Potent vasorelaxant analogs from chemical modification and biotransformation of isosteviol

Article abstract of DOI:10.1016/j.ejmech.2013.01.022

Isosteviol (1) has been reported to exhibit moderate vasorelaxant activity. In order to enhance the bioactivity of this compound, chemical modification of 1 to the dihydro analog, ent-16β-hydroxybeyeran-19-oic acid (2), was undertaken. Compound 2 was then converted to the corresponding acetate derivative, ent-16β-acetoxybeyeran-19-oic acid (3). Biotransformation of compounds 1-3 by the fungus Cunninghamella echinulata NRRL 1386 was investigated and the metabolites 4-9 were obtained. The substrates and their metabolites were subjected to in vitro rat aorta relaxant activity evaluation. The metabolite 4, ent-7α-hydroxy-16-ketobeyeran-19-oic acid, exhibited the most highly potent activity, with EC₅₀ of 3.46 nM, whereas the parent compound 1 showed relatively low activity (EC₅₀ 57.41 nM). A 17-fold increase in vasorelaxant activity of the analog 4 relative to compound 1 is of particular significant. Compound 4 exerted vasorelaxant activity at particularly low concentration and the vasorelaxant profile reached maximum at relatively low concentration, especially when compared with acetylcholine, the positive control.

Full text of DOI:10.1016/j.ejmech.2013.01.022

European Journal of Medicinal Chemistry 62 (2013) 771e776
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Potent vasorelaxant analogs from chemical modification
and biotransformation of isosteviol
a
b
b
Orawan Wonganan , Chainarong Tocharus , Chonticha Puedsing ,
c
d
a,
*
Sureeporn Homvisasevongsa , Oratai Sukcharoen , Apichart Suksamrarn
Department of Chemistry and Center of Excellence for Innovation in Chemistry, Faculty of Science, Ramkhamhaeng University, Bangkok 10240, Thailand
Department of Anatomy, Faculty of Medicine, Chiang Mai University, Chiang Mai 50200, Thailand
Division of Physical Science, Faculty of Science and Technology, Huachiew Chalermprakiet University, Samutprakarn 10540, Thailand
Department of Biotechnology, Faculty of Science, Ramkhamhaeng University, Bangkok 10240, Thailand
a
b
c
d
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 14 September 2012
Received in revised form
Isosteviol (1) has been reported to exhibit moderate vasorelaxant activity. In order to enhance the
bioactivity of this compound, chemical modification of 1 to the dihydro analog, ent-16b-hydroxybeyeran-
19-oic acid (2), was undertaken. Compound 2 was then converted to the corresponding acetate deriv-
12 January 2013
ative, ent-16b-acetoxybeyeran-19-oic acid (3). Biotransformation of compounds 1e3 by the fungus
Cunninghamella echinulata NRRL 1386 was investigated and the metabolites 4e9 were obtained. The
substrates and their metabolites were subjected to in vitro rat aorta relaxant activity evaluation. The
Accepted 15 January 2013
Available online 1 February 2013
metabolite 4, ent-7a-hydroxy-16-ketobeyeran-19-oic acid, exhibited the most highly potent activity, with
Keywords:
Isosteviol
Vasorelaxant activity
Biotransformation
EC50 of 3.46 nM, whereas the parent compound 1 showed relatively low activity (EC50 57.41 nM). A 17-
fold increase in vasorelaxant activity of the analog 4 relative to compound 1 is of particular significant.
Compound 4 exerted vasorelaxant activity at particularly low concentration and the vasorelaxant profile
reached maximum at relatively low concentration, especially when compared with acetylcholine, the
positive control.
Ó 2013 Elsevier Masson SAS. All rights reserved.
1. Introduction
sweeter than sucrose [5]. Several studies suggested that the non-
sweetening agent 1 possesses a variety of biological activities
Hypertension is one of the most common cardiovascular diseases
that can cause coronary disease, myocardial infarction, stroke and
sudden death and is the major contributor to cardiac failure and
renal insufficiency. There are several classes of anti-hypertensive
including reducing blood pressure and cardioprotective effect [6e
9], anti-hyperglycemic [10] and potential anti-tumor effects [11].
Previous reports of 1 and stevioside [12e16] on cardiovascular
and related effects led us to investigate antihypertensive action of
this type of compounds. This class of compounds is of special
interest, since stevioside could be obtained from S. rebaudiana in
large quantity [5,17] and compound 1 in turn was obtained in
good yield from this glycoside by acid hydrolysis [4,9]. Moreover,
the non-toxic or less toxic nature of this class of diterpenoids
has prompted us to investigate antihypertensive activity of
compound 1. It has been reported that this non-sweetening
compound exhibited moderate vasorelaxant activity [6], which
might not be sufficiently potent for further drug development
study. It was therefore of interest to see whether structural
modification of 1 would give rise to analog(s) with considerably
high vasorelaxant activity. The present work deals with chemical
modification and microbial transformation of isosteviol (1) to
analogs 2e9 (Scheme 1), some of which exhibited high vaso-
relaxant activity.
drugs, diuretics, b-blockers, calcium channel blockers, angiotensin-
converting enzyme inhibitors, angiotensin II receptor blockers and
vasodilators [1]. These different classes of drugs have both advan-
tages and disadvantages and some of them have adverse side effects
[2,3]. It would therefore be beneficial for patients with hypertension
if novel antihypertensive agents with lesser side effects are available.
Isosteviol (1, Scheme 1), a tetracyclic diterpenoid from the
beyerane series, was obtained from acid hydrolysis of stevioside
[
4], a diterpenoid glycoside from the leaves of Stevia rebaudiana
(
Bertoni) Bertoni. This non-caloric sweetening agent is 300 times
*
Corresponding author. Tel.: þ662 3190931; fax: þ662 3191900.
