. Fr. 1-1-2
205 mg) was separated by preparative thin layer chromatography (P-TLC, eluting with
CHCl /MeOH, 30:1, v/v) to gain 7 (14 mg).
Fr. 2 (9 g) was applied to Silica gel CC (5.0 × 100 cm) and eluted in a gradient of
MeOH/H O (50→75%, v/v) to yield four fractions (Fr. 2-1−2-4). Fr. 2-1 (3 g) was separated by
(
3
2
Silica gel CC (5.0 × 70 cm, eluted with CHCl
3
/MeOH, 40:1, v/v), to afford five minor fractions
(
Fr. 2-1-1−2-1-5). Fr. 2-1-2 (330mg) was subjected to chromatography over RP C-18 (2.0 × 20 cm,
2
eluted with MeOH/H O, 40-60%, v/v) to obtain 2 (10 mg) and 5 (8 mg). Fr. 2-2 (2 g) was
chromatographed on a Silica gel column (5.0 × 50 cm), eluted with CHCl
yield four minor fractions (Fr. 2-2-1−2-2-4). Then, Fr. 2-2-3 (400mg) was separated by P-TLC,
eluted with CHCl /MeOH (30:1, v/v) to gain minor fractions, then separated by reverse-phase
3
/MeOH (20:1, v/v) to
3
3 2
semi-preparative HPLC (CH CN/H O: 40→70%, 30 min, flow rate = 3.0 mL/min, UV 220, 290
R R R
nm) to get 3 (11 mg, t = 11.5 min), 6 (9 mg, t = 14.7 min), and 8 (8 mg, t = 23.4 min). Fr. 2-3 (3
g) was separated by use of Sephadex LH-20 (5.0 × 200 cm, eluted with MeOH, 100%, 2 L) and
divided into three fractions (Fr. 2-3-1−2-3-3), then Fr. 2-3-1 (506 mg) was applied to Silica gel CC
(
2.0 × 50 cm), eluted with CHCl
-3-1-1−2-3-1-4). Then, Fr. 2-3-1-2 (88 mg) was treated by reverse-phase semi-preparative HPLC
R
CH CN/H O: 45→80%, 35 min, flow rate = 3.0 mL/min, UV 220, 290 nm) to get 1 (20 mg, t =
3
/MeOH (10:1, v/v) to afford four minor fractions (Fr.
2
(
3
2
1
2
R
5.5 min), 4 (9 mg, t = 21.3 min).
.4.1. Caffarolide A (1)
25
White amorphous powder, [α]
D 3 3
‒246.9 (c = 0.1, CHCl ); UV (CHCl ) λmax (log ε): 239
(
1.59), 201(1.04),193 (1.05) nm; IR (KBr) vmax: 3511, 3422, 3088, 3922, 1767, 1719, 1654, 1453,
8

-1
+
1
293, 1161, 943, 875 cm ; HREIMS m/z 586.4225 [M] (calcd for C36
58 6
H O , 586.4205); 1DNMR
data shown in Tables 1 and 2.
2.4.2. Caffarolide B (2)
25
White amorphous powder, [α]
D 3 3
‒164.8 (c = 0.1, CHCl ); UV (CHCl ) λmax (log ε): 239
(
1.63), 218 (1.27), 201 (1.21) nm; IR (KBr) vmax: 3440, 2926, 2854, 1736, 1654, 1451, 1239, 1066,
-
1
+
1
105, 928, 753 cm ; HREIMS m/z 610.4249 [M] (calcd for C38
H
58
O
6
, 610.4233); 1D NMR data
shown in Tables 1 and 2.
2.4.3. Caffarolide C (3)
25
White amorphous powder, [α]
D 3 3
‒149.8 (c = 0.1, CHCl ); UV (CHCl ) λmax (log ε): 239
(
2.43), 207(0.93), 200 (0.96) nm; IR (KBr) vmax: 3440, 3423, 2954, 2925, 2853, 1767,1720, 1629,
-
1
+
1
454, 1378, 1342, 1293, 1242, 1162, 1107, 927, 618 cm ; HREIMS m/z 612.4384 [M] (calcd for
38 60 6
C H O , 612.4390); 1D NMR data shown in Tables 1 and 2.
2.4.4. Caffarolide D (4)
25
White amorphous powder, [α]
D 3 3
‒67.8 (c = 0.1, CHCl ); UV (CHCl ) λmax (log ε): 275 (1.05),
-1
2
38 (1.12), 217 (0.97) nm; IR (KBr) vmax: 3441, 3424, 2956, 1735, 1630, 1378, 1127, 620 cm ;
+
HREIMS m/z 602.4201 [M] (calcd for C36
H
58
O
7
, 602.4183); 1D NMR data shown in Tables 1
and 2.
2.4.5. Caffarolide E (5)
25
White amorphous powder, [α]
D 3 3
‒155.4 (c = 0.1, CHCl ); UV (CHCl ) λmax (log ε): 239
(
1.26), 217 (0.98) nm; IR (KBr) vmax: 3516, 3451, 2919, 2850, 1748, 1650, 1388, 1239, 1160, 928
-
1
+
cm ; HREIMS m/z 616.4170 [M] (calcd for C37
60 7
H O , 616.4186); 1D NMR data shown in Tables
2
and 2.
9

