Lathyrol diterpenes as modulators of P-glycoprotein dependent multidrug resistance: Structure-activity relationship studies on Euphorbia factor L3 derivatives

Article abstract of DOI:10.1021/acs.jmedchem.5b00058

Five series of 37 new acylate and epoxide derivatives (3-39) of Euphorbia factor L₃, a lathyrol diterpene isolated from Euphorbia lathyris, were designed by modifying the hydroxyl moiety of C-3, C-5, or C-15. Chemoreversal effects of the acylat

Full text of DOI:10.1021/acs.jmedchem.5b00058

Subscriber access provided by UNIV OF NEBRASKA - LINCOLN
Article
Lathyrol Diterpenes as Modulators of P-Glycoprotein
Dependent Multidrug Resistance: Structure-Activity
Relationship Studies on Euphorbia factor L3 Derivatives
Wei Jiao, Zhongmin Wan, Shuang Chen, Runhua Lu, Xiaozhen Chen, Dongmei
Fang, Jiufeng Wang, Shengcai Pu, Xin Huang, Haixiang Gao, and Huawu Shao
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00058 • Publication Date (Web): 09 Apr 2015
Downloaded from http://pubs.acs.org on April 14, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Lathyrol Diterpenes as Modulators of P-Glycoprotein Dependent
Multidrug Resistance: Structure−Activity Relationship Studies on
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Euphorbia factor L Derivatives
3
*,1
1
1
2
1
Wei Jiao, Zhongmin Wan, Shuang Chen, Runhua Lu, Xiaozhen Chen, Dongmei
1
1
1
1
2
*,1
Fang, Jiufeng Wang, Shengcai Pu, Xin Huang, Haixiang Gao, Huawu Shao
1
Chengdu Institute of Biology, Chinese Academy of Sciences, Chengdu 610041, China
2
Department of Applied Chemistry, China Agricultural University, Beijing 100194, China
Keywords: Lathyrol diterpene; Structure–activity relationship; Euphorbia factor L3,
Multidrug resistance; P-glycoprotein; Euphorbia lathyris
ABSTRACT:
Five series of 37 new acylate and epoxide derivatives (3-39) of Euphorbia factor L , a
3
lathyrol diterpene isolated from Euphorbia lathyris, were designed by modifying the
hydroxyl moiety of C-3, C-5 or C-15. Chemoreversal effects of the acylates on multidrug
resistance (MDR) were evaluated in breast cancer multidrug-resistant MCF-7/ADR cells
that overexpress P-glycoprotein (P-gp). Eight derivatives exhibited greater chemoreversal
ability than verapamil (VRP) against adriamycin (ADR) resistance. Compounds 19 and
2
5 exhibited 4.8 and 4.0 times, respectively, more effective reversal ability than VRP
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
against ADR resistance. To determine the key characteristics of Euphorbia factor L3
derivatives that contribute to MDR reversal, we conducted a structure–activity
relationship study of these compounds. The simulation studies indicated different
possible mechanisms and revealed the important influence of hydrophobic interactions
and hydrogen bonds in the flexible cavity of P-gp.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
Recent estimates from the International Agency for Research on Cancer indicate a sharp
increase in breast cancer incidence, with 14.1 million new cancer cases in 2012. Breast
cancer incidence has increased by more than 20%, whereas mortality has increased by
1
1
4%, to estimates in 2008. In cancer treatment, multidrug resistance (MDR) has been
recognized as a major cause for the failure of chemotherapy. Additionally, MDR is an
extensive problem implicated in numerous diseases, such as HIV infection and parasitic
2
diseases. Anthracyclines (e.g., adriamycin (ADR), daunomycin, and epirubicin) are the
3
most common oncology drugs used in combination therapy for breast cancers. However,
the response rates remain suboptimal, and curability is complicated by subclinical
metastases and, eventually, distant metastases from MDR to breast cancer chemotherapy.
In metastatic breast cancer and other metastatic cancers, drug resistance to chemotherapy
4
is thought to result in treatment failure in greater than 90% of patients.
The ABC (ATP binding cassette) transporter P-glycoprotein (P-gp, ABCB1) is one of the
principal multidrug transporters in humans and has been intensively characterized in
2
ACS Paragon Plus Environment

