Synthesis of jaspaquinol and effect on viability of normal and malignant bladder epithelial cell lines

Article abstract of DOI:10.1016/j.bmcl.2004.03.055

The synthesis of jaspaquinol 1, a monocyclic diterpene-benzenoid, is reported. Two synthetic routes to this natural product have been developed. The first, utilises a difunctional terpene derivative containing different leaving groups, facilitating the selective introduction of the cyclohexenyl and benzenoid fragments. The alternative route employs a regiospecific Stille cross-coupling reaction to introduce the cyclohexenyl fragment, which occurs without allylic transposition. Preliminary data shows the cell viability of 1 against normal and malignant human bladder epithelial cell lines.

Full text of DOI:10.1016/j.bmcl.2004.03.055

Bioorganic & Medicinal Chemistry Letters 14 (2004) 2883–2887
Synthesis of jaspaquinol and effect on viability of normal
and malignant bladder epithelial cell lines
Alexandre Demotie,a Ian J. S. Fairlamb,a,* Feng-Ju Lu,a Nicola J. Shaw,b
Peter A. Spencerb and Jennifer Southgateb
aDepartment of Chemistry, University of York, Heslington, York YO10 5DD, UK
bJack Birch Unit of Molecular Carcinogenesis, Department of Biology, University of York, York YO10 5YW, UK
Received 9 December 2003; accepted 15 March 2004
Abstract—The synthesis ofjaspaquinol 1, a monocyclic diterpene-benzenoid, is reported. Two synthetic routes to this natural
product have been developed. The first, utilises a difunctional terpene derivative containing different leaving groups, facilitating the
selective introduction ofthe cyclohexenyl and benzenoid rfagments. The alternative route employs a regiospecific Stille cross-
coupling reaction to introduce the cyclohexenyl fragment, which occurs without allylic transposition. Preliminary data shows the cell
viability of 1 against normal and malignant human bladder epithelial cell lines.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
diphosphate (FDP) and geranylgeranyl diphosphate
(GGDP), the former, which is an important substrate
for squalene synthase (SQS)3 and protein-farnesyl
transferase (PFTase),4 and the latter a substrate for
protein geranylgeranyl transferase (PGTase, types I and
II).5 The recent synthesis ofhydroxyaromatic FDP
analogues by Jarman and co-workers6 highlights this
potential.
Jaspaquinol 1,1 isolated from two distinct marine
sponges, Suberea sp. (order Verongida, family Aplysi-
nellidae) and Jaspis splendens (order Choristida, family
Jaspidae), has been reported by Crews and co-workers2
and represents the first example ofa monocyclic diter-
pene-benzenoid isolated from a natural source (Fig. 1).
It is the most potent human 15-lipooxygenase (15-HLO)
inhibitor (IC50 ¼ 0.3 lM) isolated from their studies, and
the proposed mode ofaction of 1 lies with reduction of
an active Fe(III) enzyme to an inactive Fe(II) enzyme. It
is highly plausible that the hydroquinol residue of 1
reduces iron,2 however to date this has not been proven.
The structure of 1 resembles rather closely farnesyl
Following on from our own interest7 in the development
ofnew PFTase and PGGTase inhibitors, we herein
report two synthetic routes to 1. Retrosynthesis of 1
involves two disconnections at the C(7)–C(8) and C(15)–
C(16) positions (Scheme 1).
Key to the synthesis of 1 is difunctional protected ter-
pene intermediate 2, which allows incorporation ofthe
OH
Jaspaquinol, 1
7
15
HO
OH
O
P
O
P
16
8
1
O
O
O
HO
O
O
FDP
GGDP (dotted lines)
Y
+
+
X
Figure 1.
OR
2
Br
3
4
RO
* Corresponding author. Tel.: +44-(0)1904-434091; fax: +44-(0)1904-
432516; e-mail: ijsf1@york.ac.uk
Scheme 1. Retrosynthetic disconnection ofjaspaquinol 1.
