Synthesis and Biological Studies of (+)-Liquiditerpenoic Acid A (Abietopinoic Acid) and Representative Analogues: SAR Studies

Article abstract of DOI:10.1021/acs.jnatprod.8b00884

The first semisynthesis and biological profiling of the new abietane diterpenoid (+)-liquiditerpenoic acid A (abietopinoic acid) (7) along with several analogues are reported. The compounds were obtained from readily available methyl dehydroabietate (8), which was derived from (-)-abietic acid (1). Biological comparison was conducted according to the different functional groups, leading to some basic structure-activity relationships (SAR). In particular, the ferruginol and sugiol analogues 7 and 10-16 were characterized by the presence of an acetylated phenolic moiety, an oxidized C-7 as a carbonyl, and a different functional group at C-18 (methoxycarbonyl, carboxylic acid, and hydroxymethyl). The biological properties of these compounds were investigated against a panel of six representative human tumor solid cells (A549, HBL-100, HeLa, SW1573, T-47D, and WiDr), five leukemia cellular models (NALM-06, KOPN-8, SUP-B15, UoCB1, and BCR-ABL), and four Leishmania species (L. infantum, L. donovani, L. amazonensis, and L. guyanensis). A molecular docking study pointed out some targets in these Leishmania species. In addition, the ability of the compounds to modulate GABA_A receptors (α₁β₂γ_2s) is also reported. The combined findings indicate that these abietane diterpenoids offer a source of novel bioactive molecules with promising pharmacological properties from cheap chiral-pool building blocks.

Full text of DOI:10.1021/acs.jnatprod.8b00884

Article
pubs.acs.org/jnp
Cite This: J. Nat. Prod. XXXX, XXX, XXX−XXX
Synthesis and Biological Studies of (+)-Liquiditerpenoic Acid A
(Abietopinoic Acid) and Representative Analogues: SAR Studies
†
§
§
‡
Damir Hamulic, Marco Stadler, Steffen Hering, Jose M. Padron, Rachel Bassett,∥ Fatima Rivas,∥
́
́
́
⊥
#
,†
́
Marco A. Loza-Mejía, M. Auxiliadora Dea-Ayuela, and Miguel A. Gonzalez-Cardenete*
†
̀
Instituto de Tecnología Química (UPV-CSIC), Universitat Politecnica de Valencia-Consejo Superior de Investigaciones Científicas,
Avenida de los Naranjos s/n, 46022 Valencia, Spain
§Department of Pharmacology and Toxicology, University of Vienna, Althanstrasse 14, A-1090 Vienna, Austria
‡
́
́
́
BioLab, Instituto Universitario de Bio-Organica “Antonio Gonzalez” (IUBO-AG), Centro de Investigaciones Biomedicas de
Canarias (CIBICAN), Universidad de La Laguna, C/Astrofísico Francisco Sanchez 2, La Laguna 38200, Tenerife, Spain
∥Department of Chemical Biology and Therapeutics, St. Jude Children’s Research Hospital, Memphis, Tennessee 38105, United
States
⊥
́
́
Facultad de Ciencias Químicas, Universidad La Salle Mexico, Avenue Benjamín Franklin 45, Condesa, 06140 Ciudad de Mexico,
Mexico
#Departamento de Farmacia, Universidad CEU Cardenal Herrera, Avenida Seminario s/n, 46113 Moncada (Valencia), Spain
S
* Supporting Information
ABSTRACT: The first semisynthesis and biological profiling of the
new abietane diterpenoid (+)-liquiditerpenoic acid A (abietopinoic
acid) (7) along with several analogues are reported. The compounds
were obtained from readily available methyl dehydroabietate (8), which
was derived from (−)-abietic acid (1). Biological comparison was
conducted according to the different functional groups, leading to some
basic structure−activity relationships (SAR). In particular, the ferruginol
and sugiol analogues 7 and 10−16 were characterized by the presence of
an acetylated phenolic moiety, an oxidized C-7 as a carbonyl, and a
different functional group at C-18 (methoxycarbonyl, carboxylic acid,
and hydroxymethyl). The biological properties of these compounds were investigated against a panel of six representative
human tumor solid cells (A549, HBL-100, HeLa, SW1573, T-47D, and WiDr), five leukemia cellular models (NALM-06,
KOPN-8, SUP-B15, UoCB1, and BCR-ABL), and four Leishmania species (L. infantum, L. donovani, L. amazonensis, and L.
guyanensis). A molecular docking study pointed out some targets in these Leishmania species. In addition, the ability of the
compounds to modulate GABAA receptors (α1β2γ2s) is also reported. The combined findings indicate that these abietane
diterpenoids offer a source of novel bioactive molecules with promising pharmacological properties from cheap chiral-pool
building blocks.
