Biocatalytic and semisynthetic optimization of the anti-invasive tobacco (1S,2E,4R,6R,7E,11E)-2,7,11-cembratriene-4,6-diol

Article abstract of DOI:10.1016/j.bmc.2007.12.056

Tobacco cembranoids were reported to inhibit tumorigenesis. Biocatalysis of (1S,2E,4R,6R,7E,11E)-2,7,11-cembratriene-4,6-diol (1) using the symbiotic Bacillus sp. NC5, Bacillus sp. NK8, and Bacillus sp. NK7, isolated from the Red Sea sponge Negombata magn

Full text of DOI:10.1016/j.bmc.2007.12.056

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 2886–2893
Biocatalytic and semisynthetic optimization of the anti-invasive
tobacco (1S,2E,4R,6R,7E,11E)-2,7,11-cembratriene-4,6-diol
Khalid A. El Sayed,a,* Surat Laphookhieo,a Hany N. Baraka,a Muhammad Yousaf,a
Anne Hebert,a Danielle Bagaley,b Frederick A. Rainey,b A. Muralidharan,a
Shibu Thomasa and Girish V. Shaha
aDepartment of Basic Pharmaceutical Sciences, College of Pharmacy, University of Louisiana at Monroe,
700 University Avenue, Monroe, LA 71209, USA
bDepartment of Biological Sciences, Louisiana State University, Baton Rouge, LA 70803, USA
Received 24 November 2007; revised 27 December 2007; accepted 28 December 2007
Available online 3 January 2008
Abstract—Tobacco cembranoids were reported to inhibit tumorigenesis. Biocatalysis of (1S,2E,4R,6R,7E,11E)-2,7,11-cembratriene-
4,6-diol (1) using the symbiotic Bacillus sp. NC5, Bacillus sp. NK8, and Bacillus sp. NK7, isolated from the Red Sea sponge Negom-
bata magnifica, afforded two new and four known hydroxylated metabolites 3–8. The use of symbiotic marine bacteria as biocata-
lysts for bioactive natural product scaffolds is very rare. Cembranoid 1 carbamate analogs 9–11 were prepared by its reaction with
corresponding isocyanates. Cembranoid 1 and its bioconversion and carabamate products show anti-invasive activity against the
human highly metastatic prostate PC-3M cancer cell line at 10–50 nM doses in Matrigel assay.
ꢀ 2008 Elsevier Ltd. All rights reserved.
1. Introduction
in rats, suggesting a possible use of these compounds for
the treatment of nicotine addiction.6 Tobacco cembra-
noids also show insecticidal, prostaglandin, plant
growth, and fungal spore germination inhibitory activi-
ties.1,2,5,7 Cembranoid 1 and its C-4 epimer from ciga-
rette smoke condensate were reported as anticancer
ingredients in 1985.8 Later, both cembranoids were re-
ported to inhibit the induction of Epstein–Barr virus
(EBV) early antigen by the phorbol ester, 12-O-tetra-
decanoylphorbol-13-acetate (TPA), in lymphoblastoid
Raji cells.9 They also inhibited the induction of orni-
thine decarboxylase (OD) activity by TPA in the mouse
epidermis and inhibited the promotional effects of TPA
on skin tumor formation initiated by 7,12-dimethyl-
benz[a]anthracene (DMBA) in mice.10
The leaf and flower cuticular wax of most Nicotiana
species contain high amounts of cembranoid diter-
penes.1 Cembranoids possess 14-membered macrocy-
clic rings substituted by an isopropyl residue at C-1
and by three symmetrically disposed methyl groups
at positions C-4, C-8, and C-12. The two epimeric to-
bacco cembranoids (1S,2E,4R,6R,7E,11E)-2,7,11-cem-
bratriene-4,6-diol (1) and its C-4 epimer were among
the earliest reported.1–4 The structure and absolute
stereochemistry of 1 was based on X-ray crystallogra-
phy.1–3 Both cembranoids are key flavor ingredients in
tobacco.1,2 Biodegradation of 1 and its epimer during
flue-cured fermentation of tobacco leaves produces a
range of flavor compounds.1,2,5
Biocatalysis is the use of growing microbial cultures, en-
zymes, or immobilized cells to enhance the bioactivity of
a starting material through induction of stereospecific
reactions.12–14 Although biocatalysis has been used to
generate new chiral derivatives and increase the efficacy
of drugs by metabolic activation is well documented, the
use of marine microorganisms as biocatalysts for a bio-
active natural product scaffold has been rarely re-
ported.12–16
Tobacco cembranoids exert a wide range of bioactivity.
They block the expression of the behavioral sensitization
to nicotine and inhibit neuronal acetylcholine receptors
Keywords: Tobacco cembranoids; Biocatalysis; Symbiotic marine
microbials; Semisynthesis; Anti-invasive.
