Bioorganic & Medicinal Chemistry Letters 27 (2017) 5382–5386

Contents lists available at ScienceDirect

Bioorganic & Medicinal Chemistry Letters

journal homepage: www.elsevier.com/locate/bmcl

Identification of a diverse synthetic abietane diterpenoid library and

insight into the structure-activity relationships for antibacterial activity
Wei Hou a,e, Guanjun Zhang b,e, Zhi Luo c, Di Li a, Haoqiang Ruan a, Benfang Helen Ruan a, Lin Su a,⇑,
Hongtao Xu d,⇑
a
College of Pharmaceutical Science, Collaborative Innovation Center of Yangtze River Delta Region Green Pharmaceuticals, Zhejiang University of Technology, Hangzhou 310014,
PR China
b
College of Chemical Engineering and Materials Science, Tianjin University of Science & Technology, Tianjin 300457, PR China
c
Shanghai Evergene Biotech Co., Ltd., Shanghai 201499, PR China
d
Shanghai Institute for Advanced Immunochemical Studies, ShanghaiTech University, Shanghai 201210, PR China

a r t i c l e i n f o a b s t r a c t

Article history: A diverse natural product-like (NPL) synthetic abietane diterpenoid library containing 86 compounds
Received 30 July 2017 were obtained and the SARs were studied based on their antibacterial potential. Further in vitro cytotoxic
Revised 6 November 2017 and in silico drug-like properties evaluation showed that the potent antibacterial compound 84 had good
Accepted 8 November 2017
drug-like properties and displayed low cytotoxicity toward noncancerous mammalian cells, indicating
Available online 9 November 2017
the study of AA and DHAA might be a good starting point for the search of novel antimicrobial molecules.
Future work should be focused on the optimization of their potency and selectivity.
Keywords:
Ó 2017 Elsevier Ltd. All rights reserved.
Diterpenoid
Natural product-like (NPL)
Dehydroabietic acid (DHAA)
Abietic acid (AA)
Click chemistry
Antibacterial

Despite our growing knowledge about infectious diseases and
the availability of various antibiotics and chemotherapeutics, the
rapid development of resistance in pathogenic bacteria towards
available antibiotics remains a global public health threat. In par-
ticular, a rapidly increasing number of multidrug-resistant strains
have emerged.1 As reported by the American Center for Disease
Control and Prevention (CDC), more than 1.2% out of 2 million peo-
ple die from infections with multidrug-resistant bacteria (MDRB)
in the United States each year.2 Therefore, the discovery and devel-
opment of new effective antimicrobial agents with novel chemical
structures and mechanisms to address the drug resistance and
improve the antimicrobial potency is of prime interest.
Natural products (NPs) are structural-complex and biologically
validated molecules that have a profound impact upon drug dis-
covery efforts for the treatment of human diseases.3 Their diverse
three-dimensional shapes, functionalities, stereochemistries as
well as various interesting biological activities have always pro-
vided medicinal chemists with a reliable source in their search
⇑ Corresponding authors.
E-mail addresses: sulin8023@zjut.edu.cn (L. Su), xuht@shanghaitech.edu.cn,
htxu2010@hotmail.com (H. Xu).
e
These authors contributed equally to this work.
for new drug-like molecules. For example, terpenoids, as the lar-
gest class of NPs, are the most abundant and widely distributed
naturally occurring interesting metabolites containing approxi-
mately 25,000 chemical structures. Therefore, extensive pharma-
ceutical activities were exhibited.4 Among them, abietic acid (AA,
1) and dehydroabietic acid (DHAA, 5) are naturally occurring resin
acids that can be readily obtained from pinus rosin or dispropor-
tionated rosin (Fig. 1). It is reported that AA, DHAA and their
semisynthetic derivatives possess a broad spectrum of biological
properties such as anti-inflammatory, antiviral, antitumor, BK
channel-opening, anticonvulsant, antiprotozoal, antifungal,
antileishmanial and especially antimicrobial activities.5 Moreover,
due to their abundance and commercial availability, AA and DHAA
have been widely used as starting materials for the construction of
several NPs and structurally diverse natural product-like (NPL)
molecules library.5f,g,6
In light of these findings and in our continuing efforts on the
structural modification of bioactive natural products and construc-
tion of natural product-like (NPL) library from resources-rich
NPs5f,7, herein, we report the identification of a diverse synthetic
abietane diterpenoid library with antibacterial activity. The novel
small NPL library contains 86 compounds, which were obtained
during our total synthesis and structure-activity relationships