E-mail addresses: s_apichart@ru.ac.th, asuksamrarn@yahoo.com
A. Suksamrarn).
(
0223-5234/$ e see front matter Ó 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2013.01.022

772
O. Wonganan et al. / European Journal of Medicinal Chemistry 62 (2013) 771e776
Scheme 1. Chemical modification of ent-16-ketobeyeran-19-oic acid (isosteviol, 1) to the dihydro analog 2 and the corresponding acetate 3, and microbial transformation of
compound 1 to the hydroxylated analog 4, compound 2 to the hydroxylated analogs 5, 6 and 7, and compound 3 to the hydroxylated analogs 8 and 9. a: NaBH
4 2
/THF; b: Ac O/
pyridine; c: Cunninghamella echinulata NRRL 1386.
by HMBC analysis. The H and 13C NMR spectra of 4 were consistent
1
2
. Chemistry
with the reported values of ent-7a-hydroxy-16-ketobeyeran-19-oic
2.1. Chemical modification
acid [22e24].
Incubation of 2 with the same fungus yielded three metabolites,
5e7. Compound 5 was identified as ent-7 ,16 -dihydroxybeyeran-
Isosteviol (1) was chosen as the parent compound for structural
a
b
modification to its analogs for vasorelaxant evaluation. We aimed
to modify the keto function at the 16-position to the corresponding
dihydro analogs. Acid hydrolysis of stevioside yielded 1 in 81%.
19-oic acid by comparison of the NMR and MS data with literature
[18,19]. The conversions of 1 and 2 by C. echinulata NRRL 1386 to the
corresponding 7b-hydroxylated analogs 4 and 5, respectively, have
Reduction of 1 with NaBH
reduction occurred exclusively from the
presence of carbinol proton at
4
gave the dihydro analog 2 in 97%. The
revealed the regio and stereoselectivity of the enzyme of this fungus.
ꢀ
b
-face of the molecule. The
Compound 6 displayed the [MꢀH] ion at m/z 335.2208 in the
ꢀ
d
4.10 (br d, J ¼ 6.7 Hz) confirmed the
HR-TOFMS (ESI ), compatible with a molecular formula C20
32
H O
4
.
conversion of the keto group of 1 into the hydroxyl group in 2. The
The IR spectrum suggested the presence of hydroxyl groups at 3624
1
ꢀ1
ꢀ1
13
spectroscopic ( H NMR and mass spectra) data of 2 were consistent
and 3414 cm and carboxyl group at 1701 cm . The C NMR and
DEPT spectra revealed the presence of eight CH and seven CH/CH
indicating the presence of a proton geminal to a new hydroxyl
with the reported values [18,19]. Acetylation of compound 2 with
acetic anhydride and pyridine gave the corresponding acetate 3 in
2
3
6%. The 1H NMR spectrum of 3 showed a three-proton singlet
group at
revealed a new resonance at
correlations with C-5 ( 54.6), C-6 (
and C-15 ( 34.3), thus confirming that hydroxylation has taken
d
75.7 in the C NMR spectrum. The H NMR spectrum of 6
3.70 and the HMBC spectra showed
32.6), C-8 ( 49.0), C-14 ( 51.0)
13
1
7
signal at
d
2.02, thus indicating that acetylation has taken placed.
d
A downfield shift of the carbinolic proton at C-16 also confirmed
that acetylation has taken placed at the C-16 hydroxyl group.
d
d
d
d
d
1
placed at C-7. In the H NMR spectrum, a double doublet (J ¼ 11.5
and 3.6 Hz) of H-7 was observed. The large coupling constant of H-7
indicated that it was in the axial position. The large J value resulted
from axialeaxial coupling between H-6ax and H-7. The hydroxyl
group therefore adopted the equatorial orientation. This was in
agreement with the reported product 4a obtained from microbial
transformation of 1 [21,24]. Furthermore, the orientation of the OH
group at C-7 was also confirmed by the ROESY experiments. Thus,
2
.2. Microbial transformation
The limited number of functional groups in isosteviol (1) has
prevented us from further chemical modification. Microbial trans-
formation is a powerful method for the regioselective and stereo-
selective introduction of hydroxyl group at un-activated position of
beyerane diterpenoids including isosteviol [9,18e24]. Microbial
transformation of isosteviol (1) with Cunninghamella echinulata
NRRL 1386 produced the more polar metabolite 4 (see Scheme 1),
which showed the [2MeH] ion at m/z 667 consistent with the
molecular formula C20
revealed the presence of a hydroxyl group, which was located at C-7
H-7 showed cross-peak with H-5 (
1.08 and 2.46). On the basis of the spectroscopic data above led to
the identification of 6 as ent-7 ,16 -dihydroxybeyeran-19-oic acid.