*
Corresponding author. Tel.: +1 318 342 1725; fax: +1 318 342
1737; e-mail: elsayed@ulm.edu
0968-0896/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.12.056

K. A. El Sayed et al. / Bioorg. Med. Chem. 16 (2008) 2886–2893
2887
18
O
O
HO
R1
OH
OH
Cl
O
O
6
4
N
H
a
N
H
1'
2'
3'
R5
19
1
15
13
b
16
R2
O
R4
O
4'
10
6'
20
O
R3
R6
N
2
H
8'
c
R1
R2
H
H
R3
H
H
R4
R5
H
OH
H
H
H
H
R6
H
H
H
OH
H
H
H
H
H
1
3
4
5
6
7
8
9
OH
OH
OH
OH
OH
OH
OH
a
H
H
H
H
β-OH H
H
H
H
H
H
H
H
H
α-OH H
β-OH H
H
H
H
H
α-OH H
H
H
H
H
H
H
10
11
b
c
H
Studies examining the biotransformation of 1 and its
epimer started in 1987, mainly to generate flavor-
enhancers that can be added to tobacco during its fer-
mentation.2,3,11,12 The terrestrial bacterium Bacillus
megaterium NH5 bioconverted tobacco cembranoids
to C-10- and C-20 hydroxy analogs, improving the to-
bacco natural odor and taste and reducing its irritating
effects on the throat and lungs.12,17,18 Plant cell cultures
Tripterygium wilfordii and Nicotiana sylvesteris also bio-
transformed 1, its C-4 epimer, and other minor tobacco
cembranoids.12,19–21 These reactions mainly afforded 2,
the 11S, 12S epoxide analog of 1, the epoxide analog,
and several minor 10-, 12-, and 13-enzymatically
hydroxylated and rearranged triols of the C-4 epimer
of 1.12,19–21
2. Results and discussion
High amounts of tobacco a- and b-2,7,11-cembratriene-
4,6-diols were isolated from fresh Nicotiana tabacum leaf
powder. Their identification was based on extensive
analyses of their NMR data and comparison with the
literature.1–3,11 Several symbiotic bacteria were cultured
and isolated from the Red Sea sponge Negombata mag-
nifica. The identity of the bacteria was revealed by 16S
rRNA sequence analysis.22 Thirty growing symbiotic
bacterial species were screened for their ability to bio-
convert 1. Of these, Bacilli species NC5, NK8, and
NK7 were selected for the scale-up bioconversion. This
selection was based on the TLC diversity of the gener-
ated metabolites.
There is a single report on the structure–activity rela-
tionship study aimed to optimize the anticancer activity
of tobacco cembranoids.10 Esterification of C-6 of 1
and its C-4 epimer using acetic, propionic, butyric,
valeric, and benzoic acid anhydrides reduced the activ-
ity in the Epstein–Barr virus early antigen induction
model by increasing the IC50 values and cytotoxicity,
suggesting the importance of a free C-6 hydroxy group
for the activity.10 Cytotoxicity increased with the in-
crease of acyl carbon number.10 Oxidation of C-6 alco-
hol of 1 and its C-4 epimer to b-hydroxyketones
slightly reduced the activity by increasing the IC50,
which further confirmed the importance of a hydrogen
bonding donor pharmacophore at C-6.10,12 The pur-
pose of the present study is to optimize the anticancer
activity of 1 using biocatalytic and semisynthetic
approaches.
Biocatalysis of 1 using Bacillus sp. NC5 afforded two
new hydroxylated metabolites 3 and 4, along with the
known metabolites (1S,2E,4R,6R,7E,11Z)-2,7,11-cem-
bratriene-4,6,20-triol
(5)
and
(1S,2E,4R,6-
R,7E,11E,10R)-2,7,11-cembratriene-4,6,10-triol (6).5,23
Fermentation of 1 for 14 days, using Bacilli species
NK7 and NK8 afforded the known (1S,2E,4R,6-
R,7E,11E,10S)-2,7,11-cembratriene-4,6,10-triol (7) and
(1S,2E,4R,6R,7E,11E,13R)-2,7,11-cembratriene-4,6,13-
triol (8), respectively.5,23
The HREIMS of metabolite 3 showed a molecular ion
peak at m/z 345.2406 [M+Na]+ corresponding to the
1
molecular formula C20H34O3. The H and 13C NMR
data (Table 1) suggested a monohydroxy derivative of
1. The methylene proton signals at
d 3.57 (d,
J = 11.1 Hz) and d 3.51 (d, J = 11.1 Hz) were assigned
for the newly oxygenated C-19 (Table 1). They showed
a 3J-HMBC correlation with the olefinic methine carbon
C-7 (d 130.2). Proton H-7 in turn showed a COSY cor-
relation with the oxygenated methine H-6 (d 4.82), con-
firming the assignment of segment C-6/C-8. The
Biocatalysis was conducted with the use of several novel
symbiotic marine bacteria. Semisynthetic carbamoyla-
tion of C-6 in 1 targeted the addition of more hydrogen
bonding donor and acceptor pharmacophores at this
key position to enhance the binding affinity of the prod-
ucts toward its target protein(s).
3
methylene H2-19 also showed a J-HMBC correlation
with the methylene C-9 (d 38.0). Thus, metabolite 3

2888
K. A. El Sayed et al. / Bioorg. Med. Chem. 16 (2008) 2886–2893
Table 1. 13C and 1H NMR Data of Metabolites 3–4a
Position
3a
4a
dC
dH
dC
dH
1
2
47.5, CH
131.8, CH
137.3, CH
72.1, qC
1.63, m
46.5, CH
131.8, CH
136.3, CH
71.5, qC
1.51, m
5.24, ddq (15.8, 8.8, 1.8)
5.40, d (15.8)
—
5.21, dd (15.8, 8.8)
5.35, d (15.8)
—
3
4
5
52.2, CH2
64.4, CH
130.2, CH
136.2, qC
38.0, CH2
22.5, CH2
128.0, CH
136.2, qC
27.9, CH2
32.0, CH2
32.8, CH
19.2, CH3
20.7, CH3
28.9, CH3
60.0, CH2
16.1, CH3
2.15, m 1.95, m
4.82, dd (9.8, 9.0)
5.47, d (9.8)
—
52.2, CH2
64.4, CH
126.4, CH
133.2, qC
70.9, CH
38.1, CH2
124.7, CH
136.8, qC
27.9, CH2
32.0, CH2
32.8, CH
19.2, CH3
20.7, CH3
28.9, CH3
15.1, CH2
16.1, CH3
2.00, m 1.88, m
4.86, dd (9.6, 9.2)
5.62, d (9.9)
—
6
7
8
9
2.25, m, 2.20, m
2.33, 2H, m
5.28, m
4.23, brs
10
11
12
13
14
15
16
17
18
19
20
2.40, m 2.32, m
4.99, brt (3.6)
—
—
2.27, m, 1.90, m
1.42, 2H, m
1.51, m
2.04, m, 1.66, m
1.40, m 1.31, m
1.51, m
0.79, 3H, d (6.6)
0.84, 3H, d (6.6)
1.39, 3H, s
0.78, 3H, d (6.6)
0.81, 3H, d (6.6)
1.41, 3H, s
1.53, 3H, s
1.69, 3H, s
3.57, d (11.1), 3.51, d (11.1)
1.76, 3H, s
a In CDCl3, 400 MHz for 1H and 13C NMR. Coupling constants (J) are in Hz.