Page 3 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
,6
recent decades. In breast cancer, the levels of P-gp expression strongly correlated with
7
drug resistance in the examined clinical specimens. Intrinsic or acquired overexpression
of P-gp dramatically reduces drug response, leading to poor clinical outcomes following
chemotherapy. This protein is generally expressed at higher levels in epithelial cell
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8
surfaces throughout the body, such as in the liver, pancreas, kidney, colon, and jejunum,
as well as in blood-tissue barriers such as blood-mammary tissue, blood-brain,
9
-11
blood-testis, and blood-nerve barriers.
In addition to MDR in chemotherapy, P-gp is
also closely associated with oral drug bioavailability, drug excretion, and brain
1
2
permeation. P-gp acts as an ATP-dependent drug efflux pump, analogous to a
“
hydrophobic vacuum cleaner” that removes numerous structurally dissimilar
1
3
chemotherapeutic agents from the cell. Recently, the inward-facing conformation of
murine P-gp, which shares 87% sequence identity to human P-gp, has been refined to 3.8
Å resolution, which represents the highest resolution structure of a mammalian multidrug
1
4
transporter to date. The flexible internal cavity formed by two bundles of six TM
transmembrane) helices encloses a large volume of ~6000 Å within the lipid bilayer to
which hydrophobic drugs can bind in various orientations and at different locations. As
(
1
5,16
17
predicted from previous photoaffinity labeling studies
and cross-linking analyses,
the crystal structure exhibits a uniform drug-binding domain, and this large and flexible
binding pocket hinders the development of new modulators of P-gp.
In the past decade, P-gp has been considered the most important MDR-modulating drug
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
target, and numerous reversal agents have been developed. Currently, a reversal agent is
clinically unavailable. Modulators of P-gp generally do not directly kill MDR cells; rather,
6
when combined with a cytotoxic agent, these agents restore chemotherapeutic efficacy.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
However, most MDR modulators suffer clinically due to intrinsic toxicity, low affinity or
undesired pharmacokinetic properties. In this regard, recent studies suggest natural
products as potential MDR modulators due to their various structure types. As a
promising example, macrocyclic diterpenes derived from Euphorbia have attracted
1
8,19
20-25
considerable interest,
with a focus on two types of diterpenes: jatrophane
and
2
5-28
lathyrane.
Polycyclicity and polyestericity are two significant characteristics in these
two skeleton types. These diterpenes are substituted with a variety of acyl groups and
possess 5/12- or 5/11/3-membered ring systems. For lathyrane diterpenes, the effect of
2
2,24,28
the acyl side chains is thought to be a critical factor in MDR reversal.
Moreover, the
2
5,26
fused epoxy ring within the skeleton variation also appears to play an essential role.
However, synthetic methods of selective acylation and intramolecular epoxidation have
not been completed to enrich the limited variation of lathyrane diterpenes. In our previous
studies, five different structure types of lathyrane diterpenes as potential P-gp modulators
2
9
(lathyrol, jokinol, isolathyrol, epoxylathyrol and 7-hydroxylathyrol) were investigated.
All compounds exhibited reversal effects in MCF-7/ADR breast cancer cell lines in vitro.
Euphorbia factor L , which is a lathyrol diterpene in the lathyrane family, exhibited
3
substantial reversal activity in our previous studies. This diterpene is a principal
4
ACS Paragon Plus Environment

Page 5 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
secondary metabolite from the seeds of Euphorbia lathyris, which is a common
3
0
traditional Chinese herb whose seed oil contains 3-5% lathyrane diterpene esters.
Compared to other lathyrane types, lathyrol diterpenes possess an identical skeleton of a
/11/3-membered ring system and three ester substituents characterized by a 6,17-double
bond outside of the macrocycle. In addition to Euphorbia factors L and L , only six
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
3
8
lathyrol diterpenes have been identified to date, and none have been screened for MDR
3
1-35
reversal activity.
Therefore, we conducted a structure–activity relationship (SAR)
investigation of Euphorbia factor L derivatives.
3
Herein, beginning from Euphorbia factor L , five series (I-V) of new acylates and
3
epoxides were designed with modifications at the hydroxyl of C-3, C-5 or C-15. The
absolute configuration and X-ray structures of representatives from each series were
determined for definitive structural confirmation. These derivatives were evaluated in
vitro using a chemoreversal assay and the rhodamine 123 (Rh123) accumulation assay in
MCF-7/ADR cells, which are breast cancer multidrug resistant cells that overexpress
P-gp. The interaction patterns of 14, 19 and 25 with a homology model of human P-gp in
a lipid bilayer membrane environment were investigated using molecular modeling
approaches.
RESULTS AND DISCUSSION
Chemistry. The known macrocyclic diterpene, Euphorbia factor L (1), was purified
3
using a chromatographic method from the ethanol/acetone extract of E. lathyris seeds.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
We selected Euphorbia factor L as the initial structure for the design of various esters,
3
3−39, to assess substituents at positions C-3, C-5 and C-15 of the epoxide. The ester
groups (R in Scheme 1) were selected considering the size, hydrophobicity, and
aromaticity. The diverse set of substituents included aliphatic acyclic and aromatic
groups.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Alkaline hydrolysis of Euphorbia factor L afforded lathyrol 2 with three free hydroxyl
3
groups. Esterifications of 2 were performed by treatment with the related acid (RCO H),
2
N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide
hydrochloride
(EDCI),
and
4
-dimethylaminopyridine (DMAP) to produce 3−9 and 11−14. The esters 10, 15 and 19
were achieved by treatment with benzoic anhydride, acetic anhydride or benzoyl chloride
together with Et N and DMAP, depending on the different reactivity. It is difficult to
3
distinguish the reactivity between OH-3 and OH-5; thus, several synthetic strategies were
employed. The synthesis of 16-18 was conducted beginning from 2, which was treated
with a related acid anhydride and BF ·Et O. The acylation of 2 with RCO H, EDCI and
3
2
2
DMAP yielded 20-22. The derivatization of the related monosubstituted ester at C-3
using acid anhydride or ROCl and BF ·Et O afforded 23-27. BF ·Et O was successfully
3
2
3
2
used to address different substituents of C-3 and C-5 in Euphorbia diterpenes for the first
time. Groups III (28) and IV (29-31) were prepared by acid-catalyzed intramolecular
oxa-Michael addition with the Lewis acid TMSOTf. Group V was synthesized by
epoxidation with AcCl and DMAP, affording 32-37 and 39. The one-pot synthetic route
6
ACS Paragon Plus Environment

Page 7 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
used to prepare 38 involved treatment with cinnamoyl chloride and BF ·Et O at 0 °C.
3
2
Two new synthetic paths of intramolecular epoxidation on lathyrol diterpenes were
discovered through the preparation of Group III, IV and V derivatives. Interestingly,
selective intramolecular oxa-Michael addition under different conditions resulted in two
C-13 epimers (Groups IV and V). These epoxides, which possess various ester groups,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
5,26
were further investigated, not only for the effect of the epoxy ring
but also for
3
6
cytotoxic activity in a similar skeleton. The X-ray structures of representative molecules
are shown in Figure 1. The absolute configurations of these derivatives were determined
according to the X-ray analysis of 17.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 1. Synthesis of Euphorbia factor L derivatives
3
8
ACS Paragon Plus Environment