2.4.6. Caffarolide F (6)
2
0
White amorphous powder, [α]
D 3 3
‒248.0 (c = 0.1, CHCl ); UV (CHCl ) λmax (log ε): 240.4
-1
(
2.28) nm; IR (KBr) vmax: 3485, 3398, 2964, 1721, 1651, 1370, 1240, 1032, 826 cm ; HRESIMS
+
m/z 677.9326 [M + Na] (calcd for C40
H
62
O
7
, 654.4506); 1D NMR data shown in Tables 1 and 2.
2.4.7. Caffarolide G (7)
25
White amorphous powder, [α]
D 3 3
‒138.5 (c = 0.1, CHCl ); UV (CHCl ) λmax (log ε): 270
(
2.34), 232 (2.18) nm; IR (KBr) vmax: 3497, 3108, 2922, 1741, 1612, 1390, 1297, 1177, 1089, 983,
-
1
+
8
56 6
73, 566 cm ; HREIMS m/z 584.4068 [M] (calcd for C36H O , 584.4077); 1D NMR data shown
in Tables 1 and 2.
2.4.8. Caffarolide H (8)
20
White amorphous powder, [α]
D 3 3
‒113.5 (c = 0.1, CHCl ); UV (CHCl ) λmax (log ε): 269.8
-
1
(
2.02), 239.2 (1.97) nm; IR (KBr) vmax: 3485, 3398, 2964, 1721, 1651, 1370, 1240, 1032, 826 cm ;
+
HRESIMS m/z 659.9114 [M + Na] (calcd for C40
H
40
O
6
, 636.4478); 1D NMR data shown in
Tables 1 and 2.
2.5 Mild alkaline hydrolysis of compounds 1-8.
Caffarolides A ‒H (1‒8, each for 3.0 mg, respectively) were dissolved in 1 mL of
methanol–tetrahydrofuran (1:1, v/v) and treated with 0.1 M NaOH (0.5 mL) at room temperature
for 2 h (Tamaki et al., 2008). After that, 0.01 M HCl (~ 5.0 mL) was added to the reaction
mixtures for neutralization. In order to obtain the fatty acid esters, the solutions were extracted
with ethyl acetate and then evaporated. The residues were purified by Silica gel column
3
chromatography (1.0 × 10 cm) and eluted in a step gradient manner with CHCl /MeOH (100:1→
3
10:1, v/v). The fatty acid methyl esters were obtained at 80:1 (CHCl /MeOH, v/v) fractions.
10

2.6 GC comparison analysis for the fatty acid of compounds 1-8.
HP-5 capillary column (30 m × 0.53 mm, 1.0 μm) was used in GC. Column temperature was
held initially at 60 ℃ for 1 min, increased by 10 ℃/min to 150 ℃ and held for 5 min, then by
℃/min to 250 ℃, then by 20 ℃/min to 300 ℃ and held for 7 min. The carrier gas is helium
5
with the flow rate of 1.1 mL/min and the inlet temperature was set at 200 ℃. Then, compared the
retention time of the fatty acid methyl esters with those of standards, the GC analysis confirmed
R
that 1, 4, 5, 7 esterified with palmitic acid (16:0, t = 23.5 min) and 2, 3, 6, 8 were esterified with
R R R
linoleic acid (18:2, t = 29.8 min), oleic acid (18:1, t = 27.1 min), nonadecadienoic acid (19:2, t
=
R
30.2 min), eicosadienoic acid (20:2, t = 31.7 min), respectively (Zhao et al., 2014).
2
2
.7 Platelet Aggregation Assays.
.7.1. Preparation of platelets
blood samples from the rabbit ear central artery (Japanese big ear rabbits, Kunming Chu
Shang technology co. Ltd., China) were collected into plastic tubes, anticoagulated with 3.8 %
sodium citrate acid (9:1, v/v). Platelet-rich plasma (PRP) and platelet-poor plasma (PPP) were
obtained by centrifuging the blood at 200 × g (10 min) and 2400 × g (20 min), respectively. Then,
8
-1
cell counter was adjusted to 5 × 10 mL before the measurement (Shen et al., 2003).
2.7.2. Platelet aggregation in vitro
platelet aggregation in PRP was measured as by turbidimetry method (Born,1964). Firstly,
ten disposable cuvettes were prepared. 250 μL PRP and 2.5 μL DMSO were added to each cuvette,
similarly, another cuvette with 250 μL PPP and 2.5 μL DMSO was prepared to adjust baseline of
aggregometer. Then, all the cuvettes incubated at 37 ℃ for 5 min. After incubation, eight new
-4
-6
compounds (each 3 × 10 g/mL) and positive control ADP (4.3 × 10 g/mL, Chrono-log
11

Corporation, USA) were added to PRP cuvettes, respectively. Finally, the maximal aggregation
was monitored by a chrono-log 700 aggregometer (Chrono-log Corporation, USA).
3.
Every sample was analyzed five times, and data are reported as the mean ± standard
deviation expressed as platelet maximum aggregation rate (%). P values less than 0.05 were
considered significant, less than 0.01 highly significant, and less than 0.001 extremely
significant.Result and discussion
3.1 Phytochemical investigation
Caffarolide A (1) was isolated as white amorphous powder, whose molecular formula was
+
determined to be C36
H
58
O
6
from the [M] ion peaks at m/z 586.4225 (calcd for C36
H
58
O
6
,
-
1
5
86.4205) in the HREIMS. The IR spectrum indicated that 1 possessed hydroxyl (3422 cm ), and
-
1
1
α,β-unsaturated lactone (1767 cm ) groups. The H-NMR (Table 1) spectrum showed the presence
of one singlet methyl (δ 0.75, s, H -20), one triplet methyl (δ 0.82, t, J = 6.8 Hz, H -16’), one
H
3
H
3
H
oxymethylene [δ 4.52 (d, J = 11.2 Hz, H-17a), 4.68 (d, J = 11.3 Hz, H-17b)], one olefinic
methine (δ
H
H
5.80, s, H-18), and huge methylene proton signals at δ 1.20-1.25 (m, H-2’-15’). The
13
C-DEPT spectra (Table 2) displayed a total of 36 carbon resonances. Except for the
characteristic huge methylenes, one methyl and one ester carbonyl signals for fatty acid, the
remaining resonances were assigned to be one methyl, eight methylenes (one oxygenated), four
methines (one olefinic), and seven quaternary carbons (including one olefinic, one carbonyl and
two oxygenated). The aforementioned information indicated that compound 1 could be a diterpene
ester and the diterpene moiety was exactly similar with tricalysiolide B (Nishimura et al., 2006),
H C C
which was confirmed by the HMBC correlations of δ 5.80 (H-18) with δ 171.4 (C-19), δ 173.6
(
C C H 2 C C
C-4), δ 105.5 (C-3) and δ 47.3 (C-5), of δ 4.68 (H -17) with δ 79.3 (C-16), δ 46.3 (C-13)
12