The molecular formula of compound 7 was determined as
d 1.32), H-9 (d 1.29) and H-14 (d
d
ꢀ
b
b
30 4
H O . Analysis of 1D and 2D NMR spectra
þ
þ
C
20
H
32
O
4
from its HR-TOFMS (ESI ) at m/z 359.2180 [MþNa] . The

O. Wonganan et al. / European Journal of Medicinal Chemistry 62 (2013) 771e776
773
IR spectrum showed the absorption band of the hydroxyl and
3. Results and discussion
ꢀ
1
carboxyl groups at 3452 and 1696 cm , respectively. Analysis of
13
the C NMR spectrum, compared to that of compound 2, a DEPT
experiment showed eight CH and seven CH/CH for 7 whilst for 2
there are nine CH and six CH/CH , respectively. Therefore, one
3.1. Vasorelaxant activity
2
3
2
3
The modification of isosteviol (1) at 16-keto group to its dihydro
analog 2 and the corresponding acetate derivative 3 was achieved
by chemical reactions. The substrates 1e3 were subjected to
microbial transformation by C. echinulata NRRL 1386. The metab-
olite 4 was obtained from the bioconversion of 1. Compound 2 was
biotransformed to the metabolites 5, 6 and 7. The metabolites 8 and
9 were produced from biotransformation of 3. The advantages of
microbial transformation in this work are hydroxylation has taken
placed at the un-activated position and the acetate group in the
substrates remained intact. C. echinulata NRRL 1386 regiose-
lectively and stereoselectively hydroxylated the beyerane diterpe-
methylene group of 2 was converted into a hydroxymethine
group, suggesting 7 to be a hydroxylated product of 2. Comparison
of the H and C NMR spectra of 7 with those of compounds 2, 5
and 6 revealed that an additional OH group in 7 should be located
at C-14, which was confirmed by a detailed analysis of the HMBC
1
13
data. The 1H NMR chemical shift at
d
3.49 clearly showed con-
34.2), C-15 ( 42.3) and C-16
78.8). The ROESY experiments revealed the correlations be-
tween H-14 and H-7ax, H-9 and H-12ax (see Fig. 1). This led to
a conclusion that H-14 is in the -orientation. The structure of 7
was thus established as ent-14 ,16 -dihydroxybeyeran-19-oic
nectivities with C-9 (
d
56.1), C-12 (
d
d
(
d
b
b
b
noids at C-7
b
and C-14
a. The exception was for compound 6 that
acid. Microbial hydroxylation of dihydroisosteviol (2) at the 14-
position by Mortierella isabellina has been reported [18], but the
spectroscopic data of the metabolite are different from those of
compound 7 isolated by our group and their NOESY experiments
indicated that the structure of the isolated compound was dif-
ferent from compound 7.
hydroxylation also occurred on the
a
-face of C-7. The vasorelaxant
activity of these compounds was evaluated using rat aorta rings
(see Fig. 2). The parent compound 1 and the dihydro analog 2
showed EC50 values of 57.41 and 29.07 nM, respectively (Table 1).
The result has indicated that reduction of the C-16 keto group of 1
to the corresponding hydroxyl group resulted in 2-fold increase in
activity. Acetylation at the C-16 hydroxyl group of 2 afforded the
acetate 3, with the EC50 value of 4.88 nM, which was 6-fold more
active than compound 2, or approximately 12-fold more active than
compound 1. Compound 4, the biotransformation product of 1,
exhibited the most active vasorelaxant effect, with the EC50 value of
3.46 nM, or approximately 17-fold more active than compound 1.
However, compound 2, upon introduction of a hydroxyl group to
the C-7 position to yield the isomeric metabolites 5 and 6, did not
increase the activity of the analogs. The activity in going from the
acetate 3 to 8 seemed to follow the same trend. The presence of
both the hydroxyl function at the 7-position and the hydroxyl or
acetoxyl function at the 16-position decreased the vasorelaxant
activity. Introduction of a hydroxyl group to C-14 position of 2 to
yield the metabolite 7 resulted in sharp decrease in activity. The
similar trend was also observed for the bioconversion of the acetate
3 to the analog 9. The assay results have indicated that, for
Incubation of the substrate ent-16
b-acetoxybeyeran-19-oic acid
(3) with C. echinulata NRRL 1386 resulted in the production of two
new metabolites, 8 and 9, which have been identified by spectro-
ꢀ
scopic techniques. Compound 8 showed the [MeH] ion at m/z
ꢀ
3
77.2298 in the HR-TOFMS (ESI ), corresponding to C22
H
34
O
5
. The
1
H NMR data were similar to those of the parent compound 3. The
significant different was the presence of the carbinolic proton (H-7)
at 3.56 which correlated to C-7 ( 76.5) in the HMQC spectrum.
The acetoxyl group was still present as a singlet signal at 2.04. The
HMBC correlations of H-7 with C-5 ( 47.2), C-6 ( 28.8), C-9 ( 49.7)
and C-15 ( 39.3) confirmed the location of the hydroxyl group. The
small coupling constant of H-7 (W
¼ 6.5 Hz) of 8 was consistent
with the -orientation of the 7-hydroxyl group. The spectroscopic
data (see experimental part) led to the identification of 8 as ent-
d
d
d
d
d
d
d
½
b
1
6
b
-acetoxy-7
a
-hydroxybeyeran-19-oic acid.
þ
Compound 9 showed the [MþNa] ion at m/z 401.2298 in the
þ
HR-TOFMS (ESI ), which was in agreement with the molecular
1
formula C22
The significant different was the presence of the acetoxyl group at
2.01 and the downfield shift (0.12 ppm) of H-16. The carbinolic
34 5
H O . The H NMR data of 9 were similar to those of 7.
d
C
100
4
proton at
4.7), C-12 (
the location of the hydroxyl group at the 14-position. The
d
3.04 (H-14) showed HMBC correlations with C-9 (
d
8
1
5
7
5
d
33.3), C-15 ( 38.3) and C-16 (
d
d
81.1), thus confirming
b
-ori-
*
entation of H-14 was also deduced from the ROESY experiments by
the same analogy to that of compound 7. The structure of 9 was
80
6
9
3
*
thus established as ent-16
acid. The production of the 14
indicated that another enzyme of C. echinulata NRRL 1386 regio-
selectively and stereoselectively hydroxylated at the 14-position of
the beyerane diterpenoids.
b
-acetoxy-14
b
-hydroxybeyeran-19-oic
*
*
2
*
*
*
*
*
*
a-hydroxy analogs 7 and 9 have
60
40
20
0
Compound
Compound
Compound
Compound
Compound
Compound
Compound
Compound
Compound
1
2
3
4
5
6
7
8
9
C
Acetylcholine
-
10
-9
-8
-7
-6
Log [conc] (M)
-5
-4
-3
ꢀ
9
Fig. 2. Effect of various concentrations of isosteviol (1) and analogs 2e9 (10 to
ꢀ
4
ꢀ6
10
M) on the relaxation of phenylephrine (10 M)-induced contraction in rat aorta.
Fig. 1. Selected ROESY correlations of compound 7.
*p < 0.05, significant difference compared to control (acetylcholine).