he study of the chemistry of terpenoids started nearly 200
are several ongoing clinical trials on related abietanes, such as
Tyears ago with the analysis of turpentine oil, including the tanshinones and triptolide.6 For example, ferruginol (5), the
simplest phenolic abietane diterpenoid, has recently shown
antitumor activity through inhibiting non-small-cell lung
cancer growth by inducing caspase-associated apoptosis. In
fact, intraperitoneal administration of ferruginol (5) signifi-
cantly suppressed the growth of subcutaneous CLI-5
xenografts.7 Ferruginol (5) also exhibits antiproliferative effects
in human ovarian cancer cells by inducing apoptosis and
inhibiting cancer cell migration.8 In addition, ferruginol (5) has
shown promising antileishmanial activity (IC50 12.2 μM)
against Leishmania donovani promastigotes.9 The related C-7
oxidized sugiol (6, Figure 1) displayed potent anti-inflamma-
tory and hepatoprotective activities,10 in vitro cytotoxicity
investigation of the first resin acid, abietic acid (1, Figure 1)
from pine oleoresin, that was isolated in an impure form in
1824.1 Abietic acid (1) occurs in plants of the genus Abies2 and
is the first member of the abietane-type diterpenoids. This class
is characterized by a tricyclic ring system and has shown a wide
range of chemical diversity and biological activity, including
antitumor and anti-infective properties.3,4 To date, new
abietanes are still being discovered and are of increasing
interest to scientists in multiple disciplines. For example,
medicinal chemists have studied derivatives of the two readily
available materials dehydroabietic acid (2) and dehydroabietyl-
amine (3), also called leelamine.4 To date, there is only one
commercial drug based on abietane-type diterpenoids, namely,
Ecabet (4, ecabet sodium). Ecabet is used for the treatment of
reflux esophagitis and peptic ulcer disease.5 Nevertheless, there
Received: October 22, 2018
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.8b00884
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
are presumed to occur by inhibition of the STAT3 pathway.13
STAT3 is a promising molecular target for the treatment of
various cancers, as it is an upregulated oncogenic protein.
Recently, a series of β-amino alcohol analogues of sugiol (6a)
showed improved biological activity against human solid tumor
cell lines, in comparison to that of the parent sugiol (6).14
The promising biological activity of abietane-type diterpe-
noids, the isolation of a new sugiol analogue, (+)-liquid-
iterpenoic acid A (7) (abietopinoic acid), in 2014 by Hua and
co-workers from the resin of Liquidambar formosana15 and by
Kuo and co-workers from Pinus massoniana16 in two
independent studies, and the lack of synthetic and biological
studies of this molecule prompted the synthesis and assess-
ment of the biological properties of compound 7 and
analogues with different C-18 functional groups and oxidation
pattern at C-7.
Finally, it should be noted that compounds such as 12-
hydroxydehydroabietic acid (15, Scheme 1) have recently
attracted attention after being patented for protein tyrosine
phosphatase 1B (PTP1B) inhibitory activity, following
promising in vitro tests. The potential of compound 15 as
drug in the prevention, delay, or treatment of PTP1B-mediated
diseases such as diabetes and obesity, or treatment of the
complications thereof, thus merits further study.17
Figure 1. Abietane numbering system and some typical examples.
In the present study, the semisynthesis of (+)-liquid-
iterpenoic acid A (7, or abietopinoic acid) and related
analogues (9−16) from methyl dehydroabietate (8), their
evaluation against a cellular panel of six representative human
solid tumors, five leukemia cellular lines, and four Leishmania
species (including molecular docking analysis of potential
biological targets of Leishmania), and their modulating activity
of GABAA receptor subtype (α1β2γ2s) are reported.
against human pancreatic (MIAPaCa-2, IC50 17.9 μM) and
melanoma (MV-3, IC50 34.1 μM) tumor cellular models,11 and
antitumor-promoting activity against EBV-EA activation.12
Additionally, studies of sugiol (6) demonstrated in vivo
antitumor activity in a DU145 prostate xenograft murine
model. The effects of sugiol (6) on tumor growth inhibition
Scheme 1. Synthesis of Liquiditerpenoic Acid A (7) and Related Compounds 9−16 from Methyl Dehydroabietate (8)
B
DOI: 10.1021/acs.jnatprod.8b00884
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 1. Antiproliferative Activity (GI50) of Liquiditerpenoic Acid (7, Abietopinoic Acid) Analogues 9, 10, and 12 against
a
Human Solid Tumor Cells
cell line (origin)
compound
A549 (lung)
HBL-100 (breast)
HeLa (cervix)
SW1573 (lung)
T-47D (breast)
WiDr (colon)
b
8
15 2.6
18 2.3
11 1.8
9.4 2.6
1.5 0.3
4.9 0.2
19 0.2
15 1.2
15 3.9
9.7 2.7
1.4 0.1
1.9 0.2
15 3.0
11 1.0
4.4 1.1
5.8 0.5
3.3 1.6
1.8 0.5
22 6.2
11 4.6
12 3.2
11 2.0
15 1.5
2.7 0.4
16 4.7
10 1.5
14 2.9
19 3.8
22 5.5
10 3.6
17 2.4
16 2.5
23 4.4
23 3.1
23 4.3
9
10
12
etoposide
cisplatin
17 3.3
a
b
Values are given in μM and represent the mean standard deviation of at least two independent experiments. Values taken from ref 22: Stadler,
M.; et al. Planta Med. Int. Open 2017, 4, e89−e92.
THF. Compound 13 was subsequently acetylated (Ac2O in
pyridine), oxidized at C-7 (CrO3 in HOAc), and deacetylated
(K2CO3 in MeOH) to afford the new compound 18-
hydroxysugiol (16) (Scheme 1).