and δ
determined to be palmitic acid (16:0), which was further confirmed by the alkaline hydrolysis,
then, comparing to the retention time of standard methyl palmitate in GC analysis (t = 23.5 min)
Novaes, & Neto, 2015). Additionally, the esterification position was located at C-17, which was
C
53.9 (C-15) (Figure 2). Moreover, on the basis of the molecular weight, the fatty acid was
R
(
proved by the key HMBC correlation of δ
C-16) (Figure 2).
The relative configuration of 1 was established by ROESY spectrum as shown in Figure 2.
With the solvent of DMSO instead of C D (Supporting Information), the ROESY correlations
H 2 C C
4.68 and 4.52 (H -17) with δ 173.6 (C-1’) and δ 79.3
(
5 5
N
2
of H-13/H -17 and OH-3/H-5 assigned OH-16 and OH-3 as α-orientation and β-orientation,
respectively. The absolute configuration at C-3 was established by the CD spectrum (a π−π*
transition in the α,β-unsaturated-γ-lactone moiety). The observed negative Cotton eff ect at 225 nm
revealed the 3R configuration (Supporting Information), which was supporting by literatures
(
Shen et al., 2015; Yin, Luo & Kong, 2013). Finally, the structure of 1 was estabilshed as
caffarolide A.
Caffarolide B (2) was obtained as white amorphous powder. The molecular formula
+
38 58 6 58 6
C H O was deduced from the molecular ion peak at m/z [M] 610.4249 (calcd for C38H O ,
610.4233) in HREIMS. The NMR spectra of 2 were similar to those of 1 implying that 2 has the
same structure as 1 containing a diterpenoid moiety and a fatty acid fraction. Comparison of the
13
C-DEPT spectra between 2 and 1 showed that the diterpenoid moiety of 2 was also tricalysiolide
B (Nishimura et al., 2006). Similarly, the molecular weight of 2 suggested the presence of
2
1
8-carbon fatty acid in 2. Futhermore, four additional sp methine carbons (δ
C
128.0-130.5) in 2
3
H
combined with the HMBC correlations of these aromatic signals (δ 5.35, m) with huge sp
13

1
1
C
methylenes (δ 22.0-30.0) of fatty acid (Figure 2), together with the H- H COSY correlations of
H-10’/H-11’/H-12’, indicated that the fatty acid of 2 could be linoleic acid (18:2). Subsequently,
the hydrolysis experiment and GC comparison unambiguously determined the fatty acid to be
R H 2
linoleic acid (18:2, t = 29.8 min ). In the HMBC experiment, δ 4.22 and 4.27 (H -17) showed a
C
correlation with δ 174.3 (C-1’) demonstrated the esterification position of 2 was at C-17. The
relative configuration of 2 was the same as that of 1 by the ROESY and CD spectra. Then, the
structure of 2 was established and named as caffarolide B.
60 6
The HREIMS data for caffarolide C (3) indicated a molecular formula of C38H O , with two
unites more than 2. Detailed analysis of the 1D NMR data of 3 and 2 revealed that 3 has the same
basic diterpene structure as 2, expect that the fatty acid in 3 possessed only one double bond. We
thus deduced that the fatty acid of 3 could be oleic acid (18:1). The futher evidence was
2
3
2 H
established by the HMBC correlation of the H -sp methine signals (δ 5.34, m) with huge sp
methylenes (δ
C
22.5-31.9) of fatty acid, along with GC comparison analysis (t
R
= 27.1 min).
Moreover, the esterification location of 3 was at C-17, according to the key HMBC correlation of
with C-1’. Its relative configuration was determined to be the same as that of 1 from the ROESY
and CD spectra. Therefore, the structure of 3 was determined.
On the basis of HREIMS spectrum, caffarolide D (4) was assigned a molecular formula of
C
36
H
58
O
7
. The 1D NMR data of 4 were similar to those of 1 with a palmitoyl group, except that
the methylene at C-1 in 1 was replaced by an oxymethine in 4. The observed HMBC correlations
of the singlet methyl δ 0.88 (H -20), the methylene proton δ 2.37 (H-2a) and methine proton δ
H
H
3
H
C H C
2.06 (H-9) with the oxygenated methine δ 71.1, of oxymethine δ 3.84 (H-1) with C-2 (δ 40.9)
1
1
C
and C-3 (δ 107.6), together with the H- H COSY correlation of H-1/H-2a indicated that a
14