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.03.055

2884
A. Demotie et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2883–2887
cyclohexene 3 and benzenoid 4 fragments by appropri-
ate allylic substitution chemistry. The chosen protecting
group strategy is crucial to the synthesis of 1 and has to
be done in such a way, so as to avoid unwanted disub-
stitution and moreover proceed selectively.8;9 Ifthis
challenge can be met, then there is great potential in the
range ofsubstituents/rfagments that may be incorpo-
rated via transition metal mediated allylic substitution
reactions.
Protection of 6 as its tetrahydropyranyl ether (98%
yield) and acetate hydrolysis using K2CO3 in MeOH
gave key intermediate 7 in quantitative yield (55%
overall yield from 5).
The nucleophilic displacement ofallylic tetrahydropyr-
anyl ethers, such as 7, with Grignard reagents (<5 equiv)
under Cu(I) mediated conditions has been investigated
in detail.8 This method can be used here with slight
modification. Initial reaction ofthe Grignard reagent
with the alcoholic proton is key to the success ofthe
selective SN2 displacement, through carbomagnesiation
ofthe magnesium alkoxide to the distal olefin, which
blocks SN20 attack. In order to keep the synthetic route
efficient, we felt that reaction of 7 with ethyl magnesium
bromide (1 equiv) followed by reaction with the Grig-
nard reagent (2.1 equiv, 8b) generated from known
allylic bromide13 8a, would give the required product
selectivity. Under these conditions we were able to dis-
place the tetrahydropyranyl moiety selectively in an SN2
fashion to give the coupled product 9 in a modest 41%
yield.
The synthesis ofderivatives, such as 2, has received a
vast amount ofattention, particularly where X ¼ OAc,
OTBDMS, OBn, OTHP and Y ¼ OH.10 The E-selective
allylic oxidation ofthe appropriately protected geranyl
t
derivative using catalytic SeO2 (5 mol %) and BuO2H
(3.6 equiv) in CH2Cl2, developed in the early 1970’s by
Sharpless and Umbreit,11 is still the method ofchoice
(Scheme 2). However, there are some limitations when
alternative protecting groups are employed that is,
where X ¼ OTBDMS, OBn, OTHP or OMe.12 We chose
the reactive acetate substrate 5. Oxidation of 5, under
the conditions given above, followed by reduction of the
crude material with NaBH4, gave pure alcohol 6 in 56%
yield. Alternatively, geranyl benzoate may be employed,
which gives the oxidised product in similar yield (55%).
The chloride 10 and bromide 11 derivatives were pre-
pared using a Corey et al. procedure.14 So, 9 was reacted
with N-bromosuccinimide/dimethyl sulfide and N-chlo-
rosuccinimide/dimethyl sulfide in CH2Cl2, respectively,
to give 10 and 11, in good yield.
i
RO
OAc
X
(56%)
5
The Stille coupling of 10 and 11 was next investigated.
Reaction of 10 with arylstannane 1215 in the presence of
Pd2dba3ÆCHCl3 and PPh3 gave the SN2 product 13 in
92% yield, whereas use ofthe bromide 11 gave the SN2
product in a lower 81% yield.
