Synthesis and in vitro evaluation of ferutinol aryl esters for estrogenic activity and affinity toward cannabinoid receptors

Article abstract of DOI:10.1007/s00044-015-1319-7

Ferutinin (1), the major constituent of Ferula hermonis and other Ferula species, is a sesquiterpene ester with remarkable estrogenic activity, beside other valuable medicinal properties. To investigate the influence of chemical modification of the feruti

Full text of DOI:10.1007/s00044-015-1319-7

MEDICINAL
CHEMISTRY
RESEARCH
Med Chem Res
DOI 10.1007/s00044-015-1319-7
ORIGINAL RESEARCH
Synthesis and in vitro evaluation of ferutinol aryl esters
for estrogenic activity and affinity toward cannabinoid receptors
•
Ahmed M. Galal Osman Ehab A. Abourashed Desmond Slade
•
•
•
•
•
Safwat A. Ahmed Waseem Gul Shabana I. Khan Tarek Abo Elfadl
•
•
Olivia R. Dale Afeef S. Husni Stephen J. Cutler Mahmoud A. ElSohly
•
•
Received: 12 August 2014 / Accepted: 26 December 2014
Ó Springer Science+Business Media New York 2015
Abstract Ferutinin (1), the major constituent of Ferula
hermonis and other Ferula species, is a sesquiterpene
ester with remarkable estrogenic activity, beside other
valuable medicinal properties. To investigate the influ-
ence of chemical modification of the ferutinin structure
on its estrogenic effect and binding affinity toward the
cannabinoid CB1 and CB2 receptors, twelve derivatives
of 1 were prepared and evaluated in vitro, together with
the parent compound, for the respective bioactivities,
based on the recent evidence for estrogen–endocannabi-
noid interaction. Nine of the prepared derivatives (3–11)
are new semisynthetic esters of 1. The parent compound
ferutinin (1) exhibited the highest level of estrogenic
activity (EC50 0.3 lM and a percent maximal 17b-
estradiol response of 90 % at 1 lM). Compound 6 was
found to be a selective agonist for CB2 receptor (EC50
0.051 lM, Ki 0.025 lM), with much less affinity for
CB1 receptor (EC50 97 lM, Ki 48.5 lM). Compound 8
was a selective agonist for CB1 (EC50 62, Ki 0.031 lM)
with no affinity toward CB2.
Keywords Ferutinin ꢀ Ferula hermonis ꢀ
Semi-synthetic derivatives ꢀ Estrogenic activity ꢀ
Cannabinoid receptors
T. A. Elfadl
Department of Pharmaceutical Chemistry, Asuit University,
Asuit, Egypt
A. M. G. Osman (&)
National Center for Natural Products Research, School of
Pharmacy, The University of Mississippi, University,
MS 38677, USA
O. R. Dale
National Center for Natural Products Research, The University
of Mississippi, University, MS 38677, USA
e-mail: amgalalv@olemiss.edu
E. A. Abourashed
Department of Pharmaceutical Sciences, Chicago State
University, Chicago, IL 60628, USA
A. S. Husni ꢀ S. J. Cutler
Division of Medicinal Chemistry, Department of Biomolecular
Science, University of Mississippi, University, MS 38677, USA
D. Slade
MRI Global, 425 Volker Boulevard, Kansas City, MO, USA
M. A. ElSohly (&)
Department of Pharmaceutics, University of Mississippi,
University, MS 38677, USA
e-mail: melsohly@olemiss.edu
S. A. Ahmed
Department of Pharmacognosy, Faculty of Pharmacy, Suez
Canal University, Ismailia 41522, Egypt
W. Gul ꢀ M. A. ElSohly
ElSohly Laboratories, Inc., 5 Industrial Park Drive, Oxford,
MS 38655, USA
S. I. Khan
National Center for Natural Products Research and Department
of Biomolecular Science, School of Pharmacy, The University of
Mississippi, University, MS 38677, USA
123

Med Chem Res
Introduction
and may, thus, have a modulating effect on colon cancer cell
growth (CB1) or osteoblast differentiation (CB2) (Proto et al.,
2012), (Sophocleous et al., 2011). Ligands for CB1 and CB2
receptors have emerged as lead compounds in developing
therapeutic agents for a number of health disorders, such as
diabetes, drug dependence, obesity, neurodegenerative dis-
eases, including multiple sclerosis (Diaz et al., 2008; Grein-
eisen and Turner, 2010; Guindon and Hohmann, 2008;
Malfitano et al., 2013; Pryce et al., 2014; Ribeiro et al., 2013;
Rivers and Ashton, 2010). The recent discovery of b-caryo-
phyllene as a non-cannabinoid CB2 selective agonist
(Ki = 0.155 lM) of plant origin warrants further investiga-
tion to discover natural products with better or comparable
activity (Gertsch et al., 2010). Based on the above-mentioned
literature data, in particular the recently discovered relation-
ship between estrogens and the endocannabinoid system, the
goal of this work was to prepare a number of ferutinin deriv-
atives, establish their identity via spectroscopic analysis, pri-
marily 1Dand 2DNMR, in additionto UV and HRESMS, and
examine their estrogenic activity as well as their interaction
with the cannabinoid receptors. The structural modification of
ferutinin (1) was mainly targeting the aromatic core, since this
part of the molecule has been proven to significantly con-
tribute to the estrogenic activity (Rasulev et al., 2007) and
might also affect the affinity toward the cannabinoid recep-
tors. Thus, compounds 3–9 (Scheme 1) were prepared with p-
substituents conceivably capable of forming stronger hydro-
gen or halogen bonds with the estrogen/cannabinoid receptors
in lieu of the p-hydroxyl group, which would presumably
enhance their effect. Compound 10 (Scheme 2) was prepared
to evaluate the effect of introducing a multifunctional aro-
matic core, capable of forming hydrogen bonds, in an attempt
to enhance the binding affinity. Compounds 11 and 12
(Scheme 2) were also prepared to introduce oxygen-carrying
Ferutinin (1), the major constituent of Ferula hermonis Boiss.
