Synthesis, cytotoxic activity evaluation and HQSAR study of novel isosteviol derivatives as potential anticancer agents

Article abstract of DOI:10.1016/j.ejmech.2016.03.009

A series of novel isosteviol derivatives bearing amino alcohol and thiourea fragments have been stereo-selectively synthesized and screened for their in vitro cytotoxic activities against three human cancer cell lines (HCT-116, HGC-27 and JEKO-1). The results demonstrated that these compounds exhibited prominent cytotoxicities. Especially, the compound Iw displayed the most potent anticancer activities against HCT-116 cell with IC₅₀ value of 1.450 μM. On the basis of this bioassay results, these derivatives were further investigated by the hologram quantitative structure-activity relationship (HQSAR) technique. The optimal HQSAR model with q² Combining double low line 0.663, r² Combining double low line 0.895, SEE Combining double low line 0.179 was generated using A/B/H/Ch as fragment distinction parameters and 4-7 as fragment size. This model was employed to predict the cytotoxic activities of test set compounds, and the predicted values were in good agreement with the experimental results. The contribution maps derived from the optimal model explained the individual atomic contribution to the total activity of single molecule.

Full text of DOI:10.1016/j.ejmech.2016.03.009

Accepted Manuscript
Synthesis, cytotoxic activity evaluation and HQSAR study of novel isosteviol
derivatives as potential anticancer agents
Cong-Jun Liu, Shu-Ling Yu, Yan-Ping Liu, Xing-Jie Dai, Ya Wu, Rui-Jun Li, Jing-Chao
Tao
PII:
S0223-5234(16)30183-0
10.1016/j.ejmech.2016.03.009
EJMECH 8434
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 22 January 2016
Revised Date: 2 March 2016
Accepted Date: 3 March 2016
Please cite this article as: C.-J. Liu, S.-L. Yu, Y.-P. Liu, X.-J. Dai, Y. Wu, R.-J. Li, J.-C. Tao, Synthesis,
cytotoxic activity evaluation and HQSAR study of novel isosteviol derivatives as potential anticancer
agents, European Journal of Medicinal Chemistry (2016), doi: 10.1016/j.ejmech.2016.03.009.
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.

ACCEPTED MANUSCRIPT
Graphical Abstract:
Synthesis, cytotoxic activity evaluation and HQSAR study of
novel isosteviol derivatives as potential anticancer agents
Cong-Jun Liu, Shu-Ling Yu, Yan-Ping Liu, Xing-Jie Dai, Ya Wu, Rui-Jun Li*,
Jing-Chao Tao*

ACCEPTED MANUSCRIPT
Synthesis, cytotoxic activity evaluation and HQSAR study of
novel isosteviol derivatives as potential anticancer agents
Cong-Jun Liu a,b, Shu-Ling Yu c, Yan-Ping Liu b, Xing-Jie Dai a, Ya Wu a, Rui-Jun Li*a, Jing-Chao
Tao*a
a
College of Chemistry and Molecular Engineering, New Drug Research & Development Center,
Zhengzhou University, 75 Daxue Road, Zhengzhou, Henan 450052, China
b
College of Chemical Engineering and Food Technology, Zhongzhou University, Zhengzhou,
Henan 450044, China
c
Key Laboratory of Natural Medicine and Immune-Engineering of Henan Province, Henan
University, Kaifeng, Henan 475004, China
Abstract: A series of novel isosteviol derivatives bearing amino alcohol and thiourea
fragments have been stereo-selectively synthesized and screened for their in vitro cytotoxic
activities against three human cancer cell lines (HCT-116, HGC-27 and JEKO-1). The results
demonstrated that these compounds exhibited prominent cytotoxicities. Especially, the compound
Iw displayed the most potent anticancer activities against HCT-116 cell with IC50 value of
1.450ꢀM. On the basis of this bioassay results, these derivatives were further investigated by the
hologram quantitative structure–activity relationship (HQSAR) technique. The optimal HQSAR
model with q2 = 0.663, r2 = 0.895, SEE = 0.179 was generated using A/B/H/Ch as fragment
distinction parameters and 4-7 as fragment size. This model was employed to predict the cytotoxic
activities of test set compounds, and the predicted values were in good agreement with the
experimental results. The contribution maps derived from the optimal model explained the
individual atomic contribution to the total activity of single molecule.
Key words: Isosteviol; Synthesis; Anticancer activity; HQSAR.
* Corresponding author. Tel/Fax: +86-371-67767200. Email address: jctao@zzu.edu.cn.
1. Introduction
Cancer is a major public concerned health hot-spot in all over the world. The number of global
new cancer cases is projected to rapidly increase from 14 million in 2012 to 24 million in 2035[1].
Despite great breakthroughs in recent decades, state of the art cancer therapy still has many
limitations, e.g. severe side effect of traditional chemotherapeutic drugs and high costs of scare
1

ACCEPTED MANUSCRIPT
effective anticancer agents [2-4]. Therefore, it is urgent and indispensable to develop novel
compounds as inexpensive anticancer agents with higher bioactivities and lower side effects to
improve life expectancy and quality of life of patients [5].
Natural products and their derivatives have historically been invaluable as a source of anticancer
agents and the unique metabolites produced by diterpenoids have increasingly become major
players in anticancer drug discovery [6-8]. Isosteviol, an ent-beyeran diterpenoid, could be easily
obtained by acid hydrolysis of natural available stevioside[9, 10]. Previous studies of isosteviol
were mainly focused on antihypertension[11], antihyperglycemic[12], antioxidative[13],
antidiarrheal[14], cardio- and neuroprotective effects[15, 16]. As well as the cytotoxic activities of
isosteviol deviratives, obtained by microbial transformation and chemical modification, have
attracted much attention in recent years[17-22]. Especially, hundreds of novel isosteviol
derivatives have been readily synthesized by chemical modification of one or two reactive groups
of isosteviol, and some of those compounds exhibited good cytotoxic activity as potential drug
molecules. Li and co-workers reported that compound A (Fig. 1) with α-methylene
cyclopentanone fragment in D-ring of isosteviol displayed remarkable anticancer activity against
MDA-MB-231 cell line with IC50 value of 1.58ꢀM[23]. Huang et al. found that ureido moiety
containing isosteviol derivative B (Fig. 1), prepared by replacing the 4-COOH with the ureido
moiety, showed inhibitory potency against the hepatitis B virus (HBV) in HepG2 2.2.15 cells[24].
A number of results explicitly demonstrated that some particular subunits of isosteviol derivatives
played an important role in inhibiting the growth of cancer cells, thus allowing new possibilities
for quantitative structure–activity relationship (QSAR) studies associated with structure-based
good bioactivity molecule design strategies[25-27].
Amino alcohol is a crucial building block of many natural products or synthetic molecules
exemplified by the compound C (Fig. 1) and exhibits wide-ranging biological activities[28]. In
our previous work, a crucial fragment of amino alcohol in the D-ring of isosteviol was built up and
series of compounds were obtained with significantly improved anticancer activities[29]. As well
as the thiourea derivatives were regarded as one of the most promising classes of anticancer agents
with good inhibitory activity against topoisomerase II, protein tyrosine kinases and human sirtuin
type proteins 1 and 2[30-32]. Such as hydroxyurea[33]has been described as a clinically useful
drug for the treatment of a wide range of solid tumors. Rao et al.[34, 35]reported that the
2

ACCEPTED MANUSCRIPT
anticancer activity was greatly improved when substituted thiourea moieties were coupled with the
A-ring of dehydroabietic acid D (Fig. 1). However, few reports have been focused on construction
of an anticancer compound which contains amino alcohol group and/or thiourea subunit
simultaneously in the isosteviol skeleton. Furthermore, to the best of our knowledge, hologram
quantitative structure–activity relationship (HQSAR) studies on isosteviol derivatives have been
scarcely reported hitherto. Based on these facts and in continuation of our previous work, novel
isosteviol derivatives with amino alcohol and thiourea subunits were rationally designed and
synthesized. Their cytotoxic activities against three human cancer cell lines were evaluated.
Correlating the novel molecule structures with their cytotoxic activity in HQSAR generated the
robust and predictive model which would be further taken advantage of designing new anticancer
molecules with improved potency.
Figure 1: Chemical structures of compounds A-D
2. Results and Discussion
2.1. Chemistryꢀ
Our synthetic efforts were initially focused on novel amino alcohol type derivative of isosteviol,
ethyl ent-15α-hydroxymethyl-16β-aminobeyeran-19-oate Ia[36], and the synthetic route was
outlined in Scheme 1.
17
12
13
11
20
O
C 14
O
OH
OH
1
D 16
9
ii
2
3
i
A 10
B
8
iii
15
4
7
5
6
19COOH
1
COOEt
2
18
COOEt
3
O
NOH
OH
NH2
OH
iv
v
OH
COOEt
5
COOEt
4
COOEt
Ia
Scheme 1. Reagents and conditions: (i) CH3CH2Br, DMSO, KOH, rt, 6h, 98%; (ii) HCHO, Na,
C2H5OH, 55oC, 8 h, 90%; (iii) TCC, CH2Cl2, rt, 0.5 h, 82%; (iv) NH2OH.HCl, NaHCO3, EtOH,
reflux, 3h, 98%; (v) NaBH4, MoO3, MeOH, rt, 10h, 85%.
3

ACCEPTED MANUSCRIPT
Isosteviol 1 was obtained through the sulfuric acid hydrolysis of stevioside, which was then
treated with CH3CH2Br and KOH in DMSO to afford the corresponding isosteviol ethyl ester 2.
Compound 3 was stereo-selectively synthesized via a one-pot Tollens’ reaction. The hydroxy
group at C-16 could be selectively oxidized by the trimethylammonium chlorochromate (TCC)
[37] to afford compound 4 in good yield. After treatment of compound 4 with hydroxylamine
hydrochloride in the presence of sodium bicarbonate in ethanol under reflux condition, the
corresponding oxime 5 was obtained as a white powder. Compound 5 was further converted to the
unique 1, 3–amino alcohol product Ia by reduction of the oxime group with NaBH4/MoO3 in
MeOH in 85% yield. The structure of compound Ia was confirmed by IR, NMR and HRMS,
respectively. In its IR spectrum, broad bands at 3328 and 3262 cm-1 were assigned to ν (N-H)
groups. The appearance of signals at δH 2.41 and δC 49.28 in the NMR spectra, and the
disappearance of signal at δC 172.12 in the 13C NMR spectrum further demonstrated that the
C=N-OH group had converted to the CHNH2 fragment. In accord with the molecular formula
C23H39NO3, an [M + H]+ peak at m/z 378.3006 in the HRMS was afforded. The absolute
configuration of compound Ia was confirmed by X-ray crystallographic analysis of its N,O-
acetylated derivative, indicating an anti direction of the hydroxymethyl group at C-15 and the
amino group at C-16.
In order to investigate the structural effects of the D-ring region on cytotoxic activities, such as,
the different orientations of the amino group at C-16, the synergy effect of the 15-hydroxymethyl
with the 16-amino group and the case of the D-ring opening status, and simultaneously to provide
diverse molecule structures convenient for the subsequent HQSAR study, compounds IIa, IIIa
and IVa were also designed and synthesized. The synthetic routes towards the target compounds
were depicted in Scheme 2.
4

ACCEPTED MANUSCRIPT
O
NOH
NH2
i
ii
COOEt
2
COOEt
6
COOEt
IIa
N
NH2
OH
OH
OH
ii
COOEt
7
COOEt
COOEt
3
IIIa
iii
NH
O
NH2
OH
iv
COOEt
8
COOEt
IVa
Scheme 2. Reagents and conditions: (i) NH2OH.HCl, NaHCO3, EtOH, reflux, 3h, 95%; (ii)
.
NaBH4, MoO3, MeOH, rt, 10h; (iii) toluene, BF3 OEt2, 110 oC, 1h, 95%; (iv) Raney Ni/H2, 1MPa,
MeOH, 60 oC, 8h, 83%.
Compound IIa, lacking of the hydroxymethyl group at C-15 of compound Ia, was readily
synthesized to verify further the synergy effects of 15-hydroxymethyl and 16-amino via an
oximation of compound 2 and then stereo-selectively reduction with NaBH4/MoO3 as described in
our previous work[38]. The D-ring opening compound 7 was synthesized according to the
procedure reported in the literature[39]. Compound IIIa, with hanging aminomethyl and vinyl
groups, was obtained as a white powder by reduction of compound 7 with NaBH4 in anhydrous
methanol. Compound 8 was afforded by means of an intramolecular 1, 3-dipolar cycloaddition of
.
compound 7 in the presence of a catalytic amount of BF3 OEt2 at 110oC with a yield of 95%.
Compound IVa, an epimer of compound Ia, was then produced through the catalytic
hydrogenolysis of the N-O bond cleavage with Raney Ni/H2 in 83% yield. The stereo-structure of
compound IVa was confirmed by X-ray crystallographic analysis (Fig. 2), demonstrating a β
orientation of the amino group at C-16 [36].
Figure 2: X-ray structure of compound IVa
Some attempts were carried out for further modification of the amino group at C-16 in compounds
5