hydroxyl was loacted at C-1 (Figure 3). Herein, the planar structure of the cafestol-type lactone
moiety in 4 was established. Meanwhile, combined the results of the molecular weight, hydrolysis
experiment and GC comparison analysis, the fatty acid of 4 was determined to be palmitic acid
(
16:0). Furthermore, the esterification position of 4 was located at C-17, which was proved by the
key HMBC correlations of δ 4.22 and 4.28 (H -17) with δ 176.0 (C-1’). Its relative
H
2
C
configuration was determined to be the same as that of 1, and the OH-1 group was assigned as
β-oriented as indicated by the ROESY correlation of H-1/H-20 (Figure 3). Finally, the structure of
4
was elucidated as shown.
60 7
Caffarolide E (5) was assigned a molecular formula of C37H O . Compared the 1D NMR
C
data of 5 with those of 4, showed that a methoxy (δ 58.9) at C-1 in 5 replaced the hydroxyl in 4.
H 3
The deduction was proved by the key HMBC correlations of the methyl δ 0.86 (H -20), the
H H
methylene proton δ 2.55 (H-2a) and methine proton δ 2.01 (H-9) with the oxygenated methine
C H C C
δ 80.4, of oxymethine δ 3.43 (H-1) with C-2 (δ 35.3) and C-3 (δ 102.9). The fatty acid of 5
was assigned as palmitic acid (16:0) by the molecular weight and GC comparison analysis (t
R
=
23.5 min). The esterification location of 5 was also at C-17, which was confirmed by the HMBC
spectrum. In the ROESY spectrum of 5, the correlations of H-1/H-20 assigned OMe-1 as
β-orientation. Therefore, the structure of compound 5 was elucidated as shown.
62 7
Caffarolide F (6) was assigned a molcular formula as C40H O from its HRESIMS spectrum.
Analyzed the 1D NMR data of 6 and 5 indicated that 6 has the same diterpene lactone moiety as 5,
2
but differing at the fatty acid moiety. Four additional sp methine singals were also observed in the
downfield region of 6. Thus, based on the molecular weight, we deduced that the fatty acid of 6
2
could be nonadecadienoic acid (19:2). The HMBC correlations of these H
4
-sp methine signals (δ
H
15

3
5
C R
.34) with huge sp methylenes (δ 22.7-29.9), together with GC comparison analysis (t = 30.2
2
min) proved the hypothesis. Moreover, H -17 showed the HMBC correlation with C-1’,
demonstrated that the esterification location of 6 was at C-17. The relative comfigurations of 6
was determined by the ROESY spectrum , which indicated that OMe-1 were β-orientation. Thus,
we deduced the structure of compound 6 as caffarolide F.
For support of the absolute configurations assigned for compounds 2−6, the CD spectra were
measured. The CD spectra obtained were in close agreement with that of 1 at 225 nm, thus
indicating that the absolute configuration of C-3 in 2−6 is R, the same as that of 1.
56 6
The HREIMS data for caffarolide G (7) indicated a molecular formula of C38H O . The 1D
NMR data of 7 were generally similar to those of 4, but an additional double bond was presented
at the diterpene lactone moiety in 7. The HMBC correlation of H-2 (δ 5.72, dd, J = 6.0 and 1.5 Hz)
C C C H C
with C-1 (δ 69.5), C-3 (δ 152.0), C-4 (δ 159.1), of H-18 (δ 5.76) with C-3 (δ 152.0) and C-19
(
δ
C
H C C C
170.0), of H-1 (δ 4.20, d, J = 6.0 Hz) with C-20 (δ 16.0), C-2 (δ 107.1) and C-3 (δ 152.0),
1
1
together with the H- H COSY correlations H-1/H-2 (Figure 3), demonstrated the double bond
was located at C-2 and C-3. Hereto, the planar structure of the diterpene moiety in 7 was
determined as tricalysiolide F (Nishimura et al., 2006). Similarly, the fatty acid of 7 was assigned
as palmitic acid (16:0) by molecular weight and GC comparison analysis (t
was confirmed as the esterification loacation of 7 by the HMBC spectrum. The key ROESY
correlations of H-1/H-20 and H-13/H -17 were assigned OH-1 and OH-16 as β-orientation and
R
= 23.5 min). C-17
2
α-orientation, respectively (Figure 3). Therefore, the structure of 7 was determined as caffarolide
G.
40 6
The molecular formula of caffarolide H (8) was assigned to be C40H O . Detailed
16

comparison of 1D NMR data of 8 and 7 showed that the only difference bewteen them was fatty
acid fraction. On the basis of the molecular weight, the fatty acid contains twenty carbons.
2
Meanwhile, four sp methine singals were also observed in the downfield region of 8. Therefore,
we assumed the fatty acid of 8 could be eicosadienoic acid (20:2), which was proved by
R
hydrolysis experiment and GC comparison method (t = 31.7 min). The esterification location of 8
was also at C-17 which was establishehed by the HMBC spectrum. Based on the ROESY
spectrum, OH-1 and OH-16 were assigned as β-orientation and α-orientation, respectively. Thus,
we deduced the structure of compound 8 as caffarolide H.
3.2 Platelet Aggregation Assay.
The activity of caffarolides A‒H (1‒8) inducing platelet aggregation in vitro were evaluated
by turbidimetry method, and the results are showed in Table 3. Compared to control (DMSO, 3.4
1.1 %), compounds 3, 4 and 6 showed platelet aggregation activity at the concentration of 3 ×
±
-4
1
0 g/mL, with induction rate of 11.4 ± 5.5, 15.8 ± 5.6 and 7.8 ± 3.3 %, respectively.
From these results, some primary structure-activity relationships can be deduced: the
hemiketal at C-3, the substituent groups at C-1 and the types of fatty acids can impact the platelet
aggregation effect. If the hemiketal at C-3 in compound 4 was replaced by the enol group in
compound 7, no platelet aggregation effect was shown. When the unsaturated fatty acids were
linked to the hydroxyl at C-17, compounds 3 and 6 showed platelet induction activity, whereas,
the substitute of saturated fatty acids at C-17, compounds 1 and 5 didn’t show any activity.
Moreover, the hydroxyl and methoxyl attached to C-1 enhanced the platelet induction activity (4
vs. 5 vs. 1).
4.
Conclusions
17