(family Apiaceae), is known to possess antioxidant (Ibraheim
et al., 2012), acetylcholinesterase inhibition (Dall’Acqua
et al., 2010), nitric oxide synthase enhancing activities (Col-
man-Saizarbitoria et al., 2006), and in vitro induction of
apoptosis in human RBCs (Gao et al., 2013). Ferutinin also
binds to estrogen receptors (ER) in vitro, acting as an agonist
for ERa and an agonist/antagonist for ERb, hence displaying
estrogenic and antiestrogenic activity, respectively (Appen-
dino et al., 2004; Ikeda et al., 2002; Zanoli et al., 2005). The
in vivo evaluation of ferutinin on sexual behavior of male rats
revealed that it is able to stimulate sexualbehavior in impotent
male rats after acute ingestion. It also exerted a negative
influence on the sexual capacity of potent male rats by
reducing testosterone serum levels as well as negatively
affecting appetitive and consummatory sexual behavior
(Hadidi et al., 2003; Zanoli et al., 2003; Zanoli et al., 2005).
The effect of ferutinin on sexual behavior of female rats
indicated that it did not have the capacity to alter sexual
motivation. However, it significantly inhibited female recep-
tivity, suggesting an important role for ferutinin in sexual
behavior impairment of hormone-primed female rats (Zanoli
et al., 2009; Zavatti et al., 2006). In addition, ferutinin was
demonstrated to possess antiosteoporotic effect as potent as
estradiol benzoate on bone mass, and in a number of cases, it
was reported to be superior to the human estrogen (Palumbo
et al., 2009). A published molecular modeling and QSAR
studies established that the estrogenic activity of ferutinin is
attributed to the presence of the p-hydroxybenzoyl moiety as
well as substitution of the daucane double bond (Rasulev
et al., 2007). Recent reports indicated that 17b-estradiol
interacts with the endocannabinoid receptors (CB1 and CB2)
14
10
2
9
a
b
1
3
15
4`
8
5
4
6
H
O
7
3`
H
O
H
O
H
H
H
5`
6`
H
O
O
H
O
O
12
11
R
1`
13
2`
7`
O
O
1
2
=
R
R1
=
=
=
r
=
3 R
B
I
F
6 R
Cl
,
,
1
4 R1
7 R
=
=
N
C
,
,
5 R1
8 R 1
F
C
3
,
,
1
1
=
N
9 R
,
Scheme 1 Reagents and conditions: a KOH, reflux, 2 h; b toluene, acid chloride, Et3N, DMAP, reflux
123

Med Chem Res
a
H
O
O
N
2
H
O
O
N
10
H
OO
O
+
b
1
H
O
H
O
H
H
H
O
H
O
O
O
O
O
11
12
c
1
H
O
H
H
O
O
O
13
Scheme 2 Reagents and conditions: a CH2Cl2, nalidixic acid, 0 °C, Et3N, ethylchloroformate, overnight stirring; b CH2Cl2/MeOH, meso-
tetraphenylporphine, O2, light, 13 °C, 3 h; c Pd(C)/MeOH, 24 h
functionalities in the daucane core and examining their effect
on receptor binding affinity. Compound 13 (Scheme 2) was
hitherto prepared and tested for estrogenic (Appendino et al.,
2004), but not for binding affinity at the cannabinoid CB1 and
CB2 receptors. The daucane alcohol (hydrolysis product of 1,
ferutinol or jaeschkeanadiol 2) was also evaluated along with
the other derivatives.
and biological evaluation of these new compounds. Com-
pound 10 was synthesized using nalidixic acid and ethyl-
chloroformate activation to form a mixed carboxylic–
carbonic anhydride which was then decomposed to afford
10. Compounds 11 and 12 were synthesized via singlet
oxygen photooxygenation of ferutinin, using meso-tetra-
phenylporphine as a photosensitizer and incandescent light
to afford 9-hydroperoxyferutinin and 8,9-epoxyferutinin
(Galal et al., 2001; Wasserman et al., 1986), respectively.
The epoxy compound (12) was found to be identical with a
natural one that has been previously isolated (Galal, 2000).
The 8,9-dihydroferutinin (13) (Scheme 2) was prepared
following a published procedure (Appendino et al., 2004).