was proved to be (1S,2E,4R,6R,7Z,11E)-2,7,11-cembr-
atriene-4,6,19-triol.
(d 3.60), which in turn showed COSY coupling with
H2-30 multiplet (d 3.51).
The HREIMS, 1H, and 13C NMR data (Table 1) data of
metabolite 4 also suggested a monohydroxy derivative
of 1. The oxygenated broad singlet at d 4.23, which cor-
related with a methine carbon at d 70.9, was assigned for
the newly oxygenated C-9 (Table 1). Carbon C-9
The HRMS data of 10 suggested the molecular formula,
1
C23H39NO3. The H and 13C NMR data (Table 2) were
similar to those of 9 with the replacement of the chlori-
nated terminal methylene C-30 with a methyl group. The
methyl triplet H3-30 (d 1.11) was assigned based on its
COSY coupling with H2-20 multiplet (d 3.20).
3
showed a J-HMBC correlation with the methyl proton
singlet H3-19 (d 1.53). Proton H-9 showed COSY corre-
lations with H2-10 multiplets (d 2.40 and 2.32), which in
turn showed COSY correlations with the olefinic
methine broad triplet H-11 (d 4.99), confirming the
assignment of the segment C-9/C-11. The relative stereo-
chemistry of 9-hydroxy group was assigned based on
NOESY data. The a-oriented H3-18 (d 1.41) showed a
NOESY correlation with H-9, suggesting a similar rela-
tive stereochemistry. Thus, the 9-hydroxy group should
be b-oriented and therefore, metabolite 4 was found to
be (1S,2E,4R,6R,7E,9S,11E)-2,7,11-cembratriene-4,6,9-
triol.
1
The H and 13C NMR data of 11 (Table 2) indicated a
monosubstituted benzene moiety replacing the methyl
C-30 in 10. The broad methylene singlet H2-20 (d 4.35)
showed a J-HMBC correlation with the aromatic pro-
ton doublet H-40 and H-80 (d 7.26). The latter protons
showed a COSY coupling with H-50 and H-70 doublet
of doublets (d 7.32), which in turn shows a COSY cou-
pling with H-60 (d 7.27).
3
The antiproliferative activities of cembranoids 1 and 3–
11 were evaluated against the human highly metastatic
prostate cancer cell (PC-3M) growth using the 3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
(MTT) assays. Cembranoid 1 displayed little cytotoxic
but potent antiproliferative effects at a dose of 10–
20 nM, however it was cytotoxic at doses of 50 lM or
higher (Fig. 1). Calcitonin (CT) is a potent endogenous
growth factor for prostate cancer cells and exogenous
addition of CT significantly increases proliferation and
invasion of PC-3M cells.25,26 Figure 1 also suggests that
cembranoid 1 not only attenuated the basal growth of
PC-3M cells, but almost abolished the CT-stimulated
PC-3M cell proliferation. These effects of 1 were ob-
served at all three periods of incubation (24, 48, and
72 h).
Reflux of 1 in toluene with chlororethyl, ethyl, and ben-
zyl isocyanates, in presence of catalytic amount of tri-
ethylamine, afforded the C-6 carbamates 9–11,
respectively.24 The HRMS analysis of compound 9 re-
vealed the molecular formula, C23H38ClNO3, with M
and M+2, 3:1, isotopic clusters characteristic of a mono-
chlorinated compound. The downfield shifting of C-6
carbon chemical shift in 9 (+4.9 ppm), compared with
that of the starting material 1, confirmed carbamoyla-
1
tion at this position. The H and 13C NMR data (Table
2) further supported this fact. The carbonyl carbon at d
156.5 was assigned C-10. This was based on its 3J-
HMBC correlation with the methylene multiplet H2-20

K. A. El Sayed et al. / Bioorg. Med. Chem. 16 (2008) 2886–2893
Table 2. 13C and 1H NMR Data of compounds 9–11a
2889
Position
9
10
11
dC
dH
dC
dH
dC
dH
1
46.1, CH
130.3, CH
136.3, CH
70.9, qC
1.50, m
46.0, CH
130.1, CH
136.5, CH
70.8, qC
1.55, m
46.1, CH
130.1, CH
136.4, CH
70.8, qC
1.55, m
2
5.17, dd (15.8, 9.2)
5.36, d (15.8)
—
5.16, dd (15.4, 9.2)
5.36, d (15.4)
—
5.16, dd (15.4, 8.8)
5.37, d (15.4)
—
3
4
5
51.4, CH2
69.3, CH
127.1, CH
139.6, qC
38.9, CH2
23.1, CH2
124.0, CH
133.5, qC
36.5, CH2
27.5, CH2
33.2, CH
19.5, CH3
20.5, CH3
28.3, CH3
16.0, CH3
15.0, CH3
156.5, qC
44.3, CH2
42.8, CH2
2.00, m
51.7, CH2
68.7, CH
127.3, CH
139.3, qC
38.9, CH2
23.1, CH2
124.0, CH
133.5, qC
35.9, CH2
27.4, CH2