Page 9 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Reagents and conditions: (a) 5% KOH; (b) RCO H, EDCI, DMAP; (c) acid anhydride or ROCl, Et N,
2
3
DMAP; (d) acid anhydride or ROCl, BF ·Et O; (e) TMSOTf; (f) AcCl, DMAP; 38 was afford by
3
2
treated with cinnamoyl chloride and BF ·Et O from 2.
3
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Views of the representative crystal structures: compound 17 (Group II),
compound 28 (Group III), compound 29 (Group IV), compound 35 (Group V).
Displacement ellipsoids are drawn at the 30% probability level. C27 and C28 of 17
are disordered. O4 and O6 of 28 are disordered.
Cytotoxicity Assay and Preliminary MDR Reversal Activity Screening. The cytotoxic
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
activity of Euphorbia factor L derivatives (2−39) was evaluated in MCF-7 (human
3
breast cancer) and MCF-7/ADR (an MDR subline overexpressing P-gp) cell lines using
3
7
the SRB (sulforhodamine B) assay. The IC values are presented in Table 1. All of the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
50
derivatives exhibited significantly greater cytotoxicity in MCF-7 cells than in
MCF-7/ADR cells. Certain compounds (18, 21, 22, 26, 29 and 32-39) exhibited toxicity
greater than 100 ꢀM in MCF-7/ADR cells, which is beyond the scope of this
investigation; therefore, only the trend in activity is outlined. In Group V, all of the
1
3S-epoxylathyrol analogs (32-39) exhibited low toxicity (IC50 > 100 ꢀM) in
MCF-7/ADR cells. These results suggested that the parental cells were more sensitive to
the test derivatives. For rapid evaluation of chemoreversal activity, MCF-7/ADR cells
were cotreated with ADR and test compounds at 10 ꢀM (Table 2). Nineteen derivatives
were found to be superior to verapamil (VRP) in initial screens. From these results, the
Group I (4, 6, 7, 10-14), II (19, 20, 22-27) and V (32, 36, 39) derivatives were
emphatically observed.
Table 1. Cytotoxicity Effects of 2-39 on MCF-7 Cells (PAR Cells) and MCF-7/ADR
(MDR Cells)
a
compd
IC50 (µM)
PAR cell
34.04±1.84
26.62±1.52
32.09±2.26
13.11±1.61
18.07±1.65
17.00±0.84
MDR cell
59.56±3.57
80.72±3.13
53.46±1.84
57.90±3.12
46.06±1.53
66.29±4.20
2
3
4
5
6
7
1
0
ACS Paragon Plus Environment

Page 11 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
8
9
14.78±2.09
17.26±3.78
11.60±0.36
14.82±2.46
10.99±1.59
8.24±0.92
60.77±6.57
19.27±4.63
55.27±6.63
42.90±1.53
23.45±1.70
58.79±3.90
33.67±2.51
40.96±5.06
44.98±1.01
50.86±0.96
>100
1
0
1
2
3
1
1
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
14
10.46±1.36
21.56±2.87
18.55±1.79
18.51±2.58
47.10±4.96
13.17±0.59
8.93±1.69
1
1
5
6
17
1
8
19
63.77±2.77
36.26±2.56
>100
2
2
0
1
28.69±4.01
13.36±0.60
8.47±1.00
22
>100
2
3
63.27±5.05
76.03±1.58
75.67±1.71
>100
24
9.57±1.54
2
2
5
6
9.73±1.35
4.16±0.93
27
14.18±2.80
45.78±7.59
25.25±3.09
28.50±1.41
14.15±1.49
12.67±1.44
19.14±2.70
48.15±4.60
34.93±4.45
28.47±6.08
46.76±4.47
64.22±4.33
84.87±1.00
42.12±1.92
65.16±6.16
>100
2
8
29
3
3
0
1
54.96±5.48
40.08±4.64
>100
32
3
3
>100
34
>100
3
3
5
6
>100
>100
37
>100
3
3
8
9
>100
>100
-3
-4
-2
-3
Taxol
4.42×10 ±9.72×10
4.36×10 ±4.46×10
a
Values of IC50 are the mean ± standard error (SD) of three independent experiments.
Chemoreversal Abilities of Euphorbia Factor L Derivatives against ADR Resistance.
3
The reversal abilities of the Euphorbia factor L derivatives (2−39) were evaluated using
3
MCF-7/ADR cells with gradient concentrations of the chemotherapeutic drug ADR. The
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
EC value of ADR was calculated by cotreatment with the derivatives at 10 ꢀM. The
50
reversal fold was determined by dividing the EC50 of ADR alone by the EC50 of the
combination treatment.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Multidrug Resistance Reversal Effects of 2-39 on MCF-7/ADR
b
c
group
I
compd
relative enhancement
EC (ꢀM)
fold
50
a
intensity (%)
2
3
4
5
6
7
8
9
7.3
7.1
11.63±1.89
18.49±0.17
15.18±1.33
41.59±2.18
11.45±1.03
10.62±3.81
8.41±1.08
14.78±4.17
22.39±3.19
4.66±1.20
9.89±0.21
12.76±1.02
2.15±0.36
8.37±1.71
12.11±2.26
14.49±1.20
18.51±3.86
1.00±0.12
13.13±3.60
27.04±3.40
24.93±1.88
4.22±1.12
4.23±1.48
1.21±0.65
8.66±1.44
3.53±0.76
18.61±0.97
10.62±3.10
21.03±1.57
3.0
1.9
2.3
0.8
3.0
3.2
4.1
2.3
1.5
7.4
3.5
2.7
11.3
0.1
9.1
10.1
4.7
10.3
7.5
10
11
12
9.6
11.6
9.1
13
1
4
15.5
4.7
16.0
4.1
2.8
2.4
1.9
34.4
2.6
1.3
1.4
8.2
8.1
28.6
4.0
9.8
1.9
3.2
1.6
II
15
1
6
7
4.6
1
4.0
18
5.8
19
20
21
22
23.9
9.5
6.0
10.1
12.1
9.3
23
24
25
26
27
14.4
11.1
12.1
2.4
III
IV
28
29
6.8
3
0
4.0
1
2
ACS Paragon Plus Environment