6, R = H, X = OAc
7, R = THP, X =OH
ii, iii
(98%)
8a X = Br
X
X
iv
8b X = MgBr
(41%)
O
Coordination ofthe distal olefin is predicted to favour
the SN2 pathway via either p-allyl or r-allyl intermedi-
ates (Scheme 3). The SN20 pathway requires dissociation
ofthe distal olefin16 in order for the required cis inser-
tion ofthe aryl species. The pathway is sterically hin-
dered in both the dissociated and associated forms,
although more so for the latter, and thus unlikely. Such
selectivity has been observed for the addition of soft
X
Mg
O
R
O
X = OH, 9
X = Cl, 10
X = Br, 11
Cu
v
vi
(98%)
OBn
R
(86%)
S
N2' attack
A
sterically hindered
SnBu3
(92%, using 10)
(81%, using 11)
vii
12
OBn
OBn
R
X
BnO
Oxidative
addition
13
(0)
- L
[Pd L ]
2
X
L
nBu3SnR
nBu3SnX
L
R
Pd
Pd
viii
(83%)
R
R
- L
Transmetallation
OH
"π-allyl"
γ
"σ-allyl"
γ
1
HO
α
α
Pd
Scheme 2. Reagents and conditions: (i) SeO2 (5 mol %), tBuO2H
(3.6 equiv), CH2Cl2, 25 °C, 8 h, then NaBH4, EtOH, 0 °C, 1 h; (ii) DHP
(1.2 equiv), TsOH (cat.), CH2Cl2, 25 °C, 12 h; (iii) K2CO3, MeOH,
0 °C, 6 h; (iv) EtMgBr (1 equiv), THF, 0 °C, then CuI, 8b (formed by
reaction with activated Mg in THF), THF, 50 °C, 6 h; (v) NCS, DMS,
CH2Cl2, 0 °C, 2 h; (vi) NBS, DMS, CH2Cl2, 0 °C, 2 h; (vii)
Pd2dba3ÆCHCl3 (2.5 mol %), PPh3 (7.5 mol %), 12, THF, 50 °C, 15–
18 h; (viii) LiAlH4 (3 equiv), NiCl2 (1.1 equiv), THF, reflux, 11 h.
Pd
R
R
Reductive
elimination
R'
R
Scheme 3. Selective SN2 displacement.

A. Demotie et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2883–2887
2885
stabilised nucleophiles to geranyl derivatives under Pd
or Mo catalysis.17
ditions to that reported, allowing the cross-coupled
product 19 to be synthesised, after careful chromatog-
raphy, in 47% yield. Several other products were isolated
from this reaction, as a mixture, for which accurate
characterisation was not possible. To complete the
synthetic route, deprotection ofthe benzyl groups was
accomplished using identical conditions to that given in
the first route, to give 1 in 73% yield.
The final benzyl deprotection15 was accomplished using
LiAlH4 in the presence ofstoichiometric NiCl , which
2
15
after 11 h gave jaspaquinol 1 in 83% yield. The structure
of 1 was confirmed by comparison with the natural
product (1H and 13C NMR), which is identical and
1
provides verification ofthe reported structure.
In the course ofour synthetic investigations we uf rther
studied an alternative sequence to 1 (Scheme 4).
2. Human urothelial cell viability studies
Compound 14 was prepared as reported.18 Benzylation
of 14 gave 15 in 88% yield. E-Selective oxidation ofthe
terminal methyl group of 15 was accomplished using the
SeO2/tBuO2H/CH2Cl2 system to provide, after NaBH4
reduction, alcohol 16 in 48% yield. The crude reaction
mixture contained a number ofpolar byproducts, which
were difficult to separate by flash chromatography.