https://doi.org/10.1016/j.bmcl.2017.11.014
0960-894X/Ó 2017 Elsevier Ltd. All rights reserved.
 W. Hou et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 5382–5386 5383

Fig. 1. Compounds 1–86. (The newly synthesized compounds were marked in blue (Supporting information section)).
5384 W. Hou et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 5382–5386
(SARs) studies of triptolide and click chemistry-based structural
modification of DHAA.5f,g,6,7b,8 These compounds would be
categorized into abietane acid derivatives (1–4), dehydroabietane
acid derivatives (5–30), triptophenolide derivatives (31–45), and
a series of novel C-14 1,2,3-triazole-tethered DHAA derivatives
(46–86) (Fig. 1 and Scheme 1).
The in vitro antibacterial activities of the target compounds
were evaluated for their potential against two Gram-positive bac-
teria (Bacillus subtilis (Bs) and Staphylococcus aureus (Sa)) and two
Gram-negative bacteria (Escherichia coli (Ec) and Pseudomonas flu-
orescens (Pf)) at concentrations ranging from 0.4 mg/mL to 100 mg/
mL. The results were assessed in terms of minimum inhibitory con-
centrations (MICs, mg/mL), which are defined as the lowest concen-
trations of compounds at which microbial growth was inhibited.
Amikacin sulfate (AK) and norfloxacin (Nor) were co-assayed as
positive controls against tested bacteria strains, respectively.
Experiments were performed in triplicate for each condition and
the antibacterial data were depicted in Table 1.
The data of antimicrobial screening results revealed that most
of the compounds showed varying degrees of inhibition against
both Gram-positive and Gram-negative bacteria. As illustrated in
Table 1, abietic acid 1 showed the highest in vitro activity on the
tested organisms among the AA derivatives (1–4). As for the
derivatives of DHAA (5–30), compounds 15, 19 and 29 showed
strong activities (0.8 mg/mL to 6.3 mg/mL) against both Gram-posi-
tive bacteria (B. subtilis and S. aureus) and gram-negative bacteria
(E. coli and P. fluorescens). Especially, ketene dithioacetal 29 exhib-
ited significant activity against gram-positive bacteria (B. subtilis)
with a comparable MIC value (0.8 mg/mL) to positive control ami-
kacin sulfate and norfloxacin. The majority of the triptophenolide
derivatives (31–45) showed inhibitory activities against the test
microorganisms at moderate levels, except for compound 31, 32,
33 and 45. Notably, compound 31 exhibited broad-spectrum
potent antimicrobial activities against both gram-positive bacteria
(B. subtilis and S. aureus) and gram-negative bacteria (E. coli and P.
fluorescens) (0.8 mg/mL to 1.6 mg/mL). Meanwhile, compound 32
exhibited significant inhibitory activity against Gram-negative
bacteria (E. coli and P. fluorescens) with MICs value of 0.8 mg/mL
and 1.6 mg/mL, respectively. More importantly, it was found
that most of the C-14 1,2,3-triazole-tethered DHAA derivatives
Scheme 1. Reagents and conditions: (a) alkyne, sodium ascorbate, CuSO4
5H2O, t-
BuOH-H2O (1:1), 40 °C.
(46–86) showed moderate to potent inhibitory activities against
the tested bacteria strains. Among them, compound 84 exhibited
the most potent activities against both gram-positive bacteria
(B. subtilis and S. aureus) and gram-negative bacteria (E. coli and
P. fluorescens) with comparable MICs value (1.6 mg/mL to 3.1 mg/mL)
to positive control norfloxacin.
Based on the in vitro antimicrobial data, preliminary struc-
ture-activity relationships (SARs) of compounds 1–86 were stud-
ied. Generally, the methyl esterification of C-18 carboxyl group