Page 13 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
1
8.3
13.0
7.7
21.57±5.37
1.6
V
32
3.23±0.60
11.52±2.10
4.94±1.00
20.49±3.88
16.38±3.14
29.17±2.43
9.68±2.04
25.10±0.94
4.89±0.71
10.7
3
3
3
3
3
4
5
6
3.0
6.8
7.0
4.3
1.7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11.6
6.9
2.1
37
1.2
3
8
9
2.1
3.6
3
10.4
8.4
1.4
7.1
VRP
a
Relative enhancement intensity of compounds were showed at concentration of 10
µM and calculated as equation in
b
the section of MDR Reversal Activity. EC (half maximal effective concentration) are the mean ± standard error (SD)
5
0
c
of at least three independent experiments. Reversal fold was calculated as equation in the section of MDR Reversal
Activity.
As shown in Table 2, compounds 11, 14, 19, 23-25, 27 and 32 exhibited more potent
chemoreversal ability than the positive control VRP. Compared to our previous studies,
compounds 14, 19 and 25 were more active than the parent structure, Euphorbia factor L3
(1). Notably, 19 was 4.8 times more effective than VRP in ADR reversal ability in
MCF-7/ADR cells, and it was more effective than all E. lathyris diterpenes previously
2
9
evaluated. The MDR reversal activities of compounds 23, 24, 27 and 32 were slightly
higher than that of VRP. Compounds 11 and 34 exhibited nearly an identical MDR
reversal effect as VRP. Compounds that exhibited reversal fold values greater than 15.0
were classified as potential MDR modulators. The following SAR correlations were
proposed based on the MDR reversal effect on ADR resistance.
In Group I, the monoacylated derivatives were synthesized to evaluate the substituent
effect at C-5. In this group, the enhanced size of aliphatic (3-8) and aromatic (9-14)
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
substituents did not correspond to a proportional reversal activity. The presence of the
normal alkyl chain exhibited a low effect, and acylation of the alkane isomers slightly
improved the activity, as observed for 7 and 8. Compared to the poor profiles of the
aliphatic-substituent analogs, compounds 14 (R = 1-naphthylacetyl) and 11 (R =
phenylacetyl) exhibited higher reversal fold (16.0 and 7.4, respectively), indicating the
necessity of the aromatic group at this position. However, introduction of a
nitrogen-containing aromatic group (9) resulted in a decrease in the reversal effect.
Notably, the effects were minimal when OH-5 of lathyrol (2) was replaced with aromatic
formyl (10) and propionyl (12 and 13) moieties, suggesting that the reversal activities
were also associated with the size of the aromatic side chain at this position.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In Group II, thirteen derivatives containing two ester groups at C-3 and C-5 were
investigated. Five compounds (19, 23-25 and 27) exhibited higher MDR-modulating
activities than VRP. Notably, compounds 19 and 25 were able to effectively reverse the
sensitivity to ADR by 34.4 and 28.6 fold, respectively. As expected, compounds
containing two aliphatic groups (15-18) exhibited a low reversal effect. Introduction of an
acetyl group to OH-3 in 3 led to an enhanced reversal effect, as observed for 15. The
addition of a propionyl group in 10, 11 and 14 increased the MDR reversal effect, as
observed for 23-25. The following trend in activity for the aromatic groups at C-5 was
observed: 1-naphthylacetyl > phenylacetyl ≈ benzoyl. Although the 1-naphthylacetyl
group was beneficial in 14 of Group I and 25 of Group II, the introduction together with
1
4
ACS Paragon Plus Environment