Results and discussion
Chemistry
Alkaline hydrolysis of ferutinin (1) was exploited as the
pivotal step toward the preparation of the new ester analogs
of 1. Coupling of the resulting daucane alcohol 2 to the
relevant acyl chloride or nalidixic acid afforded com-
pounds 3–10 in yields sufficient for full characterization
Bioassays
Compounds 1–13 were evaluated in vitro for estrogenic
activity and binding to cannabinoid (subtype CB1 and
123

Med Chem Res
CB2) receptors. Compound 1 exhibited the highest estro-
genic activity among all the tested compounds with an
EC50 of 0.3 lM, which is in line with earlier results
reported by Appendino et al. (Appendino et al., 2004). Of
the semisynthetic analogs, only 3, 4, and 11 exhibited weak
activities (EC50 = 310.0, 300.0, and 16.0 lM, respec-
tively, Table 1). The estrogenic activity of compound 12
was previously reported to be comparable to 1 (Appendino
et al., 2004). Overall, the estrogenic activities displayed by
the compounds included in the current study were weaker
than that of the human estrogen 17b-estradiol (EC50
0.005 lM). Modifications at the aromatic p-position,
removal of the aromatic system, and relocation of the
daucane double bond all lead to significant reduction or
loss of estrogenic activity of 1. Compound 10 did not show
any affinity despite possessing a heteroaromatic moiety
with two nitrogen atoms, which was presumed to enhance
the binding affinity through formation of hydrogen bonding
between the aromatic system and the receptor site. This
might be due to the existence of a bulk interaction between
10 and the receptor site that leads to failure of binding. In a
recent report, Proto et al. (2012) demonstrated the inter-
action between 17b-estradiol and the CB1 receptor to
suppress colon cancer cell proliferation. This finding, in
conjunction with the fact that 1 possesses pronounced
estrogenic activity, prompted us to evaluate parent com-
pound 1 and the derivatives 2–13 for their binding affinity
toward CB1 and CB2 receptors (Table 2). Compounds that
showed sufficient binding affinity to the cannabinoid
receptors were developed to the functional assays
(Table 3). Compounds 6 and 8 showed significant and
selective affinity to CB2 and CB1, with Ki 25 and 31 nM,
respectively. In the functional assays, both 6 and 8 were
shown to be agonists at CB2 and at CB1, respectively. The
parent compound 1 exhibited much weaker binding affinity
to both receptors, with Ki 7100 and 3634 nM at the CB1
and CB2, respectively. Ferutinin (1) and its 8,9-dihydrof-
erutinin derivative (13) did not show any noticeable
activity, indicating the insignificance of the double bond
for the CB1 or CB2 binding. The binding affinity of the
epoxide derivative (12) at CB1 when compared to that of
ferutinin was found to be relatively weaker. Replacement
of the hydroxyl group of the p-hydroxyphenyl residue with
bromine or iodine, in an attempt to enhance the ability of
formation of hydrogen/halogen bonds with the amino acid
residues at the receptor site to increase binding affinity,
resulted in marginal improvement. The fluoro-derivative
exhibited weaker affinity at the CB2. Likewise, replace-
ment of the hydroxyl with cyano group leads to a relative
improvement in the binding affinity for CB1. Replacement
of the p-hydroxyphenyl moiety of ferutinin with pyridine
moiety marginally enhanced the binding affinity toward
CB2 receptor. Compounds 6 and 8 displayed significant
Table 1 Estrogenic activity of compounds 1, 3, 4, and 11
Compound
Estrogenic activity
EC50 (lM)
% Maximal 17b-estradiol response
1
0.3
90 % at 1 lM
44 % at 500 lM
99 % at 500 lM
79 % at 125 lM
–
3
310.0
300.0
16.0
4
11
17b-estradiol
0.005
EC50: half maximal effective concentration
% maximal response = (max sample absorbance/max 17b-estradiol
absorbance) 9 100
Compounds 2, 5, 6, 7, 8, 9, and 10 were inactive (No estrogenic effect
up to 500 lM)
Table 2 Binding affinity of compounds 1–13 for CB1 and CB2
receptors expressed as Ki (nM)
Compound
CB1/Ki (lM)
CB2/Ki (lM)
1
7.08
130.80
2.50
1.59
NTa
3.63
52.00
0.78
0.12
4.30
4.35
NT
2
3
4
5
6
0.33
4.05
0.922
25.20
9.20
NT
7
8
0.55
0.58
0.46
NT
9
10
11
12
13
9.17
11.86
NT
4.46
a
Not tested
Table 3 Results of functional assays of compounds 3, 4, 6, 8, and 9
for CB1 and CB2 receptors
Compound
CB1/Ki (lM)
CB2/Ki (lM)
3
1.24
NAa
4
27.96
48.50
0.03
NA
6
0.025
NA
8
9
1.10
0.499
0.0021
CP-55,940
0.0008
a
No activity
binding affinities at CB1 and CB2, respectively, and were
found to be agonists for CB2 and CB1 receptors, respec-
tively, as shown in (Table 3) and (Fig. 1). Compound 11
was not tested for binding affinity to the cannabinoid
receptors owing to low yield and decomposition. In
123

Med Chem Res
Fig. 1 Dose–response curves of
compounds 3, 4, 6, 8, and 9 at
cannabinoid receptors (CB1
and/or CB2)
123

Med Chem Res
conclusion, these results warrant further investigation of
compounds 6 and 8, as cannabinoid agonists, for possible
development as potential therapeutic candidates. Further
modifications are thus needed to characterize the structural
parameters of this series of ferutinin derivatives necessary
for binding to the cannabinoid receptors. The results of
binding assays are displayed in (Table 2), whereas the
results of functional assays are shown in (Table 3) and
(Fig. 1).
triethylamine added, and the reaction mixture was stirred
until clear solution was obtained. To this solution was added
each of the following acid chlorides (two equivalents): p-
chlorobenzoyl chloride, p-bromobenzoyl chloride, p-io-
dobenzoyl chloride, p-fluorobenzoyl chloride, p-cya-
nobenzoyl chloride, p-trifluoromethylbenzoyl chloride, and
isonicotinoyl chloride hydrochloride plus a catalytic amount
of 4-dimethylaminopyridine (DMAP). The reaction mixture
was heated at 95–100 °C for 2–12 h, during which time the
progress of the reaction was monitored by TLC. The reaction
was quenched when TLC showed the disappearance or the
presence of persistent traces of the starting material. The
reaction mixture was then purified on silica gel column or by
preparative TLC to isolate the target compounds. Isonicoti-
noyl chloride hydrochloride was dissolved in triethylamine,
since it is insoluble in toluene.
Conclusion
Compounds 3–13 were synthesized from readily available
starting materials and utilizing straightforward reactions.
The new compounds in addition to 1 and 2 were evaluated
for their in vitro estrogenic activities as well as affinity to
cannabinoid receptors. Compounds 2–11 exhibited estro-
genic activities that were weaker than the parent compound
1. Nevertheless, compounds 6 and 8 showed significant
binding affinities at the cannabinoid receptors and were
found to be agonists for CB2 and CB1, respectively, as
shown in (Table 3) and (Fig. 1). The current results
necessitate further investigation for possible development
of compounds 6 and 8 as potential therapeutic candidates.