33.1, CH
19.5, CH3
20.4, CH3
28.2, CH3
15.9, CH3
15.0, CH3
156.6, qC
36.5, CH2
15.3, CH3
2.05, m
51.6, CH2
69.2, CH
127.2, CH
139.4, qC
38.9, CH2
23.1, CH2
124.0, CH
133.5, qC
36.5, CH2
27.5, CH2
33.2, CH
19.5, CH3
20.4, CH3
28.3, CH3
16.0, CH3
15.0, CH3
156.7, qC
45.1, CH2
138.4, qC
127.6, CH
128.8, CH
127.7, CH
2.05, m
6
5.60, dd (9.5, 9.0)
5.27, d (9.9)
—
5.57, dd (9.4, 9.0)
5.24, d (9.9)
—
5.62, dd (9.5, 9.1)
5.24, d (9.9)
—
7
8
9
2.05, m 2.20, m
2.10, m 2.20, m
4.91, m
2.10, m 2.25, m
2.10, m
4.91, m
2.05, m 2.25, m
2.10, m 2.30, m
4.91, brt (5.2)
—
10
11
12
13
14
15
16
17
18
19
20
10
20
30
40, 80
50, 70
60
—
—
1.90, m
1.90, m 2.05, m
1.25, m 1.60, m
1.55, m
1.95, m 2.05, m
1.25, m 1.60, m
1.55, m
1.20, m 1.55, m
1.50, m
0.79, d (6.6)
0.82, d (6.6)
1.36, s
0.79, d (6.6)
0.80, d (6.6)
1.36, s
0.77, d (6.6)
0.82, d (6.6)
1.36, s
1.48, s
1.72, s
1.48, s
1.71, s
1.48, s
1.73, s
—
3.60, m
3.51, m
—
3.20, m
1.11, t (7.0)
—
4.35, brs
—
7.26, 2H, d (8.0)
7.32, 2H, d (8.0, 7.8)
7.27, d (7.8)
4.98, brt (5.8)
–NH
5.09 brt, (6.0)
4.60, brs
a In CDCl3, 400 MHz for 1H and 13C NMR. Coupling constants (J) are in Hz.
1.2
strength of epithelial junctions can be assessed by the
measurement of transepithelial resistance (TER) and
paracellular permeability.25,26 While TER can be mea-
sured by an ohm-meter, paracellular permeability is
measured by the ability of Rhodamine-labeled antibod-
ies (TRITC)-conjugated dextran through the cell layer
of PC-3M cells.25,26 Since cembranoid 1 inhibited the
invasion of PC-3M cells in the MatrigelTM model, test-
ing whether this action was mediated by stabilization
of junctional complexes was required. Therefore, the ef-
fect of 1 on TER and paracellular permeability of PC-
3M cell cultures was tested. CT destabilizes junctional
complexes of prostate cancer cells as characterized by
a large reduction in TER and a marked increase in para-
cellular permeability of PC-3M cultures.25 CT also in-
creases the rate of cell proliferation, invasion and
tumorigenicity.25,26 Figure 3A presents TER of PC-3M
cell layers at different time points after stimulation with
and without CT. As expected, TER of PC-3M cell cul-
tures increased with time. However, addition of 50 nM
CT caused a dramatic reduction in TER (Fig. 3A). A
dose of 50 nM of 1 did not affect baseline TER, but
remarkably reduced CT-stimulated decrease in TER of
PC-3M cells (Fig. 3A). Similar to decrease in TER, in-
crease in paracellular permeability is also suggestive of
junctional destabilization.25 CT increased paracellular
permeability of PC-3M cells (Fig. 3B). In contrast, 1 sta-
bilized junctional complexes as suggested by the de-
crease of paracellular permeability of untreated and
CT-treated PC-3M cells (Fig. 3B).
0.9
Vehicle
CT (50 nM)
0.6
1 (10 nM)
1 (10 nM) + CT
0.3
1 (20 nM)
1 (20 nM) + CT
0.0
24.0
48.0
72.0
Hours
Figure 1. Antiproliferative and cytotoxic actions of 1 on PC-3M cells.
Cembranoid 1 showed a potent anti-invasive activity
against PC-3M cells using MatrigelTM assay (Fig. 2A
and B). Figure 2A depicts the micrograph of invaded
cells of each treatment group. The results show that
the addition of CT observably increased the number of
invaded cells. A dose of 5 nM of 1 caused an apparent
decrease in the invasion of basal and CT-treated PC-
3M cells. Figure 2B presents the pooled data of multiple
experiments and suggests that CT caused twofold in-
crease in the number of invaded cells. In contrast, 1 de-
creased basal as well as CT-treated PC-3M cells at a
dose of 5 nM by two to threefold.
Recent evidence suggests that the destabilization of
junctional complexes plays a crucial role in the progres-
sion of localized tumor to its metastatic form.25,26 The

2890
K. A. El Sayed et al. / Bioorg. Med. Chem. 16 (2008) 2886–2893
A
C
CT (50 nM)
1 (5 nM)
1+CT
B
75
50
25
0
C
1
1+CT
CT
Figure 2. (A) Anti-invasive activity of b-diol (1) against PC-3M cells using MatrigelTM assay, Invaded cells/400X fields. (B) Quantification of
invasion assay results.