Page 15 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
aromatic groups at C-3 in 14 resulted in a decreased reversal effect, as observed for 22 (R
= 1-naphthylacetyl) and 26 (R = benzoyl). Similarly, replacement of the OH-3 with a
phenylacetyl group in 11 reduced the MDR-modulating effect, as observed for compound
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
0. Esterification of the OH-3 in compound 10 led to the active compounds 19, 23 and 27.
The activity trend in terms of ester groups at C-3 is benzoyl > cinnamoyl > propionyl.
Compound 32 in Group V, bearing two acetyl moieties, exhibited greater reversal activity
than VRP. Compared to 15, 28 and 32, which possess the same side chain, epoxidation of
C-15 and C-12 (32) increased the activity; however, epoxidation of C-15 and C-11 (28)
decreased the activity. An identical effect was observed for 18 and 34, indicating that the
epoxy ring in aliphatic-substituted compounds of Group V may be a favorable factor for
the reversal effect. For the five-membered ring (Groups IV and V) derivatives, the
aromatic-substituted compounds exhibited minimal activity. Compounds 29 and 35
exhibited much lower reversal activity than 19, suggesting that the α, β-unsaturated
ketone moiety is a key factor for the activity. Compounds 29-31 and 35-37 possessed an
identical five-membered ring structure, with the exception of the methyl configuration at
C-13, and these derivatives exhibited minimal MDR reversal activities. These results
indicated that different configurations at C-13 did not greatly affect the MDR reversal
activity.
Inhibition of Rh123 Efflux Mediated by P-gp. To confirm our hypothesis that
Euphorbia factor L derivatives inhibit the efflux activity of P-gp, compounds 14, 19 and
3
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
5 were evaluated at several concentrations (2, 10 and 20 ꢀM) using Rh123. Compared to
the mean fluorescence intensity of VRP (170.0 at 20 ꢀM), compounds 14, 19 and 25
exhibited effective efflux inhibition of Rh123, with values ranging from 182.1 to 204.8 at
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
0 ꢀM (Table 3). With increasing concentrations, the cumulative effect of compound 25
was rapidly enhanced, whereas compound 14 maintained a similar low level throughout
the assessment. At low concentrations (2 ꢀM), compound 19 exhibited a higher
cumulative effect (FL-1 = 197.5) and remained unchanged (FL-1 = 204.8 at 10 and 20
ꢀ
M) with increasing concentration. This dose-effect result suggested that an alternate
mechanism may be involved in the activity of 19.
Figure 2. Inhibition effect of Rh123 efflux mediated by P-gp at 2 and 20 µM: (A)
compound 14, (B) compound 19, and (C) compound 25.
1
6
ACS Paragon Plus Environment

Page 17 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 3. Inhibition Effect of Rh123 Efflux Mediated by P-gp
a
b
c
d
compd
concn (
µ
M)
FSC
SSC
FL-1
SD
peak
131.1
196.8
205.9
215.4
215.4
196.8
143.5
188.2
235.8
87.3
1
4
2
134.9
148.5
138.7
146.8
144.5
148.1
138.5
137.9
131.4
142.5
112.8
137.4
96.3
102.2
94.2
141.1
168.1
182.1
197.5
204.8
204.8
149.4
167.8
204.6
108.1
217.5
170.0
61.6
72.1
77.1
90.8
92.9
88.2
66.3
75.0
89.3
43.9
76.7
66.7
1
2
0
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
19
2
102.7
104.4
102.0
94.8
1
2
0
0
2
5
2
1
0
101.2
100.9
100.6
81.1
20
MDR cells
PAR cells
VRP
215.4
157.1
20
92.2
a
b
c
d
FSC: forward scatter count of cells. SSC: side scatter count of cells. FL-1: mean fluorescence intensity of the cells.
SD: standard deviation.
Physicochemical Properties and Linear Regression Analysis. Physicochemical
parameters are summarized in Table S1, including molecular lipophilicity (LogP),
bulkiness (MW, MV and ASA), H-bond features (HBA, HBD, and TPSA), molar
3
8,39
refractivity (MR) and electrophilicity indices (ω, Table S2).
According to statistical
analysis and the pharmacophore model, P-gp inhibitors or modulators are often
4
0,41
hydrophobic, possess H-bond acceptors and tend to possess few H-bond donors.
All
of the test compounds possess a greater number of H-bond acceptors (HBA = 4-5) than
H-bond donors (HBD = 0-2). Compounds possessing a greater activity than VRP
exhibited LogP > 4 and MW > 400 (except for 32; LogP = 3.4) and were considered
4
2
potential P-gp inhibitors or modulators. For 10-14, a single negative correlation with ω
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
(
r = 0.98) can be observed, suggesting a relationship between aromatic-substituted
analogs in Group I and an electrostatic effect. Compounds 23, 24, and 25 in Group II,
2
which differ in the ester group at C-5, exhibited a positive correlation with ASA (r =
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
2
2
0
.99), MW (r = 0.96), MR (r = 0.96), MV (r = 0.93) and LogP (r = 0.92). However,
the MDR reversal activities of compounds 22, 25 and 26, which bear identical
2
substituents at C-5, decreased with increasing LogP (r = 0.92). Moreover, the analogs in
Group II containing two aromatic groups with MW ≤ 568 and MV ≤ 541.6 were more
active than VRP (19, 23-25 and 27), and those with MW ≥ 570 and MV ≥ 547.7 were less
active than VRP (20-22 and 26). This classification suggested that two bulky aromatic
groups at C-3 and C-5 may not be well tolerated by P-gp. Compounds 19, 23 and 27,
which contain identical benzoyl substituents at C-5, did not exhibit correlations with any
parameters.
Molecular Modeling. The reported crystal structure of murine P-gp (PDB ID: 3G5U) is
considered to represent the initial inward-facing stage of the transport cycle for drug
binding. The structure of murine P-gp represents the highest resolution P-gp structure to
date and was therefore selected for the homology modeling. The sequence alignment
between human P-gp (UniProtKB: P08183) and murine P-gp using BLAST indicated
8
7% overall identity and 90% identity for all residues in the TMD (transmembrane
43
domain). A Ramachandran plot of the final homology model indicated that 85.7% of the
residues were located in the most favored regions, 11.3% in the additionally allowed
1
8
ACS Paragon Plus Environment