774
O. Wonganan et al. / European Journal of Medicinal Chemistry 62 (2013) 771e776
Table 1
0.1% yeast extract, 0.1% beef extract and 0.5% glucose were inocu-
lated with freshly obtained C. echinulata cultured from the agar
slant on a rotary shaker at 200 rpm for 72 h [25].
Vasorelaxant activity of compounds 1e9.
Compound
EC50 (nM)
1
2
3
57.41
29.07
4.88
5.3. Chemical modification of isosteviol (1)
4
5
6
7
8
9
3.46
5
.3.1. Acid hydrolysis of stevioside to isosteviol (1)
41.75
39.79
303.40
59.71
19.53
65.00
Stevioside (1.0 g) was hydrolyzed with 20% H
2
SO
Cl
4
(10 ml) at 60e
and the crude
ꢁ
70 C for 2 h. The mixture was extracted with CH
2
2
extract was subjected to column chromatography eluting with
CH Cl eMeOH (98:2) to yield 1 [26] (320.4 mg, 81%). Spectroscopic
H NMR and ESMS) data were in agreement with the structure and
Acetylcholinea
2
2
1
(
a
As acetylcholine iodide.
1
the H NMR data were consistent with the reported values [26].
5
.3.2. Synthesis of ent-16
To a solution of compound 1 (150 mg) in THF (2 ml) was added
NaBH (50 mg). After stirring at ambient temperature for 1 h, water
was added and the solution was extracted with EtOAc. The EtOAc
layer was washed with water, dried over anhydrous Na SO and the
solvent was evaporated in vacuo. The crude product was purified by
column chromatography using CH Cl eMeOH (95:5) to give 2
137 mg). Spectroscopic ( H NMR and ESMS) data were in agree-
b-hydroxybeyeran-19-oic acid (2)
isosteviol (1), introduction of the 7
b
-hydroxyl group markedly
increased the vasorelaxant activity. However, for the dihydro ana-
logs of isosteviol, hydroxylations at C-7 and C-14 decreased the
activity. The activity of 4 was particularly significant (see Fig. 2).
This compound exerted vasorelaxant activity at particularly low
concentration and the vasorelaxant profile reached maximum at
relatively low concentration, especially when compared with the
positive control acetylcholine. It is interesting to note that this
compound was not toxic against the Vero cell, one of the repre-
sentatives of normal cell line (data not shown).
4
2
4
2
2
1
(
ment with the structure and the NMR data were consistent with the
reported values [18e20].
5
.3.3. Acetylation of ent-16
Acetic anhydride (0.5 ml) was slowly added to a solution of 2
150 mg) in pyridine (1 ml). The progress of the reaction was fol-
b-hydroxybeyeran-19-oic acid (2)
4
. Conclusion
(
Chemical modification of isosteviol (1) at 16-keto group to its
lowed by TLC while the mixture was kept stirring for 2 h. The re-
action was worked up with water and extracted with EtOAc. The
organic layer was washed with H O, dried over anhydrous Na SO
and the solvent evaporated in vacuo. The residue was subjected to
column chromatography eluting with CH Cl eMeOH (98:2) to
dihydro analog 2 and the corresponding acetate derivative 3
resulted in marked increase in vasorelaxant activity of the latter
analog. Its EC50 value was 4.88 nM, or 12-fold more active than the
parent compound 1. The substrates 1e3 were subjected to micro-
bial transformation by C. echinulata NRRL 1386 to the metabolites
2
2
4
2
2
yield ent-16b-acetoxybeyeran-19-oic acid (3) (130 mg).
4
e9. The analog 4 exhibited the most active vasorelaxant effect,
with the EC50 value of 3.46 nM, or approximately 17-fold more
active than compound 1. The highly active compound 4 deserves
special attention for antihypertensive drug development.
5
.3.3.1. ent-16
b-Acetoxybeyeran-19-oic acid (3). Colorless solid
ꢁ
31 ꢀ 56.7 (c ¼ 0.50 in MeOH);
(from methanol), m.p. 196e198
IR (KBr) nmax cm : 3267, 2949, 2846, 1728, 1705, 1455, 1372, 1278,
C;
a
D
ꢀ
1
1
1
214, 1150, 1033, 976, 819; H NMR (400 MHz, CDCl
H-20), 0.81 (1H, partially superimposed signal, H-1), 0.87 (3H, s, H-
7), 0.94 (1H, partially superimposed signal, H-3ax), 0.97 (1H,
3
) d 0.78 (3H, s,
5
. Experimental
1
5
.1. General
obscured signal, H-9), 1.01 (1H, br d, J ¼ 13.8 Hz, H-5), 1.04 (1H, d,
J ¼ 12.1 Hz, H-12), 1.19 (3H, s, H-18), 1.25 (1H, partially super-
imposed signal, H-12), 1.30 (1H, br d, J ¼ ca 12 Hz, H-7), 1.32 (1H, d,
J ¼ 12.1 Hz, H-14),1.37 (1H, br d, J ¼ 13.1 Hz, H-2),1.50 (1H, obscured
signal, H-7),1.53 (2 ꢂ 1H, m, 2 ꢂ H-11), 1.64 (1H, br d, J ¼ 13.8 Hz, H-
6), 1.68 (1H, obscured signal, H-1), 1.74 (1H, obscured signal, H-12),
1.80 (1H, obscured signal, H-6), 1.82 (2 ꢂ 1H, m, 2 ꢂ H-15), 1.86 (1H,
obscured signal, H-2), 2.02 (3H, s, OAc), 2.11 (1H, br d, J ¼ 13.1 Hz, H-
Melting points were determined with an Electrothermal melt-
ing point apparatus and are uncorrected. Optical rotations were
measured on a JASCO-1020 polarimeter. IR spectra were obtained
using a Bruker Tensor 27 FT-IR spectrophotometer. H and C NMR
spectra were recorded on a Bruker AVANCE 400 FT-NMR spec-
1
13
1
13
trometer, operating at 400 ( H) and 100 ( C) MHz. For the spectra
taken in CDCl and C N, the residual nondeuterated solvent
signals at 7.24 and 8.71, and the solvent signals at 77.0 and
49.90 were used as references for H and C NMR spectra,
13
3
5
D
5
3eq), 4.70 (1H, dd, J ¼ 9.9, 4.6 Hz, H-16); C NMR (100 MHz, CDCl ):
3
d
d
3
d 13.3 (C-20), 18.8 (C-2), 20.2 (C-11), 21.1 (CH -acetate), 21.5 (C-6),
1
13
1
24.8 (C-17), 29.0 (C-18), 34.5 (C-12), 37.7 (C-3), 38.1 (C-10), 39.8 (C-
1), 40.6 (C-15), 41.4 (C-7, C-13), 42.2 (C-8), 43.6 (C-4), 54.7 (C-14),
55.7 (C-9), 57.0 (C-5), 81.7 (C-16), 171.4 (CO-acetate), 184.1 (C-19);
respectively. ESMS and ES-TOFMS spectra were measured with
a Finnigan LC-Q and a Bruker micrOTOF mass spectrometer. Col-
umn chromatography was carried out using Merck silica gel 60
ꢀ
ESMS (eve): m/z (% rel. abund.) 723 [2MꢀH] (100); HR-TOFMS
ꢀ
ꢀ
(<0.063 mm) and Pharmacia Sephadex LH-20. For TLC, Merck
(ESI ): m/z 361.2383 [MꢀH] ; calcd for C
2 34 4
H O ꢀH, 361.2373.