Ferutinyl 6-p-bromobenzoate (3) Yield, 26 %, oil;
Rf = 0.7 (Hexanes-EtOAc, 80:20); UV (MeOH) kmax 235
1
sh, 250 nm; H NMR (CDCl3, 400 MHz) d 0.83 (3H, d,
J = 6.8 Hz, H-12), 0.95 (3H, d, J = 7.2 Hz, H-13), 1.10
(3H, s, H-15), 1.81 (3H, s, H-14), 2.06 (1H, d, J = 8.8 Hz,
H-10a), 2.24 (1H, d, J = 14.0 Hz, H-10b), 2.52 (1H, m,
H-11), 2.28 (1H, d, J = 14.0 Hz, H-5) 2.49 (2H, m, H-7),
1.50–1.67 (4H, m, H-2, H-3), 5.29 (1H, dt, J = 10.0,
2.8 Hz, H-6), 5.54 (1H, t, J = 6.2 Hz, H-9), 7.57 (2H, d,
J = 8.4 Hz, H-30, H-70), 7.85 (2H, d, J = 8.4 Hz, H-40,
H-60); 13C NMR (CDCl3, 100 MHz) d 44.3 (C-1), 32.1 (C-
2), 41.2 (C-3), 86.5 (C-4), 60.1 (C-5), 72.0 (C-6), 44.3 (C-
7), 133.5 (C-8), 125.5 (C-9), 41.5 (C-10), 37.5 (C-11), 18.7
(C-12), 17.7 (C-13), 26.7 (C 14), 20.4 (C-15), 166.0 (C-10),
129.6 (C-20), 131.3 (C-30), 132.0 (C-40), 128.5 (C-50), 132.0
(C-60), 131.3 (C-70); HRESIMS: m/z 443.1211 [M ? Na]?
(calcd for C22H29BrNaO3, 443.1198).
Experimental
Chemistry
General experimental procedures 1D and 2D NMR
spectra were recorded in CDCl3 on a Bruker Avance DPX-
400 spectrometer and on a Varian AS 400 spectrometer.
HRESIMS was obtained using a Bruker Bioapex FTMS in
ESI mode. LRESIMS was obtained using a 3200 Q Trap
LC/MS/MS (Applied Biosystems MDS Sciex, Foster City,
CA). TLC was carried out on aluminum-backed plates
Ferutinyl 6-p-iodobenzoate (4)
Yield 27.1 %, white
amorphous solid; Rf = 0.76 (Hexanes- EtOAc, 83:17); UV
(MeOH) kmax 255 nm; 1H NMR (400 MHz, CDCl3): d 0.83
(3H, d, J = 6.8 Hz, H-12), 0.94 (3H, d, J = 7.2 Hz, H-13),
1.10 (3H, s, H-15), 1.80 (3H, s, H-14), 2.06 (1H, d,
J = 8.8 Hz, H-10a), 2.24 (1H, d, J = 14.0 Hz, H-10b), 2.50
(1H, m, H-11), 2.27 (1H, d, J = 14.0 Hz, H-5) 2.47 (2H, m,
H-7), 1.50–1.66 (4H, m, H-2, H-3), 5.27 (1H, dt, J = 10.0,
2.8 Hz, H-6), 5.53 (1H, bt, J = 6.2 Hz, H-9), 7.70 (2H, d,
J = 8.4 Hz, H-30, H-70), 7.77 (2H, d, J = 8.4 Hz, H-40,
H-60); 13C NMR (100 MHz, CDCl3): d 44.3 (C-1), 32.2 (C-
2), 41.2 (C-3), 86.5 (C-4), 60.1 (C-5), 72.0 (C-6), 44.3 (C-7),
133.5 (C-8), 125.5 (C-9), 41.5 (C-10), 37.5 (C-11), 18.7 (C-
12), 17.7 (C-13), 26.7 (C 14), 20.4 (C-15), 166.7 (C-10),
130.2 (C-20), 131.9 (C-30), 138.0 (C-40), 101.3 (C-50), 138.0
(C-60), 131.9 (C-70); HRESIMS: m/z 491.1063 [M ? Na]?
(calcd for C22H29INaO3, 491.1059).
˚
precoated with silica gel F254 (20 9 20 cm, 200 lM, 60 A,
Merck). Visualization was accomplished by spraying with
p-anisaldehyde [0.5 mL in glacial acetic acid (50 mL) and
H2SO4 (97 %, 1 mL)] spray reagent followed by heating.
˚
Flash silica gel (40–63 lM, 60 A, Silicycle) was used for
column chromatography. Ferutinin (1) was prepared and
characterized as described elsewhere (Galal et al., 2001).
General procedure for synthesis of ferutinol esters
(3–9) The sesquiterpenediol starting material, 2, was pre-
pared via alkaline hydrolysis of 1 by refluxing with 10 %
KOH for 2 h. The liberated alcohol was subsequently
extracted with ether followed by vacuum removal of solvent
to obtain 2. The identity of 2 was confirmed by comparison of
its NMR data with the literature. Ferutinol 2 (75 mg, 0.
315 mmol) was dissolved in dry toluene (3 mL), 0.5 mL of
Ferutinyl 6-p-fluorobenzoate (5) Yield 33.4 %, white
amorphous solid; Rf = 0.62 (Hexanes- EtOAc, 85:15); UV
123

Med Chem Res
1
(MeOH) kmax 225 nm; 1H NMR (CDCl3, 400 MHz) d 0.83
(3H, d, J = 6.8 Hz, H-12), 0.95 (3H, d, J = 7.2 Hz, H-13),
1.10 (3H, s, H-15), 1.81 (3H, s, H-14), 2.06 (1H, d,
J = 8.8 Hz, H-10a), 2.24 (1H, d, J = 14.0 Hz, H-10b),
2.52 (1H, m, H-11), 2.28 (1H, d, J = 14.0 Hz, H-5) 2.49
(2H, m, H-7), 1.50–1.67 (4H, m, H-2, H-3), 5.29 (1H, dt,
J = 10.0, 2.8 Hz, H-6), 5.54 (1H, t, J = 6.2 Hz, H-9), 8.53
(2H, d, J = 8.4 Hz, H-30, H-70), 7.13 (2H, d, J = 8.4 Hz,
H-40, H-60); 13C NMR (CDCl3, 100 MHz) d 44.3 (C-1),
32.1 (C-2), 41.2 (C-3), 86.5 (C-4), 60.1 (C-5), 72.0 (C-6),
44.3 (C-7), 133.6 (C-8), 125.5 (C-9), 41.5 (C-10), 37.5 (C-
11), 18.7 (C-12), 17.7 (C-13), 26.7 (C 14), 20.4 (C-15),
165.8 (C-10), 126.9 (C-20), 132.3 (C-30), 116.6 (C-40), 161.4
(C-50), 116.6 (C-60), 132.3 (C-70); HRESIMS: m/z
383.2001 [M ? Na]? (calcd for C22H29FNaO3, 383.1998).