Page 19 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
regions, 2.1% in the generously allowed regions and 0.9% in the disallowed regions. No
outliers in the TMD of P-gp were found, a feature that is important for estimation of the
drug-binding ability (Figure 3A).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. (A) Binding positions of 14, 19 and 25 in the homology model of human
P-gp embedded in the POPC bilayer resulting from docking studies. The P-gp model
was generated based on the template structure of the murine P-gp (PDB ID, 3G5U)
and is portrayed as a cartoon. Outliers in the Ramachandran plot are indicated as
dark gray spheres. The ligands are represented as colored sticks: 14, blue; 19,
magenta; and 25, orange. POPC molecules are shown as light-green spheres. The
phosphate groups are highlighted as forest green spheres. The front lipid molecules
are not shown for clarity. All hydrogens are omitted. Binding modes of 14 (B), 19 (C)
and 25 (D) obtained after 5-ns MD simulations. Carbon atoms of the ligands are
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
represented as green sticks. The hydrogen bonds are shown as yellow dashed lines.
The P-gp receptor is shown as a gray cartoon. All hydrogens are omitted for clarity.
Carbon atoms of interacting residues are depicted as colored sticks: TM5, cyan;
TM6, yellow; TM7, orange; TM8, salmon; and TM12, magenta. Oxygen and
nitrogen atoms of all stick representations are colored red and blue, respectively.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To further understand and characterize the interaction between the derivatives and the
receptor, the three most promising ligands (14, 19 and 25) were subjected to molecular
docking with a homology model of P-gp. Because of the large and flexible drug-binding
region of P-gp, the entire hydrophobic cavity was first considered as a binding pocket,
and statistical results were subsequently used for the accurate docking. The substrates
were finally located within the TMD regions of the binding pocket, as indicated by both
1
4
biochemical and structural results. The best conformation (Figure 3A) was used as the
initial geometry for 5-ns MD calculations with explicit water and a POPC membrane.
The calculated binding free energies of 14, 19 and 25 to P-gp are shown in Table 4.
Compounds 14, 19 and 25 were primarily stabilized by van der Waals interactions
(ꢀEvdw), with values of -50.0, -51.8 and -51.0 kcal/mol, respectively. Nonpolar
contributions, including ꢀEvdw and nonpolar solvation free energies (ꢀG ), were
np
completely favorable for the binding process. For compounds 14 and 19, a significant
additional stabilization was gained through electrostatic interactions (ꢀEelec = -9.3 and
-
9.6 kcal/mol, respectively). Although the polar contribution (ꢀE ) was favorable, the
elec
2
0
ACS Paragon Plus Environment

Page 21 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
polar solvation free energies (ꢀG ) severely weakened binding of 14, 19 and 25.
PB
Therefore, the polar effect was not favorable for the binding of these lathyrol diterpenes
to P-gp.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 4. Free Energy Analysis (kcal/mol) of the Binding of Compounds 14, 19 and
2
5 to P-gp.
a
b
c
d
e
f
g
h
compd ꢀEvdw
ꢀEelec
-9.3
ꢀGPB
29.5
33.4
ꢀGnp
-6.3
-6.9
-6.7
∆Ggas
∆Gsolv ∆Gbind
SD
14
19
-50.0
-51.8
-51.0
-59.3
-61.4
-54.3
23.2
26.4
15.3
-36.1
-34.9
3.1
3.4
2.8
-9.6
2
5
-3.3
22.0
-39.0
a
b
c
ꢀEvdw: van der Waals interaction energy; ꢀEelec: the electrostatic interaction energy; ꢀG : the electrostatic
PB
d
contribution to the solvation free energy calculated by solving the linearised Poisson Boltzman (PB) equation; ꢀG :
np
e
f
nonpolar contribution to the solvation free energy; ∆G Gas phase energy (ꢀEvdw+ꢀEelec).; ∆Gsolv: solvation free
gas:
g
h
energy (ꢀG +ꢀG ); ∆Gbind: The final estimated binding free energy calculated from the terms above; SD:
np
PB
standard deviation.
Figure 4. Energy contributions of the top 10 residues from ligand−residue
interaction energy decomposition in the binding complexes of compounds 14, 19 and
2
5. Positive free energies (ignoring the negative sign) represent favorable binding.
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
As shown in Figures 3 and 4, the test structures exhibited a large number of nonbonding
interactions, which primarily resulted from hydrophobic or aromatic residues in TMs 5, 6,
7, 8 and 12. Compound 14 was bound primarily by the interactions of residues Gly300 in
TM5, Phe770 in TM8 and Gln990 in TM12 (Figures 3B, 4). The 1-naphthylacetyl group
was inserted into the gap between TM7 and TM12 (Figure 3B), which is consistent with
the speculation concerning the size of the side chain at C-5. A primary interaction was
not found for compound 19, and the binding free energies were primarily dispersed by
Phe303 and Leu304 in TM5, Asn721 in TM7, and Met986, Gly989 and Ser993 in TM12.
Two benzoyl groups of 19 protrude from the binding pocket (Figures 3C, 4), and this
feature may be related to some extent to the lower flexibility of the benzoyl group. This
feature was also found in compound 27. For compound 25, prominent hydrophobic
interactions with Phe303 and Tyr307 in TM5 and Phe343 in TM6 (Figures 3D, 4) were
observed. Furthermore, the 1-naphthylacetyl group is surrounded by TM5 and TM6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(Figure 3D), which may explain the hydrophobic trend with the reversal activities of
2
3-25.
2
2
ACS Paragon Plus Environment