In present study, a systematic chemical research was performed and resulted in the separation
of eight new diterpenes esters caffarolides A-H (1‒8) from the lipid of green Arabica coffee beans.
The structures of these new compounds were identified based on detailed spectroscopic analysis
(
NMR, HSMS, IR and CD spectra) and GC-FID analysis. Caffarolides C, D and F (3, 4 and 6)
-4
showed induction effect for platelet aggregation at 3 × 10 g/mL. The results obtained in this
study complement the current knowledge of diterpene esters in Coffea arabica.
Acknowledgements
This research work was supported financially by the National Natural Science Foundation of
China (No. 31670364). Project of Key New Productions of Yunnan Province, Centre of CHINA
(
(
No. 2015BB002 and No. 2016HE003).The STS Programme of Chinese Academy of Sciences
KFJ-SW-STS-143-8), as well as Foundation of State Key Laboratory of Phytochemistry and
Plant Resources in West China (P2015-ZZ09). Furthermore, the authors sincerely thank the
support of the analytical center and activity screening center of State Key Laboratory of
Phytochemistry and Plant Resources in West China for the analysis and platelet aggregation
activities data.
Notes
The authors declare no competing financial interest.
References
Baeza, G., Amigo-Benavent, M., Sarriá, B., Goya, L., Mateos, R., & Bravo, L. (2015). Green
coffee hydroxycinnamic acids but not caffeine protect human HepG2 cells against oxidative
stress. Food Research International, 62, 1038-1046.
Bohn, S. K., Blomhoff, R., & Paur, I. (2014). Coffee and cancer risk, epidemiological evidence,
18

and molecular mechanisms. Molecular Nutrition & Food Research, 58(5), 915-930.
Born, G. V. R. (1964). Strong inhibition by 2-chloroadenosine of the aggregation of blood platelets
by adenosine diphosphate. Nature, 202, 95-96.
Bydlowski, S. P., Yunker, R. L., Rymaszewski, Z., & Subbiah, M. T. (1987). Coffee extracts
inhibit platelet aggregation in vivo and in vitro. International Journal for Vitamin & Nutrition
Research, 57, 217-223.
Carman, A. J., Dacks, P. A., Lane, R. F., Shineman, D. W., & Fillit, H. M. (2014). Current
evidence for the use of coffee and caffeine to prevent age-related cognitive decline and
Alzheimer's disease. Journal of Nutrition Health & Aging, 18, 383-392.
Cavin, C., Holzhaeuser, D., Scharf, G., Constable, A., Huber, W. W., & Schilter, B. (2002).
Cafestol and kahweol, two coffee specific diterpenes with anticarcinogenic activity. Food and
Chemical Toxicology, 40, 1155-1163.
Chu, R., Wan, L. S., Peng, X. R., Yu, M. Y., Zhang, Z. R., Zhou, L., Li, Z. R., & Qiu, M. H. (2016).
Characterization of new ent-kaurane diterpenoids of Yunnan arabica coffee beans. Nat Prod
Bioprospect, 6, 217-223.
Chu, Y. F., Chen, Y., Black, R. M., Brown, P. H., Lyle, B. J., Liu, R. H., & Ou, B. (2011). Type 2
diabetes-related bioactivities of coffee: Assessment of antioxidant activity, NF-κB inhibition,
and stimulation of glucose uptake. Food Chemistry, 124, 914–920.
Durán, M. A., Filho, R. M., & Maciel, M. R. W. (2010). Rate-based modeling approach and
simulation for molecular distillation of green coffee oil. Computer Aided Chemical
Engineering, 28, 259-264.
Huber, W. W., Teitel, C. H., Coles, B. F., King, R. S., Wiese, F. W., Kaderlik, K. R., Casciano, D.
19

A., Shaddock, J. G., Mulder, G. J., & Ilett, K. F. (2010). Potential chemoprotective effects of
the coffee components kahweol and cafestol palmitates via modification of hepatic
N-acetyltransferase and glutathione S-transferase activities. Environmental & Molecular
Mutagenesis, 44, 265-276.
International
Coffee
Organization.
Trade
statistics
tables.
(
2018).http://www.ico.org/prices/po-production.pdf/ Accessed 06.04.18
Kalaska, B., Piotrowski, L., Leszczynska, A., Michalowski, B., Kramkowski, K., Kaminski, T.,
Adamus, J., Marcinek, A., Gebicki, J., & Mogielnicki, A. (2014). Antithrombotic effects of
pyridinium compounds formed from trigonelline upon coffee roasting. Journal of
Agricultural & Food Chemistry, 62, 2853-2860.
Kurzrock, T., & Speer, K. (2001). Diterpenes and diterpene esters in coffee. Food Reviews
International, 17, 433-450.Ludwig, I. A., Clifford, M. N., Lean, M. E., Ashihara, H., &
Crozier, A. (2014). Coffee: biochemistry and potential impact on health. Food & Function, 5,
1695-1717.
Machado-Filho, J. A., Correia, A. O., Montenegro, A. B., Nobre, M. E., Cerqueira, G. S., Neves, K.
R., Naffahmazzacoratti, M. G., Cavalheiro, E. A., Ga, D. C. B., & Gs, D. B. V. (2014).
Caffeine neuroprotective effects on 6-OHDA-lesioned rats are mediated by several factors,
including pro-inflammatory cytokines and histone deacetylase inhibitions. Behavioural Brain
Research, 264, 116-125.
Miranda, A. M., Steluti, J., Fisberg, R. M., & Marchioni, D. M. (2017). Association between
coffee consumption and its polyphenols with cardiovascular risk factors: a population-based
study. Nutrients, 9, 276-290.
20