RESULTS AND DISCUSSION
■
Chemistry. The compounds were synthesized from readily
available methyl dehydroabietate (8) (Scheme 1), which was
obtained from commercially available (−)-abietic acid (1),
following the reported method via methylation and aromatiza-
tion.18 The starting material contains the aromatic ring and
three stereocenters (C-4, C-5, C-10), which are present in the
target, (+)-liquiditerpenoic acid A (7, or abietopinoic acid),
including the C-18 carboxylic group. The aromatic diterpenoid
7 also features a phenolic moiety at C-12 and a 7-carbonyl
group, and thus, its semisynthesis can follow a synthetic
approach similar to that developed to generate (−)-sugikurojin
A.19 Based on the previous synthetic studies, a concise
synthesis of compound 10 was performed via Friedel−Crafts
acetylation (AcCl/AlCl3) followed by Baeyer−Villiger oxida-
tion (m-CPBA/TFA in DCM) to introduce the C-12 phenolic
moiety, affording the key intermediate 10 (Scheme 1) in 80%
overall yield for the two steps. Compound 10 was used to
access a series of diverse analogues and the synthesized natural
product 7. First, oxidation (CrO3 in HOAc) at the benzylic C-
7 position afforded ketone 11, which after deacetylation
(K2CO3 in MeOH) resulted in sugiol analogue 14. Compound
14 was sparingly soluble in organic solvents. Nucleophilic
methyl ester cleavage of 14 (LiI in 2,4,6-collidine) afforded the
(+)-liquiditerpenoic acid A (abietopinoic acid) (7) in 34%
overall yield for the five steps. The combined analytical data
[1H and 13C NMR in DMSO-d6 and acetone-d6 and HRMS
(calcd for C20H27O4 [M + H]: 331.1909; found: 331.1887)]
were in complete agreement with the reported data for the
natural product 7 (Scheme 1).15,16 It is important to highlight
that there is a discrepancy in the reported specific rotation.
Biology. Antiproliferative Activity. The structure−activity
relationship (SAR) of compounds 7 and 9−16 was evaluated
to determine the specific role of the introduced functional
groups. Their ability to inhibit the proliferation of the human
solid tumor cell lines A549 (lung), HBL-100 (breast), HeLa
(cervix), SW1573 (lung), T-47D (breast), and WiDr (colon)
and five leukemia cellular models (NALM-06, KOPN-8, SUP-
B15, UoCB1, and BCR-ABL) was evaluated in vitro by using
the established sulforhodamine B (SRB)20 and CellTiter-Glo21
cellular proliferation assays. The significant findings (activity
≤10 μM) are shown in Tables 1 (GI50) and 2 (EC50),
respectively (see Supporting Information for complete table
data).
Table 2. Antiproliferative Activity (EC50) of
Liquiditerpenoic Acid (7, or Abietopinoic Acid) Analogues
a
10 and 11 against Leukemia Cellular Models
cell line
NALM-
06
BCR-
ABL
SUP-
B15
compound
KOPN-8
UoCB1
BJ
10
11
STS
8.8
10.7
1.4
14.3
12.2
0.35
21.0
18.5
0.71
20.2
21.7
0.55
>40
>40
2.0
N.A.
N.A.
>40
a
Values are given in μM, N.A.: no activity at the tested
concentrations. Negative control: vehicle; positive control: staur-
osporine (STS).
Hua et al.15 reported [α]25 +80.3 (c 0.137, MeOH), while
D
The standard anticancer drugs etoposide, cisplatin, and
staurosporine were used for comparison and control. Most of
the compounds were active (GI50 < 25 μM) in the solid tumor
cells with higher activity than the parent dehydroabietic acid
(2) from previous results.22 The target molecule, (+)-liquid-
iterpenoic acid A (7, or abietopinoic acid), its deoxy acid, 12-
hydroxydehydroabietic acid (15), and the sugiol methyl ester
analogue (14) were generally inactive (GI50 > 70 μM) in the
solid tumor cells assays (Table S2, Supporting Information).
The most potent compound in general in solid tumor assays
was compound 10, characterized by an acetate group at C-12
and a methyl ester at C-18. The activity of 10 (4−16 μM) is
comparable to that of cisplatin (a currently approved drug)
and the human topoisomerase IIα catalytic inhibitor, etopo-
side. Compound 10 was more potent than etoposide and
cisplatin for the more resistant cell lines T-47D and WiDr,
indicating potential for future mechanistic studies. Compound
Kuo et al.16 reported [α]25D +21.8 (c 0.5, CHCl3). Clearly, the
difference in solvent systems and concentration may affect the
outcome. Our specific rotation data of synthesized 7 ([α]25
D
+23.8, c 0.4, MeOH) were in agreement with Kuo et al.16
However, when the optical rotation was measured in CHCl3 at
the same concentration reported by Kuo et al. (c 0.5), an
[α]24 +33.0 was obtained. It is likely that the presence of an
D
impurity in their samples affected their data. Our data were
recorded by independent investigators, and the same
conclusions were drawn. As shown by NMR data and
elemental analysis, the purity was high.
Further functional group manipulation of intermediate 10
provided, after deacetylation with K2CO3 in MeOH, phenol
12, which was subsequently treated with LiI in 2,4,6-collidine
to afford compound 15. An alternative approach led directly to
18-hydroxyferruginol (13) via reduction of 10 with LiAlH4 in
C
DOI: 10.1021/acs.jnatprod.8b00884
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 2. Cell death and cell cycle analysis upon compound treatment. Nalm06 cells were incubated with compounds for 24 h. (A) DMSO,
negative control, (B) staurosporine (2 μM), (C) compound 10 (10 μM), (D) compound 11 (10 μM). (E) Cell cycle analysis of compounds 10
and 11.
Figure 3. Cell death induced by compound treatment. KOPN-8 cells were incubated with compounds of interest for 32 h. (A) DMSO, negative
control, (B) staurosporine (1 μM), (C) compound 10 (20 μM), (D) compound 11 (20 μM), respectively.