Page 23 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Electrostatic potential of 14, 19, and 25 mapped on the electron density
surface. The scale of the electrostatic potential is negative (red) to positive (blue).
To evaluate the differences in the polar contributions among 14, 19 and 25, ab initio
calculations were performed to map the electrostatic potential of the three compounds
onto the respective molecular surfaces representing the electron densities (Figure 5). As
observed for 14, the potential H-donor at OH-15 was entirely electropositive, resulting
from an intramolecular H-bond between OH-3 and O-15. For compound 14, the most
powerful polar interaction was the H-bond between OH-15 and Gly300. Most frames
(71.5%) exhibited this H-bond effect with an average distance of 2.76 Å in the H-bond
analysis. This strong bonding interaction (-3.699 kcal/mol) coupled with an
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
intramolecular H-bond greatly restricted the direction and position of 14 (Figure 6A,
Table S3). Conversely, the hydrogen atom of OH-15 in compounds 19 and 25 revealed a
considerably less positive charge due to the participation of OH-15 in an intramolecular
H-bond with O-3. The electrostatic effect of 19 was influenced by an H-bond between the
carbonyl of the benzoyl group at C-5 and Asn721 (∆Eelec = -2.367) or between that at C-3
and Gly989 (∆Eelec = -1.945) (Table S4). No effective H-bond interaction with residues
was found in 25 during the MD simulation (Table S5). For compound 25, the formation
of an H-bond may be difficult due to the larger steric hindrance of the 1-naphthylacetyl
group.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4
ACS Paragon Plus Environment

Page 25 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 6. Superimposition of frames at 1, 3 and 5 ns in the MD trajectories of
compounds 14 (A) and 25 (B). The strong interaction residues are shown as sticks:
carbon, gray; oxygen, red; and nitrogen, blue. The ligands are depicted as colored
sticks: 1 ns, sky blue; 3 ns, forest green; and 5 ns, crimson. All hydrogens are
omitted for clarity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The classic mechanism for P-gp transport proposes that P-gp undergoes large structural
changes from an inward-facing to an outward-facing conformation during the catalytic
cycle. As demonstrated by Cys cross-linking, the packing of the TM segments in P-gp is
4
4
altered relative to the drug-free state through an induced-fit mechanism. In this large
and flexible drug-binding region, the continuous H-bond system in the complex with 14
should have constrained the adjustment ability of P-gp and ligand (Figure 6A). However,
compound 25 was not fixed in the binding pocket during the MD simulation but
interacted following variation of the flexible cavity. Even with the hydrophobic influence,
the intramolecular H-bond in 25 gradually deconstructed to better adapt to the pocket
(Figure 6B).
There are three contradictory details for the mechanism of 19 with P-gp: i) Compound 10
exhibited a dramatically low reversal effect but possesses an identical substituent at C-5
as 19; ii) The efflux inhibition effect was not improved with increasing concentrations in
the Rh123 efflux assay; and iii) The lowest binding free energy was inconsistent with the
highest reversal fold. Therefore, we propose that the mechanism of action of compound
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
9 not only includes binding to the TMD but may include a synergistic mechanism, such
4
5
as exerting a direct inhibitory effect on ATP hydrolysis, as previously reported.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CONCLUSIONS
The hydroxyl groups at C-3 and C-5 were selectively acylated to investigate the effect of
the acyl side chain on the lathyrol scaffold. Additionally, two new methods of
epoxidation at OH-15 were used to demonstrate the function of the epoxy ring in this
skeleton. Finally, 37 derivatives were newly designed and synthesized, varying in size,
hydrophobicity and aromaticity. All synthesized derivatives were evaluated for
MDR-modulating effects in MCF-7/ADR cells. In the lathyrol structure, the aromatic
group at C-5 and the α, β-unsaturated ketone moiety are necessary to maintain the MDR
reversal effect. In Group I, the size of the aromatic acetyl ester at C-5 was suitable for
maximum activity. Furthermore, a low electrophilicity is favorable for the reversal
activity. In Group II, the presence of a benzoyl group at C-5 increased the potency.
Benzoylation of the OH-3 and OH-5 of lathyrol (2) produced the most active compound
(19). Additionally, a combination of the propionyl group at C-3 and an aromatic group at
C-5 led to an improvement in potency. Accordingly, compound 25 exhibited the second
highest activity in this series. However, two bulky aromatic groups, such as phenylacetyl
and 1-naphthylacetyl, at these two positions may not be well tolerated by P-gp and
resulted in low activity. In Group V, the epoxidation of C-15 and C-11 increased the
reversal activity of aliphatic substituted derivatives. Molecular modeling was used to
2
6
ACS Paragon Plus Environment