ACCEPTED MANUSCRIPT
Ia, IIa and the aminomethyl group in compound IIIa in order to investigate whether the amino
group at C-16 playing a significant role in anticancer activities. The synthetic approaches
employed to prepare the target derivatives were outlined in Scheme 3.
NHAc
OH
NHAc
OAc
iii
COOEt
Id
COOEt
Ic
ii
NH2
NH2
OH
NH2
COOEt
IIIa
COOEt
Ia
COOEt
IIa
i
i
i
NCS
NCS
OH
NCS
COOEt
Ib
COOEt
IIb
COOEt
IIIb
Scheme 3. Reagents and conditions: (i) CS2, Boc2O, DMAP, TEA, EtOH, rt, 8 h; (ii) Ac2O,
pyridine, rt, 3h, 95%; (iii) MeOH, K2CO3, rt, 2h, 93%.
Compound Ia-IIIa was respectively reacted with carbon disulfide in the presence of Boc2O,
DMAP and TEA in EtOH to form the corresponding isothiocyanate compound Ib-IIIb in yields of
42-89%. Compound Ic, achieved through the reaction of compound Ia and acetic anhydride in
pyridine, could be easily converted to compound Id in presence of potassium carbonate under a
mild condition. The absolute configuration of compound Ic was confirmed by X-ray
crystallographic analysis (Fig. 3), indicating an α orientation of the amino group at C-16[36].
Figure 3: X-ray structure of compound Ic
With the amino alcohol derivatives and the isothiocyanates of isosteviol in hand, the following
work was focused on the introduction of thiourea group onto the skeleton of these molecules. The
synthesis of thiourea derivatives is summarized in Scheme 4. Compounds Ie–Iy containing a
hydroxymethyl group at C-15 and a thiourea fragments at C-16, simultaneously, were obtained
through condensation of compound Ia with appropriately substituted phenyl isothiocyanates in
CH2Cl2 with the final yields of 52–93%. The structures of these compounds were characterized by
6

ACCEPTED MANUSCRIPT
1
IR, NMR and HRMS spectra, respectively. In the IR, HNMR and 13CNMR spectra of Iw,
additional resonances were distinctly observed at ν 1335, δH 10.76, 6.45 and δC 181.73, indicating
the formation of thiourea subunit.
S
NH2
OH
NH
OH
NH
i
R
COOEt
Ia
COOEt
Ie-Iy
Ie: R = H (52%)
Il: R = o-Cl (82%)
Is: R = m-OCH3 (72%)
It: R = p-OCH3 (83%)
Iu: R = o-NO2 (80%)
Iv: R = m-NO2 (81%)
Iw: R = p-NO2 (88%)
Ix: R = 2,4-Cl2 (82%)
Iy: R = 2,6-Me2 (62%)
If: R = o-CH3 ( 66% ) Im: R = m-Cl (93%)
Ig: R = m-CH3 (64%)
Ih: R = p-CH3 (72%)
Ii: R = o-F (88%)
Ij: R = m-F (88%)
Ik: R = p-F (83%)
In: R = p-Cl (81%)
Io: R = o-Br (77%)
Ip: R = m-Br (81%)
Iq: R = p-Br (72%)
Ir: R = o-OCH3 (70%)
S
(65%)
IIc: R1
=
NCS
NCS
NH
NH
R1
ii
IId: R1
IIe: R1
=
=
(85%)
(82%)
COOEt
IIb
COOEt
IIc-IIf
S
NH
NH
ii
O
IIf: R1
=
(78%)
HN
COOEt
IIIb
COOEt
IIIc
Scheme 4. Reagents and conditions: (i) isothiocyanates, CH2Cl2, rt, 1h; (ii) the relative amines,
CH2Cl2, rt, 2h.
The isothiocyano group in compound IIb made it suitable for engaging condensation with various
substituted aromatic amines to form a series of thiourea containing compounds IIc-IIf. Meanwhile,
treatment of compound IIIb with aniline in CH2Cl2 produced compound IIIc as a peculiar
molecule.
2.2. Evaluation of Cytotoxic Activity
These as-synthesized isosteviol derivatives were further utilized to investigate their anticancer
activities against three human cancer cell lines: colon carcinoma (HCT-116), gastric carcinoma
(HGC-27), mantle cell lymphoma (JEKO-1). The in vitro cytotoxic activities of thirty five
compounds were determined by means of MTT assay. Cisplatin, employed in the treatment of
human cancers as one of the most effective chemotherapeutic drugs, was selected as a positive
control. The IC50 values were used to demonstrate the growth inhibition in the presence of
isosteviol derivatives and presented in Table 1.
Table 1: Cytotoxic activities of isosteviol derivatives in vitro
Cytotoxic activities (IC50, ꢀM)
Compound
HCT-116
HGC-27
JEKO-1
isosteviol l
Ia
>100
4.891±0.274
>100
5.826±0.200
>100
3.541±0.113
7

ACCEPTED MANUSCRIPT
Ib
Ic
Id
Ie
If
Ig
Ih
Ii
Ij
Ik
Il
Im
In
Io
Ip
Iq
Ir
Is
It
Iu
Iv
Iw
Ix
Iy
IIa
IIb
IIc
IId
IIe
IIf
IIIa
IIIb
IIIc
IVa
Cisplatin
4.449±0.187
20.070±2.469
15.249±1.435
5.789±0.362
4.783±0.211
4.630±0.435
5.622±0.528
4.847±0.615
4.910±0.526
4.765±0.758
7.228±0.158
2.753±0.146
4.411±0.326
20.851±1.873
4.175±0.441
2.215±0.187
5.403±0.344
3.841±0.397
4.566±0.230
11.666±0.571
1.877±0.112
1.450±0.094
4.016±0.179
4.542±0.372
5.082±0.140
6.960±0.183
8.660±0.601
82.567±6.006
6.635±0.338
15.096±1.341
2.457±0.076
4.142±0.769
3.470±0.237
4.899±0.551
3.906±0.261
7.696±1.177
20.538±1.316
18.006±0.576
5.158±0.364
5.036±0.430
4.356±0.585
5.530±0.218
7.378±0.434
4.983±0.035
4.921±0.002
8.983±0.304
5.462±0.441
3.326±0.515
15.558±2.721
9.154±0.384
2.982±0.179
9.017±0.864
7.208±0.428
7.717±0.823
22.759±1.015
2.382±0.301
2.355±0.211
4.776±0.611
8.896±0.275
2.599±0.235
11.607±0.465
13.755±1.621
92.230±3.680
18.220±0.639
26.152±2.381
3.024±0.315
11.460±2.175
6.685±1.047
7.246±1.053
2.032±0.102
30.099±1.539
60.607±2.648
61.953±3.012
27.333±1.376
24.677±1.925
20.835±0.857
30.212±1.963
35.685±0.836
28.890±1.680
23.713±2.070
40.712±1.618
20.484±1.378
5.764±0.226
14.045±1.172
27.305±0.815
6.727±0.399
28.274±0.982
31.945±0.825
40.311±1.642
26.927±0.742
5.257±0.692
3.539±0.182
27.305±1.274
21.082±0.338
2.196±0.199
38.284±0.736
21.463±1.465
94.830±4.969
15.150±2.540
56.834±1.753
2.160±0.153
5.426±0.269
2.775±0.095
2.473±0.311
2.743±0.235
As shown in Table 1, all the derivatives of isosteviol revealed better cytotoxic activities than their
corresponding precursor. Interestingly, HCT-116 and HGC-27 cells were considerable sensitive to
all the tested compounds except compound IId. While introduction of amino group into C-16,
compounds (Ia, IIa, IIIa and IVa) displayed pronounced anti-proliferative effects on all the three
cancer cell lines, which are similar to those of cisplatin. To be noteworthiness, the inhibitory
activities against HGC-27 and JEKO-1 cell lines were somewhat negatively improved by
introducing hydroxymethyl group at C-15 to construct an amino alcohol subunit (Ia, IVa vs IIa),
whereas their activities against HCT-116 were consistent. Meaningfully, the inhibitory activities of
8

ACCEPTED MANUSCRIPT
the D-ring opened compounds (IIIa, IIIb, IIIc) against the three cancer cell lines were relatively
better than their precursors, especially IIIa, which had exocyclic double bond and exocyclic
amino group, showed as the most attractive molecule. Comparing the data of Ia and IVa we can
see that there were no remarkable differences in the bioactivities of the two diastereomers. Not
surprisingly, the inhibitory activities were lowered while converting the amino group at C-16 to an
isothiocyano group (Ia vs Ib, IIa vs IIb, IIIa vs IIIb) except compound Ib which displayed the
better inhibitory activities against HCT-116 cells than compound Ia. Especially, the cytotoxicities
were prominently decreased as the amino and the hydroxyl groups in compound Ia was acylated
(Ic, Id vs Ia). In view of the performance of the tested compounds, we could arrive at the
conclusion that the amino alcohol fragment exactly played an important role in inhibiting cancer
cell proliferation.
When the substituted phenyl thiourea subunits were introduced onto the structure of compound Ia,
the cytotoxic activities against HCT-116 and HGC-27 cell lines were further improved, whereas
most of the thiourea fused deriveratives showed moderate cytotoxicity against JEKO-1 cells. In
addition, the compounds containing a para electron-withdrawing group on the benzene ring were
more potent against the three cell lines than the ortho and meta-substituted compounds (Ik vs Ii-Ij,
In vs Il-Im, Iq vs Io-Ip and Iw vs Iu-Iv). Specifically, para-nitro substituted derivative Iw
displayed the most potent anticancer activities with IC50 values of 1.450ꢀM, 2.355ꢀM and 3.539ꢀM
against HCT-116, HGC-27, JEKO-1 cells, respectively. Nevertheless, the meta-substituted
compounds containing an electron-donating group on the aromatic ring showed better inhibitory
activities against HCT-116 than the ortho and para-substituted compounds.
Comparing the anti-proliferative activities of compounds IIc and IId against the three cell lines
we could find an important difference between phenyl and cyclohexyl group in the thiourea
subunit. The cyclohexyl containg compound, IId, exhibited the poorest activity with IC50 values
of 82.567ꢀM, 92.230ꢀM and 94.830ꢀM against HCT-116, HGC-27 and JEKO-1 cell lines,
respectively, implying that the phenyl group may play a significant role in the anticancer activity.
2.3. HQSAR Model
The selection of appropriate training and test sets is of critical importance in the generation of
robust QSAR models. The data set of thirty five compounds was used for the HQSAR analysis.
Total of twenty five compounds, whose essential characteristic was that the molecule must be
9