2
precoated silica gel 60 F254 plates were used. Spots on TLC were
detected under UV light and by spraying with anisaldehydeeH
2
SO
4
5.4. Biotransformation by C. echinulata NRRL 1386
reagent followed by heating.
5.4.1. Incubation of isosteviol (1)
5.2. Microorganism, media and culture conditions
The stock culture of C. echinulata NRRL 1386 was maintained
on a potato dextrose agar slant. Twenty Erlenmeyer flask (250 ml),
each containing 100 ml of liquid medium consisting of 0.1% pep-
tone, 0.1% yeast extract, 0.1% beef extract, and 0.5% glucose, were
inoculated with freshly obtained C. echinulata cultured from the
The stock culture of C. echinulata NRRL 1386 was maintained on
a potato dextrose agar slant. Erlenmeyer flasks (250 ml), each
containing 100 ml of liquid medium consisting of 0.1% peptone,

O. Wonganan et al. / European Journal of Medicinal Chemistry 62 (2013) 771e776
775
agar slant on a rotary shaker at 200 rpm. After cultivation at
ambient temperature for 72 h, the substrate solution (100 mg of
substrate dissolved in 2.0 ml DMSO and 0.2 ml Tween 80) was
prepared [25]. Equal volume of the substrate solution was added
to each flask and the incubation continued for 3 days. Culture
control consisted of fermentation blank in which C. echinulata was
grown under identical condition but without substrate. After 3
days the culture was filtered and the broth was extracted with
EtOAc, washed with water and the solvent was evaporated in
vacuo. The crude extract (210.8 mg) was subjected to column
5.4.2.3. ent-14
solid (from methanol); m.p. 215e218 C;
b
,16
b
-Dihydroxybeyeran-19-oic acid (7). Colorless
ꢁ
29
D
a
ꢀ 29.9 (c ¼ 0.32 in
ꢀ1
MeOH); IR (KBr):
n
max cm 3452, 3411, 2927, 1696, 1462, 1260, 1155,
1
1055, 990; H NMR (400 MHz, C N)
D
5 5
d
0.97 (1H, ddd, J ¼ 12.4, 12.2,
4.2 Hz, H-1ax), 1.09 (1H, obscured signal, H-3), 1.13 (1H, obscured
signal, H-5),1.24 (3H, s, H-20), 1.27 (1H, dd, J ¼ 10.4, 6.3 Hz, H-9), 1.34
(3H, s, H-18), 1.39 (1H, m, H-12ax), 1.44 (3H, s, H-17), 1.47 (1H, m, H-
2), 1.64 (2 ꢂ 1H, m, 2 ꢂ H-11), 1.68 (1H, obscured signal, H-7ax), 1.81
(1H, br d, J ¼ 12.4 Hz, H-1eq), 2.18 (2 ꢂ 1H, m, 2 ꢂ H-6); 2.24 (1H,
partially superimposed signal, H-2); 2.30 (1H, dd, J ¼ 12.6, 5.3 Hz, H-
12eq), 2.36 (2H, d, J ¼ 7.1 Hz, 2 ꢂ H-15), 2.44 (1H, m, H-3), 2.46 (1H,
chromatography eluting with CH
increasing amount of the more polar solvent, to give 4 (45.7 mg,
4% based on the unrecovered starting material) and the starting
2 2 2 2
Cl and CH Cl eMeOH with
13
m, H-7eq), 3.49 (1H, s, H-14), 4.86 (1H, t, J ¼ 7.1 Hz, H-16); C NMR
(100 MHz, C N): 14.2 (C-20), 19.9 (C-2), 20.7 (C-11), 21.0 (C-17),
5
5
D
5
d
material 1 (20 mg). Compound 4 was the biotransformation
product of isosteviol (1) by several fungi and the NMR spectro-
scopic data of our compound were consistent with the reported
values [9,22e24].
22.3 (C-6), 29.6 (C-18), 34.2 (C-12), 37.5 (C-7), 38.8 (C-3, C-10), 40.8
(C-1), 42.3 (C-15), 44.0 (C-4), 47.4 (C-13), 47.6 (C-8), 56.1 (C-9), 57.1
(C-5), 78.8 (C-16), 91.8 (C-14), 180.4 (C-19); ESMS (eve): m/z (% rel.