Page 27 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
further demonstrate the binding modes of the derivatives with P-gp, including favorable
nonpolar interactions and detrimental H-bond interactions. These results are applicable to
similar skeletons of Euphorbia diterpenes to find chemosensitizers for future clinical
development.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
General. The ESI-MS and high-resolution ESI-MS data were recorded on a Finnigan
DECA
LCQ
spectrometer and a Bruker Bio TOF Q spectrometer, respectively. Optical
rotations were measured on an M341 polarimeter (PerkinElmer, USA). All reagents were
purchased from commercial suppliers and used without further purification unless
otherwise noted. Melting points were recorded on an X6 micro-melting point apparatus
(Beijing Fukai, China) without correction. Thin-layer chromatography was performed
using silica gel 60 GF254 precoated plates (0.20–0.25mm thickness) with a fluorescent
indicator. Visualization on TLC was achieved by UV light (254 nm) and a typical TLC
indication solution (10% sulfuric acid/ethanol solution). Column chromatography was
performed on silica gel 90, 200-300 mesh and Sephadex LH-20 (70 ꢀm, Amersham,
Sweden). NMR spectra were recorded on a Bruker Avance 600 spectrometer (Bruker
1
13
BioSpin, Switzerland) operating at 600 or 151 MHz for H and C, respectively. Spectra
were assigned using appropriate COSY, DEPT, HMQC, and HMBC sequences. NMR
chemical shifts are expressed in ppm (δ) referenced to the solvent used. Data are reported
as follows: chemical shift, multiplicity, coupling constants (Hz) and integration. All final
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
compounds are >95% pure based on AQUITY Ultra Performance liquid chromatography
(UPLC) system coupled to a photodiode array detector (PDA) (Waters Corporation,
Milford, MA, USA).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Plant Material. Seeds of Euphorbia Lathyris were collected in Sichuan province, P. R.
China, and identified by Xin Chao Long, information center of Sinotcm Corporation,
Chengdu. A voucher specimen (No. 20110812-021) was deposited at Chengdu Institute
of Biology, Chinese Academy of Sciences.
Extraction and Isolation. The powdered seeds of E. lathyris (10 kg) were exhaustively
extracted with 95% ethanol/acetone (5:1) at room temperature (4 × 30 L). Following
evaporation of the solvent under vacuum, the crude oil (3.5 kg) was subjected to column
chromatography over silica gel using petrol/EtOAc (20:1) to remove a large quantity of
lipophilic oil. The residue was chromatographed on silica gel using a petrol/EtOAc
stepwise elution to afford 5 fractions according to thin-layer chromatography. Fraction 2
was applied to silica gel using a petrol/EtOAc stepwise elution and then subjected to
Sephadex LH-20 using MeOH/H O to afford Euphorbia factor L (22.6 g).
2
3
Euphorbia
factor
L ,
3
(12E,2S,3S,4R,5R,9S,11R,15R)-5,15-Diacetoxy-3
-benzoyloxylathyra-6(17),12-dien-14-one (1). White powder; mp 152–154 °C; ESI-MS
+
−1
m/z 545 [M+Na] ; IR (KBr, ν , cm ) 3452, 2927, 2858, 1739, 1714, 1651, 1622, 1452,
max
1
1
1
369, 1277, 1223, 1110, 1068, 1008, 713; H NMR (600 MHz, CDCl ) δ 6.54 (d, J =
3
1.5 Hz, 1H, H-12), 6.21 (d, J = 10.0 Hz, 1H, H-5), 5.82 (d, J = 3.2 Hz, 1H, H-3), 5.01 (s,
2
8
ACS Paragon Plus Environment

Page 29 of 80
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
H, H-17a), 4.78 (s, 1H, H-17b), 3.53 (dd, J = 14.2, 8.4 Hz, 1H, H-1α), 2.90 (dd, J = 9.8,
3.4 Hz, 1H, H-4), 2.40–2.30 (m, 1H, H-2), 2.23–2.16 (m, 1H, H-7α), 2.05–1.98 (m, 1H,
H-7β), 1.98–1.91 (m, 1H, H-8α), 1.79–1.71 (m, 1H, H-8β), 1.72 (s, 3H, H-20), 1.70–1.64
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
m, 1H, H-1β), 1.41 (dd, J = 11.2, 8.2 Hz, 1H, H-11), 1.18 (s, 3H, H-18), 1.17 (s, 3H,
H-19), 1.20–1.13 (m, 1H, H-9), 0.94 (d, J = 6.6 Hz, 3H, H-16), 3-O-benzoyl: 8.03 (d, J =
.6 Hz, 2H), 7.58 (t, J = 7.6 Hz, 1H), 7.45 (t, J = 7.6 Hz, 2H), 5-OAc: 1.82 (s, 3H),
7
1
3
15-OAc: 2.21 (s, 3H); C NMR (151 MHz, CDCl ) δ 196.7 (C-14), 146.5 (C-12), 144.6
3
(C-6), 134.2 (C-13), 115.4 (C-17), 92.5 (C-15), 80.9 (C-3), 65.5 (C-5), 52.2 (C-4), 48.6
(C-1), 37.9 (C-2), 35.4 (C-9), 35.0 (C-7), 29.0 (C-19), 28.5 (C-11), 25.3 (C-10), 21.7
(C-8), 16.8 (C-18), 14.2 (C-16), 12.4 (C-20), 3-O-benzoyl: 166.1, 133.1, 130.1, 129.6,
1
29.6, 128.3, 128.3, 5-OAc: 169.7, 22.0, 15-OAc: 170.1, 20.9.
(12E,2S,3S,4R,5R,9S,11R,15R)-Lathyra-6(17),12-dien-3,5,15-triol-14-one
(2).
A
suspension of 1 (5 g, 9.6 mmol) in 5% KOH and MeOH (100 mL) was stirred at room
temperature for 24 h. Further purification was performed by flash column
chromatography (petroleum ether/EtOAc = 4:1) to afford 2 as a colorless crystal (2.9 g,
−
1
yield 89 %). IR (KBr, νmax, cm ) 3415, 2954, 2925, 2861, 1622, 1445, 1269, 1146, 1048,
+
1
1015, 909; ESI-MS m/z 357 [M+Na] ; mp: 158–160 °C; H NMR (600 MHz, acetone-d )
6
δ 7.10 (d, J = 11.4 Hz, 1H, H-12), 4.79 (s, 1H, H-17a), 4.68 (s, 1H, H-17b), 4.70–4.64 (m,
1
H, H-5), 4.45 (s, 1H, OH-15), 4.33 (d, J = 2.8 Hz, 1H, H-3), 4.03 (s, 1H, OH-3), 3.94 (d,
J = 6.8 Hz, 1H, OH-5), 3.04 (dd, J = 13.4, 9.1 Hz, 1H, H-1α), 2.69–2.60 (m, 1H, H-7α),
2
9
ACS Paragon Plus Environment