Moeenfard, M., Cortez, A., Machado, V., Costa, R., Luís, C., Coelho, P., Soares, R., Alves, A.,
Borges, N., & Santos, A. (2016). Anti‐angiogenic properties of cafestol and kahweol
palmitate diterpene esters. Journal of Cellular Biochemistry, 117, 2748-2756.
Moeenfard, M., Erny, G. L., & Alves, A. (2016). Variability of some diterpene esters in coffee
beverages as influenced by brewing procedures. Journal of Food Science & Technology, 53,
3916-3927.
Moreira, A. S., Nunes, F. M., Domingues, M. R., & Coimbra, M. A. (2012). Coffee melanoidins:
structures, mechanisms of formation and potential health impacts. Food & Function, 3,
903-915.
Muhammada, I., Takamatsua, S., Mustafaa, J., Khana, S. I., Khana, I. A., Samoylenkoa, V.,
Mossac, J. S., El-Feralyc, F. S., & Dunbara, D. C. (2007). COX-2 inhibitory activity of
cafestol and analogs from coffee beans. Natural Product Communications, 3, 11-16.
Muriel, P., & Arauz, J. (2010). Coffee and liver diseases. Fitoterapia, 81, 297-305.
Naito, S., Yatagai, C., Maruyama, M., & Sumi, H. (2011). Effect of coffee extracts on plasma
fibrinolysis and platelet aggregation. Japanese journal of alcohol studies & drug dependence,
4
6, 260-269.
Natella, F., Nardini, M., Belelli, F., Pignatelli, P., Di, S. S., Ghiselli, A., Violi, F., & Scaccini, C.
2008). Effect of coffee drinking on platelets: inhibition of aggregation and phenols
(
incorporation. British Journal of Nutrition, 100, 1276-1282.
Nehlig, A. (2016). Effects of coffee/caffeine on brain health and disease: What should I tell my
patients? Practical Neurology, 16, 89-95.
Nishimura, K., Hitotsuyanagi, Y., Sugeta, N., Sakakura, K. I., Fujita, K., Fukaya, H., Aoyagi, Y.,
21

Hasuda, T., Kinoshita, T., & He, D. H. (2006). Tricalysiolides A–F, new rearranged ent
kaurane diterpenes from Tricalysia dubia. Tetrahedron, 62, 1512-1519.
-
Novaes, F. J. M., Oigman, S. S., Rezende, C. M., & Neto, F. R. D. A. (2015). New approaches on
the analyses of thermolabile coffee diterpenes by gas chromatography and its relationship
with cup quality. Talanta, 139, 159-166.
Pan, M. H., Tung, Y. C., Yang, G., Li, S., & Ho, C. T. (2016). Molecular mechanisms of the
anti-obesity effect of bioactive compounds in tea and coffee. Food & Function, 7, 4481-4491.
Ranheim, T., & Halvorsen, B. (2005). Coffee consumption and human health--beneficial or
detrimental? Mechanisms for effects of coffee consumption on different risk factors for
cardiovascular disease and type 2 diabetes mellitus. Molecular Nutrition & Food Research,
49, 274.
Ratnayake, W. M., Hollywood, R., O'Grady, E., & Stavric, B. (1993). Lipid content and
composition of coffee brews prepared by different methods. Food & Chemical Toxicology, 31,
263-269.
Ren, H. L., Zhao, Z. G, Le, J. J., Xu, K. S., & Fan, H, H. (2006). Primary study on
pharmacokinetics of harp seal oil in rats with LC-MS. Chinese Journal of Pharmaceutical
Analysis, 32, 749-754.
Saab, S., Mallam, D., Ii, G. A. C., & Tong, M. J. (2014). Impact of coffee on liver diseases: a
systematic review. Liver International Official Journal of the International Association for
the Study of the Liver, 34, 495-504.
Shen, C. P., Luo, J. G., Yang, M. H., & Kong, L. Y. (2015). Cafestol-type diterpenoids from the
twigs of Tricalysia fruticosa with potential anti-inflammatory activity. Journal of Natural
22

Products, 78, 1322-1329.
Shen, Z., Dong, Z., Cheng, P., Li, L., Chen, Z., & Liu, J. (2003). Effects of plumbagin on platelet
aggregation and platelet-neutrophil interactions. Planta Medica, 69, 605-609.
Shu, Y., Liu, J. Q., Peng, X. R., Wan, L. S., Zhou, L., Zhang, T., & Qiu, M. H. (2014).
Characterization of diterpenoid glucosides in roasted puer coffee beans. Journal Agriculture
and Food Chemistry, 62, 2631-2637.
Speer, K., & Speer, I. K. (2006). The lipid fraction of the coffee bean. Brazilian Journal of Plant
Physiology, 18, 201-216.
Tamaki, N., Matsunami, K., Otsuka, H., Shinzato, T., Aramoto, M., & Takeda, Y. (2008).
Rearranged ent-kauranes from the stems of Tricalysia dubia and their biological activities.
Journal of Natural Medicines, 62, 314-320.
Urgert, R., Weg, G. V. D., Kosmeijerschuil, T. G., Bovenkamp, P. V. D., Hovenier, R., & Katan, M.
B. (1995). Levels of the cholesterol-elevating diterpenes cafestol and kahweol in various
coffee brews. Journal of Agricultural & Food Chemistry, 43, 2167-2172.
Vitaglione, P., Fogliano, V., & Pellegrini, N. (2012). Coffee, colon function and colorectal cancer.
Food & Function, 3, 916-922.
Yin, H., Luo, J. G., & Kong, L. Y. (2013). Tetracyclic diterpenoids with isomerized isospongian
skeleton and labdane diterpenoids from the fruits of Amomum kravanh. Journal of Natural
Products, 76, 237-242.
Zhao, H. X., Ren, H. L., Xu, X. L., Zhang, F. X., Ding, D. D., Yue, F. P., Ma, X. M., & Zhong, W.
K. (2014). Rapid screening for gutter oil adulteration in vegetable oil based on fatty acid
methyl esters. Food Science, 35, 148-152.
23