ACCEPTED MANUSCRIPT
orthogonal, were selected as members of the training set for model construction. The remaining
ten compounds with isothiocyano (Ib), acylamino (Id), substituted thioureido (If, Ij, In, Iq, Is, Iu,
IIc) and amino (IIIa) fragments were arranged into test set for external model validation.
Structurally diverse molecules possessing bioactivities of a wide range were taken into
consideration including in both training and test sets[40, 41]. Thus, the data sets were favorable to
HQSAR model development.
HQSAR calculations were carried out using three distinct parameters—the fragment distinction,
the fragment size and the hologram length. When the patterns of fragment counts from the training
set compounds were correlated to the cytotoxic activities against HCT-116 and HGC-27 cell lines,
respectively, then no satisfactory HQSAR model was obtained due to the subtle difference of IC50
values of compounds. However, predictable and statistical significant models were generated with
high q2, r2 values and low SEE values [42] when pIC50 (-logIC50) values against JEKO-1 cell were
used as dependent variables in HQSAR analyses. The statistical results of the PLS analyses for the
training set using different fragment distinctions in combinations with default fragment size (4-7)
and twelve default hologram lengths (53, 59, 61, 71, 83, 97, 151, 199, 257, 307, 353 and 401bins)
were summarized in Table 2. In process of selecting the most effective fragment distinction, the
highest values of q2 and r2 were not always considered because the reports suggested that models
with the highest training set accuracy alone did not afford the highest predictive accuracy of the
test set[43, 44]. Thus, the predictive capability of fourteen better models (q2>0.6) was
preliminarily assessed. The pIC50 values of the test set were predicted and predictive r2
values
pred
were also listed in Table 2. Taking into account of the predictive r2
values, A/B/H/Ch was
pred
proved as the most effective fragment distinction.
Table 2: HQSAR analysis for various fragment distinction using default fragment size (4–7)
model
Fragment distinction
q2
SEcv
r2
SEE
HL
N
r2
pred
1
2
3
4
5
6
7
8
9
10
A/B
A/B/C
A/B/H
A/C/H
A/B/Ch
A/C/Ch
A/B/DA
A/C/DA
A/B/C/H
A/Ch/DA
0.539
0.633
0.648
0.644
0.621
0.613
0.590
0.615
0.536
0.634
0.365
0.326
0.319
0.330
0.340
0.344
0.354
0.343
0.357
0.325
0.945
0.931
0.908
0.946
0.973
0.973
0.964
0.955
0.730
0.901
0.126
0.141
0.163
0.128
0.091
0.091
0.105
0.117
0.272
0.169
71
53
401
83
199
59
353
401
199
53
5
5
5
6
6
6
6
6
4
5
N
0.427
0.610
0.507
0.395
0.308
N
0.519
N
0.373
10

ACCEPTED MANUSCRIPT
11
12
13
14
15
16
17
18
19
20
21
22
A/B/H/Ch
A/B/C/Ch
A/C/H/Ch
A/B/C/DA
0.663
0.602
0.442
0.609
0.647
0.623
0.597
0.536
0.575
0.609
0.614
0.564
0.321
0.348
0.402
0.346
0.328
0.339
0.351
0.366
0.360
0.346
0.343
0.365
0.895
0.969
0.879
0.943
0.901
0.967
0.947
0.852
0.864
0.960
0.915
0.888
0.179
0.097
0.187
0.132
0.174
0.100
0.127
0.207
0.204
0.110
0.161
0.185
53
83
97
83
97
353
53
6
6
5
6
5
6
6
5
6
6
6
6
0.711
0.408
N
0.429
0.579
0.555
N
N
N
0.446
0.542
N
A/C/H/DA
A/H/Ch/DA
A/B/Ch/DA
A/B/C/H/Ch
A/B/C/H/DA
A/B/C/Ch/DA
A/C/H/Ch/DA
A/B/C/H/Ch/DA
83
151
257
97
257
q2: cross-validated correlation coefficient (LOO). SECV: cross-validated standard error. r2:
noncross-validated correlation coefficient. SEE: noncross-validated standard error. HL: hologram
length. N: optimal number of component. Fragment distinction: A-atom types, B-bond types,
C-connectivity, H-hydrogens, Ch-chirality, DA-donor and acceptor. N: not calculated.
In order to optimize statistical results, various fragment sizes were applied with A/B/H/Ch as
fragment distinction. The statistical results, listed in Table 3, displayed that there was no
improvement in the HQSAR model by altering the fragment size. The fragment size default (4-7)
offered the better statistical results in comparison with other fragment sizes. Exhilaratingly, the
optimal HQSAR model was developed by using atoms (A), bonds (B), hydrogens (H), chirality
(Ch) as fragment distinction parameters and 4-7 as fragment size demonstrating a cross-validated
r2 (q2) value of 0.663 and a non cross-validated r2 value of 0.895. Often, a high value of this
statistical characteristic (q2 > 0.5) is considered as a proof of the high predictive ability of the
model[45].Thus, this best model with a q2 greater than 0.5 should be predictive and could be used
for prediction. Subsequently, the optimal model was used to predict the cytotoxic activities against
JEKO-1 cell of these compounds, and the predicted results were listed in Table 4. The correlation
of experimental and predicted values was displayed in figure 4. The predicted pIC50 residuals of
the most test set compounds were less than 0.333 log unit, which is a strong evidence that highly
predictive QSAR model was obtained.
Table 3: HQSAR analysis for the influence of fragment sizes using the fragment distinction
model
Fragment size
q2
SEcv
r2
SEE
HL
N
r2
pred
1
2
3
4
5
6
1-3
1-4
2-4
3-5
3-7
4-5
0.558
0.539
0.535
0.628
0.576
0.512
0.367
0.375
0.377
0.337
0.360
0.386
0.925
0.950
0.953
0.968
0.888
0.954
0.152
0.123
0.120
0.099
0.185
0.119
83
6
6
6
6
6
6
N
N
N
0.463
N
151
151
53
53
257
N
11

ACCEPTED MANUSCRIPT
7
8
9
10
11
12
13
14
4-7
4-8
5-8
5-9
6-7
6-8
6-9
7-9
0.663
0.533
0.504
0.582
0.522
0.510
0.578
0.577
0.321
0.378
0.379
0.348
0.382
0.376
0.349
0.350
0.895
0.894
0.815
0.798
0.943
0.812
0.797
0.791
0.179
0.180
0.231
0.242
0.132
0.233
0.242
0.246
53
83
6
6
5
5
6
5
5
5
0.711
N
N
N
N
N
N
N
401
83
401
401
83
83
Table 4: Experimental and predicted activities (pIC50) with residual values for 35 compounds
Compd
Exp
Pred
Res
Comp
Exp
Pred
Res
Trainin
Ia
Ic
Ie
Ig
Ih
Ii
Ik
Il
Im
Io
Ip
Ir
It
Iv
Iw
Ix
Iy
IIa
IIb
IId
IIe
-1.583
-1.977
-1.180
-1.755
-0.735
-0.443
-0.393
-1.718
-1.986
-1.261
-1.674
-0.829
-0.383
-0.441
0.135
0.009
0.081
-0.08
0.095
-0.06
0.048
-0.549
-1.783
-1.437
-1.319
-1.480
-1.553
-1.375
-1.610
-1.311
-1.148
-1.436
-1.451
-1.605
-0.721
-0.549
-1.436
-1.324
-0.342
-0.508
-1.700
-1.403
-1.404
-1.357
-1.467
-1.268
-1.410
-1.307
-1.376
-1.232
-1.494
-1.381
-0.982
-0.928
-1.232
-1.491
-0.196
-0.041
-0.083
-0.034
0.085
-0.123
-0.086
-0.107
-0.200
-0.004
0.229
-0.204
0.043
-0.224
0.261
0.379
-0.204
0.167
-0.146
IIf
IIIb
IIIc
IVa
Test
Ib
Id
If
Ij
In
Iq
Is
Iu
-1.479
-1.792
-1.392
-1.461
-0.761
-0.828
-1.504
-1.430
-1.332
-0.335
-1.800
-1.789
-1.543
-1.365
-1.194
-1.114
-1.480
-1.097
-1.201
0.116
0.321
-0.00
0.151
-0.09
0.433
0.286
-0.02
-0.33
-0.13
-0.45
IIc
IIIa
Figure 4: Plot of experimental values versus predicted values of pIC50 of the training and test set
12

ACCEPTED MANUSCRIPT
An important role of HQSAR model, besides predicting the activities of untested molecules,
provides hints on what molecular fragments are directly related to biological activity[46, 47]. The
HQSAR analysis results can be graphically displayed in the form of contribution maps, in which
the color coding of each atom reflects its contribution to the activity of the whole molecule. The
colors (red, red orange and orange) at end of the spectrum represent the negative contribution (NC)
to the activity, while ones (yellow, green blue and green) of the other end denote positive
contribution (PC) to the activity. The white in the middle domain signifies intermediate
contribution (IC) to the activity.
The individual atomic contribution maps were shown in figure 5. Because our data set consisted
mainly of amino alcohol and its derivatives, Figure 5 describes three representative compounds Ia,
Id and Iw, in which compound Ia contains amino alcohol fragment, Id and Iw are derivatives of
compound Ia with the worst and best inhibitory activity against JEKO-1 cell, respectively. Except
the C2 colored the green color reflecting its positive contribution to the activity, A and B rings of
compound Ia were wholly colored red and orange indicating its negative contribution to the
activity. After introducing the phenylthiocarbamide subunit (Iw), the C ring was largely covered
by red color, whereas A, B and benzene rings were colored green and yellow as favorable
contribution, which suggests phenylthiocarbamide fragment is the pharmacologically important
group. When amino group was acylated with acetic anhydride (Id), the green color on A ring
completely disappeared, and the newly introduced acetyl amino group was full of the red and
orange colors as unfavorable contribution. More important, regions with intermediate or poor
contributions in molecule can be identified as potential targets for molecular modification and
further SAR studies.
Figure 5: The HQSAR contribution maps of compounds Ia, Id and Iw
3. Conclusions
In summary, a series of novel isosteviol derivatives containing amino alcohol and thiourea
13

ACCEPTED MANUSCRIPT
moieties have been rationally designed and synthesized with high yields, and their cytotoxic
activities against three human cancer cell lines: HCT-116, HGC-27 and JEKO-1 were evaluated.
The pharmacological assay results indicated that all of these compounds showed better
bioactivities than their corresponding precursor. Especially, compound Iw displayed the most
potent anticancer activities with IC50 values of 1.450ꢀM, 2.355ꢀM and 3.539ꢀM against HCT-116,
HGC-27, JEKO-1, respectively. HQSAR analyses were conducted on newly synthesized isosteviol
derivatives to establish the HQSAR model between cytotoxic activities against JEKO-1 cell and
molecular structures. The results demonstrated that the optimal HQSAR model was generated
using atoms (A), bonds (B), hydrogens (H), chirality (Ch) as fragment distinctions and fragment
size of 4-7 showing q2 = 0.663, r2 = 0.895, SEE = 0.179. This model was validated by external test
set of ten compounds to give the satisfactory predictive r2
value of 0.711. Meanwhile,
pred
contribution maps derived from the optimal model provided significant information of atom to the
overall molecule activity as potential targets for molecular modification and further SAR studies.
4. Experimental protocols
4.1. General procedures
All chemicals were used as received unless otherwise noted. Reagent grade solvents were
redistilled prior to use. Chromatography was performed on silica gel (100-200 mesh). All melting
points were measured with a Beijing Keyi XT5 apparatus and the temperature was not corrected.
1
The IR spectra were recorded as KBr pellets on a Thermo Nicolet (IR200) spectrometer. H and
13C NMR spectra were collected with a Bruker DPX-400 spectrometer at 400 and 100 MHz,
respectively, with TMS as internal standard. High-resolution mass spectra (HRMS) were obtained
with a Waters Micromass Q-Tof Micro™ instrument by the ESI technique. X-ray analysis was
taken on a Rigaku Saturn 724 CCD diffractomer (Mo-Kα, λ = 0.71073 Å).
4.2. The preparation of isosteviol derivatives
4.2.1. Ent-16-oxobeyeran-19-oic acid (1). Ent-16-oxobeyeran-19-oic acid was synthesized by
hydrolysis of stevioside with 10% sulfuric acid[48]. Yield 93%; Mp 228.4–230.8oC; IR (KBr):
3448, 2955, 2924, 2848, 1736, 1692, 1453, 1404, 1373, 1320, 1268, 1214, 1180, 1150, 1108, 973,
950, 796, 589cm-1; 1H NMR (400MHz, CDCl3, ppm): δ 2.64 (dd, J = 18.64, 3.76Hz, 1H), 2.17 (d,
J = 13.40Hz, 1H), 1.90–1.37 (m, 13H), 1.25 (s, 3H), 1.22–1.14 (m, 3H), 1.06–1.03 (m, 1H), 0.98
(s, 3H), 0.95–0.88 (m, 1H), 0.78 (s, 3H); 13C NMR (100 MHz, CDCl3, ppm): δ 224.22, 184.33,
14