9 was also more potent against T-47D and WiDr cell lines than
the standard of care drug, cisplatin. One clear SAR that can be
deduced is that acetylation on the hydroxy group at C-12
results in more potent compounds (Table S2, compounds 10
and 11 vs 12 and 14). In general, as found previously in similar
systems,18 the order of activity in the ferruginol series with
different functional groups at C-18 (compounds 12, 13, and
15) was ester ≥ alcohol > acid, which is a trend similar to that
of the C-19 callitrisic-derivative series.22
Previous reports on the antiproliferative activity of sugiol β-
amino alcohol analogues (6a, Figure 1)14 provided similar
cytotoxicity results against A2780, WiDr, and SW1573 cancer
cells after 48 h of exposure using the SRB assay.14 An
important difference in the β-amino alcohol analogues is the
presence of a gem-dimethyl group at C-4 in contrast with the
C-18-functionalized position in the current study. 18-
Hydroxyferruginol (13) isolated from Juniperus brevifolia has
also been described previously as a moderate cytotoxic agent
against HeLa cells.23
showed similar response to the solid tumor models, particularly
for compounds 10 and 11. Not surprisingly, compounds
showed no cytotoxicity at tested concentrations in the
glucocorticoid-resistant leukemia cell line (UoC-B1). The
acetylated compounds 10 and 11 showed the highest potency
against the NALM-06 cellular model, with EC50 values in the
low micromolar range (8.8 and 10.7 μM, respectively). These
compounds displayed no cytotoxicity against the normal
fibroblast BJ cellular model at the tested concentrations,
indicating a favorable therapeutic index (see Supporting
Information for experimental details).
Our results indicate that compounds 10 and 11 have
superior antiproliferative properties to the corresponding free
hydroxyl/acid groups, as the non-acetylated sugiol analogues 7,
14, and 16 displayed either no or little activity in the leukemia
lines tested (EC50 > 40 μM) (Table S1, Supporting
Information), confirming that the absence of the keto group
at C-7 was required for antileukemic activity.
Interestingly, the acetyl intermediate 9 (EC50 > 40 μM) and
compounds 13 and 16 with a hydroxy group at C-18 showed
poor activity in these leukemia cell lines (Table S1, Supporting
Information), in contrast to the antiproliferative activity against
Next, these compounds were evaluated against human pre-B
leukemia cellular models (NALM-06, KOPN-8, SUP-B15,
UoCB1) and murine cellular model (BCR-ABL), which
D
DOI: 10.1021/acs.jnatprod.8b00884
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 4. (A) Potentiation of GABA-evoked currents (IGABA) through α1β2γ2S receptors by indicated compounds at 10 and 100 μM. Data
represent mean SEM from at least three different oocytes and two oocyte batches. (B) Representative current traces of DHA (2), 12, and 14 in
the presence of 20 s application of a GABA EC5 (single bar) or coapplication of GABA EC5 (4−5 μM) and 10 or 100 μM compound (double bar).
Currents were recorded from Xenopus laevis oocytes expressing α1β2γ2S receptors voltage-clamped at −70 mV. GABA EC5 was determined
individually for every oocyte.
the solid tumor cell lines. Such experimental observations
illustrate the fact that different survival pathways drive solid
and hematological malignancies.
Table 3. Potentiation of IGABA in α1β2γ2S Receptors by
Compounds 7 and 9−16
compound
conc (μM)
IGABA potentiation (%)
SEM
To validate apoptosis (or programmed cell death) for
compounds 10 and 11, the appearance of phosphatidyl serine
residues on the surface of the cell, an early event in apoptosis,
was determined with annexin V conjugated to fluorescein
isothiocyanate (FITC) and propidium iodide (PI) via flow
cytometry.24 In addition DNA content was analyzed (by PI
staining) for cell cycle progression25 in Nalm06 and KOPN8
cellular models for 24 and 32 h, respectively. Representative
annexin V-FITC/PI plot images of independent experiments
are shown in Figure 2. For controls, DMSO (negative control)
and staurosporine (2 μM; positive control) are indicated along
with compounds 10 (10.0 μM) and 11 (10.0 μM) for Nalm06
(Figure 2). Fluorescent images show annexin V-FITC-positive
apoptotic cells and the PI-positive necrotic cells. No significant
cell death was observed at the tested concentration for
compound 10 (Figure 2), while compound 11 showed
progression of cell death in a dose-dependent manner (see
Supporting Information). However, both compounds indicated
cell arrest at G0/G1 and G2/M as illustrated in diagram E. We
hypothesized that these compounds display cytotoxicity at
longer incubation times in the NALM-06 cellular model.
Compounds 10 and 11 were evaluated at higher
concentrations (20 μM) and showed a profound effect on
cell death in the KOPN8 cellular model with more than 85% of
cells being in the late apoptotic state (Figure 3). Cells treated
with compound 10 displayed a similar distribution pattern as
staurosporine, while compound 11 induced significant cell
arrest at the S phase. The combined data illustrate that
compound 11 might be acting differently on these cellular
models as they have different genomic backgrounds.
DHA (2)
10
100
10
100
10
100
10
100
10
100
10
100
10
100
10
100
10
186.9
841.3
−2.0
−21.1
2.7
9.6
52.9
4.4
7.1
0.9
5.1
2.1
8.9
2.8
7
9
−2.9
5.1
10
11
12
13
14
15
16
26.3
−1.0
46.7
23.7
378.0
29.3
154.1
37.4
242.4
8.9
8.3
9.6
65.6
10.6
34.2
12.7
46.0
2.3
2.9
5.8
9.5
100
10
100
12.1
4.5
−30.3
enhancement of a GABA-evoked chloride current (IGABA),
which elicits 5% of the maximum response (∼5 μM GABA).