Zhou, J., Zhou, S., & Zeng, S. (2013). Experimental diabetes treated with trigonelline: effect on β
cell and pancreatic oxidative parameters. Fundamental & Clinical Pharmacology, 27,
279-287.
24

Figure 1. Structures of diterpene esters isolated from green Arabica coff ee beans.
1
1
Figure 2. Key HMBC (H
correlations for 1b.
C) for 1a and 2, H- H COSY (
), ROESY (H
H)
25

Figure 3. Key HMBC (H
b.
C) for 4a and 7a, ROESY (H
H) correlations for 4b and
7
26

1
a
Table 1. H-NMR Data of Compounds 1-8
b
c
d
d
e
d
d
d
d
Position
1
1
1
2
3
4
5
6
7
8
1.50 (m)
.77 (m)
.62 (dd,
0.1, 7.6)
.91 (m)
.56 (d,
.0)
1.38 (m)
.53 (m)
1.54 (m)
.61 (m)
1.20 (m)
1.80 (m)
2.35 (t,
1.50 (m)
1.58 (m)
1.32 (m)
1.88 (m)
3.84 (t, 3.0)
-
3.43 s-like
-
3.44 s-like
-
4.20 (t, 6.0)
4.20 (t, 6.0)
1
-
5.72 (dd,
6.0, 1.5)
-
-
5.72 (dd,
6.0, 1.5)
-
2
2.36 (ddd,
14.0, 14.0, 3.0)
1.88 (m)
2.34 (ddd,
14.0, 14.0, 3.0)
1.86 (m)
2
2.37 (m)
1.94 (m)
2.76 (m)
2.55 (m)
2.02 (m)
2.67 (m)
2.55 (m)
1.98 (m)
2.66 (m)
2
7.3)
1
2.17 (m)
2.26 (d,
10.7)
2
5
6
2.34 (m)
2.36 (m)
2.66 (m)
2.69 (m)
9
1.56 (m)
1.65 (m)
1.30 (m)
1.61 (m)
1.30 (m)
1.60 (m)
1.74 (m)
1.50 (m)
1.59 (m)
2.03 (s)
1.67 (m)
1.88 (m)
1.45 (m)
1.68(m)
1.49 (m)
1.64 (m)
1.72 (m)
1.83 (m)
1.29 (m)
1.65 (m)
1.91 (m)
1.64(m)
1.67 (m)
2.34 (m)
1.74 (m)
1.96 (m)
1.50 (m)
1.70 (m)
1.54 (m)
1.67 (m)
1.64 (m)
1.86 (m)
1.28 (m)
1.69 (m)
1.94 (m)
1.53 (m)
1.58 (m)
2.08 (s)
1.75 (m)
1.86 (m)
1.62 (m)
2.23 (m)
2.06 (m)
1.72 (m)
1.94 (m)
1.62 (m)
1.66 (m)
2.11 (m)
1.72 (m)
2.00 (m)
1.55 (m)
1.77 (m)
1.63 (m)
1.76 (m)
1.60 (m)
1.75 (m)
2.01 (m)
1.53 (m)
1.68 (m)
1.64 (m)
1.66 (m)
2.18 (s)
1.64 (m)
1.76 (m)
1.61 (m)
1.77 (m)
2.02 (m)
1.54 (m)
1.78 (m)
1.54 (m)
1.59 (m)
2.18 (s)
1.57 (m)
1.75 (m)
1.64 (m)
1.88 (m)
2.08 (m)
1.54 (m)
1.90 (m)
1.63 (m)
1.66 (m)
1.91 (m)
1.69 (m)
1.86 (m)
1.57 (m)
1.75 (m)
1.58 (m)
1.74 (m)
1.60 (m)
1.87 (m)
2.08 (m)
1.47 (m)
1.95 (m)
1.62 (m)
1.64 (m)
1.89 (m)
1.68 (m)
1.86 (m)
1.57 (m)
1.77 (m)
1
7
1
9
1.25 (m)
1.45 (m)
11
1
.91 (m)
1.50 (m)
.68 (m)
12
1
1
1
3
4
2.52 (s)
1.96 (m)
1.74 (m)
1.96 (m)
1.58 (m)
1.74 (m)
1.68 (m)
1.96 (m)
1.62 (m)
1.76 (m)
1.77 (m)
2.00 (m)
1.62 (m)
1.72 (m)
2
.05 (m)
1.65 (m)
.77 (m)
15
1
27