ACCEPTED MANUSCRIPT
56.79, 57.62, 56.09, 54.83, 49.73, 49.59, 44.38, 42.02, 40.42, 40.42, 38.99, 38.34, 32.80, 29.69,
22.32, 20.57, 19.55, 14.04; HRMS (ESI, m/z) calcd for C20H30O3Na [M+Na]+ 341.2093. Found:
341.2085.
4.2.2. Ethyl ent-16-oxobeyeran-19-oate (2). The ethyl ent-16-oxobeyeran-19-oate 2 was obtained
by treating isosteviol with CH3CH2Br and KOH in DMSO at room temperature in good yield
according to the literature method[48]. Yield 98%; Mp 124.7–127.4oC; IR (KBr): 2958, 2927,
2843, 1727, 1465, 1445, 1379, 1320, 1257, 1226, 1151, 1128, 1094, 1055, 1023, 974, 926, 872,
1
773, 587, 505cm-1; H NMR (400MHz, CDCl3 , ppm): δ 4.10 (q, J = 7.12Hz, 2H), 2.96 (dd, J =
18.46, 3.14Hz, 1H), 2.17 (d, J = 13.32Hz, 1H), 1.99 (d, J = 18.86Hz, 1H), 1.88–1.58 (m, 8H),
1.47–1.33 (m, 6H), 1.26 (t, J = 7.12Hz, 3H), 1.22–1.19 (m, 2H), 1.18 (s, 3H), 1.07 (s, 3H),
1.05–0.86 (m, 1H), 0.77 (s, 3H); HRMS (ESI, m/z) calcd for C22H34O3Na [M+Na]+ 369.2406.
Found: 369.2400.
4.2.3. Ethyl ent-15α-hydroxymethyl-16β-hydroxybeyeran-19-oate (3). To a solution of Na (0.069
g, 3 mmol) in ethanol (20 mL) in an ice bath, compound 2 (0.346 g, 1mmol) and 37%
formaldehyde aqueous solution (2 mL) were added. After stirring for 8 h at 55 oC, the mixture was
concentrated under vacuum, and extracted with CHCl3 and H2O, at last the organic layer was
washed with saturated NaCl aqueous solution, dried with Na2SO4 and concentrated under vacuum
to give white powder 3. Yield 90%; Mp 180.5–181.8oC; IR (KBr): 3398, 2945, 2844, 1720, 1453,
1375, 1323, 1263, 1232, 1168, 1126, 1098, 1054, 922, 851, 821, 773, 611, 572, 532cm-1; 1H NMR
(400 MHz, CDCl3, ppm): δ 4.09 (q, J = 7.10Hz, 2H), 3.98 (dd, J = 9.72, 5.00Hz, 1H), 3.63 (d, J =
4.76Hz,1H), 3.50 (t, J = 10.26Hz, 1H), 2.16 (d, J = 11.84Hz, 1H), 2.05–2.01 (m, 1H), 1.83–1.56
(m, 9H), 1.42–1.37 (m, 2H), 1.25 (t, J = 7.14Hz, 3H), 1.22–1.19 (m, 1H), 1.15 (s, 3H), 1.07–0.95
(m, 4H), 0.93 (s, 3H), 0.87–0.86 (m, 1H), 0.77 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ
177.45, 86.76, 64.98, 60.00, 57.56, 57.04, 54.21, 50.29, 43.61, 42.45, 40.86, 39.60, 38.19, 37.90,
34.80, 33.08, 28.97, 25.01, 22.19, 19.51, 18.89, 14.13, 13.22; HRMS (ESI, m/z) calcd for
C23H38O4Na [M+Na]+ 401.2668. Found: 401.2664.
4.2.4. Ethyl ent-15α-hydroxymethyl-16-oxobeyeran-19-oate (4). Compound 3 (0.378 g, 1 mmol)
was dissolved in CH2Cl2 (20 mL) and then TCC/Silica gel (0.295 g) was added. After stirring for
0.5 h at room temperature, the reaction mixture was filtered and washed with CH2Cl2. At last the
combined organic phases were washed with saturated NaCl aqueous solution, dried with Na2SO4
15

ACCEPTED MANUSCRIPT
and the filtrate was concentrated. The residue was purified by column chromatography on silica
(petroleum ether/ethyl acetate=5:1, v/v) to give product 4. Yield 82%; Mp 155.4–156.8°C; IR
(KBr): 3533, 2930, 2884, 1721, 1546, 1459, 1408, 1379, 1327, 1297, 1262, 1222, 1149, 1122,
1
1097, 1062, 1029, 966, 944, 925, 869, 831, 774, 521cm-1; H NMR (400 MHz, CDCl3, ppm): δ
4.14–4.04 (m, 2H), 3.91 (dd, J = 10.61, 5.43Hz, 1H), 3.70 (t, J = 9.64Hz, 1H), 2.51–2.47 (m, 2H),
2.17 (d, J = 13.31Hz, 1H), 1.91–1.61 (m, 8H), 1.45–1.29 (m, 3H), 1.25 (t, J = 7.11Hz, 3H),
1.21–1.20 (m, 1H), 1.19 (s, 3H), 1.17–1.14 (m, 2H), 1.02 (dd, J = 13.46, 4.17Hz, 1H), 0.98 (s, 3H),
0.90 (td, J = 13.30, 4.18Hz, 1H), 0.74 (s, 3H); 13C NMR (100 MHz, CDCl3, ppm): δ 226.20,
177.20, 60.40, 60.12, 57.11, 56.75, 52.92, 52.41, 48.50, 43.62, 40.50, 39.68, 38.28, 37.84, 36.98,
35.21, 28.95, 21.62, 19.81, 19.59, 18.86, 14.13, 13.38; HRMS (ESI, m/z) calcd for C23H37O4
[M+H]+ 377.2692. Found: 377.2688.
4.2.5. Ethyl ent-15α-hydroxymethyl-16-hydroximinobeyeran-19-oate (5). A mixture of
compound 4 (0.376 g, 1 mmol) and hydroxylamine hydrochloride (0.103 g, 1.5mmol) in C2H5OH
was stirred in presence of NaHCO3 under refluxing condition for 3 h, then the reaction mixture
was concentrated under vacuum, and extracted with CH2Cl2 and H2O. At last the organic layer was
washed with saturated NaCl aqueous solution, dried with Na2SO4 and concentrated under vacuum
to give white powder 5. Yield 98%; Mp 69.7–71.2°C; IR (KBr): 3427, 2950, 2849, 1721, 1452,
1
1382, 1321, 1235, 1180, 1152, 1098, 1049, 1027, 930cm-1; H NMR (400 MHz, CDCl3, ppm): δ
4.13–4.07 (m, 3H), 3.70 (dd, J = 10.72, 9.29Hz, 1H), 3.36 (t, J = 4.80Hz, 1H), 2.15 (d, J =
11.10Hz, 1H), 1.88–1.62 (m, 8H), 1.43–1.40 (m, 2H), 1.26 (t, J = 7.12Hz, 3H), 1.23–1.19 (m, 1H),
1.17 (s, 3H), 1.14–1.11 (m, 1H), 1.09 (s, 3H), 1.07–0.86 (m, 5H), 0.80 (s, 3H); 13C NMR (100
MHz, CDCl3, ppm): δ 177.33, 172.12, 60.88, 60.04, 57.38, 56.92, 54.34, 46.99, 43.61, 43.26,
43.11, 40.81, 39.67, 38.37, 38.11, 34.97, 28.86, 22.51, 21.36, 19.48, 18.88, 14.09, 13.46; HRMS
(ESI, m/z) calcd for C23H38NO5 [M+H]+ 392.2801. Found: 392.2803.
4.2.6. Ethyl ent-15α-hydroxymethyl-16β-aminobeyeran-19-oate (Ia). Compound 5 (0.361 g, 1
mmol) was dissolved with stirring in methanol (20 mL) and cooled to 0 oC in an ice bath. NaBH4
(0.190 g, 5 mmol) was added carefully, followed by catalytic amount of MoO3 (0.219 g, 1.5
mmol). The reaction mixture was stirred for 10 h at room temperature and then filtered. The
resultant mixture was concentrated under reduced pressure. The residue was treated with aqueous
KOH (20%) and extracted several times with dichloromethane. The combined organic layer was
16

ACCEPTED MANUSCRIPT
washed with saturated NaCl aqueous solution, dried with Na2SO4 and the filtrate was concentrated.
The residue was purified by column chromatography on silica (dichloromethane/methanol=10:1,
v/v) to give product Ia. Yield 85%; Mp 84.2–85.5°C; IR (KBr): 3356, 3328, 3262, 3165, 2940,
2845, 1722, 1592, 1465, 1378, 1326, 1264, 1232, 1178, 1149, 1094, 1054, 1023, 971, 920, 891,
1
851, 819, 772, 726, 567cm-1; H NMR (400MHz, CDCl3, ppm): δ 4.08 (q, J = 7.04Hz, 2H), 3.97
(dd, J = 9.60, 4.42Hz, 1H), 3.71 (q, J = 7.00Hz, 1H), 3.54 (t, J = 10.16Hz, 1H), 2.69 (d, J =
5.16Hz, 1H), 2.41 (s, 3H), 2.16 (d, J = 13.20Hz, 1H), 1.88–1.77 (m, 3H), 1.72–1.60 (m, 5H),
1.43–1.38 (m, 3H), 1.26 (t, J = 7.46Hz, 3H), 1.22–1.20 (m, 1H), 1.15 (s, 3H), 1.10–0.93 (m, 5H),
0.87 (s, 3H), 0.77 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 177.42, 68.62, 65.80, 59.99, 57.50,
57.02, 55.69, 49.28, 43.61, 42.08, 40.42, 39.62, 38.16, 37.97, 34.59, 32.56, 28.90, 24.65, 22.27,
19.53, 18.85, 14.13, 13.12; HRMS (ESI, m/z) calcd for C23H40NO3 [M+H]+ 378.3008. Found:
378.3006.
4.2.7. Ethyl ent-16-hydroximinobeyeran-19-oate (6). The compound 6 was obtained by the
literature method[38]. Yield 95%; Mp 42.2–44.1°C; IR (KBr): 3297, 2940, 2846, 1719, 1321,
1450, 1379, 1297, 1232, 1178, 1152, 1097, 1028, 968, 929, 850, 776, 722, 575cm-1; 1H NMR (400
MHz, CDCl3, ppm): δ 4.08 (m, 2H), 2.97 (d, J = 18.62Hz, 1H), 2.17 (d, J = 13.32Hz, 1H), 2.00 (d,
J = 18.64Hz, 1H), 1.89–1.57 (m, 7H), 1.48–1.39 (m, 4H), 1.26 (t, J = 7.20Hz, 3H), 1.23-1.20 (m,
2H), 1.18 (s, 3H), 1.10 (s, 3H), 1.09–0.84 (m, 4H), 0.77 (s, 3H); 13C NMR (100 MHz, CDCl3,
ppm): δ 177.44, 170.49, 59.99, 57.04, 56.20, 54.81, 43.78, 43.64, 40.84, 40.58, 39.89, 39.38,
38.01, 37.97, 36.75, 28.86, 22.09, 21.64, 20.35, 18.88, 14.11, 13.29; HRMS (ESI, m/z) calcd for
C22H35NO3Na [M+Na]+ 384.2515. Found: 384.2512.
4.2.8. Ethyl ent-16β-aminobeyeran-19-oate (IIa). The compound IIa was obtained by the
literature method[38]. Yield 78%; Mp 186.3–187.8; IR (KBr): 3398, 3327, 2965, 2938, 2837,
1717, 1597, 1461, 1372, 1319, 1348, 1265, 1228, 1175, 1145, 1125, 1095, 1042, 1023, 970, 924,
1
853, 810, 773, 736, 519cm-1; H NMR (400MHz, CDCl3, ppm): δ 4.08 (q, J = 7.13Hz, 2H), 2.92
(dd, J = 11.32, 6.04Hz, 1H), 2.15 (d, J = 13.28Hz, 1H), 2.10 (s, 2H), 1.82–1.51 (m, 10H),
1.42–1.31 (m, 4H), 1.24 (t, J = 7.14Hz, 1H), 1.15 (s, 3H), 1.04–0.93 (m, 5H), 0.86 (s, 3H), 0.73 (s,
3H); 13C NMR (100MHz, CDCl3, ppm): δ 177.56, 60.95, 59.89, 57.17, 56.56, 55.77, 43.69, 42.54,
41.96, 41.69, 41.52, 39.97, 38.07, 38.03, 33.36, 28.96, 24.87, 21.77, 20.53, 18.91, 14.09, 13.36;
HRMS (ESI, m/z) calcd for C22H38NO2 [M+H]+ 348.2903. Found: 348.2904.
17