Compounds 12−14, analogues of ferruginol (5) and sugiol
(6), showed considerable potentiation of GABA-evoked
currents at 100 μM. Interestingly, compounds 7 and 16 do
not potentiate, but act as weak inhibitors. Basic SAR shows
that the most potent compound is phenol 12 (IGABA 378% at
100 μM), bearing a methyl ester group at C-18. Replacement
of the methyl ester by a hydroxymethyl group (compound 13)
or the introduction of a keto group at C-7 (compound 14) led
to less active compounds. Specifically, half of the activity
(compound 13) or two-thirds (compound 14) of the parent
activity (compound 12) was provoked doing these functional
group changes. Acetylation at C-12 (compounds 10 and 11),
as well as the presence of a carboxylic acid at C-18
(compounds 7 and 15), disfavored GABAA potentiation. In
all cases, the activity was considerably lower in comparison to
the positive control, DHA (2).
GABAA Receptor Modulating Activity. In an experiment
based on the oocyte assay, the compounds 7 and 9−16 were
tested for their effects on the most abundant GABAA receptor
subtype (α1β2γ2S) by means of the two-microelectrode
voltage clamp technique in Xenopus laevis oocytes.26 All
compounds were screened at concentrations of 10 and 100 μM
and compared with dehydroabietic acid (DHA) (2) as a
positive control.26 The results are summarized in Figure 4 as a
bar diagram and tabulated in Table 3. Depicted is the
E
DOI: 10.1021/acs.jnatprod.8b00884
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 4. IC50 Leishmanicidal and Cytotoxic Effects (in μM) of C-7- and C-12-Functionalized Dehydroabietic Acid Derivatives
in an in Vitro Promastigote Assay
L. amazonensis
L. guyanensis
IC50 SD
L. donovani
IC50 SD
L. infantum
IC50 SD
macrophages J774
a
b
c
compound
IC50
SD
SI
SI
SI
SI
CC50
d
7
9
10
11
12
13
14
15
NA
NA
NA
NA
219.4 15.0
129.6 8.9
26.5 2.7
17.8 2.8
34.2 3.1
23.4 4.6
153.9 49.3
85.8 28.8
14.4 2.8
136.4 1.4
11.6 0.6
3.2 0.3
0.65 0.03
3.9 0.1
7.7 0.8
11.2
8.2
27.4
8.7
14.2 0.4
5.5 0.7
1.3 0.1
5.9 1.1
12.5 2.2
ND
71.4 1.5
16.5 0.7
18.2 0.6
9.1
4.8
13.6
5.8
14.8 0.9
8.1 0.7
4.4 0.5
9.21 0.06
13.9 0.1
ND
58.1 0.7
19.7 0.4
0.15 0.02
8.8
3.3
4.1
3.7
1.7
2.5 0.6
1.3 0.1
0.7 0.2
5.0 0.3
0.7 0.1
ND
43.6 3.4
9.7 0.8
3.4 0.6
51.8
20.4
25.4
6.8
3.0
1.9
33.4
e
ND
54.1 6.5
10.2 0.6
47.7 5.0
1.6
1.4
2.9
1.2
0.9
7.5
1.5
0.7
909
2.0
1.5
40.1
16
miltefosine
a
b
IC50, concentration of the compound that produced a 50% reduction in parasites; SD: standard deviation. Selectivity index, SI = CC50/IC50.
c
d
CC50, concentration of the compound that produced a 50% reduction of cell viability in treated culture cells with respect to untreated ones. NA,
e
no activity. ND, not determined.
According to the literature, DHA (2) induced significant
receptor modulation in the oocyte assay, with a maximal
potentiation of IGABA of 397.5 34.0% and EC50 of 8.7 1.3
μM. This was the first report of dehydroabietic acid as a
positive GABAA receptor modulator.26 Also, dehydroabietic
acid (2) exhibits large-conductance voltage- and Ca2+-activated
K+ (BK channels) channel-opening activities. However, the
structurally related 12-hydroxydehydroabietic acid (15)
showed no significant activity by other authors, concluding
that hydroxyl groups at C-12 are detrimental for BK channel-
opening activities.27
Table 5. IC50 Leishmanicidal and Cytotoxic Effects (in μM)
of Compound 9 in a in Vitro Amastigote Assay on L.
amazonensis and L. infantum
L. amazonensis
a
L. infantum
IC50 SD
37.2 2.4
macrophages J774
b
c
IC50
SD
SI
4.1
SI
CC50
31.4 6.2
3.5
129.6 8.9
a
IC50, concentration of the compound that produced a 50% reduction
b
in parasites; SD: standard deviation. Selectivity index, SI = CC50/
IC50. CC50, concentration of the compound that produced a 50%
c
reduction of cell viability in treated culture cells with respect to
untreated ones.
Antileishmanial Activity. Compounds 7 and 9−16
(Scheme 1) were evaluated for antileishmanial activity against
four different Leishmania species (L. infantum, L. donovani, L.
amazonensis, and L. guyanensis), as well as for cytotoxicity
against J774 macrophages following established procedures.28
The results on extracellular forms (promastigotes) are
summarized in Table 4. All the compounds were active against
the different species of Leishmania except compound 7. In the
case of L. donovani no compound improved the activity of the
reference drug, miltefosine. However, several compounds
proved to be more potent than the reference compound for
other species. For example, compounds 9−13 and 16
displayed better potency than miltefosine against L.
amazonensis. Interestingly, compound 11 exhibited a sub-
micromolar IC50, being 73 times more potent than the
reference drug. In the case of L. guyanensis, compounds 10−12
showed lower IC50 values than miltefosine, while compounds
9, 13, and 16 showed a similar IC50 value to that of miltefosine.