showed negative influence on antibacterial activity, for instance,
compounds 2 and 6 both with a methyl ester group at C-18
showed lower antimicrobial activity against the tested organisms
compared to the corresponding acids 1 and 5. The introduction
of amine or hydroxyl group at C-14 as well as the oxidation at
C-7 showed no significant effect on antibacterial activity, while
the introduction of methoxyl at C-14 might exert a slightly neg-
ative effect. Further, the present data indicated that the introduc-
tion of huge substituents on the C-4 b-hydroxymethyl group to
compounds such as 24 and 26 would exert positive effect on
antibacterial activity. In addition, compounds with Michael
receptor motif on ring A exhibited significantly increased
antibacterial activities, such as 19, 21, 28, 29, 30 and 33. It is
interesting that the introduction of C-5, C-6 olefinic group and
C-7 a-hydroxyl group on the triptophenolide skeleton could also
exert positive effect on antibacterial activity (45). A possible
explanation for this result was that the C-7 a-hydroxyl group
would facilitate the covalent michael addition of the compound
with its cellular target via hydrogen bonding effects.9 More
importantly, the present data also showed that the introduction
of 1,2,3-triazole ring at the C-14 position of DHAA could also
retain the potent antibacterial activities. Generally, the introduc-
tion of electron-poor aromatic ring system on the C-4 position of
the 1,2,3-triazole moiety would significantly increase the
antibacterial activities. In particular, compounds 82, 83 and 84
showed promising low MICs against the tested bacteria. While,
with the saturated alkyl-, ester- and trimethylsilyl- substituted
groups on the C-4 position of the 1,2,3-triazole moiety there
were no significant SARs.
To further assess the preliminary drug-likeness properties of
the potent compounds, in silico Molinspiration online property
calculation was conducted and the data obtained were summa-
rized in Table 2.10 As we know, for good oral bioavailability an
ideal molecule should have good intestinal absorption and
reduced molecular flexibility.11 For most tested compounds
(Table 2), the parameters like hydrogen bond acceptor (HBA),
hydrogen bond donor (HBD), rotatable bonds (nrotb), molecular
weight (MW) and Log P were all in the range of Lipinski’s ‘‘rule
of five”. Thus, it is conceivable that these compounds might have
ideal physicochemical properties.12 Particularly, for C-14 1,2,3-
triazole-tethered DHAA derivatives 82 and 84, the introduced
basic functionalities would be favorable to form hydrochloric
salts to improve the aqueous solubility.13 Further, the topological
polar surface area (TPSA), a parameter defined as the surface
sum over all polar atoms and used extensively in medicinal
chemistry to predict absorption and optimize a compound’s
membrane permeability, was calculated for the potent com-
pounds.14 It is suggested that compounds with a TPSA greater
than 140 Å2 tend to be poor at permeating cell membranes,
while a TPSA less than 60 Å2 is usually needed for molecules
to penetrate the blood-brain barrier (BBB) and thus act on the
brain and other central nervous system (CNS) tissues.14b As
shown in Table 2, the TPSAs of the 1,2,3-triazol substituted
compounds 82 and 84 suggested a beneficial intestinal
absorption (TPSA <140 Å2). Simultaneously, it seemed to be
unlikely for the two compounds to cross the BBB (TPSA >60
Å2). Based on the predicted data mentioned above, the
 W. Hou et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 5382–5386 5385