ACCEPTED MANUSCRIPT
4.2.9. Ethyl ent-8β-vinyl-13β-(hydroximino) methylpimara-19-oate (7). The compound 7 was
obtained by the literature method[39]. Yield 96%; Mp 148.5–149.6 °C; IR (KBr): 3439, 3070,
2954, 2921, 2855, 1699, 1627, 1447, 1377, 1239, 1190, 1153, 1025, 943, 628 cm-1; 1H NMR (400
MHz, CDCl3, ppm): δ 7.13 (s, 1H), 6.10 (dd, J = 17.76, 11.12Hz, 1H), 5.05 (d, J = 11.04Hz, 1H),
4.98 (d, J = 17.80Hz, 1H), 4.08–3.99 (m, 2H), 2.24 (dd, J = 13.37, 2.40Hz, 1H), 2.15–2.10 (m,
2H), 1.89–1.40 (m, 7H), 1.26 (d, J = 13.10Hz, 1H), 1.20 (t, J = 7.12Hz, 3H), 1.14 (s, 3H),
1.12–0.97 (m, 5H), 0.94 (s, 3H), 0.93–0.83 (m, 2H), 0.61 (s, 3H); 13C NMR (100 MHz, CDCl3,
ppm): δ 177.42, 159.37, 143.57, 111.27, 59.88, 58.46, 58.23, 57.78, 43.74, 40.90, 39.82, 39.65,
38.15, 38.06, 36.72, 35.40, 29.89, 28.71, 19.87, 19.09, 17.16, 13.98, 13.38; HRMS (ESI, m/z)
calcd for C23H38NO3 [M+H]+ 376.2852. Found: 376.2848.
4.2.10. Ethyl ent-8β-vinyl-13β-aminomethylpimara-19-oate (IIIa). The procedure for preparing
IIIa was exactly the same as for preparing Ia from compound 5. Yield 87%; Mp 68.7–69.3°C; IR
(KBr): 3391, 3076, 2939, 2844, 1718, 1625, 1460, 1378, 1321, 1262, 1229, 1153, 1114, 1093,
1015, 979, 900, 809, 766, 706cm-1; 1H NMR (400 MHz, CDCl3, ppm): δ 6.39 (dd, J = 18.04,
11.32Hz, 1H), 5.09–5.03 (m, 2H), 4.12–3.99 (m, 2H), 2.78–2.73 (m, 1H), 2.27–2.08 (m, 4H),
1.88–1.40 (m, 10H), 1.21 (t, J = 7.12Hz, 3H), 1.14 (s, 3H), 1.10 (dd, J = 12.56, 2.52Hz, 1H),
0.97–0.83 (m, 6H), 0.77 (s, 3H), 0.65 (s, 3H); 13C NMR (100 MHz, CDCl3, ppm): δ 177.39,
143.99, 111.02, 59.85, 58.89, 57.85, 53.48, 48.99, 43.76, 41.61, 40.76, 39.63, 38.58, 38.18, 38.11,
29.29, 28.72, 20.01, 19.10, 17.08, 14.01, 13.64; HRMS (ESI, m/z) calcd for C23H40NO2 [M+H]+
362.3059. Found: 362.3041.
4.2.11. Ethyl ent-15α,16α-isoxazolidinylbeyeran-19-oate (8). To a solution of compound 7 (0.375
.
g, 1 mmol) in toluene (10 mL), 48% BF3 OEt2 (48% solution in Et2O, 0.05 mL, 0.17 mmol) was
added dropwise and the mixture was heated under a nitrogen atmosphere at 110oC. After
completion of the reaction monitored by TLC, water (10 mL) was added to the mixture, which
was next neutralized with NaHCO3, and the organic phase was separated and dried over Na2SO4.
After evaporation in vacuo, the crude product was purified by column chromatography (petroleum
ether/ethyl acetate=3:1, v/v) to give isoxazolidine derivative 8. Yield 95%; Mp 136.5–138.1°CꢁIR
(KBr): 3206, 2978, 2943, 2872, 2855, 1704, 1467, 1382, 1326, 1267, 1238, 1178, 1149, 1113,
1052, 1029, 976, 859, 790, 732, 621cm-1; 1H NMR (400 MHz, CDCl3, ppm): δ 4.13–4.06 (m, 2H),
3.81 (s, 1H), 3.72 (s, 1H), 3.30 (d, J = 6.78Hz, 1H), 3.01–2.96 (m, 1H), 2.16 (d, J = 13.27Hz, 1H),
18

ACCEPTED MANUSCRIPT
1.83–1.50 (m, 8H), 1.43–1.26 (m, 4H), 1.25 (t, J = 7.12Hz, 3H), 1.17 (s, 3H), 1.15–0.97 (m, 3H),
0.95 (s, 3H), 0.93–0.80 (m, 2H), 0.79 (s, 3H); 13C NMR (100 MHz, CDCl3, ppm): δ 177.39, 73.05,
72.79, 59.99, 57.61, 56.32, 52.05, 51.15, 44.97, 43.63, 40.30, 39.92, 39.20, 38.27, 37.80, 35.53,
28.97, 21.72, 21.51, 19.28, 18.95, 14.10, 13.77; HRMS (ESI, m/z) calcd for C23H38NO3 [M+H]+
376.2852. Found: 376.2835.
4.2.12. Ethyl ent-15α-hydroxymethyl-16α-aminobeyeran-19-oate (IVa). Compound 8 (0.375 g, 1
mmol) was dissolved in methanol followed by addition of Raney Ni in catalytic hydrogenation
flask. The reaction proceeded 8h under 1MPa hydrogen at 60oC. Then the reaction mixture was
filtered, and the filtrate was concentrated. The residue was purified by column chromatography on
silica (dichloromethane/methanol=10:1, v/v) to give product IVa. Yield 83%; Mp 56.5–57.2°C; IR
(KBr): 3361, 3285, 3146, 2948, 2862, 1722, 1572, 1457, 1378, 1325, 1232, 1179, 1151, 1097,
1
1070, 1024, 948, 850cm-1; H NMR (400MHz, CDCl3, ppm): δ 4.13–4.03 (m, 2H), 3.76 (dd, J =
11.24, 4.96Hz, 1H), 3.67 (t, J = 11.20Hz, 1H), 3.55 (s, 2H), 3.06 (d, J = 8.32Hz, 1H), 2.51–2.45
(m, 1H), 2.15 (d, J = 13.24Hz, 1H), 1.88–1.58 (m, 7H), 1.46–1.28 (m, 4H), 1.25 (t, J = 7.14Hz,
3H), 1.17–1.16 (m, 1H), 1.15 (s, 3H), 1.11–0.96 (m, 3H), 0.93 (s, 3H), 0.88–0.82 (m, 3H), 0.77 (s,
3H); 13C NMR (100MHz, CDCl3, ppm): δ 177.40, 61.46, 60.61, 60.01, 57.30, 57.09, 53.11, 44.49,
44.16, 43.61, 41.41, 39.80, 39.14, 38.12, 37.99, 34.08, 28.87, 22.00, 19.24, 18.90, 14.15, 13.39;
HRMS (ESI, m/z) calcd for C23H40NO3 [M+H]+ 378.3008. Found: 378.3006.
Single crystals of compound IVa suitable for X-ray analysis were obtained by recrystallization
from MeOH at room temperature. Crystal data for IVa: empirical formula C23H39NO3: formula
weight: 377.55, temperature: 293(2) K, wavelength: 0.71073 A, crystal system: orthorhombic,
space group: P2(1)2(1)2(1), unit cell dimensions: a = 7.3838(15) Å, b = 10.875(2)Å, c = 27.060(5)
Å, α = 90 deg, β = 90 deg, γ = 90 deg, volume: 2173.0(8) Å3, Z = 4, calculated density: 1.154
mg/m3, absorption coefficient: 0.075 mm-1, F(000): 832, crystal size: 0.25 x 0.21 x 0.19 mm, theta
range for data collection: 2.02 to 27.88 deg, limiting indices: -9<=h<=9, -13<=k<=14,
-32<=l<=35, reflections collected / unique: 15926 / 5132 [R(int) = 0.0290], completeness to theta
= 27.88: 99.4 %, max. and min. transmission: 0.9859 and 0.9816, refinement method: full-matrix
least-squares on F2, data / restraints / parameters: 5132 / 0 / 257, goodness-of-fit on F2: 1.151, final
R indices [I>2sigma(I)]: R1 = 0.0583, wR2 = 0.1342, R indices (all data): R1 = 0.0672, wR2 =
0.1396, absolute structure parameter: 0.5(14), largest diff. peak and hole: 0.196 and -0.232 e.A-3.
19

ACCEPTED MANUSCRIPT
The data has been deposited with the Cambridge Crystallographic Date Centre as supplementary
publication number CCDC 1419387.
4.2.13. General procedure for synthesis of compounds Ib-IIIb. In a three-neck round-bottomed
flask with thermometer, compound Ia-IIIa (1 mmol), CS2 (0.750 g, 10 mmol) and Et3N (0.101 g,
1 mmol) were dissolved in ethanol (10 mL). The reaction mixture was stirred during 1h at room
temperature, and then cooled with an ice bath. Boc2O (0.218 g, 1 mmol), dissolved in ethanol (1
mL), was added, followed by the immediate addition of catalytic amount of DMAP in ethanol (1
mL). The reaction mixture was kept in the ice bath for 5 min and was then allowed to reach room
temperature. After evolution of gas from the reaction mixture had ceased, the reaction mixture was
evaporated under reduced pressure and extracted with CH2Cl2 and H2O. The organic layer was
washed with saturated NaCl aqueous solution, dried with Na2SO4 and concentrated under vacuum.
The residue was purified by column chromatography on silica (petroleum ether/ethyl acetate=8:1,
v/v) to give the corresponding products Ib-IIIb.
4.2.13.1. Ethyl ent-15α-hydroxymethyl-16β-isothiocyanobeyeran-19-oate (Ib). Yield 42%; Mp
82.1–83.7°C; IR (KBr): 3442, 2948, 2850, 2097, 1720, 1696, 1460, 1378, 1325, 1233, 1179, 1152,
1
1097, 1056, 1026, 740, 536cm-1; H NMR (400MHz, CDCl3, ppm): δ 4.09 (q, J = 7.12Hz, 2H),
3.92 (dd, J = 10.44, 4.84Hz, 1H), 3.63 (d, J = 5.36Hz, 1H), 3.55 (t, J = 9.36Hz, 1H), 2.30–2.26 (m,
1H), 2.16 (d, J = 13.32Hz, 1H), 1.88–1.65 (m, 8H), 1.55–1.39 (m, 4H), 1.32 (dd, J = 13.12,
6.20Hz, 1H), 1.25 (t, J = 7.12Hz, 3H), 1.15 (s, 3H), 1.11–1.04 (m, 2H), 1.01 (s, 3H), 0.99–0.94 (m,
13
2H), 0.86 (td, J = 13.00, 4.04Hz, 1H), 0.77 (s, 3H); C NMR (100MHz, CDCl3, ppm): δ 177.28,
129.93, 68.95, 63.20, 60.07, 57.28, 56.97, 53.66, 49.51, 43.76, 43.62, 42.70, 39.54, 38.21, 37.84,
34.40, 34.37, 28.93, 24.90, 21.96, 19.27, 18.85, 14.11, 13.24; HRMS (ESI, m/z) calcd for
C24H37NO3SNa [M+Na]+ 442.2392. Found: 442.2391.
4.2.13.2. Ethyl ent-16β-isothiocyanobeyeran-19-oate (IIb). Yield 89%; Mp 85.6–87.1; IR (KBr):
2989, 2952, 2843, 2113, 1720, 1453, 1375, 1326, 1301, 1243, 1181, 1151, 1095, 1028, 974, 926,
1
886, 853, 796, 616cm-1; H NMR (400MHz, CDCl3, ppm): δ 4.20–4.12 (m, 2H), 3.78 (dd, J =
11.48, 5.00Hz, 1H), 2.23 (d, J = 13.40Hz, 1H), 2.17–2.12 (m, 1H), 1.91–1.38 (m, 13H), 1.33 (t, J
= 7.12Hz, 3H), 1.23 (s, 3H), 1.20–1.10 (m, 3H), 1.07 (s, 3H), 1.05–1.01 (m, 1H), 0.94 (td, J =
12.86, 3.80Hz, 1H), 0.82 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 177.32, 64.34, 59.98, 56.93,
55.39, 54.45, 44.00, 43.62, 43.07, 41.03, 40.81, 39.80, 38.00, 37.89, 34.73, 28.93, 24.83, 21.57,
20