kmax 215, 225, 275 nm; H NMR (CDCl3, 400 MHz) d
0.93 (3H, d, J = 6.8 Hz, H-12), 0.98 (3H, d, J = 7.2 Hz,
H-13), 1.13 (3H, s, H-15), 1.83 (3H, s, H-14), 2.12 (1H, d,
J = 8.8 Hz, H-10a), 2.24 (1H, d, J = 14.0 Hz, H-10b),
2.60 (1H, m, H-11), 2.33 (1H, d, J = 14.0 Hz, H-5), 2.32
(2H, m, H-7), 1.25–1.60 (4H, m, H-2, H-3), 5.40 (1H, dt,
J = 10.0, 2.8 Hz, H-6), 5.60 (1H, t, J = 6.2 Hz, H-9), 8.15
(2H, d, J = 8.4 Hz, H-30, H-70), 7.75 (2H, d, J = 8.4 Hz,
H-40, H-60); 13C NMR (CDCl3, 100 MHz) d 44.0 (C-1),
32.1 (C-2), 41.3 (C-3), 86.3 (C-4), 60.1 (C-5), 72.3(C-6),
41.3 (C-7), 133.7 (C-8), 125.7 (C-9), 41.4 (C-10), 37.3 (C-
11), 18.5 (C-12), 17.3 (C-13), 26.5 (C 14), 20.5 (C-15),
165.2 (C-10), 133.3 (C-20), 130.1 (C-30), 125.8 (C-40), 135.2
(C-50), 125.8 (C-60), 130.1 (C-70), 124.9 (CF3); HRESIMS:
m/z 409.1996 [M-H]- (calcd for C23H28F3O3, 409.1991).
Ferutinyl 6-p-chlorobenzoate (6)
Yield 48 %, white
Ferutinyl 6-isonicotinate (9) Yield 40 %; white solid;
Rf = 0.15 (Hexanes-EtOAc, 80:20); UV (MeOH) kmax
255, 335 nm; H NMR (CDCl3, 400 MHz) d 0.84 (3H, d,
amorphous solid; Rf = 0.7 (Hexanes- EtOAc, 85:15); UV
(MeOH) kmax 245 nm; 1H NMR (CDCl3, 400 MHz) d 0.84
(3H, d, J = 6.8 Hz, H-12), 0.95 (3H, d, J = 7.2 Hz, H-13),
1.10 (3H, s, H-15), 1.82 (3H, s, H-14), 2.06 (1H, d,
J = 8.8 Hz, H-10a), 2.24 (1H, d, J = 14.0 Hz, H-10b), 2.52
(1H, m, H-11), 2.28 (1H, d, J = 14.0 Hz, H-5) 2.49 (2H, m,
H-7), 1.50–1.67 (4H, m, H-2, H-3), 5.30 (1H, dt, J = 10.0,
2.8 Hz, H-6), 5.54 (1H, t, J = 6.2 Hz, H-9), 7.42 (2H, d,
J = 8.4 Hz, H-30, H-70), 7.94 (2H, d, J = 8.4 Hz, H-40,
H-60); 13C NMR (CDCl3, 100 MHz) d 44.0 (C-1), 32.0 (C-2),
41.2 (C-3), 86.3 (C-4), 60.0 (C-5), 71.7 (C-6), 44.0 (C-7),
133.3 (C-8), 125.3 (C-9), 41.3 (C-10), 37.2 (C-11), 18.5 (C-
12), 17.4 (C-13), 26.4 (C 14), 20.2 (C-15), 165.6 (C-10),
129.0 (C-20), 129.0 (C-30), 132.0 (C-40), 128.8 (C-50), 132.0
(C-60), 129.0 (C-70); HRESIMS: m/z 399.1710 [M ? Na]?
(calcd for C22H29ClNaO3, 399.1703).
1
J = 6.8 Hz, H-12), 0.95 (3H, d, J = 7.2 Hz, H-13), 1.10
(3H, s, H-15), 1.82 (3H, s, H-14), 2.06 (1H, d, J = 8.8 Hz,
H-10a), 2.24 (1H, d, J = 14.0 Hz, H-10b), 2.52 (1H, m,
H-11), 2.28 (1H, d, J = 14.0 Hz, H-5) 2.49 (2H, m, H-7),
1.50–1.67 (4H, m, H-2, H-3), 5.30 (1H, dt, J = 10.0,
2.8 Hz, H-6), 5.54 (1H, t, J = 6.2 Hz, H-9), 8.04 (2H, d,
J = 8.4 Hz, H-30, H-70), 8.90 (2H, d, J = 8.4 Hz, H-40,
H-60); 13C NMR (CDCl3, 100 MHz) d 44.0 (C-1), 33.0 (C-
2), 41.2 (C-3), 86.3 (C-4), 60.0 (C-5), 73.6 (C-6), 44.0 (C-
7), 133.0 (C-8), 125.3 (C-9), 41.3 (C-10), 37.3 (C-11), 18.6
(C-12), 17.5 (C-13), 26.7 (C-14), 20.3 (C-15), 163.2 (C-10),
142.4 (C-20), 125.5 (C-30), 146.7 (C-40), 146.7 (C-60), 125.5
(C-70); HRESIMS: m/z 344.2243 [M ? H]? (calcd for
C21H30NO3, 344.2226).
Ferutinyl 6-p-cyanobenzoate (7)
Yield 30.1 %, white
Synthesis of ferutinyl 6-nalidixate (10) Nalidixic acid (1.