Again, compound 11 was the most potent (13-fold more
potent than miltefosine). With regard to L. infantum,
compounds 9−11 and 13 were more potent than miltefosine,
with compound 11 being the most potent as well for this
species. Thus, compound 11 resulted to be the most potent of
all tested compounds. The toxicity of compound 11, however,
was not negligible but showed a good selectivity index (SI) in
the range 4.1−27.4. Moreover, the best balance of activity−
selectivity was exhibited by compound 9, which displayed a
good level of antileishmanial activity with a very valuable SI
(between 8.76 and 52.49) associated with a reduced toxicity.
Therefore, compound 9 was selected for studying its activity
against amastigotes (L. amazonensis and L. infantum) (Table
5).
Ferruginol (5) isolated from Juniperus berries displayed
moderate antiparasitic activities (IC50 12.2 μM) against
Leishmania donovani promastigotes, as well as anti-Plasmodium
falciparum chloroquine-sensitive (D6; IC50 14.7 μM) and
chloroquine-resistant (W2; IC50 12.2 μM) strains.9 The
selected analogues, 7 and 9−16, also showed important
activity against L. donovani, with the sugiol analogue 11 (IC50
4.38 μM) 3 times more potent than the isolated ferruginol (5)
from Juniperus berries. The methyl ester of 12-hydroxydehy-
droabietic acid, compound 12, recently described as a new
natural product,29 has demonstrated activity against Trypano-
soma cruzi in epimastigote and axenic amastigote forms (IC50
ca. 6.0 mM).30 In our studies, compound 12 exhibited
promising antileishmanial activity with IC50 values ranging
from 3.9 to 9.21 μM and SI from 3.7 to 8.7.
Molecular Docking for Leishmania spp. Targets. To
propose a plausible mechanism of action of these compounds,
we carried out molecular docking studies on some enzymes
described as potential targets for antileishmanials (see
Supporting Information for the complete list of the analyzed
enzymes and their docking results). This methodology has
been used previously for the study of potential antiparasitics
including Leishmania31,32 and Trypanosoma.33 Docking studies
suggested UDP-glucose pyrophosphorylase, dihydroorotate
dehydrogenase, methionyl-tRNA synthetase, pteridine reduc-
tase, phosphomannomutase, and N-myristoyltransferase as
potential targets, since compounds 7−16 exhibited the highest
theoretical affinity on these enzymes. Interestingly, docking
scores on UDP-glucose pyrophosphorylase have relatively
good correlation with in vitro leishmanicidal activities in three
of the four parasites studied (r2 > 0.71), considering
F
DOI: 10.1021/acs.jnatprod.8b00884
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 6. Results of Docking Studies Carried Out on Potential Leishmanicidal Targets and Correlation with in Vitro Activity
a
(log(1/IC50)) on L. donovani
compound
log(1/IC50)
UDPGP
DHODH
MtRNAS
PtRd
S14DM
NMyrT
9
−1.1700
−0.9096
−0.6415
−0.9643
−1.1440
−1.7642
−1.2945
−87.90
−76.67
−103.75
−105.01
−82.91
−79.12
−80.21
−79.95
−88.37
0.7335
−87.73
−92.73
−96.99
−85.14
−81.21
−86.49
−86.38
−93.86
0.5990
−89.10
−97.45
−101.41
−90.39
−90.12
−97.46
−94.69
−84.49
0.4957
−84.99
−92.25
−89.75
−79.21
−85.58
−83.42
−75.17
−78.59
0.4330
−78.54
−89.16
−83.16
−75.03
−72.18
−72.76
−73.45
−83.50
0.4172
−81.78
−91.23
−91.38
−76.87
−77.08
−80.38
−82.68
10
11
12
13
15
16
average docking score
b
r2
0.3933
a
UDPGP = UDP-glucose pyrophosphorylase (PDB: 2OEG), DHODH = dihydroorotate dehydrogenase (PDB: 3MHU), MtRNAS = methionyl-
tRNA synthetase (PDB: 3KFL), PtRd = pteridine reductase (PDB: 1E7W), S14DM = sterol 14-alpha demethylase (PDB: 3L4D), NMyrT = N-
b
myristoyltransferase (PDB: 4A30). Considering compound 15 as outlier.
Figure 5. 2D (generated with Maestro35) and 3D (generated with Yasara View36) representations of predicted complexes of compounds 11 (a and
c) and 9 (b and d) in the UDPGP active site. The O-acetyl group in compound 11 interacts via hydrogen bonding with Gly83 and Gly84,
increasing theoretical affinity.
compound 15 as an outlier (Table 6). This latter compound
has an ionizable carboxylic group, which could interfere with
the access of the compound to the active site due to poor
membrane permeability, particularly in the case of dihydroo-
rate dehydrogenase, which is located inside the mitochon-
dria.34
Analysis of the interactions of compounds 9−16 in the
active site of UDPGP helps to explain our SAR observations.
Particularly, interactions with Gly83 and Gly84 seem critical
for a better binding, as the most active compounds (10 and
11) have an O-acetyl group in position 12 that interacts
through hydrogen bonds with these residues, while these
interactions are absent in the case of compounds that bear a
hydroxy or acetyl group in this position (Figure 5).