ACCEPTED MANUSCRIPT
20.04, 18.84, 14.09, 13.23; HRMS (ESI, m/z) calcd for C23H36NO2S [M+H]+ 390.2467. Found:
390.2468.
4.2.13.3. Ethyl ent-8β-vinyl-13β-isothiocyanomethylpimara-19-oate (IIIb). Yield 87%; Mp
82.5–84.3°C; IR (KBr): 3081, 2943, 2846, 2173, 2100, 1711, 1630, 1461, 1380, 1338, 1321, 1234,
1
1183, 1152, 1092, 1031, 908, 856, 775, 710, 638, 546cm-1; H NMR (400MHz, CDCl3, ppm): δ
6.25 (dd, J = 17.76, 11.08Hz, 1H), 5.15–5.10 (m, 2H), 4.06 (q, J = 6.97Hz, 2H), 3.46 (d, J =
14.28Hz, 1H), 3.31 (d, J = 14.32Hz, 1H), 2.15 (d, J = 13.32Hz, 1H), 2.08 (d, J = 12.96Hz, 1H),
1.84–1.55 (m, 8H), 1.45–1.41 (m, 1H), 1.33–1.26 (m, 1H), 1.22 (t, J = 7.12Hz, 3H), 1.14 (s, 3H),
1.11–0.96 (m, 4H), 0.93 (s, 3H), 0.86–0.73 (m, 2H), 0.68 (s, 3H); 13C NMR (100MHz, CDCl3,
ppm): δ 177.26, 142.39, 128.82, 113.30, 59.89,58.39, 57.69, 54.12, 53.46, 43.72, 42.48, 40.50,
39.64, 38.10, 38.05, 36.89, 35.92, 29.54, 28.72, 19.93, 19.04, 17.36, 14.03, 13.70; HRMS (ESI,
m/z) calcd for C24H38NO2S [M+H]+ 404.2623. Found: 404.2626.
4.2.14. Ethyl ent-15α-acetoxymethyl-16β-acetamidobeyeran-19-oate (Ic). A mixture of
compound Ia (0.377 g, 1 mmol) and acetic anhydride (0.306 g, 3 mmol) in pyridine was stirred at
room temperature. After completion of the reaction monitored by TLC, the mixture was poured
into saturated NaCl aqueous solution and extracted with CH2Cl2. The extract was washed with
water, dried, filtered and evaporated. The residue was purified by column chromatography on
silica (petroleum ether/ethyl acetate = 6:1, v/v) to afford the pure product Ic. Yield 95%; Mp
186.4–187.8°C; IR (KBr): 3298, 3072, 2942, 2848, 1741, 1716, 1641, 1548, 1462, 1368, 1290,
1
1235, 1180, 1152, 1098, 1030, 976, 919, 704, 601, 560cm-1; H NMR (400MHz, CDCl3, ppm): δ
5.44 (d, J = 9.32Hz, 1H), 4.30 (dd, J = 10.80, 5.84Hz, 1H), 4.09 (q, J = 7.08Hz, 2H), 4.03 (t, J =
9.86Hz, 1H), 3.92 (dd, J = 9.84, 6.08Hz, 1H), 2.16 (d, J = 13.24Hz, 1H), 2.03 (s, 3H), 2.00 (s, 3H),
1.84–1.28 (m, 11H), 1.25 (t, J = 6.92Hz, 3H), 1.20–1.95 (m, 1H), 1.15 (s, 3H), 1.11–0.91 (m, 6H),
0.88 (s, 3H), 0.76 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 177.28, 171.24, 169.26, 66.00,
61.76, 60.02, 57.25, 56.90, 54.79, 45.98, 43.61, 42.92, 40.89, 39.66, 38.14, 37.92, 34.68, 33.49,
28.84, 24.73, 23.72, 21.97, 21.01, 19.96, 18.86, 14.12, 13.23; HRMS (ESI, m/z) calcd for
C27H43NO5Na [M+Na]+ 484.3039. Found: 484.3037.
Single crystals of compound Ic suitable for X-ray analysis were obtained by recrystallization from
MeOH at room temperature. Crystal data for Ic: C27H43NO5 (461.64), T = 293(2) K, crystal system:
monoclinic, space group: P2(1), unit cell dimensions: a = 10.461(2) Å, b = 7.6419(15) Å, c =
21

ACCEPTED MANUSCRIPT
17.075(3) Å, α =90 deg, β = 100.68(3) deg, γ = 90 deg, volume: 1341.4(5) Å3, Z = 2, Calculated
density: 1.183 mg/m3, absorption coefficient: 0.082 mm-1, F(000) = 520, crystal size: 0.22 x 0.20 x
0.19 mm, theta range for data collection: 1.98° to 25.49°, limiting indices: -9<=h<=12, -9<=k<=9,
-20<=l<=20, reflections collected / unique: 10156 / 4908 [R (int) = 0.0531], completeness to theta
= 25.49: 99.9 %, max. and min. transmission: 0.9845 and 0.9821, refinement method: full-matrix
least-squares on F2, data / restraints / parameters: 4908 / 1 / 313, goodness-of-fit on F2: 0.980, final
R indices [I>2sigma(I)]: R1 = 0.0905, wR2 = 0.2670, R indices (all data): R1 = 0.1248, wR2 =
0.3045, absolute structure parameter: -3(3), largest diff. peak and hole: 0.204 and -0.190 e.Å-3.
The data has been deposited with the Cambridge Crystallographic Date Centre as supplementary
publication number CCDC 1419390.
4.2.15. Ethyl ent-15α-hydroxymethyl-16β-acetamidobeyeran-19-oate (Id). A mixture of the
compound Ic (0.461 g, 1 mmol) and excessive K2CO3 in methanol was stirred at room
temperature for 2h, then the reaction mixture was filtered, and the filtrate was concentrated. The
residue was purified by column chromatography on silica (petroleum ether/ethyl acetate = 3:1, v/v)
to give product Id. Yield 93%; Mp 130.6–132.4°C; IR (KBr): 3415, 3327, 2928, 2854, 1716, 1634,
1524, 1448, 1378, 1313, 1279, 1179, 1233, 1151, 1057, 1023, 1096, 983, 773, 548cm-1; 1H NMR
(400MHz, CDCl3, ppm): δ 5.80 (d, J = 9.16Hz, 1H), 4.48 (d, J = 8.80Hz, 1H), 4.06 (q, J = 6.81Hz,
2H), 3.78 (s, 1H), 3.60 (dd, J = 9.20, 5.12Hz, 1H), 3.49 (t, J = 11.08Hz, 1H), 2.15 (d, J = 13.28Hz,
1H), 2.06 (s, 3H), 1.95–1.37 (m, 13H), 1.24 (t, J = 7.12Hz, 3H), 1.15 (s, 3H), 1.11–0.93 (m, 5H),
0.89 (s, 3H), 0.76 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 177.38, 171.72, 64.47, 64.16,
60.09, 57.39, 56.91, 54.77, 51.07, 43.57, 43.42, 40.41, 39.68, 38.17, 37.97, 34.61, 33.70, 28.79,
24.81, 23.63, 21.84, 19.69, 18.80, 14.16, 13.39; HRMS (ESI, m/z) calcd for C25H42NO4 [M+H]+
420.3114. Found: 420.3116.
4.2.16. General procedure for synthesis of compounds Ie-Iy. Compound Ia (0.377 g, 1 mmol)
was dissolved in dichloromethane (5 mL). Appropriately substituted phenyl isothiocyanate (1.1
mmol) was added. The reaction mixture was stirred at room temperature. After completion of the
reaction monitored by TLC, the solvent was removed under reduced pressure. The residue was
purified by silica gel chromatography to give the thiourea derivatives Ie-Iy.
4.2.16.1. Ethyl ent-15α-hydroxymethyl-16β-phenylcarbamothioylaminobeyeran-19-oate (Ie). Yield
52%; Mp 93.9–94.8°C; IR (KBr): 3418, 3382, 3321, 3058, 2944, 2849, 1718, 1599, 1534, 1498,
22

ACCEPTED MANUSCRIPT
1
1465, 1380, 1323, 1260, 1235, 1180, 1152, 1097, 1054, 1026, 981, 754, 717, 696, 611cm-1; H
NMR (400MHz, CDCl3 , ppm): δ 8.08 (s, 1H), 7.45–7.23 (m, 5H), 6.28 (d, J = 10.0Hz, 1H),
4.50–4.39 (m, 1H), 4.13–4.05 (m, 3H), 3.92–3.76 (m, 1H), 3.63 (t, J = 9.88Hz, 1H), 2.14 (d, J =
12.96Hz, 1H), 1.89–1.48 (m, 10H), 1.39–1.36 (m, 2H), 1.24 (t, J = 6.76Hz, 3H), 1.14 (s, 3H),
1.10–0.99 (m, 4H), 0.93 (s, 3H), 0.83–0.77 (m, 2H), 0.66 (s, 3H); 13C NMR (100MHz, CDCl3,
ppm): δ 180.02, 177.30, 135.67, 130.23, 130.23, 127.73, 125.50, 125.50, 69.22, 64.12, 60.08,
57.26, 56.88, 54.63, 49.67, 43.56, 43.05, 41.64, 39.52, 38.07, 37.90, 34.68, 34.02, 28.80, 24.52,
21.83, 19.31, 18.76, 14.14, 13.15; HRMS (ESI, m/z) calcd for C30H44N2O3SNa [M+Na]+ 535.2970.
Found: 535.2968.
4.2.16.2. Ethyl ent-15α-hydroxymethyl-16β-(2'-methylphenyl)carbamothioylaminobeyeran-19-oate
(If). Yield 66%; Mp 95.2–97.1°C; IR (KBr): 3418, 3379, 3316, 3056, 2945, 2849, 1718, 1534,
1
1496, 1464, 1379, 1327, 1234, 1180, 1152, 1097, 1027, 981, 855, 754, 725, 559cm-1; H NMR
(400MHz, CDCl3 , ppm): δ 7.64 (s, 1H), 7.35–7.28 (m, 3H), 7.24 (d, J = 6.84Hz, 1H), 5.87 (d, J =
9.96Hz, 1H), 4.39 (dd, J = 9.48, 5.08Hz, 1H), 4.15–4.00 (m, 3H), 3.89–3.76 (m, 1H), 3.61 (t, J =
10.66Hz, 1H), 2.31 (s, 3H), 2.14 (d, J = 13.00Hz, 1H), 1.81–1.63 (m, 6H), 1.56–1.35 (m, 5H),
1.25 (t, J = 7.06Hz, 3H), 1.20–1.19 (m, 1H), 1.14 (s, 3H), 1.07–0.94 (m, 5H), 0.89 (s, 3H),
0.81–0.75 (m, 1H), 0.63 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 180.35, 177.26, 136.28,
133.79, 131.83, 129.12, 127.82, 127.61, 68.96, 64.10, 60.05, 57.21, 56.91, 54.58, 50.08, 43.55,
43.06, 41.35, 39.51, 38.03, 37.90, 34.68, 33.91, 28.81, 24.43, 21.80, 19.14, 18.77, 17.76, 14.14,
13.00; HRMS (ESI, m/z) calcd for C31H46N2O3SNa [M+Na]+ 549.3127. Found: 549.3127.
4.2.16.3. Ethyl ent-15α-hydroxymethyl-16β-(3'-methylphenyl)carbamothioylaminobeyeran-19-oate
(Ig). Yield 64%; Mp 92.8–93.9°C; IR (KBr): 3421, 3383, 3295, 3049, 2945, 2850, 1718, 1603,
1533, 1493, 1464, 1378, 1326, 1296, 1231, 1179, 1151, 1095, 1054, 1026, 981, 865, 779, 697,
1
570cm-1; H NMR (400MHz, CDCl3, ppm): δ 8.02 (s, 1H), 7.32 (t, J =7.12Hz, 1H), 7.13 (d, J =
6.68Hz, 1H), 7.05–7.01 (m, 2H),, 6.37 (d, J = 9.88Hz, 1H), 4.52–4.39 (m, 1H), 4.12–4.05 (m, 3H),
3.92–3.77 (m, 1H), 3.63 (t, J = 10.18Hz, 1H), 2.36 (s, 3H), 2.14 (d, J = 13.2Hz, 1H), 1.81–1.49 (m,
9H), 1.39–1.30 (m, 2H), 1.24 (t, J = 7.04Hz, 3H), 1.14 (s, 3H), 1.10–0.99 (m, 4H), 0.93 (s, 3H),
0.85–0.78 (m, 2H), 0.67 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 179.84, 177.29, 140.43,
135.56, 130.03, 128.35, 125.77, 122.17, 69.20, 64.11, 60.07, 57.29, 56.88, 54.65, 49.74, 43.57,
43.07, 41.64, 39.56, 38.07, 37.91, 34.69, 34.02, 28.78, 24.51, 21.84, 21.31, 19.31, 18.77, 14.14,
23