4
1
4
1
.52 (d,
4.13 (d,
11.4)
4.22 (d,
13.0)
4.28 (d,
13.0)
5.72 (s)
0.88 (s)
-
4.22 (d,
13.0)
4.30 (d,
13.0)
5.63 (s)
0.86 (s)
-
4.23 (d,
13.0)
4.30 (d,
13.0)
5.63 (s)
0.86 (s)
-
4.16 (d,
13.0)
4.18 (d,
13.0)
5.76 (s)
0.88 (s)
-
4.16
(d,13.0)
4.18 (d,
13.0)
5.76 (s)
0.88 (s)
-
1
7
4.23 (d, 13.0)
4.23 (d, 13.0)
4.26 (d, 13.0)
1.3)
.68 (d,
1.2)
4.22 (d,
11.3)
4.26 (d, 13.0)
1
2
8
0
5.80 (s)
5.78 (s)
0.83 (s)
7.31 (s)
4.43 (s)
5.62 (s)
5.63 (s)
0.75 (s)
0.85 (s)
0.86 (s)
OH-3
-
-
-
-
-
-
OH-16
-
-
-
-
-
3
.43
OMe-1
-
-
-
-
-
3.44 (s)
-
-
(
overlapped)
2
ʹ-8ʹ
1.28 (m)
1.28 (m)
5.35 (m)
1
.30 (m)
.34 (m)
9ʹ
1.20 (m)
5
.35 (m)
2.77 (m)
.35 (m)
1
0ʹ
5
1
1
1.20-1.25
(m)
1.21-1.30
(m)
1.30 (m)
1.28 (m)
1.20 (m)
0.78 (t, 7.0)
5.33 (m)
2.77 (m)
5.33 (m)
1
1
1
1
2ʹ
3ʹ
4ʹ
5ʹ
2.76 (m)
.34 (m)
5
5
1.28 (m)
0
.82 (t,
0.91 (t,
1.28 (m)
16ʹ
0.91 (t, 7.0)
0.86 (t, 7.0)
6
.8)
6.6)
1.30 (m)
1
1
7ʹ
8ʹ
-
-
-
-
-
-
-
-
-
1
.20 (m)
-
0.89 (t, 7.0)
0.89 (t, 7.0)
-
0
.90 (t,
1
2
9ʹ
0ʹ
-
-
-
-
-
-
-
-
-
-
-
7
.0)
-
-
-
0.78 (t, 7.0)
a
b
c
d
e
δ in parts per million, J in Hz, and obstained at 600 Hz. NMR solvent was C
5
D
5
N, DMSO, CDCl
3
, MeOD.
28

13
a
Table 2. C-NMR and DEPT Data of Compounds 1-8
b
c
d
d
e
d
d
d
d
Position
1
1
2
3
4
5
6
7
8
1
2
3
4
5
6
35.0 (t)
34.5 (t)
105.5 (s)
173.6 (s)
47.3 (d)
22.0 (t)
35.1 (t)
33.8 (t)
104.5 (s)
172.5 (s)
46.4 (d)
21.4 (t)
35.6 (t)
34.5 (t)
104.3 (s)
172.4 (s)
47.0 (d)
21.8 (t)
35.4 (t)
34.3 (t)
104.1 (s)
172.1 (s)
46.9 (d)
21.5 (t)
71.1 (d)
40.9 (t)
80.4 (d)
35.3 (t)
80.5 (d)
35.3 (t)
69.6 (d)
107.1 (d)
152.0 (s)
159.1 (s)
38.7 (d)
21.7 (t)
69.6 (d)
106.9 (d)
152.1 (s)
159.0 (s)
38.7 (d)
21.8 (t)
107.6 (s)
174.5 (s)
41.3 (d)
22.3 (t)
102.9 (s)
172.7 (s)
40.2 (d)
21.8 (t)
102.8 (s)
172.6 (s)
40.3 (d)
21.6 (t)
3
9.5
7
40.2 (t)
39.5 (t)
39.4 (t)
39.7 (t)
39.0 (t)
39.0 (t)
39.0 (t)
39.0 (t)
(
overlapped)
43.9 (s)
52.7 (s)
42.9 (s)
18.6 (t)
25.6 (t)
44.8 (d)
37.1 (t)
52.8 (t)
78.2 (s)
67.1(t)
8
9
44.7 (s)
53.6 (d)
43.7 (s)
19.3 (t)
26.3 (t)
46.3 (d)
37.9 (t)
53.9 (t)
79.3 (s)
69.0 (t)
112.5 (d)
171.4 (s)
14.4 (q)
44.7 (s)
53.3 (d)
43.7 (s)
19.3 (t)
27.4 (t)
46.0 (d)
37.6 (t)
52.9 (t)
80.2 (s)
68.3 (t)
112.9 (d)
171.1 (s)
14.4 (q)
44.5 (s)
53.2 (d)
43.5 (s)
19.2 (t)
27.2 (t)
45.8 (d)
37.5 (t)
52.7 (t)
80.0 (s)
68.1 (t)
112.8 (d)
170.7 (s)
14.4 (q)
45.6 (s)
45.3 (d)
43.7 (s)
19.5 (t)
27.0 (t)
46.8 (d)
38.3 (t)
54.3 (t)
80.8 (s)
69.3 (t)
113.7 (d)
173.2 (s)
16.0 (q)
44.4 (s)
43.6 (d)
47.5 (s)
18.9 (t)
27.3 (t)
45.9 (d)
37.5 (t)
53.3 (t)
80.0 (s)
68.3 (t)
112.4 (d)
170.8 (s)
15.5 (q)
58.9 (s)
174.3 (s)
44.5 (s)
43.6 (d)
47.6 (s)
19.0 (t)
27.4 (t)
45.9 (d)
37.6 (t)
53.3 (t)
80.0 (s)
68.3 (t)
112.4 (d)
170.7 (s)
15.5 (q)
59.0 (s)
174.2 (s)
44.3 (s)
46.1 (d)
44.0 (s)
17.8 (t)
25.9 (t)
43.9 (d)
37.1 (t)
53.0 (t)
80.1 (s)
68.3 (t)
112.3 (d)
170.0 (s)
16.0 (q)
44.3 (s)
46.1 (d)
44.0 (s)
19.3 (t)
25.9 (t)
43.9(d)
37.1 (t)
53.0 (t)
80.1 (s)
68.3 (t)
112.4 (d)
169.8 (s)
16.0 (q)
1
0
1
1
1
1
1
1
1
1
1
1
2
2
3
4
5
6
7
8
9
0
111.9 (d)
170.6 (s)
14.0 (q)
OMe-1
1’
174.0 (s)
173.2 (s)
174.3 (s)
174.1 (s)
176.0 (s)
174.3 (s)
174.2 (s)
29