0.8
A
A
1000
900
800
700
600
500
400
C
0.6
0.4
0.2
0.0
1
1+CT
CT
0
10
20
30
40
50
60
Time (hours)
B
C CT 6 6+ 11 11+ 1 1+ 7 7+ 8 8+ 9 9+ 10 10+
CT CT CT CT CT CT CT
3
2
1
0
*
500
400
300
200
100
0
B
C
CT
*
9
9+CT
10
10+CT
11
11+CT
C
1
1+CT
CT
0
12
24
36
48
60
Figure 3. (A) Effect of 50 nM dose of 1 on transepithelial resistance of
PC-3M cells. (B) Effect of 1 on paracellular permeability of PC-3M
cells.
Time (Hours)
Figure 4. (A) Antiproliferative activity of 50 nM dose of cembranoids
6–11 against PC-3M cells using MTT assay. (B) Effect of carbamates9–
11 on transepithelial resistance of PC-3M cells.
Finally, the cytotoxic, antiproliferative, anti-invasive,
and junction-stabilization effects of cembranoids 3–11
were evaluated (Fig. 4A and B). All compounds showed
reduced toxicity compared to that of 1. Biocatalytic
products 6–8 also maintained the potent anti-invasive
activity at 50 nM dose without cytotoxicity (Fig. 4A).
Therefore, this is viewed as an excellent example of po-
tential benefits of bioconversion in that while hydroxyl-
ation was stereoselectively achieved; the cytotoxicity of 1
was reduced, while the same anti-invasive potency of the
parent compound was maintained. The carbamate ana-
logs 9–11 also display potent antiproliferative activities
against PC-3M cell line at treatment doses between 5
and 50 nM (Fig. 4A). Carbamate analogs 9–11 were
much less cytotoxic than 1 at 50 nM dose in MTT assay
(Fig. 4A). The effect of carbamate analogs 9–11 on base-
line and CT-stimulated TER and paracellular perme-
ability was examined (Fig. 4B). Only carbamate 10
stabilized junctional complexes, suggesting it may be a
potential lead for future development as anti-metastatic
drug for treatment of prostate cancer.

K. A. El Sayed et al. / Bioorg. Med. Chem. 16 (2008) 2886–2893
2891
3. Conclusions
1/102, 1/103, 1/104, and 1/105 serial dilutions were made
using the previously prepared Instant Ocean solution.
About 100 ll of each concentration was inoculated on
the top of sterile tryptic soy or marine agar plates. Plates
were incubated for 72–168 h at 28 ꢁC. Symmetric fine
colonies were separated and re-inoculated on corre-
sponding agar media. Pure cultures were subjected to
PCR analysis, DNA extraction (MO Bio Laboratories,
Inc., Carlsbad, CA), and finally 16S rRNA sequenc-
ing.22 Obtained alignments were subjected to Basic Lo-
cal Alignment Search Tool (BLAST) queries. Identities
of the isolated strains were based on partial 16S rRNA
gene sequences. Strains had 16S rRNA gene sequence
similarity ꢀ 99% to the GenBank sequences listed
below.
Biocatalysis of tobacco 2,7,11-cembratriene-4,6-diol (1)
using the novel symbiotic Bacilli species afforded two
new and four known hydroxylated metabolites 3–8.
The new C-6 carbamate analogs 9–11 were also pre-
pared semisynthetically. Cembranoid 1 shows antiprolif-
erative and anti-invasive activity against the human
highly metastatic prostate PC-3M cancer cell line at
10–20 nM doses. A dose of 50 nM of 1 remarkably abol-
ished CT-stimulated decrease in TER and increase of
paracellular permeability of PC-3M cells. This indicates
that 1 stabilized tight junctional complexes of PC-3M
cells. Bioconversion and carbamate products showed
similar potent anti-invasive activity at 50 nM dose with-
out any sort of cytotoxicity. This is an example of a suc-
cessful bioconversion, where the cytotoxic starting
material bioconverted to less toxic metabolites while
maintaining nearly the same anti-invasive potency.
Cembranoids 1, 6–8, and 10 can be potential leads
appropriate for future development as possible treat-
ment for metastatic prostate cancers.
3.1.4. Biocatalysis. Biocatalytic studies were conducted
as described elsewhere.13,27 Thirty growing marine bac-
terial cultures were used for screening of 1. These were:
NB1: Bacillus sp. DF12 (AY462199), NC1: B. megateri-
um KL197 (AY030338), NA3: Pelagiobacter variabilis
(AB167354), NC4: Bacillus sp. DF15 (AY462201),
NC5: Bacillus sp. S/2 (AJ784847), NE1: B. subtilis BFAS
(AY775778), SA2: Bacillus sp. KMM3737 (AY228462),
SD1: Halobacillus sp. NT N168 (AB167053), NJ2: Jan-
ibacter marinus strain C6 (AY533561), 59A3: Bacillus
sp. GSP11 (AY553069), NK6: B. djibelorensis
(AF519467), NK8: Bacillus species Fa29 (AY131222),
NK7: Bacillus species MT21 (AY690689), EA1: B. pumi-
lus (AY112667), NG1A: Microbacterium sp. SKJH-23
(AY741722), NK4: Bacillus sp. EC2 (AY864632), EC2:
Pseudomonas sp. C127 (DQ005892), SC1: Glacial ice
bacterium G500K-17 (AF479334), SC3: Pontibacillus
marinus BH030 (AY603977), SE2: Oceanobacillus ihey-
ensis HTE831 (BA000028), SH1: Micrococcus luteus
(AB079788), NF3: B. subtilis AX20 (AY825035), NK3:
Kocuria sp. JL-72 (AY745813), SA6: B. subtilis MO2
(AY553095), 59D3: B. acidogenesis (AF547209), EF3:
Brevibacterium halotaleran (AJ620368), EF1: Virgiaba-
cillus carmonensis (AJ316302), EC2A: Microbulbifer cys-
tomdytense (AJ620879), SK2: Bacillus sp. DF49
(AY462214), ED1: B. megaterium (AJ717381). These
isolates show highest 16S rRNA gene sequence similar-
ity to the sequences of these organisms in GenBank.