ACCEPTED MANUSCRIPT
13.18; HRMS (ESI, m/z) calcd for C31H46N2O3SNa [M+Na]+ 549.3127. Found: 549.3125.
4.2.16.4. Ethyl ent-15α-hydroxymethyl-16β-(4'-methylphenyl)carbamothioylaminobeyeran-19-oate
(Ih). Yield 72%; Mp 99.5–101.3°C; IR (KBr): 3421, 3385, 3315, 3031, 2945, 2850, 1719, 1613,
1532, 1465, 1379, 1328, 1297, 1233, 1179, 1151, 1097, 1055, 1024, 981, 925, 816, 771, 721, 691,
560, 497cm-1; 1H NMR (400MHz, CDCl3, ppm): δ 7.84 (s, 1H), 7.24 (d, J = 7.68Hz, 2H), 7.12 (d,
J = 7.76Hz, 2H), 6.25 (d, J = 10.16Hz, 1H), 4.45 (dd, J = 8.88, 5.28Hz, 1H), 4.13–4.01 (m, 3H),
3.85 (t, J = 9.72Hz, 1H), 3.64 (t, J = 10.16Hz, 1H), 2.38 (s, 3H), 2.15 (d, J = 13.20Hz, 1H),
1.89–1.48 (m, 10H), 1.40–1.33 (m, 2H), 1.25 (t, J = 7.12Hz, 3H), 1.15 (s, 3H), 1.10–0.99 (m, 4H),
0.93 (s, 3H), 0.85–0.77 (m, 2H), 0.67 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 180.12, 177.30,
137.87, 132.86, 130.79, 130.79, 125.48, 125.48, 69.28, 64.13, 60.07, 57.31, 56.89, 54.65, 49.67,
43.57, 43.04, 41.61, 39.54, 38.07, 37.93, 34.70, 34.04, 28.79, 24.50, 21.84, 21.10, 19.33, 18.77,
14.14, 13.16; HRMS (ESI, m/z) calcd for C31H46N2O3SNa [M+Na]+ 549.3127. Found: 549.3126.
4.2.16.5. Ethyl ent-15α-hydroxymethyl-16β-(2'-fluorophenyl)carbamothioylaminobeyeran-19-oate
(Ii). Yield 88%; Mp 97.2–98.9°C; IR (KBr): 3421, 3385, 3317, 3063, 2945, 2849, 1718, 1624,
1600, 1535, 1499, 1462, 1330, 1262, 1232, 1180, 1151, 1096, 1055, 1026, 981, 853, 809, 753, 702,
558cm-1; 1H NMR (400MHz, CDCl3, ppm): δ 7.77 (s, 1H), 7.40–7.21 (m, 4H), 6.27 (d, J = 8.24Hz,
1H), 4.54–4.41 (m, 1H), 4.13–4.05 (m, 3H), 3.92–3.79 (m, 1H), 3.71–3.58 (m, 1H), 2.15 (d, J =
13.12Hz, 1H), 1.93–1.50 (m, 10H), 1.40–1.37 (m, 2H), 1.24 (t, J = 7.02Hz, 3H), 1.15 (s, 3H),
1.12–1.03 (m, 4H), 0.95 (s, 3H), 0.85–0.80 (m, 2H), 0.68 (s, 3H); 13C NMR (100MHz, CDCl3,
ppm): δ 180.42, 177.32, 156.28 (d, J = 252.89Hz), 129.13 127.38, 125.12, 123.76 (d, J = 14.48Hz),
117.09 (d, J = 20.01Hz), 69.22, 64.06, 60.09, 57.30, 56.91, 54.68, 49.59, 43.58, 43.12, 41.63,
39.55, 38.10, 37.90, 34.69, 33.86, 28.81, 24.39, 21.85, 19.40, 18.78, 14.13, 13.17; HRMS (ESI,
m/z) calcd for C30H43FN2O3SNa [M+Na]+ 553.2876. Found: 553.2875.
4.2.16.6. Ethyl ent-15α-hydroxymethyl-16β-(3'-fluorophenyl)carbamothioylaminobeyeran-19-oate
(Ij). Yield 88%; Mp 95.5–97.4°C; IR (KBr): 3420, 3385, 3306, 3085, 2945, 2850, 1718, 1612,
1573, 1531, 1491, 1463, 1381, 1327, 1259, 1232, 1179, 1150, 1097, 1054, 1025, 968, 860, 776,
1
714, 576cm-1; H NMR (400MHz, CDCl3, ppm): δ 8.30 (s, 1H), 7.40–6.99 (m, 4H), 6.43 (d, J =
9.04Hz, 1H), 4.53–4.41 (m, 1H), 4.09–4.07 (m, 3H), 3.93–3.79 (m, 1H), 3.64 (t, J = 9.80Hz, 1H),
2.15 (d, J = 18.00Hz, 1H), 1.94–1.49 (m, 11H), 1.40–1.37 (m, 2H), 1.24 (t, J = 6.52Hz, 3H), 1.15
(s, 3H), 1.12–1.04 (m, 3H), 0.95 (s, 3H), 0.86–0.79 (m, 2H), 0.69 (s, 3H); 13C NMR (100MHz,
24

ACCEPTED MANUSCRIPT
CDCl3, ppm): δ 179.92, 177.30, 163.30 (d, J = 249.86Hz), 137.54 (d, J = 9.75Hz), 131.41 (d, J =
8.38Hz), 120.29, 114.24 (d, J = 17.67Hz), 112.09 (d, J = 23.58Hz), 69.29, 64.08, 60.10, 57.27,
56.91, 54.65, 49.67, 43.59, 43.08, 41.72, 39.53, 38.11, 37.88, 34.68, 34.05, 28.82, 24.55, 21.85,
19.42, 18.78, 14.13, 13.16; HRMS (ESI, m/z) calcd for C30H43FN2O3SNa [M+Na]+ 553.2876.
Found: 553.2876.
4.2.16.7. Ethyl ent-15α-hydroxymethyl-16β-(4'-fluorophenyl)carbamothioylaminobeyeran-19-oate
(Ik). Yield 83%; Mp 108.8–110.5°C; IR (KBr): 3421, 3385, 3308, 3060, 2945, 2850, 1718, 1624,
1534, 1507, 1465, 1418, 1380, 1329, 1222, 1180, 1152, 1095, 1054, 1024, 981, 924, 833, 806, 725,
1
562, 507cm-1; H NMR (400MHz, CDCl3, ppm): δ 8.06 (s, 1H), 7.28–7.15 (m, 4H), 6.13 (d, J =
8.20Hz, 1H), 4.50–3.79 (m, 5H), 3.63 (t, J = 9.22Hz, 1H), 2.16 (d, J = 16.72Hz, 1H), 1.87–1.48
(m, 11H), 1.41–1.37 (m, 1H), 1.25 (t, J = 6.78Hz, 3H), 1.15 (s, 3H), 1.11–1.01 (m, 4H), 0.93 (s,
3H), 0.85–0.82 (m, 2H), 0.67 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 180.44, 177.28, 161.54
(d, J = 241.56Hz), 131.63, 127.99, 127.99, 117.15 (d, J = 22.10Hz), 117.15 (d, J = 22.10Hz),
69.22, 64.13, 60.08, 57.23, 56.90, 54.60, 49.69, 43.58, 43.05, 41.62, 39.51, 38.09, 37.89, 34.67,
34.01, 28.82, 24.51, 21.83, 19.38, 18.76, 14.13, 13.20; HRMS (ESI, m/z) calcd for
C30H43FN2O3SNa [M+Na]+ 553.2876. Found: 553.2875.
4.2.16.8. Ethyl ent-15α-hydroxymethyl-16β-(2'-chlorophenyl)carbamothioylaminobeyeran-19-oate
(Il). Yield 82%; Mp 93.4–94.8°C; IR (KBr): 3421, 3385, 3306, 3056, 2945, 2849, 1717, 1592,
1532, 1469, 1446, 1370, 1333, 1297, 1323, 1180, 1152, 1096, 1056, 1028, 981, 853, 824, 755, 732,
1
683, 563cm-1; H NMR (400MHz, CDCl3, ppm): δ 7.75 (s, 1H), 7.52–7.28 (m, 4H), 6.15 (d, J =
3.64Hz, 1H), 4.46–3.63 (m, 6H), 2.15 (d, J = 15.72Hz, 1H), 1.93–1.49 (m, 10H), 1.40–1.37 (m,
2H), 1.24 (t, J = 7.06Hz, 3H), 1.15 (s, 3H), 1.10–1.00 (m, 4H), 0.94 (s, 3H), 0.85–0.82 (m, 2H),
0.69 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 180.29, 177.29, 141.76, 133.10, 131.00, 129.04,
128.07, 127.74, 69.15, 64.06, 60.08, 57.28, 56.91, 54.64, 49.67, 43.57, 43.11, 41.56, 39.54, 38.09,
37.89, 34.68, 33.90, 28.82, 24.47, 21.84, 19.36, 18.78, 14.14, 13.17; HRMS (ESI, m/z) calcd for
C30H44ClN2O3S [M+H]+ 547.2761. Found: 547.2761.
4.2.16.9. Ethyl ent-15α-hydroxymethyl-16β-(3'-chlorophenyl)carbamothioylaminobeyeran-19-oate
(Im). Yield 93%; Mp 100.8–102.1°C; IR (KBr): 3418, 3385, 3290, 3060, 2945, 2850, 1718, 1618,
1595, 1530, 1475, 1378, 1326, 1232, 1180, 1152, 1095, 1054, 1025, 922, 865, 778, 701, 573cm-1;
1H NMR (400MHz, CDCl3, ppm): δ 8.27 (s, 1H), 7.36 (t, J = 7.22Hz, 1H), 7.28–7.26 (m, 2H),
25

ACCEPTED MANUSCRIPT
7.12 (d, J = 7.20Hz, 1H), 6.43 (d, J = 9.64Hz, 1H), 4.53–4.42 (m, 1H), 4.13–4.04 (m, 3H), 3.87 (t,
J = 8.40Hz, 1H), 3.64 (t, J = 9.36Hz, 1H), 2.16 (d, J = 16.48Hz, 1H), 1.94–1.50 (m, 10H),
1.41–1.31 (m, 2H), 1.25 (t, J = 6.42Hz, 3H), 1.16 (s, 3H), 1.12–1.04 (m, 4H), 0.96 (s, 3H),
0.86–0.79 (m, 2H), 0.70 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 179.85, 177.33, 137.30,
135.58, 131.10, 127.18, 124.71, 122.74, 69.21, 64.10, 60.11, 57.30, 56.90, 54.69, 49.68, 43.59,
43.10, 41.71, 39.54, 38.12, 37.90, 34.68, 34.02, 28.82, 24.55, 21.86, 19.44, 18.79, 14.13, 13.22;
HRMS (ESI, m/z) calcd for C30H43ClN2O3SNa [M+Na]+ 569.2581. Found: 569.2578.
4.2.16.10. Ethyl ent-15α-hydroxymethyl-16β-(4'-chlorophenyl)carbamothioylaminobeyeran-19
-oate (In). Yield 81%; Mp 101.6–103.1°C; IR (KBr): 3418, 3389, 3284, 3049, 2944, 2849, 1719,
1619, 1596, 1531, 1493, 1465, 1380, 1327, 1260, 1232, 1179, 1151, 1092, 1054, 1018, 981, 923,
1
827, 772, 723, 528cm-1; H NMR (400MHz, CDCl3, ppm): δ 8.32 (s, 1H), 7.39 (d, J = 6.76Hz,
2H), 7.17 (d, J = 7.36Hz, 2H), 6.29 (d, J = 9.72Hz, 1H), 4.52–4.39 (m, 1H), 4.10–4.07 (m, 3H),
3.92–3.79 (m, 1H), 3.63 (t, J = 8.66Hz, 1H), 2.15 (d, J = 13.32Hz, 1H), 1.91–1.48 (m, 10H),
1.41–1.30 (m, 2H), 1.24 (t, J = 6.38Hz, 3H), 1.15 (s, 3H), 1.10–0.99 (m, 4H), 0.93 (s, 3H),
0.86–0.78 (m, 2H), 0.68 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 180.00, 177.32, 134.55,
132.90, 130.20, 130.20, 126.38, 126.38, 69.21, 64.12, 60.11, 57.25, 56.88, 54.62, 49.63, 43.58,
43.04, 41.68, 39.50, 38.11, 37.88, 34.67, 34.03, 28.83, 24.55, 21.84, 19.46, 18.78, 14.13, 13.23;
HRMS (ESI, m/z) calcd for C30H43ClN2O3SNa [M+Na]+ 569.2581. Found: 569.2579.
4.2.16.11. Ethyl ent-15α-hydroxymethyl-16β-(2'-bromophenyl)carbamothioylaminobeyeran-19
-oate (Io). Yield 77%; Mp 96.6–98.3°C; IR (KBr): 3421, 3385, 3304, 3053, 2944, 2849, 1718,
1585, 1531, 1468, 1440, 1377, 1331, 1294, 1232, 1180, 1152, 1096, 1050, 1025, 981, 924, 853,
754, 729, 661, 561cm-1; 1H NMR (400MHz, CDCl3, ppm): δ 7.78(s, 1H), 7.71 (d, J =6.24Hz, 1H),
7.41–7.24 (m, 3H), 6.09 (d, J = 8.48Hz, 1H), 4.43–3.85 (m, 5H), 3.62 (t, J = 8.86Hz, 1H), 2.14 (d,
J = 13.28Hz, 1H), 1.93–1.48 (m, 10H), 1.39–1.36 (m, 2H), 1.24 (t, J = 7.08Hz, 3H), 1.14 (s, 3H),
1.11–0.98 (m, 4H), 0.92 (s, 3H), 0.84–0.80 (m, 2H), 0.67 (s, 3H); 13C NMR (100MHz, CDCl3,
ppm): δ 180.26, 177.29, 134.64, 134.17, 129.47, 128.76, 128.31, 121.67, 69.05, 64.05, 60.08,
57.27, 56.91, 54.64, 49.67, 43.56, 43.11, 41.51, 39.53, 38.09, 37.88, 34.68, 33.89, 28.83, 24.48,
21.84, 19.33, 18.78, 14.15, 13.18; HRMS (ESI, m/z) calcd for C30H44BrN2O3S [M+H]+ 591.2256.
Found: 591.2255.
4.2.16.12. Ethyl ent-15α-hydroxymethyl-16β-(3'-bromophenyl)carbamothioylaminobeyeran-19
26