ꢀ
þ
abund.) 335 [MꢀH] (100); HR-TOFMS (ESI ): m/z 359.2180
þ
[
MþNa] ; calcd for C20
32 4
H O þNa, 359.2192.
5
.4.1.1. ent-7
a
-Hydroxy-16-ketobeyeran-19-oic acid (4). Colorless
ꢁ
29
D
solid (from methanol), m.p. 230e232 C;
MeOH); IR (KBr)
a
ꢀ 90.6 (c ¼ 0.06 in
5.4.3. Incubation of ent-16b-acetoxybeyeran-19-oic acid (3)
ꢀ1
n
max cm : 3451, 3128, 2960, 2918, 2851, 1713, 1453,
Microbial transformation of compound 3 (100 mg) was carried
out in similar manner to that of compound 1. The crude extract
(240.7 mg) was subjected to column chromatography eluting with
1
371, 1337, 1260, 1241, 1182, 1149, 1118, 1050, 952, 887, 856, 783;
ꢀ
ESMS (eve): m/z (% rel. abund.) 667 [2MꢀH] (100).
CH
2
Cl
2
and CH
2
Cl
2
eMeOH with increasing amount of the more
Cl to CH Cl eMeOH (5:0.1) to give the
5
.4.2. Incubation of ent-16 -hydroxybeyeran-19-oic acid (2)
b
polar solvent from CH
2
2
2
2
Compound 2 (100 mg) was subjected to microbial trans-
formation in similar manner as that of compound 1. The crude
extract (195.7 mg) was purified by column chromatography on
metabolites 8 and 9 (35.8 and 18.3 mg, respectively). The yields of 8
and 9 were 34 and 18%, respectively.
silica gel with a stepwise elution with CH
2
Cl
2
and CH
2
Cl
2
eMeOH
5.4.3.1. ent-16
Colorless solid (from methanol); m.p. 220e221 C;
b
-Acetoxy-7
a
-hydroxybeyeran-19-oic
acid
(8).
ꢀ 38.9
ꢁ
29
a
(
2:0.1). The fractions containing the products were further puri-
D
ꢀ
1
fied by Sephadex LH-20 and crystallized with MeOH to afford the
metabolites 5, 6 and 7 (31.3 mg, 19.5 mg, 17.7 mg, respectively). The
yields of 5, 6 and 7 were 30,19 and 17%, respectively. The metabolite
(c ¼ 0.32 in MeOH); IR (KBr):
n
max cm 3505, 2992, 2944, 1718,
1
3
1444, 1372, 1253, 1059; H NMR (400 MHz, CDCl ) d 0.79 (3H, s, H-
20), 0.92 (3H, s, H-17), 0.97 (1H, partially superimposed signal, H-
1), 1.05 (1H, ddd, J ¼ 13.0, 12.0, 4.3 Hz, H-3ax), 1.20 (3H, s, H-18),
1.23 (1H, partially superimposed signal, H-12), 1.34 (2 ꢂ 1H, s,
2 ꢂ H-14), 1.41 (2 ꢂ 1H, obscured signal, H-2 and H-9), 1.58
5
was identified as ent-7a,16b-dihydroxybeyeran-19-oic acid by
comparison of the spectroscopic data with those of the reported
values [18,19]. Compounds 6 and 7 were new metabolites.
(
2 ꢂ 1H, m, 2 ꢂ H-11), 1.66 (1H, obscured signal, H-5), 1.70 (1H,
5
.4.2.1. ent-7
a
,16
b
-Dihydroxybeyeran-19-oic acid (5). White solid;
obscured signal, H-1), 1.76 (2 ꢂ 1H, obscured signal, 2 ꢂ H-15), 1.77
(1H, obscured signal, H-12), 1.85 (1H, obscured signal, H-2), 1.91
(2 ꢂ 1H, m, 2 ꢂ H-6), 2.04 (3H, s, OAc), 2.14 (1H, br d, J ¼ 13.0 Hz, H-
ꢁ
31
D
(
from methanol); m.p. 228e230 C;
a
ꢀ 23.5 (c ¼ 0.10 in MeOH);
ꢀ
1
IR (KBr): nmax cm 3476, 3412, 2978, 2952, 2889, 2845, 1701, 1456,
1
386, 1214, 1155, 1061, 1020; ESMS (eve): m/z (% rel. abund.) 671
3eq), 3.56 (1H, br s, W
½
¼ 6.5 Hz, H-7), 4.70 (1H, dd, J ¼ 10.2,
ꢀ
13
[2MeH] (100).
4.4 Hz, H-16); C NMR (100 MHz, CDCl
3
): d 13.1 (C-20), 18.8 (C-2),
1
9.8 (C-11) 21.1 (CH -acetate), 24.9 (C-17), 28.8 (C-6, C-18), 34.3
3
5
.4.2.2. ent-7
b
,16
b
-Dihydroxybeyeran-19-oic acid (6). Colorless
(C-12), 37.6 (C-3), 38.0 (C-10), 39.3 (C-15), 39.5 (C-1), 41.5 (C-13),
43.1 (C-4), 46.6 (C-8), 47.2 (C-5), 49.7 (C-9), 50.2 (C-14), 76.5 (C-7),
81.1 (C-16), 171.4 (CO-acetate), 183.1 (C-19); ESMS (eve): m/z (%
ꢁ
29
D
solid (from methanol); m.p. 226e228 C;
MeOH); IR (KBr): nmax cm 3624, 3414, 3250, 2937, 2869, 2846,
a
ꢀ 51.0 (c ¼ 0.39 in
ꢀ1
1
ꢀ
ꢀ
1701, 1650, 1455, 1260, 1193, 1069; H NMR (400 MHz, C
5
D
5
N):
rel. abund.) 755 [2MꢀH] (100); HR-TOFMS (ESI ): m/z 377.2298
ꢀ
d
0.95 (1H, ddd, J ¼ 13.0, 12.8, 3.3 Hz, H-1ax), 1.08 (1H, d, J ¼ 11.7 Hz,
[MeH] ; calcd for C22
H
34
O
5
ꢀH, 377.2322.