In Silico Simulations. The structures of tested compounds
were manually drawn in ChemDraw Professional 15.0
software, and the SMILES notation was obtained for each
molecule. To calculate the parameters of Lipinski’s rule of five
G
DOI: 10.1021/acs.jnatprod.8b00884
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(see Table 7), the SMILES notation in Molinspiration online
software (available from http://www.molinspiration.com/cgi-
carried out on the analyzed targets and SMILES codes
(PDF)
Table 7. Calculated Molecular Properties (Drug-likeness)
by Molinspiration Online Software for Compounds 7 and
9−16
AUTHOR INFORMATION
■
a
Corresponding Author
́
*E-mail (M. A. Gonzalez Cardenete): migoncar@itq.upv.es.
n-
n-
Lipinski’s
violation
Tel: +34 96 3877810. Fax: +34 96 3879444.
compound miLog P
MW
HBA HBD TPSA
ORCID
7
9
4.21
5.64
5.29
4.39
5.73
5.24
4.83
5.11
4.33
330.42
356.51
372.50
386.49
330.47
302.46
344.45
316.44
316.44
<500
4
3
4
5
3
2
4
3
2
0
0
0
1
2
1
2
74.60
43.38
52.61
69.68
46.53
40.46
63.60
57.53
57.53
0
1
1
0
1
1
0
1
0
Marco Stadler: 0000-0003-4489-6382
́
́
Jose M. Padron: 0000-0002-7488-1774
Fatima Rivas: 0000-0003-3643-2035
10
11
12
13
14
15
16
́
Miguel A. Gonzalez-Cardenete: 0000-0002-8762-0426
Notes
The authors declare no competing financial interest.
3
not
2
not
ACKNOWLEDGMENTS
■
rule of five not >5
1 violation
allowed
>10
>5
Financial support by the Spanish Government [Consejo
Superior de Investigaciones Cientificas (201680I008)] is
a
́
Values were calculated using Molinspiration Cheminformatics
software (Molinspiration, Slovensky Grob, Slovak Republic, 2015,
http://www.molinspiration.com). P = partition coefficient; MW =
molecular weight; n-HBA = number of hydrogen bond accepting
groups; n-HBD = number of hydrogen bond donating groups; TPSA
= total polar surface area.
gratefully acknowledged. M.S. thanks the support by the
doctoral program “Molecular Drug Targets” (Austrian Science
Fund FWF W 1232). F.R. thanks the American Lebanese
Syrian Associated Charities (ALSAC). M.A.D.-A. thanks the
Santander Bank for the support for her project in consolidable
groups of CEU-UCH.
bin/properties, accessed on 04/17/2018) was employed. In
view of the concern raised recently about pan assay
interference compounds (PAINS) or colloidal aggregators,37
we have also checked in several publicly available databases
such as http://zinc15.docking.org/patterns/home, http://
advisor.docking.org/, http://fafdrugs3.mti.univ-paris-diderot.
fr/, and all of our compounds were fine in these in silico tests.
In summary, we have demonstrated the first chemical
synthesis of the abietane-type natural product (+)-liquid-
iterpenoic acid A (7, or abietopinoic acid). It is accessed in a
short synthetic sequence of five steps in 34% overall yield from
the readily available methyl dehydroabietate (8). Methyl
dehydroabietate (8) is easily obtained from chiral (−)-abietic
acid, a building block available for ∼$0.4 USD/gram of purity
higher than 80%.38 During this process, parallel routes led to a
number of oxidized analogues, some either naturally occurring
(13 and 15) or new (16). Their availability allowed a
biological profiling of several analogues as well as of the target
compound. This work definitively establishes a versatile
platform for the synthesis of a number of bioactive abietanes
with the possibility of accessing even further oxidized,
unnatural analogues of these diterpenoids using similar
chemistry. The combined findings indicate that these
abietane-diterpenoid natural product analogues offer a source
of novel bioactive molecules with promising pharmacological
and drug-likeness properties.
REFERENCES
■
(1) Bhat, S. V.; Nagasampagi, B. A.; Sivakumar, M. Terpenoids. In
Chemistry of Natural Products; Springer: Berlin; 2005; pp 131−139.
(2) Talapatra, S. K.; Talapatra, B. Diterpenoids (C20). In Chemistry
of Plant Natural Products; Springer: Berlin, 2015; pp 478−490.
(3) For a review on this topic, see: Gonzalez, M. A. Nat. Prod. Rep.
2015, 32, 684−704.
(4) For a review on this topic, see: Gonzalez, M. A. Eur. J. Med.
́
́
Chem. 2014, 87, 834−842.
(5) Xu, Q. A.; Madden, T. L. In Ecabet sodium, Chapter 5. In
Analytical Methods for Therapeutic Drug Monitoring and Toxicology;
Wiley: NJ, 2011; pp 183−207, DOI: 10.1002/9780470909799.ch5.
(6) Accessed www.ClinicalTrials.gov on March 30, 2018.
(7) Ho, S.-T; Tung, Y.-T.; Kuo, Y.-H.; Lin, C.-C.; Wu, J.-H. Integr.
Cancer Ther. 2015, 14, 86−97.
(8) Xiong, W.-D.; Gong, J.; Xing, C. Mol. Med. Rep. 2017, 16, 7013−
7017.
(9) Samoylenko, V.; Dunbar, D. C.; Gafur, M. A.; Khan, S. I.; Ross,
S. A.; Mossa, J. S.; El-Feraly, F. S.; Tekwani, B. L.; Bosselaers, J.;
Muhammad, I. Phytother. Res. 2008, 22, 1570−1576.
(10) Shyur, L.-F.; Huang, C.-C.; Lo, C.-P.; Chiu, C.-Y.; Chen, Y.-P.;
Wang, S.-Y.; Chang, S.-T. Phytochemistry 2008, 69, 1348−1358.
(11) Fronza, M.; Murillo, R.; Slusarczyk, S.; Adams, M.;
Hamburguer, M.; Heinzmann, B.; Laufer, S.; Merfort, I. Bioorg.