ACCEPTED MANUSCRIPT
-oate (Ip). Yield 81%; Mp 108.4–109.8°C; IR (KBr): 3418, 3385, 3284, 3061, 2944, 2849, 1718,
1592, 1529, 1473, 1378, 1326, 1232, 1179, 1151, 1095, 1053, 1025, 914, 861, 775, 692, 571cm-1;
1H NMR (400MHz, CDCl3, ppm): δ 8.23 (s, 1H), 7.42–7.16 (m, 4H), 6.43 (d, J = 9.04Hz, 1H),
4.47–4.06 (m, 4H), 3.87 (t, J = 7.92Hz, 1H), 3.64 (t, J = 9.54Hz, 1H), 2.16 (d, J = 15.84Hz, 2H),
1.95–1.50 (m, 10H), 1.41–1.32 (m, 2H), 1.25 (t, J = 5.80Hz, 3H), 1.16 (s, 3H), 1.12–1.04 (m, 4H),
0.96 (s, 3H), 0.86–0.80 (m, 1H), 0.71 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 179.81, 177.33,
137.41, 131.39, 130.11, 127.56, 123.37, 123.22, 69.20, 64.09, 60.11, 57.31, 56.91, 54.70, 49.70,
43.59, 43.10, 41.72, 39.55, 38.14, 37.91, 34.69, 34.03, 28.82, 24.55, 21.86, 19.46, 18.81, 14.14,
13.28; HRMS (ESI, m/z) calcd for C30H43BrN2O3SNa [M+Na]+ 613.2075. Found: 613.2075.
4.2.16.13. Ethyl ent-15α-hydroxymethyl-16β-(4'-bromophenyl)carbamothioylaminobeyeran-19
-oate (Iq). Yield 72%; Mp 111.2–112.1°C; IR (KBr): 3418, 3381, 3286, 3049, 2945, 2850, 1718,
1619, 1592, 1531, 1490, 1465, 1379, 1327, 1260, 1232, 1179, 1151, 1097, 1070, 1009, 981, 923,
1
824, 772, 726, 497cm-1; H NMR (400MHz, CDCl3, ppm): δ 7.55 (d, J = 4.84Hz, 2H), 7.36 (s,
1H), 7.12 (d, J = 6.52Hz, 2H), 6.30 (d, J =9.76Hz, 1H), 4.47–3.63 (m, 6H), 2.16 (d, J = 14.16Hz,
1H), 1.91–1.49 (m, 10H), 1.42–1.32 (m, 2H), 1.25 (t, J = 6.52Hz, 3H), 1.16 (s, 3H), 1.12–1.00 (m,
13
4H), 0.95 (s, 3H), 0.87–0.80 (m, 2H), 0.69 (s, 3H); C NMR (100MHz, CDCl3, ppm): δ 179.97,
177.30, 135.04, 132.21, 132.21, 126.56, 126.56, 120.75, 69.27, 64.12, 60.11, 57.27, 56.91, 54.63,
49.62, 43.59, 43.04, 41.69, 39.52, 38.13, 37.88, 34.68, 34.06, 28.84, 24.55, 21.84, 19.48, 18.79,
14.14, 13.24; HRMS (ESI, m/z) calcd for C30H44BrN2O3S [M+H]+ 591.2256. Found: 591.2255.
4.2.16.14. Ethyl ent-15α-hydroxymethyl-16β-(2'-methoxyphenyl)carbamothioylaminobeyeran-19
-oate (Ir). Yield 70%; Mp 83.3–84.2°C; IR (KBr): 3382, 3310, 3063, 2942, 2848, 1718, 1602,
1535, 1500, 1463, 1379, 1336, 1267, 1232, 1180, 1152, 1099, 1025, 981, 924, 851, 748, 571cm-1;
1H NMR (400MHz, CDCl3, ppm): δ 7.56 (s, 1H), 7.28 (t, J =7.30Hz, 2H), 7.00 (d, J = 7.52Hz,
2H), 6.37 (d, J = 10.04Hz, 1H), 4.47–3.99 (m, 5H), 3.85 (s, 3H), 3.64 (t, J = 10.32Hz, 1H), 2.14 (d,
J = 13.52Hz, 1H), 1.89–1.49 (m, 11H), 1.39–1.30 (m, 2H), 1.24 (t, J = 7.10Hz, 3H), 1.14 (s, 3H),
1.09–0.98 (m, 4H), 0.93 (s, 3H), 0.81 (td, J = 13.08, 3.32Hz, 1H), 0.67 (s, 3H); 13C NMR
(100MHz, CDCl3, ppm): δ 180.13, 177.30, 152.92, 128.40, 125.41, 124.65, 121.00, 112.36, 69.34,
64.10, 60.04, 57.35, 56.89, 55.80, 54.70, 49.65, 43.57, 43.07, 41.62, 39.55, 38.08, 37.92, 34.70,
34.01, 28.79, 24.39, 21.84, 19.35, 18.76, 14.12, 13.19; HRMS (ESI, m/z) calcd for
C31H46N2O4SNa [M+Na]+ 565.3076. Found: 565.3074.
27

ACCEPTED MANUSCRIPT
4.2.16.15. Ethyl ent-15α-hydroxymethyl-16β-(3'-methoxyphenyl)carbamothioylaminobeyeran-19
-oate (Is). Yield 72%; Mp 85.3–86.7°C; IR (KBr): 3418, 3379, 3295, 3053, 2943, 2849, 1718,
1602, 1532, 1492, 1463, 1379, 1325, 1262, 1232, 1179, 1152, 1096, 1046, 978, 920, 855, 775, 693,
1
569cm-1; H NMR (400MHz, CDCl3, ppm): δ 7.99 (s, 1H), 7.34 (t, J = 7.92Hz, 1H), 6.85 (d, J =
8.52Hz, 1H), 6.82 (d, J =7.52Hz, 1H), 6.76 (s, 1H), 6.43 (d, J = 9.88Hz, 1H), 4.45 (dd, J = 8.92,
5.52Hz, 1H), 4.13–4.03 (m, 3H), 3.85 (t, J = 9.04Hz, 1H), 3.81 (s, 3H), 3.64 (t, J = 10.24Hz, 1H),
2.15 (d, J = 13.20Hz, 1H), 1.92–1.49 (m, 11H), 1.40–1.29 (m, 2H), 1.24 (t, J = 7.08Hz, 3H), 1.15
(s, 3H), 1.11–0.98 (m, 4H), 0.94 (s, 3H), 0.82 (td, J = 13.56, 3.88Hz, 1H), 0.67 (s, 3H); 13C NMR
(100MHz, CDCl3, ppm): δ 179.87, 177.31, 161.07, 136.83, 130.88, 116.99, 113.37, 110.66, 69.30,
64.10, 60.07, 57.32, 56.90, 55.48, 54.67, 49.67, 43.58, 43.07, 41.67, 39.56, 38.09, 37.92, 34.70,
34.07, 28.79, 24.53, 21.85, 19.38, 18.77, 14.13, 13.16; HRMS (ESI, m/z) calcd for
C31H46N2O4SNa [M+Na]+ 565.3076. Found: 565.3076.
4.2.16.16. Ethyl ent-15α-hydroxymethyl-16β-(4'-methoxyphenyl)carbamothioylaminobeyeran-19
-oate (It). Yield 83%; Mp 96.5–98.2°C; IR (KBr): 3421, 3382, 3309, 3049, 2945, 2849, 1718,
1608, 1535, 1510, 1465, 1380, 1325, 1297, 1241, 1179, 1151, 1097, 1030, 982, 925, 828, 774, 728,
1
693, 572, 518cm-1; H NMR (400MHz, CDCl3, ppm): δ 7.76 (s, 1H), 7.18 (d, J = 8.44Hz, 2H),
6.96 (d, J = 8.52Hz, 2H), 6.11 (d, J = 10.04Hz, 1H), 4.43 (dd, J = 8.88, 5.24Hz, 1H), 4.13–4.01 (m,
3H), 3.87–3.80 (m, 4H), 3.63 (t, J = 10.38Hz, 1H), 2.15 (d, J = 13.24Hz, 1H), 1.88–1.48 (m, 10H),
1.40–1.31 (m, 2H), 1.25 (t, J = 7.12Hz, 3H), 1.15 (s, 3H), 1.10–0.95 (m, 5H), 0.92 (s, 3H), 0.80
(dd, J = 13.00, 4.40Hz, 1H), 0.66 (s, 3H); 13C NMR (100MHz, CDCl3, ppm): δ 180.56, 177.30,
159.14, 127.92, 127.78, 127.78, 115.32, 115.32, 69.23, 64.15, 60.06, 57.29, 56.90, 55.58, 54.62,
49.70, 43.57, 43.04, 41.56, 39.54, 38.07, 37.92, 34.69, 34.01, 28.79, 24.49, 21.83, 19.32, 18.76,
14.14, 13.17; HRMS (ESI, m/z) calcd for C31H46N2O4SNa [M+Na]+ 565.3076. Found: 565.3076.
4.2.16.17. Ethyl ent-15α-hydroxymethyl-16β-(2'-nitrophenyl)carbamothioylaminobeyeran-19-oate
(Iu). Yield 80%; Mp 96.2–97.5°C; IR (KBr): 3421, 3392, 3330, 3056, 2944, 2849, 1717, 1608,
1
1509, 1465, 1348, 1258, 1230, 1180, 1150, 1096, 1025, 980, 856, 785, 740, 719, 573cm-1; H
NMR (400MHz, CDCl3, ppm): δ 9.60 (s, 1H), 8.08 (d, J = 8.24Hz, 2H), 7.60 (t, J = 7.60Hz, 1H),
7.26 (t, J = 7.38Hz, 1H), 6.79 (s, 1H), 4.69–3.58 (m, 6H), 2.17 (d, J = 12.40Hz, 1H), 2.08–2.03 (m,
1H), 1.85–1.54 (m, 9H), 1.45–1.37 (m, 2H), 1.25 (t, J = 7.10Hz, 3H), 1.17 (s, 3H), 1.12–1.03 (m,
13
4H), 0.99 (s, 3H), 0.95–0.82 (m, 2H), 0.75 (s, 3H); C NMR (100MHz, CDCl3, ppm): δ 180.86,
28