The viability ofnormal human urothelial NHU cells
treated with 1 at various concentrations (0–100 lM) for
24–72 h was determined by both phase contrast
microscopy and MTT assay.22 Cells exposed to doses of
0.78 lM and below remained morphologically normal,
showed no effect on growth and entered exponential
growth on day 3 (Fig. 2). Cells exposed to 1 at 100 lM
showed a reduction in the number compared to controls
by 24 h post-treatment (p < 0:0001) and were morpho-
logically slightly rounded. By day 3, cells exposed to
concentrations of25 lM and above showed evidence of
cell death, with cell debris visible (Fig. 3). In addition, a
dose-dependent decrease in cell numbers compared to
controls was observed at days 3 and 6 at concentrations
of3.125 M and above. These results suggest that 1 had
no effect on cell growth and morphology before 24 h at
concentrations below 100 lM, it then caused a reduction
in cell growth at concentrations of3.125 lM and above,
and appeared to cause cell death at concentrations of
25 M and above.
Stille cross-coupling of 17, available through reaction of
16 with NCS–DMS,14 with allylic stannane 18 was
envisaged as being an efficient way ofintroducing the
cyclohexenyl fragment. Allylstannyl reagents generally
react with allylic Pd-complexes by a different process to
that normally observed for other organostannanes, in so
much as allylic transposition in the tin reagent is
sometimes seen. To this end, Godschalx and Stille19 and
Trost and Keinan20 encountered allylic transposition in
similar reactions. Notably, Schwartz discovered that
maleic anhydride could be added to the reaction mixture
to promote the formation of the desired 1,5-diene (acts
as a p-acceptor ligand, which is proposed to accelerate
reductive elimination).21 We used similar reaction con-
Compound 1 was tested against a urothelial carcinoma
EJ cell line, which contains the Harvey ras oncogene
(Fig. 4). It was found that after 24 h, EJ cells were
morphologically unaffected by treatment with 1, with
the exception that cell death was apparent at 100 lM
concentration.
OR
R = H, 14
i
R = Bn, 15
OR
(88%)
ii
(48%)
OBn
0uM
0.390625uM
0.78125uM
1.5625uM
3.125uM
6.25uM
12.5uM
X
1.2
X = OH, 16
X = Cl, 17
iii
(73%)
1
OBn
0.8
iv
(47%)
SnBu3
25uM
50uM
100uM
18
0.6
0.4
0.2
0
OBn
OBn
19
v
(73%)
1
0
1
2
3
4
5
6
Days after treatment with jaspaquinol, 1
Scheme 4. Reagents and conditions: (i) NaH (2.5 equiv), BnBr
(3 equiv), DMF, rt, 24 h; (ii) SeO2 (5 mol %), tBuO2H (3.6 equiv),
CH2Cl2, 25 °C, 8 h, then NaBH4, EtOH, 0 °C, 1 h; (iii) NCS, DMS,
CH2Cl2, 0 °C, 2 h; (iv) Pd(p-allyl)Cl]2 (2 mol %), maleic anhydride
(5 mol %), RSnBu3 18, dry, oxygen-free THF, 60°C, 15 h; (v) LiAlH4
(3 equiv), NiCl2 (1.1 equiv), THF, reflux, 12 h.
Figure 2. Effect of 1 on NHU cell growth. MTT assays were performed
0, 1, 3 and 6 days after the start of a 0–100 lM 1 treatment range
applied to Y324 cells at P5. The mean absorbance at 570 nm (A570 nm)
is proportional to the number ofcells. Each data point is the average of
six replicates, and standard deviations are shown by error bars.

2886
A. Demotie et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2883–2887
required for cell death. Compound 1 was investigated
1.4
1.2
1
for effects against a urothelial carcinoma EJ cell line,
containing the Harvey ras oncogene. However, in this
cell line at least, compound 1 does not inhibit the ras
process. The full biological results, including the PFTase
activity of 1 and related analogues, will be determined in
due course.
*
**
0.8
0.6
0.4
0.2
0
Acknowledgements
**
We gratefully acknowledge Prof. Phillip Crews and
co-workers (University of California, USA) for the
generous gift of naturally occurring jaspaquinol. I.J.S.F.
thanks the University of York for funding. Johnson
Matthey PLC (Dr. P. Bishop, UK) is thanked for their
kind donation ofpalladium salts.
**
**
0
0.39 0.78 1.56 3.13 6.25 12.5 25 50 100
Concentration (µM) of jaspaquinol, 1
Figure 3. Effect of 1 on NHU cell growth at day 3. MTT assay at day
3. The mean relative A570 nm was determined by normalisation to a
solvent control (0 lM treatment). Each data point is the average ofsix
replicates, and standard deviations are shown by error bars.
(ꢀP < 0:05, ꢀꢀP < 0:01).
References and notes
1. Murray, L. M.; Johnson, A.; Diaz, M. C.; Crews, P. J.
Org. Chem. 1997, 62, 5638.