GenBank accession numbers are listed in parenthesis.
Bacillus sp. NC5, B. sp. NK8, and B. sp. NK7 were se-
lected for biocatalysis scale-up of 1. Each of these organ-
isms was inoculated in twenty 1000-mL flasks each
containing 250 mL tryptic soy or marine broth. After
48 h, compound 1 was added into their respective flasks
(15 mg/flask). After 14 days, the growth medium was fil-
tered and extracted with EtOAc (4 · 1000 mL). The
EtOAc layer was then concentrated under vacuum. Res-
idue obtained from biocatalysis of 1 by Bacillus species
NC5 was purified on silica gel 60 column, followed by
reversed phase Si gel medium pressure liquid chroma-
tography (MPLC) to yield compound 3 (12 mg, Rf
0.26, CHCl3/MeOH 9:1), compound 4 (4.5 mg, Rf
0.0.27, CHCl3/MeOH 9:1), compound 5 (8 mg), and
compound 6 (21 mg). Bacillus species NK8 fermentation
extract was subjected to silica gel 60 column chromatog-
raphy by n-hexane/EtOAc gradient elution followed by
MPLC on C-18 reversed phase silica gel with isocratic
3.1. Experimental
3.1.1. General experimental procedures. Measurements
of optical rotation were carried out on a Rudolph Re-
search Analytical Autopol III polarimeter. IR spectra
were recorded on a Varian 800 FT-IR spectrophotome-
ter. The 1H and 13C NMR spectra were recorded in
CDCl3, using TMS as an internal standard, on a JEOL
Eclipse NMR spectrometer operating at 400 MHz for
1H and 100 MHz for 13C. The HREIMS experiments
were conducted at the University of Michigan on a
Micromass LCT spectrometer. TLC analyses were car-
ried out on precoated silica gel 60 F254 500 lm TLC
plates, using the developing system n-hexane/EtOAc
(1:1) or CHCl3/MeOH (9.5:0.5). For CC, silica gel 60
(particle size 63-200 lm) or Bakerbond octadecyl
(C18), 40 lm was used.
3.1.2. Extraction and isolation. Fresh tobacco leaf pow-
der (Custom Blends, NY, 27.2 kg) was extracted with
hexane (130 L) in percolators three times at room tem-
perature. The hexane extract was concentrated under
vacuum and dried extract (1050 g) was vacuum liquid
chromatographed on Silica gel (200–300 mesh, 2 Kg,
Natland International Corporation) using gradient n-
hexane/EtOAc to yield crude cembranoid-containing
fraction (64.0 g) which was further chromatographed
on normal phase and finally on reversed phase silica
gel (MeOH–H2O, 2:3, isocratic) to give 1 (3.6 g) and
its C-4 epimer (17.9 g). The identity of 1 was confirmed
by extensive NMR analysis and comparison with the
literature.1,2,5,11
3.1.3. Symbiotic bacteria culture and isolation. About 2 g
of each of frozen N. magnifica voucher, which was col-
lected and kept frozen in sterile bags, was macerated
overnight in 0.5 L Instant Ocean solution and then vac-
uum filter sterilized. Other 2 g of the sponge was asepti-
cally blended in a sterile blender with 18 mL of the
sponge Instant Ocean solution. Ten milliliters of 1/10,

2892
K. A. El Sayed et al. / Bioorg. Med. Chem. 16 (2008) 2886–2893
developing system MeOH/H2O (1:2) to afford com-
pound 7 (9 mg, Rf 0.32, CHCl3/MeOH 9:1). Similarly,
Bacillus species NK7 afforded compound 8 (11 mg, Rf
0.21, CHCl3/MeOH 9:1).
could proliferate during the remaining period of incuba-
tion, leading to a slight overestimation of the final re-
sults. To correct this probability, the growth rate of
PC-3M cells was determined under identical culture con-
ditions. 25 · 103 cells were plated at hourly intervals in
six-well dishes and cultured with/without CT (50 nM)
for 1–24 h. Mean percent increase in the cell number
was determined at the end of the incubation period by
counting the net increase in the number of cells. The rel-
ative CT-induced increase of the pooled results of all
time points was found to be 1.19 (vehicle control = 1).
This correction was applied to the results of invasion
assays.25,26
3.1.5. Preparation of compounds 9–11. To solutions of 1
(16–35 mg) in toluene (2 mL), 10 lL of 2-chloroethyl
isocyanate or 35 lL of ethyl isocyanate or 27 lL of ben-
zyl isocyanate was added and separately mixed with
10 lL of Et3N.24 Each solution was separately stirred
and refluxed for 3 h. Water was then added and the
product of each reaction mixture was extracted with
EtOAc.24 Each EtOAc extract was dried over anhydrous
Na2SO4 and concentrated under reduced pressure.
Crude products were then purified by CC on silica gel
(EtOAc-hexane 2:8 or 1:9) to give compounds 9,
14 mg, 63%, 10, 7.7 mg, 13%, and 11, 20 mg, 39.8%,
respectively.
3.1.9. Measurement of transepithelial resistance. Cells
were plated at confluency and grown on six well Trans-
well filters (0.4 lm pore size) in complete medium. TER
values were measured in duplicate wells in every 12 h
using EVOM voltohmmeter (World Precision Instru-
ments). The TER values were normalized to the area
of the monolayer filter and calculated by subtracting
the blank values from the filter and bathing medium.