0 mmol) was suspended in dichloromethane (20 mL) and
stirred at 0 °C. To the stirred suspension, triethyl amine (1.
1 mmol) was added followed by drop wise addition of
ethylchloroformate (1.1 mmol) and the stirring continued
for 20 min. The resulting clear solution was then added to a
stirred solution of ferutinol (1.1 mmol) in dichloromethane
and stirring continued overnight. Solvent was evaporated
under reduced pressure, and the product was purified on
silica gel column.
amorphous solid; Rf = 0.48 (Hexanes- EtOAc, 80:20); UV
(MeOH) kmax 245 nm; 1H NMR (CDCl3, 400 MHz) d 0.80
(3H, d, J = 6.8 Hz, H-12), 0.94 (3H, d, J = 7.2 Hz, H-13),
1.04 (3H, s, H-15), 1.80 (3H, s, H-14), 2.06 (1H, d,
J = 8.8 Hz, H-10a), 2.24 (1H, d, J = 14.0 Hz, H-10b), 2.52
(1H, m, H-11), 2.28 (1H, d, J = 14.0 Hz, H-5) 2.49 (2H, m,
H-7), 1.50–1.67 (4H, m, H-2, H-3), 5.28 (1H, dt, J = 10.0,
2.8 Hz, H-6), 5.51 (1H, t, J = 6.2 Hz, H-9), 8.07 (2H, d,
J = 8.4 Hz, H-30, H-70), 7.71 (2H, d, J = 8.4 Hz, H-40,
H-60); 13C NMR (CDCl3, 100 MHz) d 44.2 (C-1), 32.1 (C-2),
41.2 (C-3), 86.5 (C-4), 60.0 (C-5), 72.5 (C-6), 44.2 (C-7),
133.4 (C-8), 125.6 (C-9), 41.5 (C-10), 37.5 (C-11), 18.7 (C-
12), 17.7 (C-13), 26.7 (C 14), 20.3 (C-15), 164.9 (C-10),
134.6 (C-20), 130.2(C-30), 132.5 (C-40), 116.5 (C-50), 132.5
(C-60), 130.2 (C-70), 118.1 (CN); HRESIMS: m/z 390.2039
[M ? H]? (calcd for C23H29NNaO3, 390.2045).
Ferutinyl 6-nalidixate (10) Yield 90 %; white solid;
Rf = 0.59 (Hexanes–acetone, 70:30); UV (MeOH) kmax
1
250 sh, 255, 325 nm; H NMR (CDCl3, 400 MHz) d 0.87
(3H, d, J = 6.8, H-12), 0.98 (3H, d, J = 7.2, H-13), 1.10
(3H, s, H-15), 1.81 (3H, s, H-14), 2.04 (1H, d, J = 8.8,
H-10a), 2.11 (1H, d, J = 14.0, H-10b), 2.52 (1H, m, H-11),
2.11 (1H, d, J = 14.0, H-5) 2.41 (2H, m, H-7), 1.50–1.62
(4H, m, H-2, H-3), 5.01 (1H, dt, J = 10.0, 2.8, H-6), 5.50
(1H, t, J = 6.2, H-9), 8.34 (1H, s, H-20), 8.53 (1H, d,
Ferutinyl 6-p-trifluoromethylbenzoate (8) Yield 44.4 %,
wax; Rf = 0.66 (Hexanes-EtOAc, 80:20); UV (MeOH)
123

Med Chem Res
J = 7.6 Hz, H-50), 7.20 (1H, d, J = 6.7 Hz, H-60), 2.61
(3H, s, H-100), 4.36, 4.47 (2H, q, J = 7.2, H-110), 1.44 (3H,
t, J = 7.2, H-120); 13C NMR (CDCl3, 100 MHz) d 44.3 (C-
1), 32.2 (C-2), 41.8 (C-3), 86.1 (C-4), 60.8 (C-5), 73.4 (C-
6), 44.3 (C-7), 133.9 (C-8), 125.1 (C-9), 41.9 (C-10), 37.1
(C-11), 19.0 (C-12), 17.8 (C-13), 27.1 (C 14), 20.2 (C-15),
147.2 (C-20), 114.1 (C-30), 174.5 (C-40), 137.0 (C-50), 121.3
(C-60), 163.1 (C-70), 148.8 (C-80), 121.2 (C-90), 25.3 (C-
100), 46.7 (C-110), 15.4 (C-120), 166.6 (C-130); HRESIMS:
m/z 453.2747 [M ? H]? (calcd for C27H37N2O4,
453.2753).
the assay, 1.5 mL of the culture and 250 lL of chloro-
phenol red b-D-galactopyranoside (CPRG, 10 mg/mL)
were added to 25 mL of growth medium. Test compounds
(10 mM) or 17b-estradiol (1 mM, positive control) were
dissolved in DMSO. Serial dilutions were made, and 10 lL
of the diluted samples were transferred from the dilution
plate to the assay plate and mixed with 190 lL of yeast cell
suspension containing CPRG. The highest DMSO con-
centration was 5 %. The plate was incubated at 30 °C for
48 h, and the absorbance was measured at a dual wave-
length of 540/630 nm. Dose curves were generated, and
half maximal effective concentration (EC50) values were
obtained. Highest effective dose was determined for each
compound showing estrogenic response, and the percent
maximal response at the highest effective dose was cal-
culated in comparison to 17b-estradiol.
Synthesis of 9-hydroperoxyferutinin (11) and 8,9-epoxyf-
erutinin (12) Ferutinol (100 mg) was dissolved in 25 mL
of a mixture of dichloromethane/ethanol (1:1) in a Dudley
tube fitted in a cooling bath and then was added the photo-
sensitization dye meso-tetraphenylporphine (*2.0 mg).