Table 1
Hot map of antibacterial activity of compounds 1–86.
MIC (µg/mL) MIC (µg/mL)
No. Gram-positive Gram-negative No. Gram-positive Gram-negative
Sa Bs Ec Pf Sa Bs Ec Pf
1 25 50 50 25 45 3.1 6.3 3.1 3.1
2 50 75 75 50 46 25 75 25 50
3 75 75 100 75 47 50 50 75 25
4 75 100 75 100 48 75 50 100 25
5 25 50 75 50 49 25 50 25 75
6 75 100 100 100 50 50 75 50 50
7 50 75 50 25 51 50 75 6.3 75
8 75 75 100 75 52 75 50 75 50
9 25 50 50 50 53 75 100 75 100
10 50 50 25 50 54 50 12.5 25 75
11 75 100 75 100 55 50 50 75 50
12 100 100 75 100 56 25 75 50 75
13 100 75 100 25 57 50 25 25 50
14 50 25 50 75 58 25 50 25 12.5
15 3.1 6.3 3.1 3.1 59 12.5 6.3 25 12.5
16 75 100 100 75 60 50 25 50 50
17 50 25 50 25 61 75 50 75 25
18 100 100 75 75 62 50 75 75 50
19 3.1 6.3 6.3 3.1 63 6.3 50 12.5 25
20 100 75 100 100 64 100 75 75 75
21 25 12.5 6.3 25 65 50 50 75 75
22 100 100 100 100 66 25 100 75 75
23 75 100 75 50 67 6.3 6.3 3.1 6.3
24 25 50 25 25 68 6.3 12.5 12.5 25
25 100 75 100 75 69 25 50 50 75
26 25 50 25 25 70 25 50 75 50
27 100 100 100 75 71 25 50 75 12.5
28 25 12.5 3.1 6.3 72 25 12.5 12.5 25
29 6.3 0.8 6.3 1.6 73 25 12.5 12.5 25
30 50 25 75 25 74 12.5 12.5 6.3 25
31 1.6 0.8 1.6 1.6 75 25 25 25 50
32 50 25 0.8 1.6 76 50 12.5 12.5 25
33 6.3 6.3 3.1 6.3 77 25 25 12.5 12.5
34 100 75 100 75 78 25 12.5 25 6.3
35 100 100 100 100 79 75 50 25 25
36 75 100 100 100 80 50 75 25 50
37 100 100 50 100 81 25 12.5 12.5 12.5
38 50 100 100 100 82 3.1 12.5 3.1 6.3
39 75 50 100 75 83 0.8 6.3 25 3.1
40 100 100 100 100 84 1.6 3.1 1.6 1.6
41 100 100 100 100 85 12.5 25 6.3 12.5
42 50 75 50 75 86 12.5 6.3 50 6.3
43 100 100 100 100 AK 0.8 0.8 1.6 0.8
44 75 100 75 100 Nor 1.6 1.6 3.1 1.6

Table 2
In silico drug-like properties evaluation of the representative compounds.