All cell culture media were supplemented with 25 mM
Hepes, pH 7.4, and the integrity and cell density of mon-
olayers were carefully monitored during TER measure-
ment studies.25,26
3.1.6. Proliferation of PC-3M MTT assays. The growth
of prostate cell line PC-3M was measured by MTT kit
(ATCC). Exponentially growing cells were plated in a
96-well plate at a density of 8 · 103 cells per well and al-
lowed to attach for 24 h. Complete growth medium was
replaced with 100 lL of RPMI serum-free medium (GIB-
CO-Invitrogen, NY) and culture continued at 37 ꢁC un-
der 5% CO2. After 24 h of culture, the cells were treated
with MTT solution at 37 ꢁC for 4 h.24,25 The color reac-
tion was stopped by the addition of solubilization/stop
solution (100 lL/well), and the incubation at 37 ꢁC con-
tinued to completely dissolve the formazan product.
Absorbance of the samples was determined at 595 nm
with an ELISA plate reader (Bio-Rad, Hercules, CA).
3.1.10. Permeability assay (paracellular diffusion of
dextran). Cells were plated at confluency and grown
on 12-well Transwell filters in complete medium for
optimum TER development. Tetramethyl-rhodamine-
dextran, with an average molecular mass of 4 kDa (Sig-
ma, St. Louis, MO) dissolved in Hanks’ Balanced Salt
Solution (HBSS) to a concentration of 1 mg/mL, was
added to the upper chamber. The lower chamber was re-
placed with HBSS. At various time intervals, 100 lL ali-
quots were collected from the lower chamber and
luminescence assayed using spectrophotometer using
excitation 530 and emission 590.25,26
3.1.7. Invasion assay. These experiments were conducted
in 24-well, two compartmented, MatrigelTM invasion
chambers (Becton Dickinson, Bedford, MA).25,26 Expo-
nentially growing prostate cancer cell line was serum-
starved for 24 h with basal RPMI medium containing
no serum or growth factors (but containing 0.1% BSA,
10 mM Hepes, 4 mM L-glutamine, 100 IU/mL penicillin
G, and 100 mg/mL streptomycin). The cells were then
harvested and seeded at a density of 25 · 103 cells/well
in the upper insert of the MatrigelTM invasion chamber.
The lower chamber received the chemo-attractant med-
ium, which consisted of 90% basal RPMI medium and
10% conditioned medium from the cultures of PC-3 M
cells constitutively expressing active Gas protein. The
incubations were carried out for 24 h, after which the
MatrigelTM (along with non-invading cells) was scraped
off with cotton swabs, and outer side of the insert was
fixed and stained using Diff Quick staining (Dade Beh-
ring Diagnostics, Aguada, Peurto Rico). The number
of cells migrated on the outer bottom side of the insert
was counted under the microscope in six or more ran-
domly selected fields (magnification: 100·). The final re-
sults were expressed as means SEM per 100· field.
Each experiment was conducted in triplicates, and the
experiment was repeated twice.25,26
3.1.11. (1S,2E,4R,6R,7Z,11E)-2,7,11-cembratriene-4,6,
25
D
19-triol (3). Colorless oil, ½aꢁ +58.2 (c 0.1, CHCl3); IR
tmax (CHCl3) 3602, 3420, 2928, 2856, 1462, 1379, 1082,
908 cmꢂ1 1H and 13C NMR, see Table 1; HREIMS
;
m/z 345.2395 [M+Na]+ (calcd for C20H34O3Na,
345.2406).
3.1.12. (1S,2E,4R,6R,7E,9S,11E)-2,7,11-cembratriene-
25
D
4,6,9-triol (4). Colorless oil, ½aꢁ +68.0 (c 0.1, CHCl3);
IR tmax (CHCl3) 3421, 2930, 2855, 1461, 1374, 1096,
1
907 cmꢂ1; H and 13C NMR, see Table 1; HRESMS
m/z 345.2438 [M+Na]+ (calcd for C20H34O3Na,
345.2406).
25
3.1.13. Compound 9. Colorless oil; ½aꢁ +58.9 (c 0.50,
D
CHCl3); IR mmax (neat) 3685, 3612, 3450, 3020, 2929,
1
1703, 1513, 1428, 1221, 1045 cmꢂ1; H and 13C NMR
see Table 2; HRESMS m/z 434.2427 [M+Na]+ (calcd
for C23H38ClNO3Na, 434.2438).
25
3.1.14. Compound 10. Colorless oil; ½aꢁ +50.3 (c 0.06,
3.1.8. Growth correction. Since some of tumor cell lines
exhibit high proliferation rate, it is likely that the cells
migrated during early part of the 24 h incubation period
D
CHCl3); IR mmax (neat) 3685, 3450, 3021, 2928, 1966,
1
1602, 1518, 1425, 1213 cmꢂ1; H and 13C NMR see Ta-

K. A. El Sayed et al. / Bioorg. Med. Chem. 16 (2008) 2886–2893
2893
ble 2; HRESMS m/z 400.2830 [M+Na]+ (calcd for
C23H39NO3Na, 400.2828).
7. Olsson, E.; Holth, A.; Kumlin, E.; Bohlin, L.; Wahlberg, I.
Planta Medica 1993, 59, 293.
8. Saito, Y.; Takayawa, H.; Konishi, S.; Yoshida, D.;
Mizusaki, S. Carcinogenesis 1985, 6, 1189.
9. Saito, Y.; Nishino, H.; Yoshida, D.; Mizusaki, S.; Ohni-
shi, A. Oncology 1988, 45, 122.
10. Saito, Y.; Tsujino, Y.; Kaneko, H.; Yoshida, D.; Mizu-
saki, S. Agric. Biol. Chem. 1987, 51, 941.