The wine-red reaction mixture was then subjected to 500 W
incandescent light, while oxygen was gently bubbled
through, and the temperature of the bath was maintained at
13 °C. After 3 h of irradiation, the reaction was complete
(disappearance of ferutinol on TLC) (Wasserman et al.,
1986). The solvent was distilled off from the reaction mix-
ture to leave an olive green residue which was purified by
normal phase preparative scale TLC to afford 9-hydroper-
oxyferutinin (11) and the known 8,9-epoxyferutinin (12).
Cell culture HEK293 cells (ATCC #CRC-1573) were
stably transfected via electroporation with full-length
human recombinant cDNA for cannabinoid receptor sub-
types 1 and 2 (obtained from Origene). These cells were
maintained in a Dulbecco’s modified Eagles’s medium/F-
12 (50/50) nutrient mixture supplemented with 10 % fetal
bovine serum, 1 % penicillin/streptomycin, and either 1 %
G418 sulfate (Geneticin), depending on the cell line. Per-
centages are based on a total media volume of 500 mL.
Both cannabinoid cell lines were kept at 37 °C and 5 %
CO2. Membranes were prepared by scraping the cells in a
50 mM Tris–HCl buffer, homogenized via sonication, and
centrifuged for 40 min at 13,650 rpm at 4 °C. These were
kept at 80 °C until used for binding and functional assays.
Protein concentration was determined via Bio-Rad protein
assay (Bradford, 1976).
9-Hydroperoxyferutinin (11)
Yield 11.4 %; oil;
Rf = 0.33 (Hexanes–EtOAc, 70:30); UV (MeOH) kmax
255 nm; 1H NMR (CDCl3, 400 MHz) d 0.83 (3H, d,
J = 6.8 Hz, H-12), 0.89 (3H, d, J = 7.2 Hz, H-13), 1.14
(3H, s, H-15), 1.88 (3H, s, H-14), 2.06 (1H, d, J = 8.8 Hz,
H-10a), 2.11 (1H, d, J = 14.0 Hz, H-10b), 2.40 (1H, m,
H-11), 2.11 (1H, d, J = 14.0 Hz, H-5) 1.50–1.67 (4H, m,
H-2, H-3), 5.50 (1H, bs, H-6), 4.45 (1H, bs, H-9), 5.83 (1H,
d, J = 10.4 Hz, H-7), 7.90 (2H, d, J = 8.4 Hz, H-30, H-70),
6.86 (2H, d, J = 8.4 Hz, H-40, H-60); 13C NMR (CDCl3,
100 MHz) d 44.5 (C-1), 32.0 (C-2), 42.1 (C-3), 87.1 (C-4),
54.1 (C-5), 73.0 (C-6), 128.0 (C-7), 137.2 (C-8), 83.5 (C-
9), 42.4 (C-10), 37.1 (C-11), 18.6 (C-12), 17.7 (C-13), 24.5
(C 14), 20.2 (C-15), 167.5 (C-10), 121.8 (C-20), 132.4 (C-
30), 115.7 (C-40), 161.5 (C-50), 115.7 (C-60), 132.4 (C-70);
HRESIMS: m/z 413.1958 [M ? H]? (calcd for
C22H30NaO6, 413.1940).
Radioligand binding for cannabinoid receptor sub-
types In the primary bioassay screen, compounds were
tested at a final concentration of 10 lM for competitive
binding to the respective receptor. For the cannabinoid
receptor assays, test compounds were added into a 96-well
plate followed by 0.6 nM [3H]CP-55,940 and 10 lg of
cannabinoid membrane resuspended in 50 mM Tris (pH 7.
4), 154 mM NaCl, and 20 mM Di-Na-EDTA supple-
mented with 0.02 % BSA. The cannabinoid assay was
allowed to incubate at 37 °C for 90 min. The reaction was
then terminated by rapid filtration using GF/C (presoaked
in 0.3 % BSA) and washed with the buffer. Dried filters
were then covered with scintillant and measured for the
amount of radioligand retained using a PerkinElmer Top-
count (PerkinElmer Life Sciences Inc., Boston, MA, USA).
Nonspecific binding, which was determined in the presence
of 1 lM CP-55,940 for cannabinoid receptors, was sub-
tracted from the total binding to yield the specific-binding
values. Compounds showing competitive inhibition of the
In vitro biological assays
Estrogenic assay The estrogenic activity was determined
by the YES (yeast estrogen screen) assay as described
earlier with some modifications (Beresford et al., 2000;
Tabanca et al., 2004).
Saccharomyces cerevisiae cells expressing the estrogen
receptor alpha (ER-a) were cultured at 30 °C for 24 h. For
123

Med Chem Res
Guindon J, Hohmann AG (2008) Cannabinoid CB2 receptors: a
therapeutic target for the treatment of inflammatory and
neuropathic pain. Br J Pharmacol 153:319–334
Hadidi KA, Aburjai T, Battah AK (2003) A comparative study of
Ferula hermonis root extracts and sildenafil on copulatory
behaviour of male rats. Fitoterapia 74:242–246
labeled ligand to bind to the receptor at 50 % or greater
were tested in a dose–response curve with concentrations
of the test compound ranging from 300 lM to 1.7 nM. Ki
and IC50 values were calculated using GraphPad Prism 5.
[35S]-GTP-cS binding In the functional assay for can-
nabinoid receptor binding, membranes (20 lg/well) were
incubated with 0.5 nM [35S]-GTP-cS and test compound in
50 mM Tris–HCl, 0.2 mM EGTA, 9 mM MgCl2, 150 mM
NaCl, 50 lM GDP, and 1.4 mg mL-1 BSA. The reaction
incubated for 2 h at 30 °C and was terminated by rapid
vacuum filtration with cold 10 mM Tris–HCl in a Perkin-
Elmer harvester through GF/B filters. Nonspecific binding
was determined by 40 lM of GTP-cS.