H-14a), 1.10 (1H, obscured signal, H-3), 1.13 (3H, s, H-17), 1.22 (3H, s,
H-20), 1.29 (1H, obscured signal, H-9), 1.32 (1H, obscured signal, H-
5
5.4.3.2. ent-16
Colorless solid (from methanol); m.p. 218e220 C;
b
-Acetoxy-14
b
-hydroxybeyeran-19-oic
acid
(9).
ꢁ
29
), 1.35 (3H, s, H-18), 1.49 (1H, partially superimposed signal, H-12),
a
ꢀ 43.5
D
ꢀ1
1
.50 (1H, br d, J ¼ ca 14 Hz, H-2), 1.72 (1H, obscured signal, H-11),
(c ¼ 0.32 in MeOH); IR (KBr):
nmax cm 3509, 3205, 2949, 2880,
1
1.78 (1H, br d, J ¼ 13.0 Hz, H-1eq), 1.93 (1H, br d, J ¼ 12.0 Hz, H-15a),
3
1718, 1456, 1373, 1253, 1000; H NMR (CDCl ) d 0.77 (3H, s, H-20),
2
2
2
2
.23 (1H, obscured signal, H-11), 2.26 (1H, obscured signal, H-2),
.27 (1H, obscured signal, H-12), 2.40 (1H, obscured signal, H-6ax),
.46 (1H, br d, J ¼ 11.7 Hz, H-14b), 2.47 (1H, br d, J ¼ 13.0 Hz, H-3),
.57 (1H, br d, J ¼ 11.5 Hz, H-6eq), 3.00 (1H, dd, J ¼ 12.0, 10.6 Hz, H-
0.82 (1H, ddd, J ¼ 13.3, 13.0, 3.9 Hz, H-1), 0.94 (3H, s, H-17), 0.95
(1H, partially superimposed signal, H-3), 1.03 (1H, dd, J ¼ 10.4,
6.3 Hz, H-9), 1.15 (3H, s, H-18), 1.20 (1H, partially superimposed
signal, H-5), 1.25 (1H, obscured signal, H-12), 1.28 (1H, obscured
signal, H-7), 1.35 (1H, br d, J ¼ 13.3 Hz, H-2), 1.47 (2 ꢂ 1H, m, 2 ꢂ H-
11),1.58 (1H, br d, J ¼ 13.0 Hz, H-6),1.72 (2 ꢂ1H, obscured signal, H-
1 and H-12), 1.79 (1H, obscured signal, H-15a), 1.82 (2 ꢂ 1H, par-
tially superimposed signal, H-2 and H-6),1.86 (1H, obscured signal,
H-7), 2.01 (3H, s, OAc), 2.03 (1H, partially superimposed signal, H-
1
4
2
5b), 3.70 (1H, dd, J ¼ 11.5, 3.6 Hz, H-7), 4.25 (1H, dd, J ¼ 10.6,
.2 Hz, H-16); 13C NMR (100 MHz, C
0.8 (C-11), 25.9 (C-17), 29.5 (C-18), 32.6 (C-6), 34.3 (C-15), 34.6 (C-
2), 38.6 (C-10), 38.9 (C-3), 40.3 (C-1), 42.2 (C-13), 43.8 (C-4), 49.0
5 5
D N): d 14.0 (C-20), 19.7 (C-2),
1
(
C-8), 51.0 (C-14), 54.6 (C-5), 55.9 (C-9), 75.7 (C-7), 80.0 (C-16),
ꢀ
1
80.3 (C-19); ESMS (eve): m/z (% rel. abund.) 671 [2MꢀH] (100);
15b), 2.08 (1H, br d, J ¼ 13.3 Hz, H-3), 3.04 (1H, s, H-14), 4.98 (1H,
ꢀ
ꢀ
13
HR-TOFMS (ESI ): m/z 335.2208 [MꢀH] ; calcd for C20
32
H O
4
ꢀH,
dd, J ¼ 10.3, 3.9 Hz, H-16); C NMR (C
5 5
D N, 100 MHz): d 13.3 (C-
3
35.2217.
3
20), 18.8 (C-2), 19.4 (C-17), 19.6 (C-11), 20.8 (C-6), 21.0 (CH -

776
O. Wonganan et al. / European Journal of Medicinal Chemistry 62 (2013) 771e776
acetate), 28.9 (C-18), 33.3 (C-12), 35.5 (C-7), 37.7 (C-3), 38.1 (C-10),
Appendix A. Supplementary data
3
9
8.3 (C-15), 40.0 (C-1), 43.3 (C-4), 46.0 (C-13), 46.4 (C-8), 54.7 (C-
), 56.3 (C-5), 81.1 (C-16), 90.8 (C-14), 171.4 (CO-acetate), 183.0 (C-
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2013.01.022.
ꢀ
1
9); ESMS (eve): m/z (% rel. abund.) 755 [2MꢀH] (100); HR-
þ
þ
TOFMS (ESI ): m/z 401.2298 [MþNa] ; calcd for C22
H
34
O
5
þNa,
4
01.2304.
References
[
1] L.L. Brunton, K.L. Parker, D.K. Blumenthal, I.L.O. Buxton, Goodman & Giman’s
Manual of Pharmacology and Therapeutics, McGraw-Hill, New York, 2008.
2] F.H. Messerli, S. Bangalore, S. Julius, Circulation 117 (2008) 2706e2715.
5
5
.5. Vasorelaxant assay
.5.1. Animals
[
[3] J. Varon, Drugs 68 (2008) 283e297.
[
[
[
4] A.G. Avent, J.R. Hanson, B.H. de Oliveira, Phytochemistry 29 (1990) 2712e2715.
5] J.M.C. Geuns, Phytochemistry 64 (2003) 913e921.