2. Carroll, J.; Jonsson, E. N.; Ebel, R.; Hartman, M. S.;
Holman, T. R.; Crews, P. J. Org. Chem. 2001, 66, 6847.
3. Biller, S. A.; Neuenschwander, K.; Ponpipom, M. M.;
Poulter, C. D. Curr. Pharm. Des. 1996, 2, 1.
4. Leonard, D. M. J. Med. Chem. 1997, 40, 2971.
5. Lebowitz, P. F.; Prendergast, G. C. Oncogene 1998, 17,
1439.
6. Marriott, J. H.; Moreno Barber, A. M.; Hardcastle, I. R.;
Rowlands, M. G.; Grimshaw, R. M.; Neidle, S.; Jarman,
M. J. Chem. Soc., Perkin Trans. 1 2000, 4265.
7. (a) Fairlamb, I. J. S.; Pike, A. C.; Ribrioux, S. P. C. P.
Tetrahedron Lett. 2002, 43, 5327; (b) Fairlamb, I. J. S.;
Dickinson, J. M.; Pegg, M.; Cohen, L. H.; van Thiel, C.
Bioorg. Chem. 2003, 31, 80.
8. Mechelke, M.; Wiemer, D. F. J. Org. Chem. 1999, 64,
4821.
9. Mechelke, M.; Wiemer, D. F. Tetrahedron Lett. 1998, 39,
783.
10. For recent references see: (a) Radetich, B.; Corey, E. J. J.
Am. Chem. Soc. 2002, 124, 2430; (b) Liu, X.-H.; Prestwich,
G. D. J. Am. Chem. Soc. 2002, 124, 20.
11. Sharpless, K. B.; Umbreit, M. A. J. Am. Chem. Soc. 1977,
99, 5526, and references cited therein.
1.2
**
**
1
0.8
0.6
0.4
0.2
0
**
*
*
**
**
**
**
0
0.39 0.78 1.56 3.13 6.25 12.5 25 50 100
Concentration (µM) of jaspaquinol, 1
Figure 4. Effect of 1 on EJ cell growth at day 3. MTT assay at day 3.
The mean relative A570 nm was determined by normalisation to a
solvent control (0 lM treatment). Each data point is the average ofsix
replicates, and standard deviations are shown by error bars.
(ꢀP < 0:05, ꢀꢀP < 0:01).
12. Yields for these reactions also tend to vary depending on
the purity and supplier ofSeO 2, see: Fairlamb, I. J. S.;
Dickinson, J. M.; Pegg, M. Tetrahedron Lett. 2001, 42,
2205.
13. Available in three steps from b-ionone, see: Crombie, B.
S.; Smith, C.; Varnavas, C. Z.; Wallace, T. W. J. Chem.
Soc., Perkin Trans. 1 2001, 206, and references cited
therein.
After 3 days, EJ cells showed no morphological differ-
ences in response to doses of25 lM and under. Treat-
ment with 50 lM of 1 caused some cell death, whilst
there was increased cell death in response to 100 lM of
1. The MTT data reflects these observations as there was
a sharp drop-off in cell viability in response to doses
over 25 lM.
14. Corey, E. J.; Kim, C. U.; Takeda, M. Tetrahedron Lett.
1972, 13, 4339.
15. Barrero, A. F.; Alvarez-Manzaneda, E. J.; Chahboun, R.;
ꢀ
Dıaz, C. G. Synlett 2000, 1561.
In conclusion, we have identified two synthetic routes to
jaspaquinol 1. Each route has potential for the intro-
duction ofa range ofhydrophobic and hydrophilic
moieties. The viability ofnormal human urothelial
NHU cells treated with 1 has been studied, which
demonstrates that relatively high concentrations of 1 are
16. Various Pd-complexes ofgeranyl chloride demonstrate
that the distal olefin forms a stable chelated system (NMR
and X-ray structure) see: (a) Akermark, B.; Hansson, S.;
Krakenberger, B.; Vitagliano, A.; Zetterberg, K. Organo-
metallics 1984, 3, 679; (b) Akermark, B.; Vitagliano, A.