Compd MW HBDa HBAb LogPc Nrotbd TPSAe

15 360.5 0 4 4.99 4 36.9
19 256.3 0 2 3.58 1 26.3
29 450.7 0 4 4.97 8 44.7
31 316.4 1 5 1.09 1 72.8
33 284.3 0 3 2.96 1 35.5
45 340.4 1 4 2.12 2 55.8
67 571.7 2 9 4.39 8 115.2
82 515.6 1 8 4.13 7 99.0
84 516.6 1 9 3.42 7 111.9
a
HBA: number of hydrogen bond acceptors.
b
HBD: number of hydrogen bond donors.
c
LogP: logarithm of partition coefficient of compound between n-octanol and water.
d
nrotb: number of rotatable bond donors.
e
TPSA: topological polar surface area.
5386 W. Hou et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 5382–5386
Table 3
In vitro cytotoxic activity of the representative compounds.
Compd CC50 (mg/mL)a
HL-7702
15 8.2 ± 1.5
19 9.1 ± 2.0
29 11.4 ± 0.7
31 >50
33 25.2 ± 2.5
45 9.2 ± 0.8
67 12.2 ± 1.2
82 48.2 ± 3.7
84 41.8 ± 4.4
a
The drug concentration required for 50% inhibition of cell viability. Each
experiment was run for at least three times independently with nine concentra-
tions, and the results are presented as mean ± S.D.
calculated TPSA values for compounds 82 and 84 would indicate
a possible better oral bioavailability with a lower chance of
central nervous system toxicity.15
Considering that the concomitant toxicity to noncancerous
mammalian cells is one of the major obstacles to the druggability
optimization of many novel compounds with potent antibacterial
activities, we chose the selected compounds with low MICs for fur-
ther evaluation of their cytotoxic activities on human normal fore-
skin fibroblast (HFF) and liver (HL-7702) cells. As illustrated in
Table 3, compounds 15, 19, 29, 45 and 67 showed CC50 values in
the same order of magnitude with their MICs against bacterial
cells, suggesting that these compounds had no clear selectivities
between microbial and mammalian cells. Pleasingly, compounds
31, 82 and 84 only showed very low cytotoxicities (CC50 31.7
mg/mL) against human normal foreskin fibroblast (HFF) and liver
(HL-7702) cells, implying that there was a therapeutic window to
use these compounds.
In summary, a diverse NPL abietane diterpenoid library con-
taining 86 compounds were obtained and their antibacterial
SARs were studied. The results showed that some compounds
such as 15, 19, 29, 45 and 67 possessed good antimicrobial prop-
erties, although they were poorly selective towards bacterial
cells and mammalian cells. To our delight, compounds 31, 82
and 84 revealed impressive antimicrobial activities without obvi-
ous toxicities to mammalian cells. Notably, in silico evaluation
indicated that compound 84, with 1,2,3-triazole moiety at C-14
position, would have good drug-like properties. Collectively,
these results highlighted the great potential of AA and DHAA
as starting points for the search of novel antimicrobial mole-
cules. The mechanism study, further optimization of their
potency and selectivity is ongoing in our laboratory and will
be reported in due course.
Acknowledgments
This work was partially supported by the Scientific Start-up
funding from Zhejiang University of Technology, China (Number:
2017116002229) and China Postdoctoral Science Foundation,
China (Number: 2017M612024).
A. Supplementary data
Supplementary data associated with this article can be found, in
HFF the online version, at https://doi.org/10.1016/j.bmcl.2017.11.014.
5.4 ± 0.9
9.3 ± 1.1 References
13.2 ± 2.4
47.5 ± 7.9 1. (a) Levy SB, Marshall B. Nat Med. 2004;12:122;
9.9 ± 1.7 (b) Aksoy DY, Unal S. Clin Microbiol Infect. 2008;14:411;
10.3 ± 0.9 (c) Demain AL, Sanchez S. J Antibiot. 2009;62:5.