11. Wahlberg, I.; Eklund, A. M.. In Progress in the Chemistry
of Organic Natural Products; Herz, W., Kirby, G. W.,
Moore, R. E., Steglich, W., Tamm, C. H., Eds.; Springer-
Verlag Wien: New York, 1992; Vol. 59, pp 142–293.
12. ElSayed, K. A.; Sylvester, P. W. Drugs 2007, 16, 877.
13. Clark, A. M.; McChesney, J. D.; Hufford, C. D. Med. Res.
Rev. 1985, 5, 231.
14. Thayer, A. N. Chem. Eng. News 2006, 84, 16.
15. Sanchez-Porro, C.; Martin, S.; Mellado, E.; Ventosa, A. J.
Appl. Microbiol. 2003, 94, 295.
16. Fukano, Y.; Ito, M. Appl. Environ. Microbiol. 1997, 63,
1861.
17. Yamazaki, Y.; Mikami, Y. Jpn. Kokai Tokkyo Koho 85-
266107, 1987; Yamazaki, Y.; Mikami, Y. Chem. Abstr.
1987, 110, 132381.
25
3.1.15. Compound 11. Yellowish oil; ½aꢁ +52.1 (c 0.60,
D
CHCl3); IR mmax (neat) 3449, 2929, 1711, 1511, 1455,
1259, 1127 cmꢂ1 1H and 13C NMR see Table 2;
;
HRESMS m/z 462.2975 [M+Na]+ (calcd for C28H41
NO3Na, 462.2984).
Acknowledgments
This publication was made possible by the support of
Philip Morris USA Inc. and Philip Morris International
as well as the NIH Grant No. P20RR16456 from the
BRIN Program of the National Center for Research Re-
sources and R01 Grant No. R01CA96534 (G.V.S.). Its
contents are solely the responsibility of the authors
and do not necessarily represent the official views of
NIH. FAR and DB were supported by the Governor’s
Biotechnology Initiative of the Louisiana Board of Re-
gents (BOR#021-Moving an Established Marine Bio-
technology Program to the Next Level: Natural
Product Screening and Development). Louisiana Board
of Regents is also acknowledged for the support of the
NMR spectrometer probe purchase (LEQSF (2007–
08)-ENH-TR 80 to KAE).
18. Yamazaki, Y.; Mikami, Y. Jpn. Kokai Tokkyo Koho 86-
72567, 1987; Yamazaki, Y.; Mikami Chem. Abstr. 1987,
110, 111980.
19. Arnarp, J.; Chu, W. L. A.; Enzell, C. R.; Hewitt, G. M.;
Kutney, J. P.; Li, K.; Milanova, R. K.; Nakata, H.; Nasiri,
A.; Okada, Y. Acta Chem. Scand. 1993, 47, 683.
20. Arnarp, J.; Chu, A. W. L.; Enzell, C. R.; Hewitt, G. M.;
Kutney, J. P.; Li, K.; Milanova, R. K.; Nakata, H.; Nasiri,
A.; Tsuda, T. Acta Chem. Scand. 1993, 47, 689.
21. Arnarp, J.; Chu, A. W. L.; Enzell, C. R.; Hewitt, G. M.;
Kutney, J. P.; Li, K.; Milanova, R. K.; Nakata, H.; Nasiri,
A.; Tsuda, T. Acta Chem. Scand. 1993, 47, 793.
22. Rainey, F. A.; Ward-Rainey, N.; Kroppenstedt, R. M.;
Stackebrandt, E. Int. J. Systematic Bacteriol. 1996, 46,
1088.
References and notes
1. Wahlberg, I.; Enzell, C. R. Beitrage zur Tabakforschung
International 1984, 12, 93.
2. Wahlberg, I.; Eklund, A. M. In Trends in Flavour
Research; Maarse, H., Heji, V., Eds.; Elsevier Science
B.V.: Oxford, UK, 1994; pp 449–462.
3. Wahlberg, I.; Forsblom; Vogt, C.; Eklund, A. M.;
Nishida, T.; Enzell, C. R.; Berg, J. E. J. Org. Chem.
1985, 5, 4527.
23. Forsblom, I.; Berg, J. E.; Wahlberg, I. Acta Chem. Scand.
1993, 47, 80.
24. Basheer, A.; Rappoport, Z. J. Org. Chem. 2006, 71,
9743.
25. Sabbisetti, V. S.; Chirugupati, S.; Thomas, S.; Vaidya, K.
S.; Reardon, D.; Chiriva-Internati, M.; Iczkowski, K. A.;
Shah, G. V. Int. J. Cancer 2005, 117, 551.
26. Shah, G. V.; Rayford, W.; Noble, M. J.; Austenfeld, M.;
Weigel, J.; Vamos, S.; Mebust, W. K. Endocrinology 1994,
134, 596.
4. Dauben, W. G.; Thiessen, W. E.; Resnick, P. R. J. Am.
Chem. Soc. 1962, 84, 2015.
5. Wahlberg, I.; Olsson, E.; Berg, J. E. In Prog. Flavour
Precursor Stud. Proc. Int. Conf., Schreier, P., Winterhal-
ter, P., Eds; Allured Publishing Corporation: Carol
Stream, Illinois, 1993; pp 83–95.
27. El Sayed, K. A.; Hamann, M. T.; Waddling, C. A.; Jensen,
C.; Lee, S. K.; Dunstan, C. A.; Pezzuto, J. M. J. Org.
Chem. 1998, 63, 7449.
6. Ferchmin, P. A.; Hao, J.; Perez, D.; Penzo, M.; Maldo-
nado, H. M.; Gonzalez, M. T.; Rodriguez, A. D.; Jean, D.
V. J. Neurosci. Res. 2005, 82, 631.