Med. Chem. 2011, 19, 4876−4881.
(12) Iwamoto, M.; Minami, T.; Tokuda, H.; Ohtsu, H.; Tanaka, R.
Planta Med. 2003, 69, 69−72.
(13) Jung, S.-N.; Shin, D.-S.; Kim, H.-N.; Jeon, Y. J.; Yun, J.; Lee, Y.-
J.; Kang, J. S.; Han, D. C.; Kwon, B.-M. Biochem. Pharmacol. 2015, 97,
38−50.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.8b00884.
■
S
́
́
́
́
(14) Cordova, I.; Leon, L. G.; Leon, F.; San Andres, L.; Luis, J. G.;
́
Padron, J. M. Eur. J. Med. Chem. 2006, 41, 1327−1332.
(15) Shang, H.-J.; Li, D.-Y.; Wang, W.-J.; Li, Z.-L.; Hua, H.-M. Nat.
Prod. Res. 2014, 28, 1−6.
(16) Mohamed, H. A.; Hsieh, C.-L.; Hsu, C.; Kuo, C.-C.; Chang, H.-
S.; Lee, C.-K.; Lee, T.-H.; Wu, J.-B.; Chang, C.-I.; Kuo, Y.-H. Helv.
Chim. Acta 2014, 97, 1146−1151.
Complete experimental details for the synthesis of
1
compounds 7 and 9−16, including copies of H NMR
and 13C NMR spectra, and for all the biological assays;
list with the complete results of the docking studies
(17) Hu, J.; Hu, C.; Xiong, J.; Yang, G. Patent number CN
107021942, application number CN 2016−10068135, 2017.
H
DOI: 10.1021/acs.jnatprod.8b00884
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
́
(18) Gonzalez, M. A.; Perez-Guaita, D.; Correa-Royero, J.; Zapata,
B.; Agudelo, L.; Mesa-Arango, A.; Betancur-Galvis, L. Eur. J. Med.
Chem. 2010, 45, 811−816.
́
́
(19) Gonzalez-Cardenete, M. A.; Zaragoza, R. J. Synthesis 2017, 49,
1315−1318.
(20) Monks, A.; Scudiero, D.; Skehan, P.; Shoemaker, R.; Paull, K.;
Vistica, D.; Hose, C.; Langley, J.; Cronise, P.; Vaigro-Wolff, A.; Gray-
Goodrich, M.; Campbell, H.; Boyd, M. J. Natl. Cancer Inst. 1991, 83,
757−766.
(21) Mitachi, K.; Salinas, Y. G.; Connelly, M.; Jensen, N.; Ling, T.;
Rivas, F. Bioorg. Med. Chem. Lett. 2012, 22, 4536−4539.
́
́
(22) Stadler, M.; Padron, J. M.; Gonzalez-Cardenete, M. A. Planta
Med. Int. Open 2017, 4, e89−e92.
(23) Moujir, L.; Seca, A. M. L.; Silva, A. M. S.; Barreto, M. C. Planta
Med. 2008, 74, 751−753.
(24) Gautam, L. N.; Ling, T.; Lang, W.; Rivas, F. Eur. J. Med. Chem.
2016, 113, 75−80.
(25) Ling, T.; Lang, W.; Feng, X.; Das, S.; Maier, J.; Jeffries, C.;
Shelat, A.; Rivas, F. Eur. J. Med. Chem. 2018, 146, 501−510.
̈
(26) Rueda, D. C.; Raith, M.; De Mieri, M.; Schoffmann, A.; Hering,
S.; Hamburger, M. Fitoterapia 2014, 99, 28−34.
(27) Cui, Y.-M.; Yasutomi, E.; Otani, Y.; Yoshinaga, T.; Ido, K.;
Sawada, K.; Ohwada, T. Bioorg. Med. Chem. Lett. 2008, 18, 5201−
5205.
́
́
(28) Dea-Ayuela, M. A.; Bilbao-Ramos, P.; Bolas-Fernandez, F.;
́
Gonzalez-Cardenete, M. A. Eur. J. Med. Chem. 2016, 121, 445−450.
(29) Kinouchi, Y.; Ohtsu, H.; Tokuda, H.; Nishino, H.; Matsunaga,
S.; Tanaka, R. J. Nat. Prod. 2000, 63, 817−820.
(30) Ramírez-Macías, I.; Marín, C.; Chahboun, R.; Messouri, I.;
́
Olmo, F.; Rosales, M. J.; Gutierrez-Sanchez, R.; Alvarez-Manzaneda,
́
E.; Sanchez-Moreno, M. Am. J. Trop. Med. Hyg. 2012, 87, 481−488.
(31) Ogungbe, I. V.; Erwin, W. R.; Setzer, W. N. J. Mol. Graphics
Modell. 2014, 48, 105−117.
(32) Ogungbe, I.; Setzer, W. Molecules 2013, 18, 7761−7847.
(33) Ogungbe, I.; Setzer, W. Molecules 2009, 14, 1513−1536.
(34) Arakaki, T. L.; et al. Mol. Microbiol. 2008, 68, 37−50.
̈
(35) Schrodinger LCC. Maestro. Available at https://www.
schrodinger.com/maestro.
(36) YASARA Biosciences GmbH. Yasara View. Available at http://
yasara.org/viewdl.htm.
(37) Aldrich, C.; Bertozzi, C.; et al. ACS Cent. Sci. 2017, 3, 143−147.
(38) Available from TCI Europe at http://www.tcichemicals.com/
en/eu/.
I
DOI: 10.1021/acs.jnatprod.8b00884
J. Nat. Prod. XXXX, XXX, XXX−XXX