6] K.-L. Wong, P. Chan, H.-Y. Yang, F.-L. Hsu, I.M. Liu, Y.-W. Cheng, J.-T. Cheng,
Life Sci. 74 (2004) 2379e2387.
Male Sprague-Dawley rats (age 8 weeks) were obtained from
the National Laboratory Animal Center, Mahidol University, Nakhon
Pathom, Thailand. The animals were housed under a 12:12-h lighte
ꢁ
[7] K.L. Wong, H.Y. Yang, P. Chan, T.H. Cheng, J.C. Liu, F.L. Hsu, I.M. Liu, Y.W. Cheng,
dark cycle and maintained at 24 ꢃ 1 C. Animal feed and water was
J.T. Cheng, Planta Med. 70 (2004) 108e112.
supplied ad libitum.
[
[
8] D. Xu, Y. Li, J. Wang, A.K. Davey, S. Zhang, A.M. Evans, Life Sci. 80 (2007)
69e274.
9] F.-L. Hsu, C.-C. Hou, L.-M. Yang, J.-T. Cheng, T.-C. Chi, P.-C. Liu, S.-J. Lin, J. Nat.
2
5.5.2. Smooth muscle tone determination
Prod. 65 (2002) 273e277.
Aortic rings 4 mm in length were cut and immediately placed in
[
[
[
10] J. Ma, Z. Ma, J. Wang, R.W. Milne, D. Xu, A.K. Davey, A.M. Evans, Diabetes Obes.
100% oxygen-saturated HEPES-buffer physiological salt solution
Metab. 9 (2007) 597e599.
11] Y. Mizushina, T. Akihisa, M. Ukiya, Y. Hamasaki, C.M. Nakai, I. Kuriyama,
T. Takeuchi, F. Sugawara, H. Yoshida, Life Sci. 77 (2005) 2127e2140.
12] P. Chan, D.-Y. Xu, J.-C. Liu, Y.-J. Chen, B. Tombson, W.-P. Huang, J.-T. Cheng, Life
Sci. 63 (1998) 1679e1684.
(HPSS: 140 mM NaCl, 5 mM KCl, 2 mM CaCl , 1 mM MgCl , 5 mM
2
2
HEPES,11 mM glucose, pH 7.4). The ring was equilibrated for 2 h with
several changes of HPSS. The changes in isometric force were recor-
ded on a computer system using the Chart version 7 PowerLab
ADInstruments (2009) program. The resting tension was pre-
contracted with phenylephrine (PE). The % relaxation of the aortic
[13] P. Chan, B. Tomlinson, Y.-J. Chen, J.-C. Liu, M.-H. Hsieh, J.-T. Cheng, Br. J. Clin.
Pharmacol. 50 (2000) 215e220.
[
14] D. Xu, S. Zhang, D.J.R. Foster, J. Wang, Clin. Exp. Pharmacol. Physiol. 34 (2007)
488e493.
ꢀ9
ꢀ4
rings against log concentration of the compounds at 10 to 10 M is
presented inFig. 2. The EC50 of compounds1e9 ispresented inTable 1.
[15] M.-H. Hsieh, P. Chan, Y.-M. Sue, J.-C. Liu, T.H. Liang, T.-Y. Huang, B. Tomlinson,
M.S.S. Chow, P.-F. Kao, Y.-J. Chen, Clin. Ther. 25 (2003) 2797e2808.
[
16] L.A.F. Ferri, W. Alves-Do-Prado, S.S. Yamada, S. Gazola, M.R. Batista,
R.B. Bazotte, Phytother. Res. 20 (2006) 732e736.
5
.5.3. Statistical analysis
Statistical analysis was performed with one way analysis of
[17] G.I. Kovylyaeva, G.A. Bakaleinik, I.Y. Strobykina, V.I. Gubskaya, R.R. Sharipova,
0
V.A. Al fonsov, V.E. Kataev, A.G. Tolstikov, Chem. Nat. Compd 43 (2007) 81e85.
[
18] C.-L. Lin, S.-J. Lin, W.-J. Huang, Y.-L. Ku, T.-H. Tsai, F.-L. Hsu, Planta Med. 73
2007) 1581e1587.
[19] L.-M. Yang, F.-L. Hsu, J.-T. Cheng, C.-H. Chang, P.-C. Liu, S.-J. Lin, Planta Med. 70
2004) 359e363.
variance (ANOVA). The differences were considered statistically
significant when compared to normal control at P < 0.05.
(
(
[
[
[
[
20] M.S. Ali, J.R. Hanson, B.H. de Oliveira, Phytochemistry 31 (1992) 507e510.
21] B.H. de Oliveira, R.A. Strapasson, Phytochemistry 43 (1996) 393e395.
22] B.H. de Oliveira, M.C. Dos Santos, P.C. Leal, Phytochemistry 51 (1999) 737e741.
23] T. Akihisa, Y. Hamasaki, H. Tokuda, M. Ukiya, Y. Kimura, H. Nishino, J. Nat.
Prod. 67 (2004) 407e410.
Acknowledgments
This work was supported by Higher Education Research Pro-
motion (HERP) Program, Office of the Higher Education Commis-
sion. Partial supports from National Science and Technology
Development Agency (NSTDA), The Thailand Research Fund and
Center of Excellence for Innovation in Chemistry (PERCH-CIC) are
gratefully acknowledged.
[
[
[
24] S.-F. Chang, L.-M. Yang, C.-H. Lo, J.-H. Liaw, L.-H. Wang, S.-J. Lin, J. Nat. Prod. 71
(2008) 87e92.
25] C. Changtam, O. Sukcharoen, B. Yingyongnarongkul, N. Chimnoi,
A. Suksamrarn, Tetrahedron 64 (2008) 2626e2633.
26] M. Korochkina, M. Fontanella, A. Casnati, A. Arduini, F. Sansone, R. Ungaro,
S. Latypov, V. Kataev, V. Alfonsov, Tetrahedron 61 (2005) 5457e5463.