Organometallics 1985, 4, 1275.

A. Demotie et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2883–2887
17. Malkov, A. V.; Baxendale, I. R.; Mansfield, D. J.;
2887
constant at 0.1% and a solvent control was included in all
experiments. Concentrations of1 at 0.39, 0.78, 1.56, 3.125,
6.25, 12.5, 25, 50 and 100 lM were used. NHU cells from
cell line Y324 at passage 5 were seeded in 96-well plates
at 3 ꢁ 103 cells/well, and EJ cells were seeded at
1.5 ꢁ 103 cells/well and the cells were grown for 16 h. The
growth medium was aspirated, and 200 lL ofmedium
containing the appropriate concentration of 1 was added
to 96-well plates in replicates of 6. The cells were exposed
to the agent for 24 or 72 h (day 1 and day 3 time points,
respectively) at 37 °C; the medium was replaced and the
cells were cultured for a further 3 days (day 6 time point).
Cell growth was assessed using an MTT ([3-(4,5-dimethyl-
thiazol-2-yl]-2,5-diphenyltetrazolium bromide) assay on
days 0, 1, 3, and 6 after the start of the treatment. 200 lL
of 0.5 mg/mL (for NHU) or 0.25 mg/mL (for EJ) MTT
(Sigma) was added to the wells ofthe plate, and plates
were incubated at 37 °C for 4 h. During this time, the
yellow MTT tetrazolium salt was converted to insoluble,
purple formazan crystals by mitochondrial dehydrogenase
in viable cells. Nonconverted MTT and media were
discarded, and the formazan crystals were dissolved in
200 lL DMSO. The absorbance was read at 570 nm
relative to a DMSO blank. The data are displayed
graphically as a mean value with error bars representing
the standard deviation (SD). One way ANOVA was
performed using GraphPad InStat to test for significant
differences between the treatments and the controls.
ꢁ
ꢀ
Kocovsky, P. J. Chem. Soc., Perkin Trans. 1 2001, 1234.
18. Harwood, L. M.; Oxford, A. J.; Thomson, C. J. Chem.
Soc., Chem. Commun. 1991, 1303.
19. Godschalx, J.; Stille, J. K. Tetrahedron Lett. 1980, 21,
2595.
20. Trost, B. M.; Keinan, E. Tetrahedron Lett. 1980, 21, 2595.
21. (a) Goliaszewski, A.; Schwartz, J. J. Am. Chem. Soc. 1984,
106, 5028; (b) Goliaszewski, A.; Schwartz, X. Organomet-
allics 1985, 4, 417.
22. Normal human urothelial (NHU) cell lines were estab-
lished from surgical specimens taken from patients with no
history ofurothelial malignancy, as previously described,
see: (a) Southgate, J.; Hutton, K. A.; Thomas, D. F.;
Trejdosiewicz, L. K. Lab. Invest. 1974, 71, 583; (b) NHU
cells were maintained in keratinocyte serum-free medium
(KSFM) containing bovine pituitary extract and recombi-
nant human epidermal growth factor at the manufac-
turer’s recommended concentrations (Invitrogen) and
30 ng/mL cholera toxin (Sigma). The urothelial carcinoma
EJ cell line, which contains the Harvey ras oncogene, was
obtained from the European Collection of Cell Cultures
(Porton Down) and was maintained in a 1:1 mixture of
DMEM and RPMI 1640 (Invitrogen) with 5% (v/v) FBS
(Harlan Sera-Lab), see: Parada, L. F.; Tabin, C. J.; Shih,
C.; Weinberg, R. A. Nature 1982, 297, 474, Cultures were
maintained at 37 °C in a humidified atmosphere of5%
(NHU) or 10% (EJ) CO2 in air and were passaged by
trypsinisation. The inhibitor, 1, was dissolved fresh in
DMSO. The concentration ofthe DMSO carrier was kept
P values of0.05 and below were considered to be
significant.