2. (a) CDC: Antibiotic resistance threat in the United States. Technical Report
9.7 ± 1.1
2013, 77.;
47.9 ± 3.8
(b) Sanabria-Ríos DJ, Rivera-Torres Y, Rosario J, et al. Bioorg Med Chem Lett.
31.7 ± 2.5
2015;25:5067.
3. (a) Newman DJ. J Med Chem. 2008;51:2589;
(b) Harvey AL. Drug Discovery Today. 2008;13:894;
(c) Newman DJ, Cragg GM. J Nat Prod. 2012;75:311.
4. Huang M, Lu J-J, Huang M-Q, Bao J-L, Chen X-P, Wang Y-T. Expert Opin Investig
Drugs. 2012;21:1801.
5. (a) González MA, Pérez-Guaita D, Correa-Royero J, et al. Eur J Med Chem.
2010;45:811;
(b) Cui YM, Yasutomi E, Otani Y, et al. Bioorg Med Chem Lett. 2008;18:5201;
(c) Yao GY, Ye MY, Huang RZ, et al. Bioorg Med Chem Lett. 2013;23:6755;
(d) Huang R-Z, Liang G-B, Huang X-C, et al. Eur J Med Chem. 2017. https://doi.
org/10.1016/j.ejmech.2017.07.020;
(e) González MA. Eur J Med Chem. 2014;87:834;
(f) Xu H, Liu L, Fan X, Zhang G, Li Y, Jiang B. Bioorg Med Chem Lett. 2017;27:505;
(g) Hou W, Luo Z, Zhang G, et al. Eur J Med Chem. 2017. https://doi.org/10.1016/
j.ejmech.2017.07.049;
(h) Li F-Y, Wang X, Duan W-G, Lin G-S. Molecules. 2017;22:1087;
(i) Gu W, Wang S, Jin X, et al. Molecules. 2017;22:1154;
(j) González MA, Correa-Royero J, Agudelo L, Mesa A, Betancur-Galvis L. Eur J
Med Chem. 2009;44:2468;
(k) Olmo F, Guardia JJ, Marin C, et al. Eur J Med Chem. 2015;89:683;
(l) Talevi A, Cravero MS, Castro EA, Bruno-Blanch LE. Bioorg Med Chem Lett.
2007;17:1684;
(m) Pertino MW, Vega C, Rolón M, Coronel C, De Arias AR, Schmeda-
Hirschmann G. Molecules. 2017;22:1;
(n) Dea-Ayuela MA, Bilbao-Ramos P, Bolás-Fernández F, González-Cardenete
MA. Eur J Med Chem. 2016;121:445;
(o) Wang H, Nguyen TTH, Li S, Liang T, Zhang Y, Li J. Bioorg Med Chem Lett.
2015;25:347;
(p) Gu W, Wang S. Eur J Med Chem. 2010;45:4692;
(q) Gu W, Wu R, Qi S, Gu C, Si F, Chen Z. Molecules. 2012;17:4634;
(r) Gu W, Qiao C, Wang S-F, Hao Y, Miao T-T. Bioorg Med Chem Lett.
2014;24:328;
(s) Liu ML, Pan XY, Yang T, et al. Bioorg Med Chem Lett. 2016;26:5492;
(t) Helfenstein A, Vahermo M, Nawrot DA, et al. Bioorg Med Chem. 2016;25:132;
(u) Zhang WM, Yang T, Pan XY, et al. Eur J Med Chem. 2017;127:917;
(v) Berger M, Roller A, Maulide N. Eur J Med Chem. 2017;12:6937.
6. (a) Xu H, Tang H, Yang Z, Feng H, Li Y. Tetrahedron. 2014;70:3107;
(b) Rafferty RJ, Hicklin RW, Maloof KA, Hergenrother PJ. Angew Chem Int Ed.
2014;53:220;
(c) Xu H, Tang H, Feng H, Li Y. ChemMedChem. 2014;9:290.
7. (a) Xu HT, Tang HY, Feng HJ, Li YC. Eur J Med Chem. 2014;73:46;
(b) Xu HT, Chen Y, Tang HY, Feng HJ, Li YC. Bioorg Med Chem Lett. 2014;24:5671;
(c) Xu H, Fan X, Zhang G, et al. Biomed Pharmacother. 2017;93:1004.
8. (a) Xu HT, Tang HY, Feng HJ, Li YC. J Org Chem. 2014;79:10110;
(b) Xu HT, Tang HY, Feng HJ, Li YC. Tetrahedron Lett. 2014;55:7118.
9. (a) Dinkova-Kostova AT, Massiah MA, Bozak RE, Hicks RJ, Talalay P. Proc Natl
Acad Sci. 2001;98:3404;
(b) Jackson PA, Widen JC, Harki DA, Brummond KM. J Med Chem. 2017;60:839.
10. S.R. Molinspiration Chemoinformatics Brastirlava, Available from: http://
www.molinspiration.com/cgi-bin/properties.
11. Refsgaard HHF, Jensen BF, Brockhoff PB, Padkjaer SB, Guldbrandt M,
Christensen MS. J Med Chem. 2005;48:805.
12. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Adv Drug Del Rev. 2012;64:4.
13. Chen H, Yang Z, Ding C, et al. Eur J Med Chem. 2014;82:195.
14. (a) Clark DE, Pickett SD. Drug Discovery Today. 2000;5:49;
(b) Cardoso MFC, Rodrigues PC, Oliveira MEIM, et al. Eur J Med Chem.
2014;84:708.
15. Desplat V, Vincenzi M, Lucas R, et al. Eur J Med Chem. 2016;113:214.