Ibraheim ZZ, Abdel-Mageed WM, Dai H, Guo H, Zhang L, Jaspars M
(2012) Antimicrobial antioxidant daucane sesquiterpenes from
Ferula hermonis Boiss. Phytother Res 26:579–586
Ikeda K, Arao Y, Otsuka H, Nomoto S, Horiguchi H, Kato S, Kayama
F (2002) Terpenoids found in the Umbelliferae family act as
agonists/antagonists for ERa and ERb: differential transcription
activity between ferutinine-liganded ERa and ERb. Biochem
Biophys Res Commun 291:354–360
Malfitano AM et al (2013) Effects on immune cells of a new 1,8-
naphthyridin-2-one derivative and its analogues as selective CB2
agonists: implications in multiple sclerosis. PLoS One 8:e62511
Palumbo C et al (2009) Influence of ferutinin on bone metabolism in
ovariectomized rats. I: role in preventing osteoporosis. J Bone
Miner Metab 27:538–545
Proto MC et al (2012) Interaction of endocannabinoid system and
steroid hormones in the control of colon cancer cell growth.
J Cell Physiol 227:250–258
Acknowledgments The authors thank the United States Department
of Agriculture, Agriculture Research Service Specific Cooperative
Agreement No. 58-6408-7-012, for partial support of this work. The
authors are also grateful to Dr. Bharathi Avula for helping with the
HRESIFTMS analyses.
Pryce G et al (2014) Control of spasticity in a multiple sclerosis
model using central nervous system-excluded CB1 cannabinoid
receptor agonists. FASEB J 28(117–130):1. doi:10.1096/fj.1013-
239442
References
Rasulev BF, Saidkhodzhaev AI, Nazrullaev SS, Akhmedkhodzhaeva
KS, Khushbaktova ZA, Leszczynski J (2007) Molecular mod-
elling and QSAR analysis of the estrogenic activity of terpenoids
isolated from Ferula plants. SAR QSAR Environ Res
18:663–673
Ribeiro R, Yu F, Wen J, Vana A, Zhang Y (2013) Therapeutic
potential of a novel cannabinoid agent CB52 in the mouse model
of experimental autoimmune encephalomyelitis. Neuroscience
254:427–442
Rivers JR-J, Ashton JC (2010) The development of cannabinoid CBII
receptor agonists for the treatment of central neuropathies. Cent
Nerv Syst Agents Med Chem 10:47–64
Sophocleous A, Landao-Bassonga E, van’t Hof RJ, Idris AI, Ralston
SH, Landao-Bassonga E, Landao-Bassonga E (2011) The type 2
cannabinoid receptor regulates bone mass and ovariectomy-
induced bone loss by affecting osteoblast differentiation and
bone formation. Endocrinology 152:2141–2149
Tabanca N et al (2004) Estrogenic activity of isolated compounds and
essential oils of Pimpinella species from Turkey, evaluated using
a recombinant yeast screen. Planta Med 70:728–735
Wasserman HH, Lu TJ, Scott IA (1986) The total synthesis of
tetracycline. J Am Chem Soc 108:4237–4238
Zanoli P, Benelli A, Rivasi M, Baraldi C, Vezzalini F, Baraldi M
(2003) Opposite effect of acute and subchronic treatments with
Ferula hermonis on copulatory behavior of male rats. Int J Impot
Res 15:450–455
Zanoli P, Rivasi M, Zavatti M, Brusiani F, Vezzalini F, Baraldi M
(2005) Activity of single components of Ferula hermonis on
male rat sexual behavior. Int J Impot Res 17:513–518
Zanoli P, Zavatti M, Geminiani E, Corsi L, Baraldi M (2009) The
phytoestrogen ferutinin affects female sexual behavior modulat-
ing ERa expression in the hypothalamus. Behav Brain Res
199:283–287
Zavatti M, Montanari C, Zanoli P (2006) Role of ferutinin in the
impairment of female sexual function induced by Ferula
hermonis. Physiol Behav 89:656–661
Appendino G, Spagliardi P, Sterner O, Milligan S (2004) Structure-
activity relationships of the estrogenic sesquiterpene ester feruti-
nin. modification of the terpenoid core. J Nat Prod 67:1557–1564
Beresford N, Routledge EJ, Harris CA, Sumpter JP (2000) Issues
arising when interpreting results from an in vitro assay for
estrogenic activity. Toxicol Appl Pharmacol 162:22–33
Bradford MM (1976)
A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the
principle of protein-dye binding. Anal Biochem 72:248–254
Colman-Saizarbitoria T, Boutros P, Amesty A, Bahsas A, Mathison
Y, Garrido MDR, Israel A (2006) Ferutinin stimulates nitric
oxide synthase activity in median eminence of the rat. J Ethno-
pharmacol 106:327–332
Dall’Acqua S, Maggi F, Minesso P, Salvagno M, Papa F, Vittori S,
Innocenti G (2010) Identification of non-alkaloid acetylcholin-
esterase inhibitors from Ferulago campestris (Besser) Grecescu
(Apiaceae). Fitoterapia 81:1208–1212
Diaz P, Xu J, Astruc-Diaz F, Pan H-M, Brown DL, Naguib M (2008)
Design and synthesis of a novel series of N-Alkyl isatin
acylhydrazone derivatives that act as selective cannabinoid
receptor 2 agonists for the treatment of neuropathic pain. J Med
Chem 51:4932–4947
Galal A (2000) Sesquiterpenes from Ferula hermonis Boiss.
Pharmazie 55:961–962
Galal AM, Abourashed EA, Ross SA, ElSohly MA, Al-Said MS, El-
Feraly FS (2001) Daucane sesquiterpenes from Ferula hermonis.
J Nat Prod 64:399–400
Gao M, Wong SY, Lau PM, Kong SK (2013) Ferutinin induces
in Vitro eryptosis/erythroptosis in human erythrocytes through
membrane permeabilization and calcium influx. Chem Res
Toxicol 26:1218–1228
Gertsch J, Pertwee RG, Di Marzo V (2010) Phytocannabinoids beyond
the Cannabis plant—do they exist? Br J Pharmacol 160:523–529
Greineisen WE, Turner H (2010) Immunoactive effects of cannab-
inoids: considerations for the therapeutic use of cannabinoid
receptor agonists and antagonists. Int Immunopharmacol
10:547–555
123